Pharisaism and the Gospels
Full text of "Studies in Pharisaism and the Gospels : First series"
STUDIES IN PHARISAISM
AND THE GOSPELS
CAMBRIDGE UNIVERSITY PRESS
C. F. CLAY, MANAGER
Honfcon : FETTER LANE, E.G.
<?Rrittiwrgt) : 100 PRINCES STREET
#to gorfe: G. P. PUTNAM'S SONS
Bomfcag, Calcutta antj flSatorag: MACMILLAN AND CO., LTD.
Toronto : J. M. DENT AND SONS, LTD.
Eofcgo: THE MARUZEN-KABUSHIKI-KAISHA
All rights reserved
STUDIES IN PHARISAISM
AND THE GOSPELS
BY
I. ABRAHAMS, M.A.
READER IN TALMUDIC, UNIVERSITY OF CAMBRIDGE,
FORMERLY SENIOR TUTOR, JEWS* COLLEGE, LONDO
FIRST SERIES
fctl
Cambridge :
at the University Press
1917
PREFACE
TN 1909 Mr C. G. Montefiore published what may without
exaggeration be termed an epoch-making Commentary on the
Synoptic Gospels in two volumes. It was intended that I should
have the honour of contributing a third volume, containing
Additional Notes. This plan has not been fulfilled. The reason
is simple. I had promised more than I could perform. The
problems proved so many, so intricate, that I have found it beyond
my capacity to deal with them all.
But if the original design could not be fully carried out, neither
was it entirely abandoned. A saying of Rabbi Tarphon seemed
appropriate to the situation. " It is not thy part to complete the
work, yet art thou not altogether free to desist from it." On this
principle, Notes were from time to time written and printed,
until by the year 1912 the contents of the present book were in
type. Most of the Notes were actually written between the years
1908-1911. I have recently gone through the proofs carefully, and
have added some references to later literature, but substantially
the Notes remain as they were written several years ago. The
abandonment, for the present at least, of the hope to do much
more has impelled me to publish what I have been able to do.
The circumstance that this volume was designed as an Appendix
to Mr Montefiore 's work accounts for the inclusion of subjects of
unequal importance. Certain Notes, natural and necessary to a
consecutive Commentary, would hardly have suggested themselves
V] PREFACE
for a series of independent Studies. Moreover, some of the
Chapters in the present book, though possibly they might pass as
exegetical comments, are quite inadequate as essays. It must be
remembered that it was purposed to supplement several of these
Notes by further Notes on other aspects of the same problems as
they presented themselves in the course of the Synoptic narratives.
The author is not without hope that he may be able before long
to issue a second series of Studies in which some of the omissions
are rectified. In point of fact several Studies on other matters
are practically written, and others definitely planned. Among the
subjects to be discussed in this second Series would be: certain
aspects of " Life under the Law," the " Yoke of the Command-
ments," "Ritual Purity," the "Traditions of the Elders," the "Last
Supper," "Rabbinic Conceptions of Sacrifice and Prayer," the
"Trial of Jesus," the "Am Ha-ares," the "Two Ways," the "Psy-
chology and Liturgy of Confession," and above all the " Kingdom
of God," " Pharisaic Eschatology," and the " Jewish Apocalypses."
This being the case, I have deferred for a later occasion any
general appreciation of the Gospel teachings. Nor do I think it
necessary to justify at any length the intrusion of a Jewish student
into the discussion of the Synoptic problem. Mr Montefiore, as is
admitted on all hands, rendered a conspicuous service both to
Jewish and Christian scholars by his frank and masterly exami-
nation of the Gospels from a professedly Jewish stand-point.
Undoubtedly a (though not the) real Synoptic problem is : how to
hold the balance truly between the teaching of Jesus on the one
hand and of Pharisaic Judaism on the other. Obviously, then,
Jewish students have both the right and the duty to attempt a
contribution to this balanced judgment. Apart from the fact that
their studies in Pharisaic literature are inevitably more intimate,
there is another very important consideration. Pharisaism was
not a mere historical phase ; it has remained a vital force, it has
gone on without a moment's break from the centuries before the
Christian era to the twentieth century of that era. It has been
PREFACE Vll
put to the test of time and of life. It has survived throughout
an experience, such as no other religious system has undergone.
Hence the Jewish student is able to apply to current criticisms
of Pharisaism not merely literary tests, but also the touchstone
and possibly the corrective of actual experience.
There is perhaps room for yet another suggestion. Jewish
students of the Old Testament have gained much from the re-
searches made by Christian scholars, not merely philologically and
in the archaeological field, but also theologically. For the Jew f
has so ingrained a belief in the organic union of ritual with religion,
is so convinced that the antithesis of letter and spirit is mistaken
psychologically, that he needed the analytical criticism to enable
him to appreciate historically the difference between the prophetic
and the priestly strata in the Hebrew Bible, between the abiding
principles and Messianic dreams of religion and those detailed
rules of ritual and maxims of conduct by which it is sought
to realize those principles and dreams in actual life. But it is
just because of this that the Jew may be able to return the
compliment, and help Christians to understand certain phases of
the Gospels. Many modern Christians seem torn between two
sides of the teaching of Jesus his prophetic-apocalyptic visions of
the Kingdom and his prophetic-priestly concern in the moral and
even ritual life of his day, in which he wished to see the Law
maintained in so far as it could be applied under existing circum-
stances. The Christian scholar, impregnated with Paulinism,
sometimes appears to find these two aspects of the Gospel teachings
inconsistent. Hence we have the disturbing phenomenon of waves
in Christian thought, the humanists who regard Jesus as almost
exclusively a moralist, and the apocalyptists who treat him as
almost exclusively a visionary. The Jew sees nothing inconsistent
in these two aspects. The very causes which make Christian
commentaries useful for the Jew if he would understand the Old
Testament, may make Jewish commentaries helpful to the Christian
for understanding some aspects of the New Testament.
Vlll PREFACE
I am well aware of the many imperfections of the Studies here
presented. But I do claim that I have not written apologetically.
Still less have I been moved by controversial aims. Only on rare
occasions have I directly challenged the picture of Pharisaism
drawn in Germany by Prof. Schurer and in England by Canon
Charles. I have preferred to supplement their views by a positive
presentation of another view. In this sense only are these Studies
apologetic and controversial. At all events, though I acknowledge
that I have fallen far below Mr Montefiore in the faculty of un-
prejudiced judgment, I have never consciously suppressed defects
in the Pharisaic position, nor have I asserted in behalf of it more
than the facts, as known to me, have demanded. I am confident
that those who are best acquainted with the difficulties of the
problems discussed will be the most lenient critics of my errors
and misconceptions.
I. A.
December, 1916.
CONTENTS
PAGE
I THE FREEDOM OF THE SYNAGOGUE ... 1
II THE GREATEST COMMANDMENT . . . . 18
III JOHN THE BAPTIST 30
IV PHARISAIC BAPTISM 36
V THE DOVE AND THE VOICE 47
VI LEAVEN 51
vVII PUBLICANS AND SINNERS 54
VIII "GIVE UNTO CAESAR" 62
v IX FIRST CENTURY DIVORCE 66
X WIDOWS' HOUSES 79
, XI THE CLEANSING OF THE TEMPLE. ... 82
XII THE PARABLES . . . ' . , . /
XIII DISEASE AND MIRACLE . . . . . . 108
XIV POVERTY AND WEALTH 113
XV THE CHILDREN * 118
XVI FASTING . . . . . ^ . ., 121
XVII THE SABBATH . . .... - ., 129
XVIII THE PERSONAL USE OF THE TERM MESSIAH . 136
XIX GOD'S FORGIVENESS % 139
XX MAN'S FORGIVENESS 150
XXI THE LIFE OF THE RESURRECTION . . . 168
INDEX (I) OF NAMES AND SUBJECTS . . . 171
(II) OF NEW TESTAMENT PASSAGES . . 177
It may be well to indicate the relation of the present Chapters
to the Additional Notes referred to in Mr Montefiore's work.
The correspondence is as follows:
Additional Notes in Chapters in
Mr Montefiore's work. the present volume.
1 xviii.
2 iii.
3 iv.
4 v.
6 i.
7 xiii.
8 xix., xx.
9 vii.
11 vii.
12 xvi.
13 xvii.
14 xii.
15 xiv.
16 xiv.
17 vi.
18 ix.
19 xv.
20 xiv.
21 xi.
22 viii.
23 xxi.
24 ii.
25 x.
I. THE FREEDOM OF THE SYNAGOGUE.
The Synagogue, that most gracious product of Jewish legalism
cannot have been the invention of the Hellenistic diaspora (as is
maintained, without adequate evidence, by M. Friedlander, Introd.
to Synagoge und Kirche, 1908). If it was due to a diaspora at all, it
must be attributed to the exile in Babylon. This is no modern guess,
for we have the statement of Justin (Dialogue with Trypho 17) that
Jews applied Malachi i. n, 12 to the prayers of the Israelites then in
dispersion. We may confidently assert (with W. Bacher, Hastings'
Dictionary of the Bible, s.v.; G. A. Smith, Jerusalem i. 364) that the
Synagogue was a Palestinian institution of the Persian period. It was
an institution momentous for the history of religion. "Their (the
Jews') genius for the organisation of public religion appears in the
fact that the form of communal worship devised by them was adopted
by Christianity and Islam, and in its general outline still exists in the
Christian and Moslem worlds" (C. Toy, Introduction to the History
of Religions, 1913, p. 546).
In the Greek diaspora the Synagogue undoubtedly became of special
importance. But its connection with Palestinian models is clear.
Philo's account of the services in the Greek synagogues points to
the two features which distinguished the Palestinian system; the
reading and interpretation of the Scriptures, and the recitation of
passages to which the assembly responded by terms of liturgical
assent (cf. Cambridge Biblical Essays, 1909, p. 190). These features
are shown in Ezra and Chronicles, and in all the Palestinian records
that have come down to us (as in Sirach). True, the Maccabean
history makes no direct reference to the Synagogue, but the main
interest in that history was Jerusalem and the Temple. None the
less, the books of the Maccabees prove most clearly that the people
were in possession of copies of the Scroll of the law from which they
read publicly (i Mace. i. 57, iii. 48), were in the habit of gathering
A. 1
2 I. THE FREEDOM OF THE SYNAGOGUE
for prayer (iii. 44), and above all of singing hymns with such refrains
as " His mercy is good, and endureth for ever " (iv. 24).
That there is little allusion in the Books of the Maccabees to places
of worship is intelligible though the silence is not absolute. It must
not be overlooked that (iii. 46) Mizpah is described not as an ancient
shrine or altar but as " a place of prayer " (TOTTOS Trpoo-cv^?). But the
fact seems to be that the institution of the Synagogue was earlier than
the erection of places of worship. In the Temple itself, the reading
of the Law was conducted by Ezra in the open courts, which remained
the scene of the prayer-meetings to the end, as the Rabbinic sources
amply demonstrate (e.g. Mishnah Sukkah chs. iv v; cf. Sirach 1.
5 21 ; i Mace. iv. 55). So, too, with the first prayer-meetings in the
"provinces." The meetings were probably held in the open air; and
that this was the most primitive form is shown by the fact that the
assemblies on occasions of national stress, even in the last decades of
the existence of the temple, were held in the public thoroughfares
(Mishnah Taanith ii. i). By the first century A.D. Synagogue
buildings were plentiful both in the capital and the provinces. They
probably came into being under the favourable rule of Simon. It
must always, however, be remembered that Synagogue buildings in
various parts of Palestine are possibly referred to in Psalm Ixxiv. 8,
usually assigned to the early years of the Maccabean age.
This is not the place to discuss the whole question, but one supreme
fact must not be omitted. From first to last, there was an organic
relation between Temple and Synagogue (though Friedlander, loo. cit. t
denies this). That there were prayers in the Temple is of course
certain (Mishnah Tamid v ; Philo on Monarchy vi). Isaiah's phrase
(Ivi. 7) a "house of prayer" (LXX. OIKOS Trpotrevx^s) applied to the
Temple was fulfilled to the letter. It is probable that all the Greek
words used in the diaspora for the Synagogue (that word itself,
Proseuche and place of instruction, the last occurs in the Hebrew
Sirach) were derived from Hebrew or Aramaic equivalents. Certain is
it that, in Palestine, 110 Greek terms were imported to describe the
Synagogue. The real model for Palestine and the diaspora was the
Temple. It was a true instinct, therefore, which identified the " smaller
sanctuary" of Ezekiel xi. 16 with the Synagogue (T. B. Megillah 29 b).
The very word Abodah used of the Temple service became an epithet
for the service of prayer (the " Abodah of the heart," Sifr Deut. 41).
The link between Temple and Synagogue was established in Palestine
by the system in accordance with which local delegacies accompanied
I. THE FKEEDOM OF THE SYNAGOGUE 3
the priests during their course of service in Jerusalem, while at home
there were simultaneously held public readings of the law (Mishnah
Taanith iv. 2).
The evidence from the Greek sources points in the same direction.
Agatharchides of Cnidos (second century B.C.) records how the Jews
spend their Sabbath in rest, and " spread out their hands and pray
(evx^Oai) till the evening." The whole context of the passage (as
cited in Josephus Against Apion I. 22) shows that Agatharchides was
referring to Jerusalem. That, however, in Egypt the Synagogue
imitated the Palestinian methods is clear from Philo. Even Philo's
Egyptian Therapeutae have their analogue, and possibly exemplar,
in the Palestinian Essenes. As regards Alexandria, Philo gives
unmistakable proof of the dependence of the Synagogue on the Temple
method. His account, though its force has not been adequately realized,
entirely depends on the Palestinian model. He tells us how (n. 630)
1 'the multitude listens in silence, except when it is customary to say
words of good omen by way of assent to what is read." This can only
refer to the recitation of passages (chiefly no doubt Psalms) by one
while the rest answer by "Amen" and similar ancient liturgical
responses, such as were used in the Temple. That this must refer
to prayers and not to reading the law is certain, for Philo then
proceeds to describe the Scriptural readings and the expositions. Very
instructive as to the connection between the Synagogues of the diaspora
and the Temple is Philo's further statement that the exposition of the
Scriptures was delivered by one of the priests who happened to be present
(rtov iepwv Sc TIS 6 7rapo>i/) or by one of the elders (^ ru>v ycpovrwi/).
This picture of the activity of the priests in teaching the law is
a remarkable testimony to the truth that though the Temple was
essentially the home of the sacrificial ritual, its influence on life was
far-reaching and beneficial. Had it been otherwise, Philo would not
have eulogised the Temple and priesthood as he does in many places.
Perhaps nothing could more piquantly show how completely Jerusalem,
its Temple and its services, contrived to harmonise sacrificial ritual
with prayer and a manifold activity, than the quaint report given by
one who lived in Jerusalem during the existence of the Temple and
survived its fall. R. Joshua b. Hananya said: "When we rejoiced
(during Tabernacles) at the Joy of the Water-drawing we saw no sleep
with our eyes. How so ? The first hour, the morning Tarn id (sacrifice),
and thence to the prayer ; thence to the musaph (additional) offering,
thence to the musaph prayer ; thence to the House of Study, thence to
12
4 I. THE FREEDOM OF THE SYNAGOGUE
the meal ; thence to the afternoon prayer, thence to the evening Tamid ;
thence onwards to the joy of the water-drawing " (T. B. Sukkah 53 a).
The Synoptists draw a pleasing picture of the freedom of teaching
permitted by the Synagogue. Jesus performed this function throughout
Galilee. The Fourth Gospel and Acts confirm the Synoptic record as
to the readiness of the " rulers of the Synagogue " to call upon any
competent worshipper to interpret and expound the Scriptures that
had been read. Such instruction was usual in the Synagogue long
before the time of Jesus as Zunz has shown (Die gottesdienstlichen
Vortrdge der Juden, ch. xx.), and the evidence is admirably marshalled
and supplemented by Schiirer (Geschichte des jiidischen Volkes etc. n 4 .
pp. 498 seq.). Philo (u. 458) describes how one would read from the
book, while another, " one of the more experienced " (TWI/ c/xTmporaTwi'),
expounded. In Palestine, too, the only qualification was competence,
just as for leading the services experience (cf. the b'O") of the Mishnah
Taanith ii. 2) was a chief requisite. As the discourses grew in
length the locale for the sermon seems to have been transferred from
Synagogue to School, and the time sometimes changed from the morning
to the afternoon or previous evening. We find later on both customs
in force together (T.J. Taanith, i. 2 etc.). But at the earlier period,
when the discourse was brief, it must have been spoken in the Synagogue,
and immediately after the lesson from the Prophets.
The only two occasions of which we have a definite account of
teaching in the Synagogue are, curiously enough, treated by Schiirer v
(n 4 . 533 n. 123) as exceptions. His reason fordoing so is derived from
a purely philological argument. In the two cases, Luke iv. 17 and
Acts xiii. 15, it is specifically recorded that the address followed the
reading from the Prophets. In the first instance Jesus speaks after
reading a couple of verses from Isaiah; in the second, we are explicitly
told that in the Synagogue of Antioch, after the reading of the Law
and the Prophets, the rulers of the Synagogue sent to them [Paul and
his company], saying, "Brethren, if ye have any word of exhortation for
the people, say on." We may note in passing that whereas Jesus both
reads the lesson and expounds it, Paul does not seem to have read the
lesson. This indicates an interesting difference in practice, for which
there is other evidence. Rapoport (Erech Millin, 168) concludes from
various Rabbinical passages that in the second century the reader of
the Prophetical lesson was, in general, one who was able also to preach.
It may be that this custom existed side by side with another method
which encouraged the children to read the lessons in Synagogue (cf.
I. THE FEEEDOM OF THE SYNAGOGUE 5
Blau, Revue des Etudes juives, LV. 218). The two customs can be
reconciled by the supposition (based on Soferim, xii. 7, xiv. 2) that
when a preacher was present, he read the Prophetical lesson, and in
the absence of such a one the children read it, perhaps at greater
length. For the Prophetical reading was by nature a sermon, and as
the service concluded with a sermon, the Prophetical lesson concluded
the service when no preacher was present. It is clear from the
narrative in T.B. Beza, 15 b, that the homily of the Rabbi was the
end of the service, and it follows that the homily was given after the
reading from the Prophets, But Schiirer holds that as a general rule the
discourse followed on the Pentateuchal lesson, and that the Prophetical
reading without explanation concluded the service. True it is that
the Prophetical lesson was named haftara (mtSQn or mtDDtf), a word
corresponding to demissio, i.e. the people was dismissed with or after
the reading from the Prophets. But this surely is quite compatible
with a short discourse, and the dismissal of the people might still be
described as following the Prophetical lesson. Moreover, it may well
be that the term haftara refers to the conclusion not . of the whole
services but of the Scriptural readings, the Prophetical passage being
the complement of the Pentateuchal section. This was the view of
various medieval authorities as cited in Abudarham and other liturgists.
(It is accepted by I. Elbogen in his treatise Der judische Gottesdienst
in seiner geschichtlichen JZntwicklung, Leipzig, 1913, p. 175)-
The oldest Prophetical lessons were most probably introduced for
festivals and the special four or five Sabbaths in order to reinforce
and interpret the Pentateuchal lessons, and (in the view of some) to
oppose the views of schismatics. The Pharisees, owing to the con-
flicting theories of the Sadducees, attached to the sections from the
Law such readings from the other Scriptures (particularly the " Earlier
Prophets" who offered historical statements) as supported the Pharisaic
exposition of the festival laws. (Of. Biichler, J. E. y vi. 136 a. The
same writer there cites T.B. Megilla, 25 b, T.J. Megilla, iv. 750, Tosefta,
iv. 34 as Talmudic evidence that the reading of the haftara on the
Sabbath had already been instituted in the nrst century of the common
era). According to Abudarham, the author of a famous fourteenth
century commentary on the Synagogue liturgy, the Prophetic readings
grew up in a time of persecution, and were a substitute for the
Pentateuchal readings when these were interdicted. On the other
hand, L. Venetianer has lately suggested (Z. D. M. G. vol. 63, p. 103)
that there were no specific readings from the Prophets till the end of
6 I. THE FREEDOM OF THE SYNAGOGUE
the second century, and that the Prophetical lectionaries were chosen
polemically in reply to lectionaries and homilies in the early Christian
Church. But it seems far more probable that the haftaras were
chosen for other reasons : (a) to include some of the most beautiful
parts of the Scriptures, (b) to reinforce the message of the Pentateuch,
and (c) to establish firmly the conviction that the whole of the
canonical Scriptures (which, when the haftaras were first appointed,
did not yet include the hagiographa) were a unity. (Cf. Bacher,
Die Proomien der alien judischen Homilie, 1913, Introduction.)
There does not seem to have been any interval between the two
readings, in fact the reciter of the haftara previously read a few
verses from the Pentateuchal lesson (T.B. Megilla, 2 3 a). The sermon
often dealt with the substance of the Pentateuchal lesson, and the
preacher frequently took his text from it. But it is initially unlikely
that the sermon should precede the haftara, seeing that the latter was
introduced to help the understanding of the Law. We are not, how-
ever, left to conjecture. For we possess a large number of discourses
which were specifically composed round the haftara. Many of the
homilies in the Pesiqta Rabbathi are of this class ; they are of course
not, as they now stand, so early as the first century, but they represent
a custom so well established as to point to antiquity of origin. The
famous fast-day discourse reported in the Mishnah Taanith ii. i is
based on two texts from the prophets (Jonah iii. 10 and Joel ii. 13),
both of which passages were eminently suitable as the lesson for
such an occasion. Of the forty-seven chapters in the Pesiqta (most
of which are compounded of many discourses) in Friedmann's edition,
more than twenty are based on haftaras ; in the Pesiqta of JR. Cahana
there are eleven such chapters. That these discourses followed the
reading from the Prophets is shown by the recurrence of such a phrase
as : " As he has read as haftara in the Prophet " (80333 &b&f\& niO
Friedmann, i b) when quoting the text expounded. (The verb D^fc?
is equivalent in this context to ntOBNj just as NJilD^ is another word
for rntSQn, and it must signify to complete the lesson rather than to
dismiss the congregation.) Similar evidence that the discourse
was preceded by the actual reading of the haftara is derivable
from Friedmann's edition, pp. 29 a, 42 a (tf^n D^tPPIG? HDIo), 54 a,
i42b (pjjn 3n2BP HDD -"As he has written in the passage read"),
i49b (fcorun pajn INnpt? n), 179 a (p$n pnpP POD). Perhaps the
most instructive passage of all is on 1 7 2 a. Here the discourse is on
the Pentateuchal text Leviticus xxiii. 24 read on the New Year
I. THE FREEDOM OF THE SYNAGOGUE 7
festival : " In the seventh month, in the first day of the month, shall
be a solemn rest unto you, a memorial of blowing of trumpets." At
the end of the last Pisqa the homily runs : " Says the Holy One,
blessed be He, in this world, through the trumpet (shofar) I have had
compassion on you, and so in time to come I will be merciful to
you through the trumpet (shofar) and bring near your redemption.
Whence? From what we have read in the lesson of the Prophet
(N*3J3 pjjn iaopB> HDD ?pa): Blow ye the trumpet in Zion...for
the day of the Lord cometh (Joel ii. i)." In this case it is quite clear
that the discourse on the Pentateuchal text followed the haftara.
I have been at some pains to show that the New Testament
accounts of the preaching in the Synagogues refer to the normal
and not to the exceptional, because these accounts are the most
precise we possess and it is important to know that we may rely
on them completely. What then can we exactly infer as to the extent
of freedom which the worshippers enjoyed not only with regard to
teaching but also with regard to the selection of passages on which
to speak? I do not find it possible to accept the view that the
homilist was allowed a perfectly free hand, that he might open the
Prophet or Prophets where he willed, read a verse or two and then
address the congregation. That the readings from the Law and the
Prophets were in the time of Jesus very short is fairly certain.
The rule that at least 21 verses were read from the Law and the
Prophets was, as Buchler shows (J. Q. 7?., v. 464 seq. ; vi. i4seq., 45),
late. In the Massoretic divisions we find Sabbath lessons (Sedarim)
which contain seven, eight and nine verses, and there are many in-
dications that the oldest haftara often comprised very few verses.
This follows indeed from the very nature of the haftara. It originally
corresponded in substance with, and agreed often in its opening word with
the opening word of, the Pentateuchal lesson. But this correspondence
mostly only concerns a single verse or two, not long passages. Thus
the reading Isaiah Ixi. i 2 (Luke iv. 16) was possibly the whole of
the haftara. Later on, it became usual to round off the reading by
skipping until a suitable terminating verse was reached.
Let us try to define exactly what it is that Luke describes. Jesus
stood up to read. Then " there was delivered unto him a book of the
prophet Isaiah." The verb used for " delivered up " (lir&oO-rj) might
be interpreted "was delivered unto him in addition." In that case
Jesus would have first read a verse of the Pentateuchal lesson (perhaps
Deut. xv. 7) and then proceeded with the haftara. But it is impossible
8 N I. THE FREEDOM OF THE SYNAGOGUE
to press the Greek verb in this way. Yet it is at all events clear
that the prophet was not Jesus' choice ; it was handed to him. More-
over, the wording in Luke makes it almost certain that just as the
book of Isaiah was not Jesus' own choice, so the passage from Isaiah
was not chosen by Jesus himself. " He opened the book and found
the place where it was written." The word "found" (ev/oev) does not
mean he looked for it and chose it, but he " found " it ready. This is
implied by a change in the verbs which has I think been overlooked.
We are simply told that Jesus "opened" (avoias) the book. Jesus
does not unroll it, as he would have done had he searched for a text.
(The reading aVa7rrv'as is rejected by W.H., Nestle etc.) Luke on
the other hand tells us that when he had finished the reading he
"rolled it up." The A.V. "he closed the book" does not give the
force of the Greek (TTTV^). Thus when he has finished Jesus rolls up
the scroll which he did not unroll, for it was given to him already
unrolled, so that he only opened it at the place already selected and
found the passage in Isaiah ready for him to read. In fact, while the
Pentateuch was read in an unbroken order, the haftara might be
derived from any part of the Prophets, provided always that one
condition was fulfilled : the passage was bound to resemble in subject-
matter the Torah portion just read. As Dr Biichler well puts it :
"This is clear from the origin of the institution itself; and moreover
the examples quoted by the Mishna, Boraitha and Tosefta, bear un-
mistakable testimony to the existence of this condition" (.7. Q. R.,
vi. 12).
It has often been pointed out that Jesus sat down (Luke iv. 20) to
expound the Scriptures, and that this accords with Rabbinic custom.
There is no contradiction in Acts xiii. 16, where "Paul stood up."
Though Paul's exhortation follows Jewish lines in its structure, it is
not an explanation of the Law. For, though the address may be due
more to Luke's hand than to Paul's, it resembles the exhortations in
the Books of the Maccabee,s ; and, at all events, so far from expounding
the Law, it is an ingenious eulogy of it up to a point, and thence an
argument against its sufficiency. The climax of Paul's whole speech is
reached in verse 39, and the opposition which followed, from those
who venerated the Law against one who proclaimed its insufficiency,
cannot be regarded as any breach in that freedom of the Synagogue
which he had previously enjoyed. On the other hand, Jesus expounded
the Scriptures, applying Isaiah Ixi. i, 2 to himself. He seems to have
combined Iviii. 6 with Ixi. i. The right to "skip" while reading the
I. THE FREEDOM OF THE SYNAGOGUE 9
Prophets was well attested (Mishnah Megilla iv. 4). Being written
on a Scroll, the two passages might easily be open together, and Jesus,
in accordance with what at all events became a usual Rabbinic device,
intended to use both texts as the key to his exposition. Such skipping
to suitable passages may be noted in the Geniza fragments of haftaras
in the triennial cycle.
If the view here taken of the incident in Luke be correct, then we
have distinctly gained evidence that, at the opening of the public
teaching of Jesus, the Synagogue lectionary was becoming fixed at
all events in its main principles. That this was the case with the
essential elements of the service is very probable. There is no reason
whatever to doubt the tradition (T.B. Berachoth, 33 a) which ascribed
the beginnings of the order of service to the "Men of the Great
Synod," the successors of the three post-exilic prophets, Haggai, Zecha-
riah, and Malachi. The doubts which Kuenen threw on the reality of
this body doubts which for a generation caused the " Great Synod "
to be dismissed as a myth are no longer generally shared, and
Dr G. Adam Smith in his Jerusalem has fairly faced the absurd
position in which we are placed if we deny, to a highly organised
community such as Ezra left behind him, some central legislative and
spiritual authorities in the Persian and Greek periods. The two
functions were afterwards separated, and it may well be (Biichler
Das Synhedrium in Jerusalem, 1902) that two distinct Synhedria,
one with civil the other with religious jurisdiction, existed in the last
period before the fall of the Temple. As regards the Synagogue
service, it probably opened with an invocation to prayer, must have
included the Shema (Deut. vi. 4 9, xi. 13 21 ; to which was added
later Numbers xv. 37 41), a doxology and confession of faith, the
eighteen benedictions in a primitive form, readings from Pentateuch
and Prophets, and certain communal responses. With this Schiirer
(loc. cit.) is in substantial agreement. The actual contents of the
liturgy long remained fluid; the fixation of the Synagogue prayers
was the work of the post-Talmudic Gaonim of the seventh century
onwards.
Attention should be paid to a remarkable difference of language with
regard to prayer and study of the Law. Nothing better brings out the
real character of Pharisaism. It relied on rule and based much con-
fidence on the effect of good habits. But it left free the springs of emotion
and the source of communion. While, then, Shammai urged (Aboth i.
15) " Make thy Torah a fixed thing " (mp "jniin nEty), Simon a disciple
10 I. THE FREEDOM OF THE SYNAGOGUE
of Johanan b. Zakkai proclaimed (ib. ii. 18) "Make not thy prayer
a fixed thing " (mp -jn^Qn CWn ^)- Study was to be a habit, prayer
a free emotion. The true tradition of Pharisaism from beginning to
end of the first century is seen from Hillel, through Johanan, to his
disciples one of whom in answer to Johanan's problem : " Go forth
and see which is the good way to which a man should cleave " said :
" A good heart." And the master approved this solution as the right
one (Aboth ii. 13). No fixation of a liturgy changed this attitude.
Prayer might be, as time progressed, ordained to follow certain forms,
but within those forms freedom prevailed, as it still prevails in the
most conservative Jewish rituals.
With regard to reciting the Scriptures, the public reading of the
Law for occasions was certainly instituted by Ezra, and continued by
his successors in authority; the passages read were translated into
the vernacular Aramaic (Targum). We know that the Palestinian
custom, when finally organised, provided for a cycle of Sabbath lessons
which completed a continuous reading of the Pentateuch once in
every three years (T.B. Megillah, 29 b). As to the antiquity of the
beginnings of this Triennial Cycle Dr Biichler's epoch-making Essays
leave no doubt (J. Q. JR., v. 420, vi. i). The strongest argument
for this supposition is of a general character, but it is reinforced by
many particular facts. Many events in the Pentateuch which are
left undated in the original are dated with exactitude in the Rabbinic
tradition. This is amply accounted for by the simple fact that these
events are contained in the Sabbath lessons which fell normally to be
read on certain dates, which Tannaitic tradition thereupon associated
with those events. This argument enables us to work backwards and
assume a somewhat early origin for the fixation of the readings on
those particular dates.
It may here be of interest to interpolate one or two instances
of the light thrown on passages in the N.T. by the Cycle of lessons.
Dr King (Journal of Theological Studies, Jan. 1904) has ingeniously
shown that the association (in the second chapter of the Acts) of
the Gift of Tongues with Pentecost falls in admirably with the
Triennial Cycle. The first year of the Cycle began on Nisan i,
and the opening verses of Genesis were then read. The eleventh
chapter of Genesis was reached at the season of Pentecost. This
chapter narrated the story of Babel, i.e. the Confusion of Tongues.
The Gift of the Spirit is a " reversal of the curse of Babel." A second
instance may be found in the Fourth Gospel. The discourse of Jesus
I. THE FREEDOM OF THE SYNAGOGUE 11
regarding the Manna must have occurred in the spring, although the
date "the Passover was near" (John vi. 4) is justly held to be a
suspicious reading. But the note of time "there was much grass in
the place" (verse 10) is confirmed by the " green grass " of Mark vi. 39.
John particularly specifies that the five loaves were made of " barley "
(verses 9, 13). The new barley would certainly not be available till a
few weeks after the Passover, and the poor would not have possessed
a store of the old barley so late as the spring. Everything points,
then, to a date soon after the Passover. Now in the second year of
the Triennial Cycle the lessons for the first weeks in lyyar (end
of April or beginning of May) were taken from Exodus xvi., the
very chapter in which the miracle of the Manna is reported. Of
course the dates of both Acts and the Fourth Gospel are uncertain.
But such coincidences as these (to which others could easily be added)
point to the use of good and old sources, and they at least confirm the
view that, in its initial stages, a Cycle of lessons may have been already
in vogue in the first century.
Some obscure arguments in the Gospels might lose their difficulty if
we were acquainted with the Scriptural readings with which they were
possibly associated. Thus in the Sabbath incident (Matthew xii.),
the argument would be more logical if Numbers xxviii. 9 10 and
i Sam. xxi. i 10 had been recently read in the Synagogues. "Have
ye not read what David did?" and "Have ye not read in the Law?"
(Matthew xii. 3, 5) would have a sharp sarcastic point in that case.
It may well be, again, that the Parable of the Prodigal Son was spoken
during the weeks when Genesis xxv. onwards formed the Sabbath
lessons. There is distinct indication from Philo (see below Note on
Parables) that the idea conveyed in the Parable alluded to was con-
nected with the story of Esau and Jacob. Another instance is yet
clearer. The discourse in the Fourth Gospel (vii. 37, 8) belongs to
Tabernacles. " As the scripture hath said, out of his belly shall flow
rivers of living waters. But this spake he of the Spirit." The reference
probably is to Zechariah xiv. 8 (now read in the Synagogues on the first
day of Tabernacles, possibly under the Triennial Cycle read later in the
festival week). Zechariah indeed has : "living waters shall go out
from Jerusalem." But as in Rabbinic tradition (T. B. Sanhedrin 37 a,
Ezekiel xxxviii. 8, Jubilees viii.) Jerusalem was situated in the navel
of the earth, John may be using belly as a synonym for Jerusalem.
Even more significant are the words that follow: "But this spake he
of the Spirit." The Ceremony of the Water-drawing (already referred
12 I. THE FREEDOM OF THE SYNAGOGUE
to above), which occurred on Tabernacles, was interpreted to mean the
draught of the Holy Spirit (Genesis Rabba, ch. 70). Some far-reaching
suggestions as to the nature of the teaching of Jesus, as found in the
Fourth Gospel, in relation to the ideas of the Doreshe Reshumoth (on
whom see a later Note), may be found in G. Klein's Der alteste christ-
liche Katechismus und die jiidische Propaganda- Liter atur (Berlin 1 909).
See especially the section (pp. 4961) entitled "Jesu Predigt nach
Johannes." My own general impression, without asserting an early
date for the Fourth Gospel, is that that Gospel enshrines a genuine
tradition of an aspect of Jesus' teaching which has not found a place
in the Synoptics.
There is no reason to suppose that the freedom of teaching in the
Galilean Synagogues was ever denied to Jesus. So important and
dramatic an incident as such a denial must have found a mention
in the Synoptists. Yet they are agreed in their silence as to an event
of that nature ; of course John (xviii. 20) represents Jesus as through-
out, and to the last, teaching in synagogue. The cessation of references
to such teaching in Mark after the sixth chapter may be best explained
on the supposition that Jesus voluntarily changed his method when he
found that he no longer carried the Synagogue audiences with him.
The turning point is clearly given by Mark in his account of the
experience of Jesus at Nazareth. The prophet found no honour in his
own country, and this loss of sympathy appears to have induced
Jesus to abandon the Synagogue discourses in favour of more in-
formal teaching in the villages and in the open air, reverting indeed
to the older practice. Prof. Burkitt (The Gospel History and its
Transmission, p. 68) holds that the final rupture occurred with the
religious authorities in Galilee in consequence of the healing of
the man with a withered hand in the Synagogue on a Sabbath
(Mark iii. i). The Pharisees are said thereupon to have taken counsel
with the Herodians to accuse and destroy Jesus. This was the
definite breach (iii. 6). Prof. Burkitt with brilliant skill works out
a scheme which accounts for Jesus spending the eight months in
territory in which the jurisdiction of Herod A.ntipas did not run.
During the greater part of the year before the last Passover Jesus
"lives a wandering life in exile from Galilee or in concealment, and his
chief work is no longer that of Revivalist but of the Pastor pastorum "
(op. cit., p. 89). This theory makes it necessary to explain as excep-
tional not only the later attempts to teach in the Nazareth Synagogue
(where the failure is certainly not due to Pharisaic hostility), but also
I. THE FREEDOM OF THE SYNAGOGUE 13
the subsequent teaching in the villages recorded in general terms (Mark
vi. 6 " And he went round about the villages teaching " ; cf. the
parallels in Matthew and Luke) and the teaching of the crowd (Mark
vi. 34). Moreover the language of Mark viii. 27 points to public
teaching (outside Galilee), and (x. i) where he enters the borders
of Judsea "multitudes come together unto him again, and as he was
wont he taught them again." That the death of John the Baptist
greatly influenced Jesus in avoiding Galilee is highly probable ; and
there may have been some growing suspicion of him in the official
circles of the Synagogues. But it cannot be said that there is any
evidence at all that Jesus ever attempted to teach in any synagogue
and was met with a refusal.
Still less is there any ground for holding that " the influence of the
Sanhedrin everywhere haunted " Jesus and his disciples. Prof. G. A.
Smith (Jerusalem, I. 416-7) strongly maintains that this was so, though
Schweitzer, Quest, p. 362, is of another opinion. My own conviction is
that most of the controversies between Jesus and the Pharisees
occurred in Jerusalem and not in Galilee. If the tradition of the
Galilean scene be authentic, the Pharisees were Priests who had been in
Jerusalem and had returned to their Galilean homes after serving
their regular course. The references to Pharisees or scribes who came
from Jerusalem (Mark iii. 22, Matthew xv. i) do not point to deputa-
tions from the capital. The language of Mark vii. i is the most
explicit : " And there were gathered together unto him the Pharisees
and certain of the scribes which had come from Jerusalem and had
seen that some of his disciples ate their bread with denied, that is
unwashen hands." This looks very much as though the Pharisees were
there in quite a normal manner; it is forcing the words, here and
in the other passages cited, to represent them as "deputations" or
as dogging the footsteps of Jesus. Herod Antipas may have had some
such designs, but the Sanhedrin of Jerusalem had neither power nor
motive to take action until the scene was transferred to the capital.
With regard to the effect of Jesus' discourses in the Synagogues,
we are told that "he taught as one having authority" (Mk i. 22 ; Mt.
vii. 29 ; Luke iv. 32). If the only version of this record were Luke's,
the reference would obviously be to the authority with which the words
of Jesus " came home to the consciences of his hearers " (Plummer).
But the other two Synoptists agree in contrasting this "authority"
with the manner of the Scribes. H. P. Chajes suggests that the real
meaning is that Jesus taught in Parables (see Note on Parables below).
14 I. THE FREEDOM OF THE SYNAGOGUE
This would possibly have to be compared with Philo's remark that the
teaching in the Alexandrian Synagogues was by way of allegory (oid
<nvA/JoAa>i/ ii. 630). More acceptable is A. Wiinsche's explanation of
the claim that Jesus spoke <o? egovo-iav l^wi/ (Neue Beitrage zur Erlaii-
terung der Evangelien, Gottingen, 1878, p. no). The phrase recalls the
Rabbinic idiom of speaking "from the mouth of power" (rmnjn *B),
connoting the possession of direct divine inspiration. The Pharisaic
teachers certainly laid no general claim to the dignity. But the remark
" he taught as one having authority " is usually explained by referring
to the Rabbinical method as unfolded in the Talmud with all its
scholastic adhesion to precedents, and its technical and complicated
casuistry. But this reference is not quite relevant.
For the Talmudical method was the result of long development
after the age of Jesus, and the question is : to what extent can we
reasonably assert that the method was already prevalent before the
destruction of the Temple and the failure of the Bar Cochba War of
Independence (135 A.D.) drove the Rabbis into their characteristic
scholasticism ? There was, moreover, all along a popular exegesis
besides the scholastic, a form of homily specially intended for the
edification and instruction of the simple and unlearned ; and it would
thus be improper to contrast the simplicity and directness of Jesus
with the sophistication and precedent citations of the Rabbis even
if the latter features were earlier than we have evidence of. Hillel,
the greatest of the predecessors of Jesus, taught almost without
reference to precedent ; he only once cites an earlier authority. Hillel's
most characteristic utterances are as free as are those of Jesus from
the bonds of scholastic tradition. He, too, exemplifies the prophetic
independence of conventions. Naturally, the appeal to and reliance
on precedents presupposes an accumulation of precedents to appeal to
and rely on. Such a mass of previous rule and doctrine would only be
built up gradually. (See T. J. Pesahim 39 a, where Hillel cites his
teachers. In the Babylonian Talmud Pesah. 66 the citation, however,
is omitted. Cf. Bacher Tradition und Tradenten in den Schulen
Paldstinas und Babyloniens, Leipzig, 1914, p. 55.) It was mainly the
Amoraim of the third century onwards that made the appeal to
precedent, and naturally as the precedents accumulated so appeal to
them would increase, as in the modern English legal experience with
regard to the citation of illustrative "cases." The earlier Jewish
teaching certainly goes to the Scriptures, but so does Jesus ; and this
earlier teaching (like that of Jesus) uses the Scriptures as a general
I. THE FREEDOM OF THE SYNAGOGUE 15
inspiration. It is only later (in the middle of the second century)
that we find a strict technical reliance on chapter and verse, and
in point of fact Jesus (in the Synoptists) appeals in this way to
Scripture quite as much as does any of the earlier Rabbis. It was
perhaps just his eclecticism, his independence of any particular school,
that is implied by the contrast between Jesus' teaching and that of
the Scribes.
The solution may be found in the supposition that Jesus taught at
a transition period, when the formation of schools of exegesis was in
process of development. Hillel's famous contemporary, Shammai, does
seem to have been a stickler for precedent, and his school was certainly
distinguished from that of Hillel by this very characteristic. If it be
the truth, further, that Shammai (as Dr Biichler conjectures) was a
Galilean, then it is possible that especially in Galilee there was growing
up in the age of Jesus a school which taught with close reference to
particular rules and views with which Jesus had little in common.
The ordinary Galilean Jew would then feel that there was a difference
between the conventional style of the local scribes and that of Jesus,
who did not associate himself with any particular school. On some
points, however, such as his view of divorce, Jesus (if the text of
Matthew xix. 9 be authentic) appears to have been a Shamraaite. It
is by no means improbable (Bloch, Memorial Volume, ^niTl "IDD Hebrew
Section, pp. aiseq.) that at the time of Jesus the views of Shammai
were quite generally predominant, the school of Hillel only gaining
supremacy in Jewish law and custom after the fall of the Temple.
If that be so, Jesus, in departing from the Shammaite method, might
well seem to be one who taught with authority and not as one of the
Scribes. At a later period the question as to the school to which a
scholar belonged would no doubt influence his admissibility as preacher
in a particular place.
Jesus spoke without reference to any mediate authority. To the
Scribes it became an ever more sacred duty to cite the original authority
for any saying, if it were consciously derived from another teacher.
Such reference was an obligation which attained even Messianic import.
" He who says a word in the name of its author brings Redemption to
the world " ( Aboth Chapter of R. Meir vi., Megilla 1 5 a). Verify
your quotations, is C. Taylor's comment (Sayings of the Jewish Fathers,
1897, Additional Note 54). The saving-power of literary and legal
frankness goes deeper than that. Such punctiliousness assuredly
cannot be attributed to the Scribes as aught but a virtue, which if it
I
16 I. THE FREEDOM OF THE SYNAGOGUE
encouraged scholasticism, also encouraged honesty. It did more. It
promoted the conception of a continuous tradition, which conception
while it obscured the facts of history, and required constant criticism
by those facts and also by appeal to ultimate principles as distinct from
derived rules, nevertheless gave harmony to the scheme of doctrine.
The view that Jesus was an original eclectic, that like Horace
though in a far from Horatian sense he was " nullius addictus jurare
in verba magistri," is confirmed by the difficulty of " placing " Jesus with
regard to the schools of his age. The fact is not to be minimised that
we are imperfectly acquainted with those schools ; we have only the sure
knowledge (which is derivable from Philo and Josephus) that an amazing
variety of religious grouping was in progress in the first century. But
even as far as we know these schools Jesus seems to belong to none of
them. It is undeniable that certain features of his teaching are Essenic.
But he did not share the Essenic devotion to ceremonial ablutions.
Further, he was an Apocalyptic, but he was also a powerful advocate of
the Prophetic Judaism. Then, again, it is plausible to explain much of
the gospel attack on the Scribes as due to contempt of the Sadducean
priesthood. But R. Leszynsky (Die Sadducder, Berlin 1912, ch. in.)
finds it possible to claim Jesus as a Sadducee !
It is sometimes thought that the teaching with authority is shown
by Jesus' frequent phrase " but I say unto you " (J. Weiss on Mk i. 21).
But this use of the phrase needs interpretation. The most interesting
passage in which it occurs is Mt. v. 43 4: "Ye have heard that it
was said, Thou shalt love thy neighbour and hate thine enemy, but
I say unto you, Love your enemies." Now it is obvious that nowhere
in the O.T. are men told to hate their enemies. But in the exegetical
terminology of R. Ishmael (end of first century) there is a constantly
recurring phrase which runs thus : " The text reads so and so. I hear
from it so and so : but other texts prove that this is not its true
meaning" (-10^ nWl...'3K JHW). If this as Schechter (Studies in
Judaism) suggests (though Bacher Die dlteste Terminologie der judischen
Schriftauslegung, i. 190 dissents on inadequate grounds), underlies the
passage just cited from Matthew, then Jesus' phrase: "Ye have heard...
but I say unto you" would be parallel to the Rabbinic idiom. It
removes the main difficulty in regard to the hating of one's enemy, for
Jesus would not be referring to any text enjoining hatred, but to a
possible narrowing of the meaning of the text enjoining love. In that
case, Jesus' " but I say unto you " differs from the usual Rabbinical
formula in that it introduces a personal element, but as with them,
I. THE FEEEDOM OF THE SYNAGOGUE 17
Jesus' exegesis really leads up to the citation and interpretation of
another text (in this case: '"Ye shall be perfect as your heavenly
father is perfect " ') which takes a wider sweep and illumines the
particular matter under discussion. This is in full conformity with
the Rabbinic method. They, too, derived the ideal of man's character
from the character of God. "Be ye holy for I the Lord am holy"
(Leviticus xix. 2, of which the turn in Matthew is a reminiscence) was
with the Rabbis the ground text of the idea of the Imitation of God.
It was with them the highest motive for lovingkindness and charity.
(Sifra on Levit. xix. 2).
II. THE GREATEST COMMANDMENT.
The combination of the commandments to love God and to love
one's neighbour is "highly striking and suggestive." Commentators
rightly see that the Scribe's question as to the Greatest Commandment
was not captious, but (as Gould puts it) the Pharisee thought : " Here
is possibly an opportunity to get an answer to our standing question,
about the first commandment." For practical purposes of ethical
monition, the enunciation both of Love God and Love thy fellow man
is necessary. But on a profounder analysis the second is included in
the first, as is shown in the Midrash. Man being made in the image of
God, any misprision of man by man implies disregard of Him in
whose image man is made (Genesis Rabbah xxiv. last words). It there-
fore is not at all unlikely that such combinations as we find in the
Synoptics were a common-place of Pharisaic teaching. It is true that
Wellhausen oblivious of the occurrence of the combination in the
Testament of the Twelve Patriarchs (Isaachar v. 2, vii. 5, Dan v. 3)
holds that "the combination of commandments was first effected in this
way by Jesus." That excellent student of Rabbinics, Dr C. Taylor, was
not so certain on this point. It will perhaps be interesting to cite
what he says on the subject in one of his earlier works (The Gospel
in the Law, 1869, p. 276):
It might seem that our Lord's teaching was novel in respect of its exhibiting the
twofold Law of Love as the sum of Old Testament morality. Thus, in Matt. xxii. 40,
Christ is represented as answering to the lawyer's question : * Thou shalt love the
Lord thy God with all thy heart, and with all thy soul, and with all thy mind.
This is the first and great commandment. And the second is like unto it, Thou
shalt love thy neighbour as thyself. On these two commandments hang all the law
and the prophets. 1 But the addition in St Mark's account (xii. 32) : ' Master, Thou
hast said the truth,' might imply that the answer to that oft-mooted question was
no new one, but rather that which was recognised as true. In another passage
introductory to the Parable of the Good Samaritan ' a certain Lawyer ' gives the
two commandments, To love God, and, To love one's neighbour, as a summary of the
law. He is asked : ' What is written in the law ? how readest thou ? ' And he
II. THE GREATEST COMMANDMENT 19
answers : ' Thou shalt love the Lord thy God with all thy heart, and with all thy
soul, and with all thy strength, and with all thy mind ; and thy neighbour as thy-
self ' (Luke x. 26, 27). But the fact that St Paul grounds this equivalence on reason
solely, goes far to prove that he did not regard the mere statement of it as a
characteristic novelty in the Christian scheme. 'Love,' writes the Apostle,
* worketh no ill to his neighbour : therefore love is the fulfilling of the law '
(Bom. xiii. 10). In John xiii. 34 the words, 'A new commandment I give unto you,
That ye love one another,' might seem to imply that the law of mutual love was
put forward as new. But the words following explain wherein lay the novelty :
' As I have loved you, etc.'
It is not clear why a "lawyer" (VO/AIKOS) is introduced in Matthew;
Luke's frequent use of the word is more intelligible. But it seems
probable that the word had become acclimatised in Hebrew though
there is only one instance recorded of it. Jose b. Halafta (second
century) was so famed as a profound and ready exponent of the Law that
it was said of him " his information as to the Law is ever with him "
(iDy IplDi), where several authorities see th.e Greek i/o/u/o? (sc. eTrio-rr/ju,^).
Cf. Levy and Krauss s.v.; Bacher Agada der Tannaiten ii. 155.
Jastrow s.v. takes another view.. In support of the identification, it
may be pointed out that NO/UKO'S had become a proper name in the first
century. Joesdros, son of Nomikos, was one of the four orators who
were sent to attack Josephus (2 War, xxi. 7). For the suggestion
that the VO/UKO'S of the Synoptics was a Sadducean lawyer, see J. Mann
in J.Q.R. Jan. 1916, p. 419. Possibly the use of the term should be
sought in another direction. In the primitive account of the incident,
the questioner may have been, not a born Jew, but a Gentile I/O/UKO'S
inclined to become, or who had recently become, a proselyte to Judaism.
As will be shown, at the end of this note, such summaries of the Law
were naturally made in the literature of propaganda or catechism.
Aqiba attached, as every Jew did, the highest importance to the
text in Deut. vi. 4, and he died with it on his lips (T.B. Berachoth
6 1 b). He further saw in martyrdom the fulfilment of the law bidding
Israel love God with all his soul or life. The various terms of this law
are differently rendered in the LXX, Deut. vi. 5 and 2 Kings xxiii. 25,
and this fact goes far to explain the dissimilar versions of the
Deuteronomic text in the three Synoptics. Chajes aptly suggests
(Markus-Studien p. 67) that the LXX in Deut. was influenced by
Rabbinic exegesis. It there uses Siavoc'as for KapoYas, and it elsewhere
employs the former word in rendering yeser (Gen. viii. 21 n^ -^ 'O
JH DIKH, on eyKCirat 17 Siavoia TOV, I Chr. xxix. 18 33^ ni2?n)0 W^i ev
Siai/oia KapSias, Gen. vi. 5 13? JYl3B>nD 1V > fel, *at TCXS T ts 8iaj/oiemu tv ry
22
20 II. THE GREATEST COMMANDMENT
avrov). Now the Rabbinic interpretation of Deut. vi. 5 also
introduced the yeser (-p3^ ^33 : " With all thy heart," i.e. with thy
two yesers, 7-^ VJEO, Sifre on Deut. vi. 5, ed. Friedmann 73 a).
Similarly though in 2 Kings xxiii. 25 the LXX renders nKD by icr^v?
in Deut. vi. 5 it uses the term Swa/us, a word which, as the LXX
of Ezek. xvii. 18, 27 shows, may correspond to the sense substance
(jin), which was precisely the Rabbinic interpretation of -pKD in Deut.
vi. 5 (Sifre, loc. cit.\ Ber. 61 b).
A well-known passage of the Sifra (on Leviticus xix. 18, ed. Weiss,
p. 89 a) runs thus : " Thou shalt love thy neighbour as thyself: R. Aqiba
said, This is the greatest general principle in the Law (min3 ^113 A>D Ht).
Ben Azzai said : This is the book of the generations of man (Genesis v. i)
is a greater principle than that (riTD ^nj fe)-" There is no difference
between these Tannaim on the question itself : love of one's fellow-
man is fundamental, but while Aqiba derives the conclusion from
Leviticus xix. 18, Ben Azzai points back to the story of the creation, to
the book of the generations of man, as the basis of the solidarity of the
human race, and the obligation that accrues to every man to love his
fellow. Aqiba himself elsewhere traces the same duty to another
phrase in the Genesis story (Mishnah, Aboth iii. 14, in Taylor iii. 21) :
"Beloved is man in that he was created in the image of God"
(Genesis ix. 6, cf. the quotation from Genesis Rabbah above). As
Taylor remarks on this last passage in the Mishnah (Sayings of the
Jewish Fathers, ed. 2, p. 56) : "Man is beloved by God in whose image
or likeness he was created; and he should be beloved by his fellow-men
as a consequence of this love towards God himself." The text cited
(Genesis ix. 6) runs in full : " Whoso sheddeth man's blood, by man
shall his blood be shed : for in the image of God made he man." As
R. Aqiba comments (Genesis Rabba xxxiv.) : "If one sheds blood it is
accounted to him as though he diminished the likeness." The same
idea is also attributed (Aboth d. . Nathan, xxxix. ed. Schechter, p. 118)
to one of Aqiba's most noted disciples Meir while another of his
disciples Nehemiah (op. cit. xxxi. p. ijo), on the basis of Genesis v. i,
declares "A single human life is equal to the whole work of
creation," jvt$>fcO3 H^yD *?D n3 blpt? inK D1K (with Aqiba's saying
in Aboth iii. 14, especially the latter part of the Mishnah, cf. i Ep.
John iii. i).
These citations, it will be observed, are from Jewish authorities
of the end of the first or the beginning of the second century. But,
as is well known, the idea that forbearance to one's fellow-man is the
II. THE GKEATEST COMMANDMENT 21
basis of the Mosaic law goes back to Hillel (T.B. Sabbath 31 a ; Abotk
de JR. Nathan ii. 26). The mere formulation of the "Golden Rule" in
the negative version is far older than Hillel. So far as Jewish sages
are concerned it may ultimately rest on such phrases as Psalm xv. 3,
where the man who sojourns in the Lord's tent is he that doeth no evil
to his neighbour (njn injrfe PIE>JJ J-6). The actual maxim of Hillel is
found in Tobit iv. 15 (o /xio-cis /xr/Sevi TrowyV^s). This version points to
the conclusion that when Hillel used the word "pin^ ("par6 ^D "J^jn
najm $h, " What-to-thyself is-hateful to-thy-fellow thou shalt not do "),
he meant by it fellow man. In the Aramaic text of Tobit (Neubauer,
Oxford, 1878, p. 8) the reading is Tajm ^ TW"6 "fr ^KDIl (the
Hebrew text, ibid. p. 24, runs Dnn&6 rTOn &6 T.PBA &O2TI IBW).
Hillel elsewhere (Abotk i. 12) uses the widest possible term : he speaks
of love for one's fellow-creatures (nVIlH HX anis). As is well known,
the negative form of the Golden Rule not only preceded Jesus it sur-
vived him. It underlies Romans xiii. 10. St Paul's remark runs :
TOV TrXtjcruov crov <os creavroi'. rj d-yamy TU> TrXrjffiov KCLKOV OVK
thus the Apostle explains or rather justifies Leviticus xix. 18
by the negative form of the Golden Rule (practically as in Ps. xv. 3).
Curiously enough this is paralleled by the Targum Jer. on Leviticus
xix. 1 8 (ed. Ginsburger, p. 206), for the Targum actually inserts the
negative Rule as an explanation of "thou shalt love thy neighbour as
thyself " (mK JOT -j-or6 rvomrn *py oaS nan jniM *6i popj pinn vb
" NJN n^ noyn N(? ^ ^D) Philo (ap. Eusebius, P. viii. 7), too, has
the negative Rule, though his phraseology (d ns TraOtiv IxOaLpei, py
Troitiv OLVTOV) is not verbally derived either from Tobit or from the
source employed in the Didache (iravra. Be o<ra eav OeXyo-ys /AT) yti/ecr^at
o-ot, Kai (TV aAA.w fivj TTotet). But Philo's source can easily be suggested.
It is not Jewish at all. Isocrates (Nicocles 39 c) has the maxim : a
iratr^ovres v<j> erepoov opyt^ctr^e, TO.VTO. rots aXXots pr) Trotetrc. Moreover,
a similar saying is quoted from the Confucian Analects (Legge, Chinese
Classics I. Bk. xv. 23). Jacob Bernays, on the other hand, holds that
Isocrates had no thought of a general moral application of the principle,
and believes that Philo was drawing on a Jewish source (Gesammelte
Abhandlungen, Berlin, 1885, Vol. I. ch. xx.). Bernays cites Gibbon's
quotation of Isocrates in his account of the Calvin-Servetus episode
(Decline and Fall, ch. liv. n. 36).
Here it may be pointed out that the contrasts drawn between the
negative and positive forms of the Golden Rule are not well founded.
One cannot share the opinion of some Jewish scholars (such as
22 II. THE GREATEST COMMANDMENT
Hamburger) that there is no difference between the negative and
positive formulations. But Bischoff (Jesus und die Rabbinen, p. 93)
is equally wrong in asserting that Hillel's maxim differs from that of
Jesus just as "Neminem laede" differs from "Omnes juva," or as
Clough puts it in his fine satirical version of the Decalogue : " Thou
shalt not kill, but needst not strive officiously to keep alive." Augustine
(Confessions I. xviii.) saw no objection to paraphrase the positive of
Matthew vii. 1 1 into the negative id se alteri facere quod nolit pati.
For the Old Testament commands in " thou shalt love thy neighbour as
thyself" (Leviticus xix. 18) and "ye shall love the stranger" (Deut. x. 19)
are positive enough, and Hillel himself elsewhere (Aboth i. 12), as already
cited, uses a quite positive (and general) phrase when he accounts as
one of the marks of the peace-loving disciples of Aaron "love for fellow
creatures." It would be absurd to maintain that Philo, who also, as
has been seen uses the negative form, teaches a negative morality.
Similarly with Tobit. The negative rule occurs in a chapter full of
positive rules of benevolence : Give alms of thy substance ; Love thy
brethren ; Give of thy bread to the hungry, and of thy garments to
them that are naked ; bless the Lord thy God always and so forth.
Why should Hillel not have satisfied himself with citing the text of
Leviticus xix. iSI One suggestion is given below. But a profounder
answer may lie in the thought that the negative form is the more
fundamental of the two, though the positive form is the fuller expres-
sion of practical morality. Hillel was asked to summarise the Torah,
and he used that form of the Golden Rule from which the Golden
Rule itself is a deduction. The axiomatic truth on which the moral
life of society is based is the right of the unimpeded use of the individual's
powers, the peaceful enjoyment of the fruit of his labours, in short, the
claim of each to be free from his fellow-man's injury. When we
remember how great is our power of evil, how relatively small our
power for good, how in Sir Thomas Browne's words, " we are beholden
to every man we meet that he doth not kill us," how " the evil that
men do lives after them, the good is oft interred with their bones," it
is at least a tenable theory that the negative Rule goes deeper into the
heart of the problem. "Do as you would be done by" is less funda-
mental than Hillel's maxim, just as it is less full than the Levitical law
of neighbourly love, for love is greater than doing (cf. the writer's
remarks in Aspects of Judaism, ch. vi). This criticism does not dispute,
however, that the Gospel form is a splendid working principle which
has wrought incalculable good to humanity. The persistence, however,
II. THE GREATEST COMMANDMENT 23
of the negative after the pronouncement of the positive form, itself
argues that the former is more basic.
But neither Tobit nor Philo, nor any other sources cited, do more
than formulate the Golden Rule. Hillel not only formulates it, he
describes it as the essence of the Torah, Sabb. 31 a : nSlD minn SD ton IT
("this is the whole law") and in the Aboth d. R. Nathan, loc. cit.:
nnyn *& ~r\irh "p-iA JD JIKT no mm ^ nfe Kin ("This is the
principle, substance, of the law : what thou hatest for thyself do not to
thy fellow "). This is on the same line with the famous saying of
R. Simlai (third century), but it goes beyond it. Simlai said (T.B.
Makkoth 23 b 24 a) : "Six hundred and thirteen precepts were im-
parted to Moses, three hundred and sixty-five negative (in correspondence
with the days of the solar year) and two hundred and forty-eight
positive (in correspondence with the number of a man's limbs). David
came and established them (lit. made them stand, based them, j-poyn) as
I eleven, as it is written (Ps. xv.) : Lord, who shall sojourn in thy tent,
who shall dwell in thy holy mountain ? (i) He that walketh uprightly
and (ii) worketh righteousness and (iii) speaketh the truth in his heart,
(iv) He that backbiteth not with his tongue, (v) nor doeth evil to his
neighbour, (vi) nor taketh up a reproach against another; (vii) in
whose eyes a reprobate is despised, (viii) but who honoureth them that
fear the Lord, (ix) He that sweareth to his own hurt, and changeth
not ; (x) he that putteth not out his money to usury, (xi) nor taketh
a bribe against the innocent. He that doeth these things shall never
be moved. Thus David reduced the Law to eleven principles. Then
Isaiah came and established them as six (xxxiii. 15) : (i) He that
walketh in righteousness and (ii) speaketh uprightly; (iii) he that
despiseth the gain of deceits, (iv) that shaketh his hands from holding
of bribes, (v) that stoppeth his ears from hearing of blood, and
(vi) shutteth his eyes from looking upon evil. Then came Micah and
established them as three (Micah vi. 8) : What doth the Lord require
of thee but (i) to do justice, (ii) to love mercy, and (iii) to walk humbly
with thy God? Once more Isaiah established them as two (Is. Ivi. i) :
Thus saith the Lord : (i) Keep ye judgement, and (ii) do righteousness.
Then came Amos and established them as one (Amos v. 4) : Thus saith
the Lord, Seek ye me and ye shall live, or (as R. Nahman b. Isaac
preferred) : Habakkuk came and made the whole Law stand on one
fundamental idea (Habakkuk ii. 4) : The righteous man liveth by his
tlaith."
Such attempts to find a basic principle for the whole of the Law
24 II. THE GREATEST COMMANDMENT
can thus be traced clearly from Hillel through Aqiba to the days
of Simlai. Simlai, it will be observed, quotes the prophets as the
authors of attempts in this direction, and it is interesting to note
(cf. Giidemann, Ndchstenliebe, Vienna, 1890, p. 23) that while Hillel
contents himself with concluding "this is the whole Law," Jesus
(Matthew xxii. 40) adds the words "and the prophets." Naturally
there was no intention in the Pharisaic authorities who thus reduced
the Law to a few general rules, to deny the obligation to fulfil the rest
of the law. Hillel's reply to the would-be proselyte, who asked to be
taught the Law while he stood on one foot, runs : " That which
thou hatest (to be done to thyself ) do not to thy fellow ; this is the
whole law ; the rest is commentary ; go and learn it." Yet, the
person so addressed might omit to go and learn it. Hence in Jewish
theology an objection was raised to such summaries just because they
would tend to throw stress on part of the Torah to the relative
detriment of the rest. This feeling has always lain at the back of the
reluctance to formulate a Jewish creed; even the famous attempt of
Maimonides failed to effect that end. Could the legalistic spirit of an
earlier period permit a thoroughgoing distinction between important
and unimportant laws? When Aqiba and Ben Azzai spoke of
neighbourly love as the greatest fundamental law (^nn ^fe) they meant
such a general or basic command from which all the other commands
could be deduced. Thus (as Giidemann rightly argues, op. cit. p. 21),
the Tannaitic Hebrew (^nj fe) does not correspond to the Synoptic
Greek (/w-eyaA.?/ evroXij). The Rabbi was not discriminating between the
importance or unimportance of laws so much as between their
fundamental or derivative character. This is probably what Jesus was
asked to do or what he did ; the Greek obscures the exact sense both
of question and answer. That a Hebrew original underlies the Greek
is probable from the use of the positive : TTOLO. ei/roA?) /AeyaA.?; ei/ TCO
vo'jao) 1 It is more natural in Hebrew (cf. Giidemann, op. cit. p. 23) to
find the positive thus used as superlative (Aqiba's mini ^113 ^>D = the
greatest fundamental law in the Torah). But the passage from the one
idea to the other is easy. Easy, but not inevitable, whether by the
logic of thought or the ethics of conduct. For Pharisaism created just
that type of character to which do these and leave not the. others undone
(Matthew xxiii. 23) admirably applies a type which against all logic
effected a harmony between legislative punctiliousness as to detailed
rules and the prophetic appeal to great principles. The same second
century Rabbi (Ben Azzai) who said (Aboth, iv. 5) "Hasten to a light
II. THE GREATEST COMMANDMENT 25
precept " also maintained that the text relating the common origin of
all the human kind was the fundamental text of the Torah (Sifra ed.
Weiss, p. 89 a) and that the love of God was to be shown even unto
death (Sifre, Deut. 32). The Hebrew prophets, however, did dis-
criminate between the moral importance of various sides of the religious
and social life, and there may have been those who in Jesus' day desired
such a discrimination, and welcomed its reiteration by Jesus.
In a sense, estimations of the varying importance attaching to
precepts must have been in vogue at the beginning of the Christian
era. If Matthew v. 19 20 be admitted as genuine, Jesus differentiated
the precepts in this way (" one of the least of these commandments "),
while exhorting obedience to all precepts alike. Philo in the context
already quoted (Eusebius P. E. viii. 7) very distinctly occupies the
same position (Gifford's translation, p. 389).
But look at other precepts besides these. Separate not parents from children,
not even if they are captives ; nor wife from husband, even if thou art their master
by lawful purchase. These, doubtless, are very grave and important command-
ments; but there are others of a trifling and ordinary character. Rifle not the
bird's nest under thy roof: reject not the supplication of animals which flee as it
were sometimes for protection : abstain from any harm that may be even less than
these. You may say that these are matters of no importance ; but at all events
the law which governs them is important, and is the cause of very careful
observance; the warnings also are important, and the imprecations of utter
destruction, and God's oversight of such matters, and his presence as an avenger
in every place.
Some aspects of this problem especially with regard to the lawful-
ness and even obligation to sacrifice some precept in the interests of
fulfilling others will be discussed later in the Note on the Sabbath.
Here it must be enough to point out the continuity of the theory, that
while the precepts could be divided between 'light' and 'heavy,'
obedience to all was equally binding. While, however, Philo bases
this general obligation on the punishment for disobedience, the
Pharisaic tradition rested on the reward for obedience, and placed
that reward in the life after this (much as in Matthew v. 19). When
we reach the latter part of the second century, we find R. Jehuda
Ha-nasi definitely teaching : " Be heedful of a light precept as of a
grave one, for thou knowest not the grant of reward for each precept "
(Aboth, ii. i). But the very terms of the caution that ODC command-
ment is light (rA>p) while another is heavy (miftn), admit the differentia-
tion. Rabbi Jehuda, it will be noted, asserts that all the commandments
must be equally observed, because the reward for each is unknown.
26 II. THE GREATEST COMMANDMENT
This last clause is to be explained by the parable which is to be found
in Debarim Rabba, ch. vi. and in parallel Midrashim (on the text,
Deut. xxii. 7).
A King hired some labourers and sent them into his Pardes (garden, estate).
At eve, he inquired as to the work of each. He summoned one. " Under which
tree didst thou labour? " " Under this."" It is a pepper plant, the wage is a gold
piece." He summoned another. " Under which tree didst thou labour?" "Under
this." "It is a white-flowered tree (almond), the wage is half a gold piece." He
summoned a third. "Under which tree didst thou labour?" "Under this."
"It is an olive tree, the wage is two hundred zuzim." They said : " Shouldst
thou not have informed us which tree would earn the greatest reward, that we
might work under it?" The King answered : "Had I so informed you, how would
my whole Pardes have been worked ? " Thus the Holy One did not reveal the
reward except of two commandments, one the weightiest of the weighty honour
of parents (Exod. xx. 12), the other the lightest of the light letting the mother-
bird go (Deut. xxii. 7) [note the parallel here with Philo], in both of which is
assigned the reward, length of days.
Underlying the parable (as indeed is to some extent implied by the
form of the Parable in the Tanhuma) must have been a more primitive
one in which all the labourers receive the same reward (cf. Matt. xx.
10), in accordance with the famous saying (end of T.B. Menahoth),
that not the amount of service but its motive is the decisive quality.
So, too, with regard to the very two precepts alluded to in the Parable,
we have the view of R. Jacob (middle of the second century) as given
in the Talmud (Qiddushin, 39 b).
K. Jacob held that the reward for the performance of the precepts is not in
this world. For he taught : Whenever, side by side with a Precept written in the
Torah, the reward is stated, the future life (resurrection) is concerned. Of the
honour to father and mother it is written (Deut. v. 16) " that thy days may be
prolonged and that it may be well with thee." Of the letting go of the mother-
bird it is written ' ' that it may be well with thee, and that thou mayest prolong
thy days" (Deut. xxii. 7). Behold, a father bade his son, Ascend the tower (birah)
and bring me some young birds. The son ascended, let the mother go, and took
the young. In the act of descending, he fell and died. How was it well with
him, and where his length of days ? But the meaning is, that it may be well with
thee in the world which is all good, and that thy days may be prolonged in a world
whose duration is eternal.
Gradation of precepts was, nevertheless, admitted. Certain of
them were described as essential, corpora legis (mm MJ, Aboth end
of ch. iii., Hagigah i. 8, see Dictionaries, s.v. ppj), others as less essen-
tial. This difference perhaps concerned rather the question as to the
ease or difficulty of arriving at the Scriptural basis. Certain of these
essentials related to the ritual laws committed to the (Aaronite ?) Am-
II. THE GREATEST COMMANDMENT 27
haares (T.B. Sabbath, 32). Other views of gradation concerned the
moral laws: thus in one famous enumeration (i) the most important
rewardable performances were honouring parents, the exercise of loving-
kindness, effecting reconciliation between man and his fellow, and the
study of the Torah ; and (2) the most serious punishable offences were
idolatry, incest, bloodshedding, and slander ; for the former there was
reward, for the latter punishment, in this world and in the next
(Aboth de R. Nathan, I. ch. xl., ed. Schechter, p. 120). Again, the
seven " Noachide " precepts were regarded as the fundamental demands
of ethics (on these see Jewish Encyclopedia, vol. vii. p. 648). Further,
the obligation of the priest to disregard the laws of ritual purity when
engaging in the burial of the dead for whose obsequies no one else was
available (rn^D J1D, on which see J. Mann, loc. cit.), the discussions
as to the relative worth of studying the Torah and of performing
the commandments ; the evaluation of the import of fear of sin and
wisdom ; the supersession of the honour of parents by the higher law
of reverencing God when the parents urged actions opposed to that
reverence ; the metaphorical contrast of root and branch, meet us
throughout the first and second centuries (cf. several citations in
Mishnah Aboth, and Sifra on Leviticus xix.). This range of ideas
reaches its culmination in the decision made by the famous assembly
at Lydda after the Hadrianic persecutions of 135. What were the
limits of conformity to the Roman demands? Rather than commit
idolatry, murder, or incest a Jew must die ! (T.B. Sanhedrin, 74 a).
We may suppose, however, that just as there were scruples in later
ages (Hagigah 1 1 b), so not everyone in the age of Jesus was willing to
admit these gradations. As Giidemann writes : " If it be asked how
it came about that a Scribe should need to ask the question of Jesus,
it may be rejoined that the endeavour to bring Judaism within one or
a few formulas would certainly not have been agreeable to the
supporters of the Zealot party. They might perceive in such an
endeavour a connivance towards what we should nowadays term the
liberal position, and it is undeniable that every generalisation easily
renders the particulars volatile. The ignorant, the Am-haares, might,
if he heard speak of a few fundamental rules, readily persuade himself
that these alone as Hillel and similarly after him Jesus expressed
themselves comprised the ' whole Law ' ; while the demand of Hillel
to regard 'the rest' as 'commentary' and to 'learn it' would be
altogether ignored." The questioner of Jesus desired an opinion as to
whether Jesus did or did not share this fear of reducing the Law to
28 II. THE GREATEST COMMANDMENT
fundamental rules. At the same time, Jesus may well have been
attaching himself to HilleFs example, while at the same time implying
a moral discrimination between law and law. Yet this last point is
not certain. In the Palestinian Talmud (Berachoth i. 8 [5]), R. Levi,
a pupil of Aqiba, cites the Shema (Deut. vi. 4 seq.) as fundamental
because the Decalogue is included within it (n^ta nnnn mBW ^QD
Di12; on the connection between the Shema and the Decalogue see
Taylor, Sayings of the Jewish Fathers, Excursus iv.). It is noticeable
(cf. Giidemann, op. cit. p. 22) that in Mark (xii. 29) the answer of
Jesus begins with the Shema, Deut. vi. 4 (^>JOE rG?)i though in
Matthew the verse is wrongly omitted. It does not seem that in any
extant Rabbinic text, outside the Testaments of the Twelve Patriarchs,
the Shema and the love of one's neighbour are associated, though there
is mention of a passage in which this combination was effected by Ben
Zoina and Ben Nanas with the strange addition that greater than any
of these texts was Numb, xxviii. 4, possibly because of the atoning
function of the daily sacrifices, or because of the association of God,
Exod. xxv. 9 etc., with the Sanctuary, the divine dwelling place on
earth (Introd. to the En Jacob see Giidemann, loc. cit., Theodor,
Genesis Rabba, p. 237). In the Nash Papyrus the Decalogue is followed
by the Shema; the two passages indeed stand close together (the
Decalogue in Deut. v. 6 18, the Shema in vi. 4 9). The Didache
(ch. i.) associates the combination as found in the Synoptics also with
the negative form of the Golden Rule : " There are two ways, one of
life and one of death, and there is much difference between the two
ways. Now the way of life is this : First, thou shalt love God that
made thee ; secondly thy neighbour as thyself; and all things whatsoever
thou wouldest should not happen to thee, neither do thou to another."
The Decalogue follows. The Jewish provenance of this passage is
indisputable. Taylor (Teaching of the Twelve Apostles) suggests that
the negative rule grew out of the Decalogue, with its many do nots.
What is the general principle of the things not to do to one's neighbour ?
Answer : " What-to-thyself is-hateful " (the >:D -]!?jn of Hillel). Hence
its description by Hillel as the sum total of the Law. One further
point only calls for remark here. It is quite natural that simplifica-
tions or systematisations of the Law would be most required for
proselytising propaganda. It would be necessary to present Judaism
in as concise a form as possible for such purposes. Hence it is not
surprising on the one hand that it is to a would-be proselyte that
Hillel's summary as well as a similar citation of the principle by Aqiba
II. THE GREATEST COMMANDMENT 29
(Aboth de R. Nathan, ed. Schechter, p. 53) is addressed and on the
other that we find it in the Didache and in connection with the doctrine
of the two ways. Nor is it without significance that Philo's citation
of the negative rule occurs in a passage in which he is selecting just
those elements of the Jewish Law which were worthy of commendation
and acceptance by the Greek world. (Of. on these and several other
matters the interesting work of G. Klein, Der Aelteste Christliche
Katechismus und die Jildische Propaganda-Liter -atur, Berlin, 1909, p. 85,
and K. Kohler in Judaica, Berlin, 1912, pp. 469 seq. The latter
points to the old Jewish Didaskalia, in his view enshrining the ethics
of the Essenes.)
III. JOHN THE BAPTIST.
The Rabbinic literature contains no reference to John the Baptist.
There is, however, an interesting passage on the subject in Josephus
(Antiquities, xvin., v. 2). Some doubt has been thrown on the
authenticity of this passage, but the suspicion has no firm basis.
Josephus gives a favourable account of John and his work. This
is & priori what we should expect, for John has decidedly Essenic
leanings and the Essenes were favourites with the Jewish historian.
John, says Josephus, was "a good man who exhorted the Jews to
exercise virtue (apenf), both as to justice (SiKaioo-vVr?) towards one
another and piety (cvcrc'/Seia) towards God, and to come to baptism
(/SaTTTter/xo) cnWi/at). For baptism (r-rjv fSdirriariv) would be acceptable
to God thus (OVTW), if they used it, not for the pardon of certain sins,
but for the purification of the body, provided that the soul had been
thoroughly purified beforehand by righteousness" (fj.rj CTTI rtvwv d/x,ap-
7rapaiTr)<rL Xpa)/Ava>v, aAA' e<' ayi/cia TOV o-w/xaro?, are Sr; KOL r/s
SiKOLLocrvvri Trpoe/cKe/caflapjaef'rjs). People, continues Josephus,
nocked to him in crowds, were stirred by his addresses, and seemed
willing to follow him in all things. Herod Antipas, fearing a popular
rising, seized John, sent him in chains to Machaerus, and had him put
to death there. When Herod's army suffered a reverse, the people
attributed the king's misfortune to God's displeasure at the ill-
treatment of John.
Both the recent editors of Josephus (Niese and Naber) admit this
passage without question. There is a natural reluctance on the part
of cautious scholars to pronounce unreservedly in its favour, mainly
because of the fact that elsewhere the text of Josephus has been
tampered with in a similar context. Thus Schiirer (i 3 . 438), after
presenting a forcible though incomplete argument in favour of the
passage, adds: "Since, however, Josephus in other places was certainly
subjected to interpolation by a Christian hand, one must not here
III. JOHN THE BAPTIST 31
place too absolute a reliance on the authenticity of the text." On the
Jewish side, though his leanings are in favour of the authenticity,
S. Krauss (Das Leben Jesu nach judischen Quellen, Berlin, 1902,
p. 257) remarks : "The question as to the genuineness of the John-
passage has not yet been decisively settled ; the passage is anyhow
open to suspicion." But, on the whole, the authenticity of the
reference is accepted by scholars, Jewish and Christian. Thus to cite
only two instances, H. St J. Thackeray (Dictionary of the Bible, Extra
Volume, p. 471) passes judgment in these words : "There is no reason
why it should not be accepted as genuine"; and K. Kohler (Jewish
Encyclopedia vn. p. 218) does not even mention the controversy, but
uses the passage without any question. The passage in Josephus
referring to John the Baptist rests, of course, on a different footing to
the "testimony to Christ" (Josephus, Antiq. xviu. iii. 3). The
authenticity of the latter has been recently maintained with much
plausibility by Profs. F. C. Burkitt (Theologisch Tijdschrift, 1913,
xlvii. pp. 135-144), A. Harnack (Internationale Monatsschrift, June,
1913, pp. 1038-1067), and W. E. Barnes (Companion to Biblical
Studies, 1916, p. 34). But it remains very difficult to accept Josephus'
"testimony to Christ" as genuine, at all events as it stands; the
reference to John the Baptist may well be so.
It seems to me that a Christian interpolator must have brought
that passage into closer accord with the Gospels. I do not refer merely
to such differences as the motive assigned for putting John to death.
Josephus assigns fear of political unrest; the Gospels, the personal
animosity of Herodias. But, as Schiirer is careful to point out, these
motives are not absolutely incompatible. Much more significant is
the silence of Josephus as to any connection between John and Jesus.
This, of itself, is almost enough to authenticate the passage. Gerlach
has called attention to this fact in his book Die Weissagungen des
Alien Testaments in den Schriften des Flavins Josephus (Berlin, 1863,
p. 113) and Origen had long ago done the same thing. Origen
(c. Celsum I. xlviii.) says: "The Jews do not associate John with
Jesus." Gerlach misuses this statement, for Origen is not making an
independent assertion, but (as the context shows, cf. op. cit. xlvii.) is
basing his generalisation on the passage in Josephus. Origen, by the
way, who cites this passage, has no knowledge of the supposed
"testimony to Christ" (see, however, Burkitt, as already cited);
the two passages stand, as said above, on quite different footings.
That Jews other than Josephus may have taken a favourable view
32 III. JOHN THE BAPTIST
of John's work is indicated also by several passages in the Gospels.
Luke, it is true, asserts (vii. 30) that the Pharisees and the lawyers
(scribes) rejected John, and refused to accept his baptism. But
this is in opposition to the statement of Matthew (iii. 7) : " [John]
saw many of the Pharisees and Sadducees coming to his baptism,"
and Mark (i. 5) implies no Jewish opposition to his call to baptism.
Moreover, all three Synoptics (Mark ii. 18; Matthew ix. 14; Luke
v. 33) represent the disciples of John as associated with the
Pharisees in fasting. Thus just as Josephus assures us that the
Pharisees were not opponents of the Essenes (as they were of the
Sadducees) so there was no violent division between John and the
Pharisees; the assumption that the Jews rejected John belongs to
the later conception (whether originating with John himself or not)
that John was the forerunner of Jesus. That John's own disciples
did not accept this conception is thus asserted by Prof. Adeney
(The Century Bible, St Luke, p. 185): "These [the disciples of John]
then hold together and keep up their customs after their master has
been removed from them, and in spite of the appearance of the new
Prophet, thus declining to follow John's own teaching in pointing on
to Christ. We meet such later at Ephesus (see Acts xviii. 25, xix. 3)."
Of. also the remarks of Prof. Lake, The Earliest Epistles of St Paul,
1911, pp. 108, etc.
Still more important is another point to which Gerlach called
attention, and to which Naber has more recently again referred.
There is a real difference between the nature of John's baptism as
described by Josephus and the Gospels. Mark (i. 4) introduces John
as proclaiming a "baptism of repentance for remission of sins" (/SaTrrtoyAa
/x.Tavoias cis a</>eo-(,v d/xaprtaJi/). But in Josephus this significance of bap-
tism is specifically dissociated from John. Not only is this deliberate,
it is clearly controversial. As Naber argues (Mnemosyne xm. 281), it
is scarcely credible that Josephus was ignorant of the Christian baptism
which was "for the remission of sins." Naber suggests, then, that in
the passage in which Josephus refers to Jesus, the historian cited the
Christian baptism with expressions of disapproval, and as this was
displeasing to Christian readers, the passage was altered. On the
other hand the John passage was left standing, and the controversial
/x.i) eVi TIVWI/ a^aprdSo)!/ TrapatT^o-ei xpu/ueiwy remained. If this be SO,
it may well be that Josephus really has preserved for us the exact
nature of John's baptism. But before saying a word on that, it is
necessary to turn to a question of language.
III. JOHN THE BAPTIST 33
In his first editions Graetz accepted Josephus' account of John
as authentic. But in his later editions of the Geschichte der Juden
he strongly contends that the passage is spurious. He urges that
Josephus would not have described John as the "Baptist" (TOV TTI-
KaXov/xeVov ySaTTTio-Tov) without further explanation. Graetz does not
see that it is possible to regard these three words as an interpolation
in a passage otherwise authentic. But it is not necessary to make
this supposition. For it is quite in Josephus' manner to use designa-
tions for which he offers no explanation (cf. e.g. the term "Essene").
And the meaning of "Baptist" is fully explained in the following
sentence, Josephus using the nouns /3a7TTns and /2a7TTio7/,o's to describe
John's activity. The terminology of Josephus, I would urge, makes
it quite unlikely that the passage is an interpolation. For, it will be
noted (a) Josephus does not use /8a7mcr/xa which is the usual N.T.
form; (6) he does use the form /SaTmcris which is unknown to the N.T.;
(c) he uses /foTrrwr/xo's in a way quite unlike the use of the word when
it does occur in Mark (vii. 4) or even in Hebrews (ix, 10). It is in
fact Josephus alone who applies the word /SaTrrur/xo's to John's baptism.
Except then that Josephus used the epithet /JaTTTicmys (which may be
interpolated) his terminology is quite independent of N.T. usage. It
is true that Josephus uses the common LXX. word A.ov<o when
describing the lustrations of the Essenes, but the verb /?a7rria> was
quite familiar to Jewish writers. It is rare in LXX. but is curiously
enough found precisely where bathing in the Jordan is referred to, in
the significant passage 2 Kings v. 14 : "Then went he down and
dipped himself (e/foTrreVaro) seven times in Jordan 1 ." Significant, too,
is the fact that Aquila, who translated under Aqiba's influence, uses
/3a7TTiu> where the LXX. uses ./SaTrrw (Job ix. 31; Psalm Iviii. 3). In
the latter place the verb is also used by Symmachus, who further
introduces it into Jer. (xxxviii. 22). To Josephus himself the verb
was so familiar that he even makes a metaphorical use of it. In
describing the masses of people "flocking into the city" he says
/3a7rrt(rai/ TTJV TTO\IV.
Another point on which a few words are necessary is John's relation
1 Cheyne, Encycl. Biblica col. 2499, represents John the Baptist "who was no
formalist" as using the Jordan in spite of the Kabbinic opinion that "the waters of
the Jordan were not pure enough for sacred uses." But the Jordan water was only
held insufficiently clean for one specific purpose : the ceremony of the Bed Heifer
(Parah viii. 9). No Babbi ever dreamed of pronouncing the Jordan unfit for the
rite of baptism.
34 HI- JOHN THE BAPTIST
to the Essenes. That Josephus means to identify him with that sect
is clear. For the very words he uses of John are the terms of entry
to the Essenic confraternity. In Wars 11. viii. 7 Josephus reports :
"If he then appears to be worthy, they then [after long probation]
admit him into their society. And before he is allowed to touch
their common food, he is obliged to take tremendous oaths, in the first
place that he will exercise piety towards God, and next that he will
observe justice towards men " (irporrov /w,ev evo-e/^orciv TO 0eiov, cTrctra TO.
irpos avfl/oowrovs SiKcua Sia<vAaeiv). The other terms used of John by
Josephus (aperr/, ayi/ia) are also used by him of the Essenes. The
Gospels attribute to John Essenic characteristics. The account of
John in Mark i. is more than merely illustrated by what Jo.sephus
says in his Life ii. : " When I was informed that a certain Bannos
lived in the desert, who used no other clothing than grew on trees, and
had no other food than what grew of its own accord, and bathed
himself in cold water frequently, both by day and by night, in order
to preserve purity (rrpbs dyvciav), I became a follower of his." John's
asceticism is not identical with this, but it belongs to the same order.
It is quite untenable to attempt, as many are now tending to do,
to dissociate John altogether from Essenism. Graetz seems right in
holding that John made a wider appeal than the Essenes did by re-
laxing some of the Essenian stringency : their communism, their
residence in separate colonies, their rigid asceticism. John, like another
Elijah, takes up the prophetic rdle. He calls to the Jews to repent,
in expectation of the Messianic judgment perhaps. Pharisaic eschato-
logy, in one of its tendencies, which rising in the first century became
dominant in the third, connects the Messianic age with repentance.
There is, however, this difference. The formula of John (or Jesus)
was : Repent for the Kingdom is at hand. The Pharisaic formula
was : Repent and the Kingdom is at hand. Pharisaic eschatology did
not, however, ally this formula to the baptismal rite. John associates
his prophetic call with baptism, partly no doubt in relation to the meta-
phorical use of the rite in many parts of the O.T., but partly also
in direct relation to the Essenic practices. He treats baptism as a
bodily purification corresponding to an inward change, not as a means
of remitting sins. Cheyne, who takes a different view as to the
Essenic connection of John, expresses the truth, I think, when he
writes as follows (Encyclopaedia Biblica, col. 2499) : " ^ e ^ them
[his followers] to the Jordan, there to give them as representatives of
a regenerate people the final purification which attested the reality of
III. JOHN THE BAPTIST 35
their inward change." Then he adds in a note : " No other exegesis
seems reasonable ; Josephus, as we have seen, sanctions it. The true
baptism is spiritual (Psalm li. 7 [9]). But it needs an outward symbol,
and Johanan [John], remembering Ezekiel xxxvi. 25, and having
prophetic authority, called those who would know themselves to be
purified to baptism. It is no doubt true that baptism was regularly
required of Gentile proselytes, but Johanan's baptism had no con-
nection with ceremonial uncleanness." It is interesting to note the
use made in Pharisaic circles of this same text in Ezekiel. " Said
R. Aqiba [end of first and beginning of second century A.D.] : Happy
are ye, O Israel ! Before whom do you cleanse yourselves 1 Who
cleanseth you 1 Your Father who is in Heaven ! As it is written,
And I will sprinkle clean water upon you and ye shall be clean."
On the question of Baptism in general see next Note. On John's
references to the Pharisees see note on Pharisees. John we are told
in a difficult passage (Matt. xi. 13; Luke xvi. 16) was the end of the
Law and the Prophets. He certainly was faithful to the Law and a
worthy, upholder of the olden Prophetic spirit. But except in the
sense that, in the Christian view, he was the last to prophesy the
Kingdom in the spirit of the Law and the Prophets, John was the end
of neither. When John died the ' Law ' was only in the first stages
of its Rabbinical development. And from that day to this there have
never been lacking in the Jewish fold men who, in accord with the
Prophetic spirit, have made a direct appeal to the hearts of their
brethren on behalf of repentance and inward virtue.
32
IV. PHARISAIC BAPTISM.
Unnecessary doubt has been thrown on the prevalence of baptism
as an initiatory rite in the reception of proselytes during Temple times.
Schiirer, while exaggerating the number of ablutions prescribed by
Pharisaic Judaism, rightly insists (in 3 . 131) that both ct priori, and
from the implications of the Mishnah (Pesahim, viii. 8), proselytes
must have been baptised in the time of Jesus. The heathen was in
a state of uncleanness and must, at least as emphatically as the Jew
in a similar state, have undergone the ritual of bathing. Only in a
state of ritual cleanness could the new-comer be received " under the
Wings of the Divine Presence " a common Rabbinic phrase for prose-
lytism (e.g. T.B. Yebamoth, 46 b) directly derived from the beautiful
terms of Boaz' greeting to Ruth, the ideal type of all sincere proselytes :
"The Lord recompense thy work, and a full reward be given thee of
the Lord God of Israel, under whose wings thou art come to trust."
So, too, Jesus, after his baptism, sees the spirit of God descending as a
dove. The symbolism of the Holy Spirit by a dove is a notion found
in Rabbinic books (see below note on "the Dove and the Voice").
But I think it is more fully explained when it is brought into con-
nection with the figure that the proselyte comes under the Wings of the
Divine Presence. Thus the fact that, in the Gospels, baptism precedes
the metaphorical reference to the bird, strengthens the argument in
favour of the early prevalence of the baptism of proselytes.
Yet it can hardly be said that the evidence so far adduced proves
the case. Schiirer (loc. cit.) and Edersheim (II. Appendix xn.) think
that the Mishnah (cited above) does establish the point. But Dr Plummer,
while conceding that "the fact is not really doubtful," asserts that
"direct evidence is not forthcoming" (Hastings, Dictionary of the Bible,
I. 239). The Mishnah cited (to which Eduyoth, v. 2 is parallel) de-
scribes a difference of view between the schools of Hillel and Shammai.
If a man has " been made a proselyte " on the fourteenth of Nisan and
IV. PHARISAIC BAPTISM 37
has then been baptised, (must he wait seven days before he is regarded
as " clean " or) may he eat the Paschal lamb the same evening ? (The
suggestion of Bengel, Ueber das Alter der jud. Proselyten-taufe, p. 90,
that the bath was not a proselyte-bath is groundless.) This Mishnah
certainly implies that the baptism of proselytes occurred while the
Paschal lamb was still being offered, i.e. during Temple times. But
the passage does not quite prove this, for it is just possible that the
discussion is merely scholastic. On turning, however, as neither Schiirer
nor Edersheim has done, to the Jerusalem Talmud and the Tosefta,
it becomes certain that we are dealing with historical fact and not
with dialectics. (See T. J. Pesahim, viii. last lines ; Tosefta, Pesahim,
vii. 13, ed. Zuckermandel, p. 167.) "Rabbi Eleazar ben Jacob says:
Soldiers were Guards of the Gates in Jerusalem ; they were baptised
and ate their Paschal lambs in the evening." Here we have an actual
record of the conversion of Roman soldiers to Judaism on the day
before the Passover (an altogether probable occasion for such a step),
and of their reception by means of baptism. This Eleazar ben Jacob
the Elder is one of the most trustworthy reporters of Temple events
and rites, which he knew from personal experience. (Of. Bacher, Die
Agada der Tannaiten, I 2 , p. 63.) "The Mishnah of R. Eleazar is a
small measure, but it contains fine flour" (T.B. Yebamoth, 49 b) was
the traditional estimate of the value of this Rabbi's traditions. The
exact date of this incident cannot be fixed. Graetz places it in the
year 67 A.D. If that be so, then we are still without direct evidence
that proselytes were baptised half a century earlier. But the prob-
ability is greatly increased by this historical record.
It is noteworthy that, according to Bacher's reading of this account,
baptism without previous circumcision seems sufficient to qualify the
heathen proselyte to eat the Paschal lamb. This is directly opposed
to the Law (Exodus xii. 48). Later on there was indeed found an
advocate for the view that baptism was sufficient (without circum-
cision) to constitute a proselyte (T.B. Yebamoth, 46 a). But it seems
more reasonable to suppose that R. Eleazar ben Jacob takes it for
granted that the Roman soldiers were circumcised before baptism.
In the corresponding Mishnah, and in the whole context in the
Tosefta, this is certainly presupposed. The predominant and almost
universal view was that in Temple times three rites accompanied the
reception of proselytes : circumcision, baptism, and sacrifice (T.B.
Kerithoth, 81 a). After the fall of the Temple the first two of these
three rites were necessary (ibid. 9b). In the case of women, when
38 IV. PHARISAIC BAPTISM
sacrifices could not any longer be brought, the sole initiatory rite was
baptism. It may be that as women were of old, as now, the more
numerous proselytes, baptism came to be thought by outside observers
as the only rite in all cases. Thus Arrian, in the second century,
names baptism as the one sufficing ceremony which completely turns
a heathen into a Jew (Dissert. Epictet. n. 9).
The baptism by John resembles the baptism of proselytes in several
points, among others in the fact that both forms of baptism are
administered, not performed by the subject himself. At all events,
the proselyte's bath needed witnessing.
In Mark i. 9 the repentant are baptized VTTO 'Icoai/i/ov. But in
Luke iii. 7, where the ordinary text (and Westcott and Hort) has
paiTTLcr&'fjvai VTT' avrov, the Western text has fta.TrTicr6v}va.i CI/WTTIOV avrov
(probably as Prof. Burkitt has suggested to me = niD"ip). In the
Pharisaic baptism of proselytes, at all events, the presence of others
was entirely due to the necessity of witnessing (Yebamoth, 47 a).
Sometimes a causative form, sometimes the kal form, of the verb
tabal is used in the Rabbinic texts; but in the case of male prose-
lytes there seems to have been no act on the part of the witnesses.
In the case of women, the witnesses (three dayanim) stood outside,
and other women "caused her to sit down" (i.e. supported her) in
the bath up to her neck. The male proselyte stood, with the water
up to his waist (Yebamoth, 46-48; Gerim, ch. i.). In all cases, the
bathing was most probably by total immersion (for the evidence see
the writer's article in the Journal of Theological Studies, xn. 609,
with the interesting contributions by the Rev. 0. F. Rogers in the
same periodical, xii. 437, xm. 411). Total immersion is clearly implied
by the Zadokite Fragment (edited by Schechter, 1910, ch. xii). If
that fragment be a genuine document of the second century B.C., its
evidence for the total immersion of the priests is of great weight.
In the Talmud the bath in such a case had to be at least of the
dimensions 1x1x3 cubits, sufficient for total immersion (1213 ^3B>
Dili r6iy, Erubin, 46). The bathing of the niddah (menstrual woman)
was by total immersion, and we have the definite statement of a
baraitha (Yebamoth, 47 b) that the rules for the bathing of proselytes
(male and female) were the same as for the niddah. In only one case
of baptism did the bystander participate actively. On entering Jewish
service, a heathen slave was baptised. If he claimed that such baptism
was for complete proselytism (nil^ D^ 1 ?) he became free. But in order
to make it clear that the baptism was not for this purpose, the owner
IV. PHARISAIC BAPTISM 39
of the slave was required to seize hold of him while in the water
(D^D IDpn!?), as a clear indication that the baptism was not a complete
proselytism ( Yebamoth, 46 a). Obviously in cases of proselytes the
baptism would be the perfectly free, unfettered and unaided act of
the proselyte himself.
But there is, it is often said, this difference between Johannine
and Pharisaic baptism : the former was a moral, the latter a physical
purification. Josephus, it has been shown, hardly regarded this con-
trast as essential. Nor, in the case of the proselyte-bath, can it be
doubted that the two ideas are welded together. In the older Rab-
Jbinical literature we do not, it is true, find any specific reference to
a baptism of repentance. The phrase first meets us in the Middle
Ages. A thirteenth century authority for the first time distinctly
speaks of the man who bathes for penitence' sake (rQ1K>n D6J
and of bathing in general, as an essential of repentance (DUB71
r6'ltD3 D'TTl). See Shibbole Halleket, 93 (ed. Venice, fol. 41 a).
Apparently this rule that " all penitents are baptised " is traced to a
passage in the Aboth de R. Nathan (see the Tanya, 72 ; ed. Venice,
p. 102 b). But though the passage in the Aboth (ch. viii.) does not
easily bear this implication (the text as we have it is certainly corrupt),
we can carry the evidence five hundred years further back than the
thirteenth century. In the Palestinian Midrash Pirke de R. Eleazar,
compiled about 830, Adam's repentance after expulsion from Eden
consists of bathing, fasting and confession (op. cit. ch. xx.). Older
still is the passage in the Apocryphal (and not obviously Christian)
Life of Adam and Eve, which represents the repentant Adam as
standing for forty days in the Jordan (Kautzsch, Pseudepigraphen
zum Alten Testament, p. 512; Charles, Apocrypha and Pseudepigrapha
of the Old Testament, Oxford, 1913, p. 134).
Earlier still is (probably) the famous passage in the fourth Sibylline
Oracle (iv. 165 seq.) which, even in its present form, must belong to
the first Christian century (c. 80 A.D.). In iii. 592 there is a reference
to the morning lustrations (cf. the morning bathers of T.B. Berachoth,
2 2 a. On this and other allied points see S. Krauss, Talmudische
Archaologie, Leipzig, 1910-1912, I. pp. 211, 217, 229, 669; n. p. 100;
III. p. 360). But in iv. 165 there is a direct association of repentance
with bathing. I quote Terry's rendering with some emendations :
Ah! miserable mortals, change these things,
Nor lead the mighty God to wrath extreme;
But giving up your swords and pointed knives,
40 IV. PHARISAIC BAPTISM
And homicides and wanton violence,
Wash your whole body in perennial streams,
And lifting up your hands to heaven seek pardon
For former deeds and expiate with praise
Bitter impiety; and God will give
Eepentance; he will not destroy; and wrath
Will he again restrain, if in your hearts
Ye all will practise precious godliness.
This, it will be noted, is an appeal to the heathen world. It falls well
within the range of the Jewish Hellenistic literature, and there is no
necessity for assuming a Christian authorship
Water was a symbol of repentance still earlier. The Targum to
i Samuel vii. 6 (cf. Midrash, Samuel and Yalkut, ad loc., and T.J.
Taanith, ii. 7) explains the action of Israel at Mizpah in that
sense. The text does indeed associate in a remarkable way a water-
rite (of which nothing else is known), fasting, and confession as
elements in repentance : " And they gathered together to Mizpah,
and drew water and poured it out before the Lord, and fasted on that
day, and said there, We have sinned against the Lord." Ascetic rites
(such as fasting) were ancient accompaniments of the confession of
sin, as in the ritual of the day of Atonement ; and the association
of asceticism with cold bathing is at least as old in Judaism as the
Essen es. In the Didache fasting precedes baptism (vii. 4), but it is
not clear how early the Synagogue introduced the now wide-spread
custom of bathing on the Eve of the Day of Atonement in connection
with the confession of sins. Talmudic is the rule "A man is bound to
purify himself at the festivals" (T.B. Rosh ffashana, 16 b), no doubt
with reference to ceremonial uncleanness. But Leviticus (xvi. 30) lays
it down : " From all your sins before the Lord ye shall be clean " on
the Day of Atonement, and the same word ("lino) which here means
spiritually clean also signifies physically and ritually clean. "Wash
you, make you clean, put away the evil of your doings" (Isaiah i. 16)
is one characteristic text of many in which the prophets make play
with the metaphor. The Sibylline call to actual baptism of the
sinning Greek world is obviously based on this very passage. Another
passage, to which great importance was justly attached in Rabbinical
thought, is Ezekiel xxxvi. 25 27 : "I will sprinkle pure water upon
you, and ye shall be clean ; from all your h'lthiness and from all
your idols will I cleanse you. A new heart also will I give you, and
a new spirit will I put within you ; and I will take away the stony
heart out of your flesh, and I will give you an heart of flesh. And
IV. PHARISAIC BAPTISM 41
I will put my spirit within you, and cause you to walk in my
statutes, and ye shall keep my judgments and do them." Here we
have, together, all the main ideas of Pharisaic baptism ; and it is
noteworthy that this passage from Ezekiel is extensively used in
Rabbinic homilies.
Such passages as these attest the early association between physical
and moral purification, such as meets us in the Johannine baptism.
And the ideas are close. Whoever invented the epigram " Cleanliness
is next to Godliness," it is a fair summary of Pharisaic conceptions on
the subject under discussion. Throughout the Psalms of Solomon " to
be clean" is identical with "to be forgiven." In Rabbinic Hebrew,
as in Biblical, the same word means physically and spiritually clean.
To "repent" is to "be purified." (Cf. the in^ 1 ? nn of T.B. Yoma,
38 b, and the phrase " before whom do you cleanse yourselves 1 " i.e.
repent of your sins, of the previous Note.) Sin is, conversely, un-
cleanness. There is no need to quote Biblical instances of the use.
In Rabbinic Hebrew the very strong word (mo) which literally means
"to be putrid" is a common term for "to sin." A very remarkable
figure of speech is attributed to Hillel. He bathed his body to keep
clean that which was made in the image of God (Levit. Rabba, xxxv.).
The connection between sin and atonement by bathing is brought out
in the Midrash on Ps. li. 4 on the text, " Wash me thoroughly from
mine iniquity." The Midrash comments : " Hence, whoever commits a
transgression is as though he was defiled by contact with a dead body,"
and he needs sprinkling with hyssop. Here the reference is clearly to
moral not to ritual transgression. In 2 Kings v. 14 we are told of
Naaman that after his leprosy was healed "his flesh came again like
the flesh of a little child"; and so the proselyte on his baptism
"became like a little child" (T.B. Yebamoth, 2 2 a, 48 b). On the
text "Be thou a blessing" (Gen. xii. 2) the Midrash (playing on
the similar words nD"Q "blessing" and rO*"l3 "pool") comments:
" As yonder pool purifies the unclean, so thou bringest near the far
off and purifiest them to their Father in Heaven" (Genesis Rabba,
xxxix. n). And those thus brought near are created anew. "He
who makes a proselyte is as though he created him" (ibid. 14)
thus conversion is a re-birth. In this sense the lustrations of Exodus
xix. 10 were regarded as physical accompaniments of the approaching
revelation on Sinai, when all the world was made anew. Man's re-
pentance is the cause, too, of the creation of the new heavens and the
new earth of Isaiah Ixvi. (Yalkut, Isaiah, 372). There are shades
42 IV. PHAEISAIC BAPTISM
of difference in this idea of renewal, especially as concerns the nature
of man. John's baptism seems to have this point in common with
the Pharisaic baptism of proselytes it was a baptism once for all.
For the proselyte had, in the Pharisaic view, adopted Judaism com-
pletely; and, like one born physically a Jew, he could not thereafter
evade the responsibilities of the religion which he had freely accepted,
just as he shared its hopes. Benedictions usually preceded the per-
formance of precepts. Not so with the tebilah, baptism, of the
proselyte. It was only as he ascended from the bath that he said :
"Blessed art thou who hast sanctified us by thy commandment and
commanded us concerning tebilah" (T.B. Pesahim 6 b). It may well
be, as Bousset states (Die Religion des Judenthums im neutestamentliche
Zeitalter ed. 2, 1906, p. 230) that there was nothing sacramental in
Pharisaic baptism. But, like the performance of the whole Law, it
was a consecration.
Pharisaic baptism, then, agreed with what seems to have been the
primitive Christian view that it was once for all, though in the case of
a revert, and of a slave seeking freedom, tebilah would be again
necessary. Tebilah, however, did not ensure sinlessness, or the
abrogation of the power to sin. That consummation was reserved
for the Messianic age. If, however, Christian baptism was the intro-
duction to the Kingdom, then no doubt baptism would carry with it
the hope of sinlessness. (On the problem of sin after Christian
baptism, and the apparent reversion to the Jewish theory of repentance,
see Prof. K. Lake, The Stewardship of Faith, London, 1915, p. 181).
John seems to imply also that the consequent change of mind (/oteravoia)
was also " once for all." In the Rabbinic theology such a permanent
amelioration of the human character was not possible, at least in
the earthly life. Men might move the stone from the mouth of the
well, but it had to be replaced, and the "evil inclination" (Yeser
hara) returned to where it had been and needed expulsion again and
again (Genesis Rabba, Ixx. 8). God will in the end destroy the evil
Yeser, but in human life the struggle is incessant and the Yeser leads
to sin daily (T.B. Qiddushin, 30 b). "In this world," says God to
Israel, " ye become clean and again unclean ; but in the time to come
I will purify you that ye never again become unclean " (Midrash, Tan-
huma, Mesora, 1718). Contrariwise (as perhaps John's baptism
intends), repentance brings the Messiah near (T.B. Yoma, 86 a, b. Cf.
Montenore, Jewish Quarterly Jteview, xvi. p. 236 and references there
given). The renewal of man's nature by repentance, unlike the re-birth
IV. PHARISAIC BAPTISM 43
by conversion, is continuous and constant. It is a regular process, not
a catastrophe. Israel is compared to the Angelic hosts. " As they
are renewed day by day, and return, after they have praised God, to
the fire from which they issued, so too the Israelites, if their evil
passions ensnare them in sin, and they repent, are forgiven by God
year by year and granted a new heart with which to fear him"
(Midrash, Rabba, Shemoth xv. 6; Echo, on v. 5).
In Ezekiel's phrase, God sprinkles pure water on Israel and puts
His spirit within him. By the middle of the second century the " last
of the Essenes," Phineas ben Jair, treats "purification" as what
Dr Schechter well calls " one of the higher rungs of the ladder leading
to the attainment of the holy spirit" (Studies in Judaism u. p. no).
But the connection between water and the Holy Spirit can be traced
much closer than this. In the Hebrew Bible the word "to pour
out" ("|SK>), properly applicable only to liquids, is applied to the
Divine Spirit. "In those days I will pour out my spirit on all
flesh " (Joel iii. i [ii. 28] ; cf. Ezekiel xxxix. 29). In Rabbinic
Hebrew the word which means "to draw" liquids (2NEJ>) is often
used of drawing the holy spirit. In Isaiah xii. 3 we have the
beautiful image : " With joy shall ye draw water from the wells of
salvation." With all of this compare Genesis JKabba, Ixx. 8 (on
Genesis xxix. 2seq.). "Behold there was a well in the field', that is
Zion ; lo there were three flocks of sheep : these are the three pilgrim
feasts ; from out of that well they drew water : from thence they drew
the holy spirit." Similarly the " Place of the Water-drawing," referred
to above in Note I., is explained as the place whence " they drew the
holy spirit" (T.J. Sukkah, v. i).
There is no ground then for the emphatic statement of Dr S. Krauss
(Jewish Encyclopedia, n. 499) that "The only conception of Baptism
at variance with Jewish ideas is displayed in the declaration of John
that the one who would come after him would not baptise with water
but with the Holy Ghost." The idea must have seemed quite natural
to Jewish ears, as is evident from the parallels quoted above. It must
be understood that some of these parallels (especially the last, which is
not older than the third century) are cited not as giving the origin of
the phrase in the Gospels, but as illustrating it. Such illustrations
may be used irrespective of their date in order to discriminate from
specifically un-Jewish ideas, those ideas which are found in the New
Testament, and are found again in Jewish circles later on. It is
important to know the ideas that recur. And, of course, the parallels
44 IV. PHARISAIC BAPTISM
may often be older than the first citation in which they are now to be
found. On the other hand, some borrowing from the Gospels must
not be dismissed as impossible or unlikely. An idea once set in
circulation would become general property, and if it fitted in with
other Jewish ideas might find a ready hospitality. It is well to make
this plain, though I do not for a moment think that in baptism we
have a case in point. The Rabbis have no hesitation in saying that
prayer replaced sacrifice, but they never hint at the thought that
baptism replaced the proselyte's sacrifice, as some writers suggest.
My main contention is that the recurrence or non- recurrence of New
Testament ideas and expressions is the surest test we have of their
essential Jewishness or non-Jewishness. The test is not perfect, for
parallels are occasionally missing to very Judaic ideas, and on the
other hand alien ideas did occasionally creep into the theology of
Judaism inadvertently. Often again, the usages and ideas of the New
Testament stand between Old Testament usages and later Rabbinic ; in
such cases they are valuable links in the chain. This is emphatically
the case with the New Testament references to Synagogue customs.
A good instance is also the metaphor of baptism with fire which,
though absent from Mark, occurs in both Matthew and Luke. Fire
in the Old Testament is not only capable of being "poured out" like
water, but its capacity in this respect becomes the basis of a second
derived metaphor : " He hath poured out his fury like fire " (Lament-
ations ii. 4). Fire is the natural element for purging, and is frequently
used in the Old Testament in the two senses of punishing and refining.
In the phrase "baptism by fire" we have thus two Old Testament
ideas combined ; fire is poured out, and it is used as a purifying and
punitive agent. Some see in the baptism by fire an allusion to
illumination. The light of day was removed by Adam's sin and
restored on his repentance (Genesis Rabba, xi. ; T.B. Aboda Zara, 8 a).
The illuminative power of repentance is already found in Philo (Cohn
and Wendland, 179): "From the deepest darkness the repentant
behold the most brilliant light." In the Testament of Gad (v. 7,
ed. Charles, p. 154) we read: "For true repentance after a godly
sort driveth away the darkness and enlighteneth the eyes." The same
illuminating function is (on the basis of Psalm xix. 8) often ascribed,
of course, to the Law, which further (with reference to Deut. xxxiii. 2)
is also typified by fire. But the context in which baptism by fire
occurs in the Gospels precludes all thought of fire as an illuminant.
In the Sibylline passage quoted above, the gracious promise of pardon
IV. PHAKISAIC BAPTISM 45
after true repentance on immersion in water has a harsher sequel. If
there be no repentance with baptism, there shall be destruction by fire.
For the Oracle continues (iv. 70) :
But if, ill-disposed, ye obey me not,
But with a fondness for strange lack of sense
Eeceive all these things with an evil ear,
There shall be over all the world a fire
And greatest omen with sword and with trump
At sunrise ; the whole world shall hear the roar
And mighty sound. And he shall burn all earth,
And destroy the whole race of men, and all
The cities and the rivers and the sea ;
All things he'll burn, and it shall be black dust.
Fiery baptism is a purging process, and in Luke (iii. 17) is associated
with the winnowing fan ("but the chaff he will burn"). The context
is equally clear in Matthew (iii. 12). This is a frequent Old Testament
usage. The idea is carried out most fully in a saying of Abbahu (end
of third century). Schottgen has already cited this parallel from T.B.
Sanhedrin, 39 a. Abbahu explains that when God buried Moses, he
bathed himself in fire, as it is written: "For behold the Lord will come
with fire" (Isaiah Ixvi. 15). Abbahu goes on to say, "By fire is the
essential baptism," and he quotes : " All that abideth not the fire ye
shall make to go through the water" (Num. xxxi. 23). Thus baptism
by fire is the divine analogue to man's baptism by water. Man could
not bear the more searching test.
One other phrase needs annotation : baptising in or into the name
of Christ. It is a difficult expression, but so are all the Rabbinic
metaphors in which the word "name" occurs. (Of. my article on
" Name of God " in Hastings, Dictionary of Religion and Ethics,
vol. ix.) Part of the significance of the Gospel expression is seen from
the corresponding late Hebrew (Gerim i. 7): "Whoever is not a
proselyte to (or in) the name of heaven (D^DS? DK^) is no proselyte."
(Cf. for the phrase, Koheleth Rabbah on Eccles. vii. 8 end.) In this
context the meaning is that the true proselyte is baptised for God's
sake, and for no personal motive. It is a pure, unselfish act of
submission to the true God. But in the Talmud (e.g. T.B. Yebamoth,
45 b, 47 b last lines) there is another phrase, which throws light on
this. Slaves, on rising to the rank of freemen, were re baptised, and
this slave baptism was termed a baptism to or in the name of freedom
(inns? DS^ or piin p Dfc6). A fine contrast and complement of
baptism in the name of freedom is the proselyte's baptism in the name
46 IV. PHARISAIC BAPTISM
of heaven, or in its Gospel form baptism in the name of Christ. The
Christian phrase, it is strongly contended by many, has a magical
connotation. But if so, (and it is hardly the case unless magical be
interpreted as equivalent to mystical), it was an acquired rather than
a primitive connotation. The explanation suggested comes near that
which regards baptism into the name as a Roman legal term, implying
that the newcomer is admitted on the roll of the patron's clients or
dependents. Never, surely, was a legal term more transfigured, both
in Church and Synagogue.
V. THE DOVE AND THE VOICE.
From two opposite sides the Rabbinic parallels to the Dove have
been minimised, by Dr Edersheim and Dr Abbott. The former, in
order to expose the "mythical theory," insists with "warmth of
language " that the whole circumstances connected with the baptism of
Jesus " had no basis in existing Jewish belief." The latter, in pursuance
of his view that the " Dove " arose from a textual misunderstanding,
argues equally that there was no extant Jewish symbolism which could
justify the figure.
But the doubt would have been scarcely possible had the two ideas,
the Dove and the Heavenly Voice, been treated together. It must not be
overlooked that in several passages the Heavenly Voice (Heb. Bath-Qol,
Daughter of the Voice) is represented as piping or chirping like a bird.
The notes of a bird coming from aloft often unseen would naturally
enough lend themselves to mystic symbolism in connection with the
communication of a divine message. There are two clear instances of
this use of the verb " chirp " with regard to the Bath-Qol in the Midrash
Qoheleth Kabbah. In one (on Eccles. vii. 9) we read : " I heard the
Daughter of the Voice chirping (nSVQVD) and saying : Return O back-
sliding children (Jer. iii. 14)." Even clearer is the second passage on
Eccles. xii. 7, though the text explained is verse 4 of the same chapter :
"And one shall rise up at the voice of a bird. Said R. Levi, For 18
years a Daughter of the Voice was making announcement and chirping
(naVDVD) concerning Nebuchadnezzar." (It is possible that in the
Jerusalem Talmud, Sabbath vi. 9, we have another instance, and that
we should correct nWlQE, which is the reading of the text there, to
nSQO). The evidence goes further. For while in these passages the
Heavenly voice is likened to the soft muttering of a bird, in one place
the BattirQol is actually compared to a dove. This occurs in the
Babylonian Talmud, Berachoth fol. 3 a) : "I heard a Bath-Qol moaning
as a dove and saying : Woe to the children through whose iniquities
I laid waste My Temple."
It is this association of the bird and the heavenly voice that may
underlie the Gospel narrative of the baptism, and at once illustrate and
48 V. THE DOVE AND THE VOICE
authenticate the symbolism of the Synoptists. There is no need to enter
here at length into the question of the Bath-Qol, for Dr Abbott (From
Letter to Spirit, Book n. and Appendix iv.) has admirably collected the
materials It is surely supercritical to question the antiquity of the
Bath-Qol in face of the evidence of Josephus (Antiquities, xm. x. 3)
and of the Rabbinic tradition concerning Hillel : " There came forth
a BathQoL and said : There is among you a certain man worthy of the
Holy Spirit, but the generation is not worthy thereof" (Jer. Sofa ix. 12,
otherwise 13). Dr Abbott aptly compares Mark i. 7. The whole
passage in Mark fits in with the belief that in the absence of the direct
inspiration of prophets by the Holy Spirit (after the death of Haggai,
Zechariah and Malachi), the Bath-Qol took its place (loc. cit.). The
Synoptists, like the Rabbis, never report a direct message from God.
In the Rabbinic literature the dove is for the most part an emblem
of Israel, its gentleness, fidelity, its persecution, its submission
(H. J. Holtzmarm, Die Synoptiker, ed. 3, p. 44, has collected some useful
materials on the symbolism of the Dove in other literatures). Here is
a characteristic Rabbinic passage (Midrash Tanhuma, p. Tesave : of. ed.
Buber, Kxod. p. 96), "Israel is compared to a dove (Canticles i.). As
the dove knows her mate and never forsakes him, so Israel, once
recognising the Holy One as God, never proves faithless to him. All
other birds, when they are about to be slaughtered, wince, but the dove
holds out its neck to the slayer. So there is no people so willing as
Israel to lay down its life for God. Just as the dove (after the flood)
brought light to the world, so God said unto Israel, who are likened
to the dove, Take olive oil and light my lamp before me." It has been
suggested (R. Eisler in the Quest, July 1912) that the Jews expected
the Messiah to be a second Noah, and that he would inaugurate the
era 1 y a punishment and a purification by a new flood. If the evidence
were sufficient to support this view (Eisler quotes Zech. xiv. 2, Joel
iii. [iv.] 18, and Ezekiel xlvii. i) we might see a Messianic reminiscence
of Noah's dove. Elsewhere other points of comparison are made
(Berachoth 530, etc.). As "the wings of a dove covered with silver
and her leathers with yellow gold" (Ps. Ixviii. 13) are the bird's means
of escape from danger, so is Israel saved by the Law, the pure words
of the Lord which are "as silver tried in a furnace of earth, purified
seven times " (Ps. xii. 7). But, as Wunsche well remarks (Neue Beitrage,
p. 501) the very comparison of suffering Israel to a dove may have
influenced the growth of the metaphor as applied to the Messiah, whose
function it was to save Israel. The "Spirit of God" of the Cosmogony
V. THE DOVE AND THE VOICE 49
in Genesis is thus sometimes (as we shall see later) compared to a dove,
sometimes to the spirit of the Messiah, who will not come until Israel
deserves the boon by Repentance (Genesis Rabba, ch. ii., ed. Theodor,
p. 17; Yalqut on Gen. i. 2). The identity is carried farther. In the
Bible God is said to have borne Israel on Eagle's wings, to protect
Israel as a parent bird protects its nest (Deut. xxxii. n); more
generally (Isaiah xxxi. 5): "as birds flying so will the Lord of hosts
protect Jerusalem." Nay more, just as the Divine Presence goes into
exile with Israel, so God himself is, with Israel, compared to a troubled
bird (though not a dove), driven from its nest (the Temple) while the
wicked prevail on earth (Mid rash on Ps. Ixxxix., Yalqut 833). It is
quite in keeping with this whole range of ideas to find the Targum
(Canticles ii. 12, etc.) interpreting as the "voice of the Holy Spirit of
Salvation" the text, the "voice of the turtle-dove is heard in our land."
(Cf. alfo Sifre on Deut. 314, with reference to Canticles ii. 8.)
Now it is obviously near at hand to find the main source of the
comparison of the Holy Spirit to a bird in Genesis i. 2, "And the
Spirit of God brooded (as a bird) upon the face of the waters." We
are happily not called upon to discuss the origin of the idea in Genesis
itself and its relation to the " world-egg." The Jewish commentators
(even on Jeremiah xxiii. 9) recognise no other meaning for the verb
used in Genesis (^1^), except brooding or moving as a bird. It is well
here to cite Rashi's note on the Genesis passage : " The Spirit of God
was moving : the Throne of Glory was standing in the air and moving
on the face of the waters by the Spirit of the Mouth of the Holy One
blessed be he, and by his Word like a dove that broods on the nest, in
French acoveter" This idea is derived by the commentator partly from
the Midrash Conen (Jellinek, Bet Hamidrash, ii. 24 : " And the holy
spirit and the holy Presence was moving and breathing on the water "),
but chiefly from the famous incident concerning Ben Zoma, a younger
contemporary of the Apostles. I have cited Rashi's adoption of it to
prove that some moderns have misread the Talmud when they regard
the Rabbis as deprecating Ben Zoma's idea. If anyone understood the
spirit of the Talmud it was Rashi, and the fact that he (like other
Jewish commentators) adopts the simile of the dove is of itself enough
to show that Ben Zoma's simile was not considered objectionable. More-
over, the passage relating to Ben Zoma is too frequently reproduced in
the Rabbinical sources for it to have been held in the disrepute which
has strangely been assigned to it by those who would like to expunge
this very clear parallel to the dove of the Synoptists, for it is obvious
A. 4
50 V. THE DOVE AND THE VOICE
that we have not only a comparison to the dove, but also to its
appearance "on the face of the waters," which fits in so well with the
baptismal scene at the Jordan, the dove descending as " Jesus, when he
was baptised, went up straightway from the water." Even without the
Ben Zoma analogue one could hardly doubt that the Synoptists must
have had Genesis i. 2 in mind.
The Ben Zoma incident is reported in the Talmud (Hagiga 15 a)
as follows : " Rabbi Joshua the son of Hananiah was standing on an
ascent in the Temple Mount, and Ben Zoma saw him but did not stand
before him. He said to him : Whence comest thou and whither go thy
thoughts, Ben Zoma? He replied, I was considering the space
between the upper waters and the lower waters, and there is only
between them a mere three fingers' breadth, as it is said, and the Spirit
of God was brooding on the face of the waters like a dove which broods
over her young but does not touch them. Rabbi Joshua said to his
disciples, Ben Zoma is still outside; for, 'and the Spirit of God was
hovering ' when was this ? On the first day. But the separation was
on the second day." There are several variants of the passage, but this
on the whole seems to me the most original in the important reference
to the dove. (Bacher, Agada der Tanaiten, ed. 2, Vol. i., p. 423, holds
the Tosefta Hagiga ii. 5 and Jer. Talm. Hagiga reading more original
because the allusion to the Temple is an anachronism.) Some of the
variants either suppress the dove or replace it by an eagle, citing
Deut. xxxii. 1 1 (where the same verb *|m is used of an eagle). Such
a harmonisation shows the hand of an editor, and the dove would not
have been introduced later. Dr Schechter (Studies in Judaism, n. p. 113)
is convinced that the dove is the original reading. Now the theory
that by the phrase " Ben Zoma is still outside " it was implied in this
" fragment of a Jewish Gnosis" (as L. Low, Lebensalter, p. 58 suggests)
that he had not yet returned to the orthodox path is quite untenable.
Other passages show that the meaning is : Ben Zoma is still out of his
senses. He had pried too closely into the problems of creation, and
had fallen into such perplexity that he confused the work of the first
with that of the second day. At all events, the figure of the dove is
not asserted to have originated with Ben Zoma, there is nothing in the
passage to imply that it was regarded as an innovation, or that Ben
Zoina's idea was unorthodox or heretical. Of course it is quite true, as
Dr Abbott urges, that the Rabbinic figure does not imply that the
Holy Spirit appeared visibly as a dove, but that the motion and action
of the Spirit were comparable to the motion of a dove over her young.
VI. LEAVEN.
The term leaven pfcf = Gk. w) is used in N.T. as a symbol of
" corruption." Something of the same idea is found in a well-known
Rabbinic passage to be discussed later. As to the O.T. conception
of leaven, an excellent account is given by A. R. S. Kennedy in
Encyclopaedia BiUica, col. 2754, "In the view of all antiquity, Semitic
and non-Semitic, panary fermentation represented a process of corruption
and putrefaction in the mass of the dough." Plutarch (Quaest. Rom. 109)
has the same idea. Philo, on the other hand, has the idea with a some-
what different nuance. To him, leaven symbolises the puffing-up of
vain self-conceit (Frag, on Exod. xxiii. 18), or the vice of insolence
(on Levit. ii. n, de offer, vi., Mangey n. 255). It is probable, too,
that the Roman satirist Persius (i. 24) also implies by fermentum
"vanity" rather than "corruption."
Later Jewish moralists (cf. Zohar on Gen. xlvii. 31) have made
extensive use of the leaven metaphor (especially with reference to
the prohibition of leavened bread ^Dn on Passover). As, however,
"leavened" bread was in itself more palatable as an article of food
than unleavened, the metaphorical use of "leaven" sometimes expresses
an improving process. Kennedy (loc. cit.) puts it rather differently :
" In the N.T. leaven supplies two sets of figures, one taken from the
mode, the other from the result, of the process of fermentation. Thus
Jesus likened the silent but effective growth of the ' Kingdom ' in the
mass of humanity to the hidden but pervasive action of leaven in the
midst of the dough " (Mt. xiii. 33). It is probable, however, that the
parable also takes account of the result; the leavened mass of humanity,
through intrusion of the leaven, attains a superior moral condition, just
as the leavened bread is a more perfect food than unleavened. Paul
applies the process in the opposite sense. Just as " evil company doth
corrupt good manners" (i Cor. xv. 33), so "a little leaven leaveneth
the whole lump!' (i Cor. v. 6 ; Gal. v. 9). The latter idea is Rabbinic
(Succah 56 b) both on this side ( Woe to the wicked, woe to his neighbour
42
52 VI. LEAVEN
N) and on the reverse side, for the righteous extends
virtue and its consequences to his neighbour (Happy the righteous,
happy his neighbour W3G$6 31D p^6 31D). But the Rabbinic idea does
not associate itself with leaven, but with the plague-spot, which appear-
ing in one house, compels the demolition of the next house (Mishna,
Negaim xii. 6; Sifra on Levit. xiv. 40; Weiss 73b). A very close
parallel to Paul's proverb (/u*pa v/x->; o\ov TO <vpa/xa v/xot) is found in
Hebrew (nbnJI HDJJ pIDIIO ETOn llK^n -?K3), but this occurs in a
fifteenth century book (Abraham Shalom b. Isaac's neve shalom xi. 2),
and is possibly a reminiscence of i Cor. But the sentence is not very
recondite, and may be independent of Paul. The permanence of the
effect of leaven in the mass is found in Yalqut Ruth 601, where the
leaven is said to cling to proselytes up to 24 generations.
Most notable of all metaphorical applications of leaven is its
association with man's evil tendencies or inclinations (jnn "IV). The
chief references in Rabbinic thought are two, both of which are
alluded to in the passage about to be quoted from Weber. The latter
(in his Judische Theologie) identified the evil inclination with the body.
On p. 221 (ed. 2 p. 229) he writes :
That the body is impure, not merely as perishable, but because it is the seat
of the evil impulse, we see from what is said in Num. Eabba xiii. (Wiinsche p. 312) :
God knew before he created man that the desire of his heart would be evil from his
youth (Gen. viii. 21). "Woe to the dough of which the baker must himself testify
that it is bad." This Jewish proverb can be applied to the Jewish doctrine of man.
Then the dough is the body, which God (the baker) worked and shaped, and the
impurity of the body is grounded in the fact that it is the seat of the yeser hara 1 ,
which is in the body that which the leaven is in the dough (nD l| JD^ Tltffc^), a
fermenting, impelling force (Berachoth 17 a).
But, as Prof. F. C. Porter rightly comments, Weber's view is not
well founded. This is Prof. Porter's criticism ("The Yeger Hara," in
Yale Biblical and Semitic Studies, p. 104).
Here the identification of the dough with the body, in distinction from the soul,
is mistaken. The dualistic psychology is supplied by Weber, not suggested by the
source. God's judgment upon man in Gen. viii. 21 is likened to a baker's con-
demnation of his own dough. The proverb is also found in Gen. Eabba xxxiv.
(Wiinsche, p. 152) as a saying of E. Hiyya the Great (Bacher, Agada der Tannaiten
ii. 530). The comparison of the evil impulse with leaven is an entirely different
saying, which should not be connected with the other. But in this case also the
dough is man, human nature, not the body. It is the prayer of E. Alexander
(Berach. 17 a) : "It is revealed and known before thee that our will is to do thy will.
And what hinders? The leaven that is in the dough and servitude to the Kingdoms.
May it be thy will to deliver us from their hand."
VI. LEAVEN 53
Matthew (xvi. 12) interprets the " leaven of the Pharisees " to mean
"teaching of the Pharisees," an interpretation which Allen (p. 175)
rightly rejects. Luke (xii. i) interprets it of "hypocrisy." Mark
(viii. 14-21) gives no explanation, but reads "beware of the leaven of
the Pharisees and the leaven of Herod." It will be seen that this
reading strangely agrees with the words of R. Alexander's prayer :
" the leaven that is in the dough (= the leaven of the Pharisees) and
servitude to the Kingdoms (= the leaven of Herod)." Two things
impede man : the evil yeser and the interference of alien rule. Both
these preventives to man's advance will vanish with the coming of the
Kingdom. With the advent of the Messiah the evil yeser will be finally
slain (see refs. in Schechter, Aspects of Rabbinic Theology, p. 290); and
in the second place with the Kingdom of heaven Israel triumphs over
Rome (Pesiqta K. 50 a; Pesiqta R. 7 5 a).
There is a striking saying attributed to R. Joshua b. Levi, who
belongs to the first half of the third century. It is obvious that the
parable of the leaven requires a favourable application of the symbol.
R. Joshua carries this application to the extent of likening leaven to
peace. "Great is peace, in that peace is to the earth as leaven to
dough; for had not God set peace in the earth the sword and the
wild-beast would have depopulated it" (Pereq ha-Shalom, beginning;
Bacher, Agada der Paldstinensischen Amorder I. 136). The exact
force of R. Joshua's comparison is not clear. He bases his idea on
Leviticus xxvi. 6 : and it is possible that he had in mind the thought
found in the Sifra on that text (ed. Weiss, p. 1 1 1 a). "I will give
peace in the land" and (in the usual translation) "I will make evil
beasts to cease." So R. Judah interprets. But according to R. Simeon
the meaning is that God will not destroy evil beasts, but will render
them innocuous ; for "the divine power is better seen when there are in
existence evils which do not injure" (comparing Isaiah xi. 6 8). In
this sense, peace would be not inert, but an active agency ; a ferment
of the good against the evil. The idea of stirring, agitating (m and
DVD), is not only applied to the evil yeser. It is also used of the good
yeser. "Let a man stir up his good yeser against his bad" (T. B.
Berachoth, 5 a) ; " rouse thy [good] yeser and thou wilt not sin " (Ruth
Rabbah, towards end). Peace is thus the leaven, stirring up the good
yeser, to strive against hostile forces. If Peace is to have her victories,
she must fight for them.
VII. PUBLICANS AND SINNERS.
The Roman taxes and custom duties and their mode of collection
are admirably described by Schiirer (T. 17) and Herzfeld (Handels-
geschichte der Juden des Alterthums, 47). The taxes proper were in
Roman times collected by state officials, but the customs were farmed
out to publicani. In maritime places these were particularly onerous,
and Herzfeld ingeniously cites the proverbial maxim ('Aboda Zara, 10 b)
11 Woe to the ship which sails without paying its dues " in illustration
of Matthew ix. 9, 10. That the demands of the publicani and their
underlings were often excessive is natural enough, and especially
when the officials were native Jews (cp. Biichler, Sepphoris, pp. 13,
40, etc.) the class was consequently the object of popular resentment.
It is not the case (as Schiirer assumes) that the Jewish authorities
connived at frauds on the regular revenue. At all events the trick
permitted in the Mishnah (Nedarim, iii. 4) was interpreted by the
Talmud (Nedarim, 28 a) as having reference not to the authorised taxes
but to the arbitrary demands of unscrupulous extorters or inventors of
dues. " The law of the Government is law " on which see Note VIII
is used on the Talmudic folio just quoted as making it impossible
that the Mishnah (which permits one to evade "murderers, robbers,
confiscators and tax-gatherers" by falsely declaring the property
coveted to be sacerdotal or royal property) can refer to lawful taxes.
We have already seen that the tax-gatherers are associated with
robbers and murderers (cp. also Baba Qama, 113 a). Hence they were
regarded as unfit to act as judges or to be admitted as witnesses
(Sanhedrin, 25 b). An early baraitha made a tax-gatherer ineligible as
haber ; in the older period the disqualification did not cease with the
abandonment of the occupation, afterwards this particular severity was
mitigated (Bechoroth, 31 a). It is clear from the last quotation that
the publican might sometimes be a man of learning. Yet this con-
demnation was not universal. Baya (or Mayan) the tax-gatherer (or
VII. PUBLICANS AND SINNERS 55
his son), who was charitable to the poor, was publicly mourned and
honoured at his death (Sanhedrin, 446; J. Hagiga, ii. 2). So, con-
cerning the father of Ze'ira (Sanhedrin, 25 b) a favourable report is
made. There is also a (late) story of Aqiba (or in another version
Johanan b. Zakkai), telling how the Rabbi with eagerness reclaimed
the son of an oppressive tax-gatherer, teaching him the Law, and
bringing peace to the father's soul (Kallah, ed. Coronel, 4 b. For other
references see Jewish Encyclopedia, vol. I. p. 310).
The association in the Gospels of the two expressions Publicans
and Sinners is parallel to the combination of " publicans and robbers "
in the Rabbinic literature. The " sinners " were thus not those who
neglected the rules of ritual piety, but were persons of immoral life,
men of proved dishonesty or followers of suspected and degrading
occupations. The Rabbis would have been chary of intercourse with
such men at all times, but especially at meals. For the meal was not
regarded simply as a satisfaction of physical needs. It was a service
as well, consecrated by benedictions ; it was also a feast of reason. The
keynote of this is struck in the saying of R. Simeon (Aboth, iii. 3):
" Three who have eaten at one table and have not said over it words
of Torah, are as if they had eaten sacrifices of the dead (idols), for
it is said : All tables are full of vomit and filthiness without place
(Maqom)." This last word is taken in its secondary sense to mean
the Omnipresent, God. "But," continues R. Simeon, "three who
have eaten at one table, and have said over it words of Torah, are
as if they had eaten of the table of God (Maqom), blessed be he, for it
is said: This is the table that is before the Lord" (Ezekiel xli. 22).
This conception is exemplified also in the table-discourses of Jesus to
his disciples, and lies, to some extent, at the bottom of institution of
the Eucharistic meal. In Jewish life this idea that the table is an
altar gained a firm hold and led to a whole system of learned readings,
devotions, and most remarkably, of hymns during meals, the Passover
home-rites being but a conspicuous example of a daily Jewish usage.
Just, then, as later on Christians would not share the Eucharistic meal
with notorious evil-livers, so the Jewish Rabbi at various periods
would (with less consistent rigidity) have objected to partake of any
meal with men of low morals. So, also, Jesus' disciples are exhorted
(Matthew xviii. 17) to treat certain offenders as "the Gentile and
the Publican" with whom common meals would be impossible. The
Essenes held a similar view as to the exclusion from their table of
those who did not share the Essenic principles.
56 VII. PUBLICANS AND SINNERS
When, then, we find that the "pure-minded in Jerusalem would
not sit down to a meal unless they knew who their table-companions
were to be" (Sanhedrin, 23 a), the motive was neither pride nor
exclusiveness, but a desire that the meal should not degenerate into
mere eating and drinking. They would wish to be assured of the
presence of fit comrades for learned and edifying discourse. They
would not readily accept invitations to banquets at all, " the student
who is always found at other people's tables profanes the name of
God" (Yoma, 866, Aboth de R. Nathan, i. xxvi.). The Rabbis were
convivial, but not gluttons ; and many of them would never eat outside
their own homes except at a " meal of duty," i.e. a semi-religious
function, such as a marriage festivity. Instructive is the incident
recorded as having occurred in Jerusalem c. 65 A.D. At the feast held
on the circumcision of Elishah b. Abuyah, among those present were
Eleazar b. Hyrqanos and Joshua b. Hananyah. While the other
guests were partaking of meat and wine, these two sat "stringing
together," like pearls on a cord, the words of the Scriptures. (Qoh. R.
on viii. 8 ; see Bacher, Prooemien etc., p. 9.) To such men, a meal was
not a mere occasion for eating and drinking. The reluctance to eat
with the 'Am ha-ares was of a different origin ; fears as to neglected
tithes etc. arose (cf. Biichler, Der Galildisch Am-haares 162, 208).
Similarly, with regard to joining the heathen at table, fear of mixed
marriage came to the fore (cf. A. Wiener, Die judischen Speisegesetze,
Breslau 1895, pp. 430 seq.; W. Elmslie on 'Abodah Zarah v. 5, with
references there given). It is clear from the context that such joint
meals did take place. But with all this there went a unique sense
of obligation to the poor and the miserable. Isaiah (Iviii. 7) had
spoken of the duty " to bring the poor that are cast out to thy house,"
and from the middle of the second century B.C. it was laid down as a
duty to entertain at meals "the children of the poor" (Aboth, i. 5), to
which category were later added "those who were distressed in soul"
(Aboth de R. Nathan, n. xiv.). It is not at all the case that a Pharisee
would have declined to receive even "sinners" at his own table. But
he might have refused an invitation to join them at their table, where
the ritual and atmosphere could hardly fail to be uncongenial.
Probably the Pharisees exaggerated the force of evil example (cf.
Hernias Mand. x. i. 4 against <i\icus efli/i/cais). We frequently find
in the second and third centuries regulations due to a sensitive
repugnance to placing oneself in a position of suspicion. (This is
the meaning of some passages quoted by Dr Biichler in his essay on
VII. PUBLICANS AND SINNEES 57
"The Political and Social Leaders of the Jewish Community of
Sepphoris," ch. iii. 5.) On the other hand especial eulogy was
expressed of those who defied suspicion and remained untainted in
an environment of temptation (Pesachim, 113 a). But for the most
part the Pharisees entertained an exaggerated fear both of the danger
of actual moral lapse, and even more of the loss of repute from
suspicion of such lapse, likely to be incurred by association with
dishonest men or unchaste women. It was, however, a defensible
theory of conduct, and one which most educationalists of the present
day accept. We sometimes find Rabbis prepared to defy suspicion and
temptation when engaged in what we now call rescue work, but such
cases are rare. Moreover, as the women who were the unchaste associ-
ates of unchaste men were chiefly foreigners, the Rabbis felt no strong
impulse towards putting their heads in the lions' dens.
But, to return to my main point, it is unnecessary to cite the
Rabbinic passages in which men are warned of the personal dangers of
associating with men or women of low morals. Some passages have
already been quoted in Note VI. (Of. also 0. Taylor's Note on
Aboth i. 8 [7].) Another common saying was that though the evil
yeser of idolatry had been slain, the evil yeser of unchastity was very
much alive (Yoma, 69 b ; 'Aboda Zara, 1 7 b). There was much lack
of courage, but less taint of self -righteousness, in the efforts of the
moralist to preserve men from temptation and contagion. Luke's
Pharisee who thanked God that he was not as the Publican (Luke
xviii. n) must have been an exceptional case, one of the weeds of
ritualism, not one of its ordinary or natural fruits. "A familiar
saying in the mouth of the Rabbis of Jabneh," says the Talmud
(Berachoth, 17 a), " was this : I (who study the Law) am a creature (of
God), and my fellow man is a creature (of God). My work is in the
city, his in the field; I rise early to my work, he rises early to his.
Just as he cannot excel in my work, so I cannot excel in his. Perhaps
thou wilt say : I do much and he does little (for the Torah). But we
have learned (Menahot, no a), He who offers much and he who offers
little are equal, provided that each directs his heart to Heaven." The
penitent publican's prayer "God be merciful to me a sinner," as
well as his gesture ("he smote upon his breast") are essentially
Pharisaic ; it is interesting to see Luke introducing this last ritualistic
touch in an attack on ritualism. The Pharisee placed the repentant
sinner on a higher pedestal than the out-and-out saint (Berachoth, 346).
This was expressed in another way by saying that God honours the
58 VII. PUBLICANS AND SINNEES
repentant. Again, "Broken vessels are a disgrace for a man to use,
but God loves the broken heart " (Midrash, Levit. Rabba, vii. 2 ; Mid.
Tehillim on xviii. 2). A penitent publican, like any other repentant
sinner (cf. the fine passage on the harlot in Philo, On Monarchy ii. 8),
would find a ready welcome to the arms of the Rabbi. True it was
held difficult for a publican to repent (Baba Qama, 94 6), but by
repent is meant in the context to 'make restitution. The victims of the
publican's oppression were not easily identifiable, and it was not in
the sinner's power to undo the wrong which he had inflicted. Besides,
the community must not connive at such plundering by manifesting
over-readiness to take back payment from ill-gotten gains. The Rabbis
would have scornfully rejected the cynical principle pecunia non olet.
But though the community might decline the preferred restitution,
God would accept ; man might justly reject, yet the sinner must do
restitution (anonymously) for God's sake. On the basis of this same
passage (Baba Qama, 94 95) Maimonides thus accurately sums up the
position : " If the robber wished to repent, and the thing actually
stolen being no longer in existence, offered to repay the value of the
stolen thing, it is an ordinance of the sages that they must not accept
the money, but they help him and pardon him, so as to make near
unto the penitent the right way ; yefc if one received the money from
him he would not forfeit the approval of the sages" (Hilchoth n^TJ,
i. 13). And even though the Scripture says the opposite (Proverbs
xxi. 27: "The sacrifice of the wicked is an abomination"), the gift-
offerings of sinners were accepted in the Temple in order to encourage
them to repent (Hullin, 5^; Pesikta R., 192 a).
There was in the Pharisaism of all ages a real anxiety to make the
return of the sinner easy. It was inclined to leave the initiative to
the sinner, except that it always maintained God's readiness to take
the first step. Jesus in his attitude towards sin and sinners was more
inclined to take the initiative. Yet, until the modern epoch of a new
humanism, society has worked by reprobation rather than attraction,
and the practical methods of Western communities in dealing with
criminals have been as harsh as the methods of any other system.
And Rabbis did often act in the same spirit as Jesus. In the first
place if a genuine Pharisee ever thanked God that he was not as the
publican, he would only have done so in the spirit of the famous
utterance: "There, but for the grace of God, goes John Baxter."
Thus a first century Rabbi (Nehunya ben Haqana) utters a prayer in
which he contrasts the happier lot of the speaker who frequents the
VII. PUBLICANS AND SINNERS 59
house of study and the less happy lot of someone else who frequents
theatre and circus (Berachoth, 28 a). This prayer is simply a grateful
recognition for good fortune ; it in no sense implies (except quite in-
directly) that the speaker prides himself on being a better man. His
lines have been cast in happier places. Such prayers and such an
attitude are moreover an encouragement to right living. They aim
at showing that virtue has its abundant reward in a sense of duty
done and in the confident hope of future bliss. And here arises the
real difficulty. Praying for sinners (i.e. for other people), fussy efforts
at rescuing outcasts (i.e., again, other people) may come very close
indeed to "pharisaic" self-righteousness. These psychological problems
are so complex that they transcend the grasp of most theologians,
and the latter are driven to look at the problems incompletely and
therefore erroneously. One might put it generally by asserting that
the Rabbis attacked vice from the preventive side ; they aimed at
keeping men and women honest and chaste. Jesus approached it from
the curative side; he aimed at saving the dishonest and the unchaste.
The Rabbis thought that God loves the prayers of the righteous ;
they held that all the divine sympathy was not expended on the
petitions of the sinner. But the association of the sinner with the
righteous in prayer and fasting was necessary to make religion a
real thing (Kerithoth, 6 b). And as regards actual, practical intrusion
into the life of the sinner, there is much in the Rabbinic literature
urging men to seek the active reclamation of the erring. "He who
does not pray for his neighbour or bring him to penitence himself will
suffer" (Midrash Jonah). As Maimonides puts it (on the basis of
several Talmudic passages, Ber. 126 etc.): "Whoever has it in his power
to prevent others from sinning, yet leaves them in their stumbling, has
no forgiveness" (Teshuba, iv. 2; Deoth vi.). So far does this counsel
go, that the Israelite is required to press his reproof and his efforts at
reclamation on the sinner though the latter revile and even strike his
monitor (Erachin, 16 b). Thoroughly in accord with Rabbinic teaching
(Sif ra on Leviticus xix. 1 7) is the Targum rendering of that same text :
" Thou shalt rebuke thy neighbour and not receive punishment for his
sin " which your active reproof might have prevented. His sin becomes
your sin. The parable of Moses and the stray sheep which he seeks in
the desert and bears in his bosom (Midrash, Shemoth Rabba, ch. ii.)
points the same moral. This idea is already found in the Psalter,
" I have gone astray like a lost sheep; seek thy servant " (Ps. cxix. 176).
So, Ps. xxxiv. 14, "Seek peace and pursue it," was held by the Rabbis
60 VII. PUBLICANS AND SINNERS
to compel men to go about the world as peace-makers. Perhaps the
most apt citation in this connection (the subject is further discussed
in the note on Forgiveness) is the manner in which Jewish homilists
set up Aaron as an ideal character. There can be no question
here but that this idealisation is earlier than the Gospel criticism
of the Pharisaic indifference to "sinners." We meet with its germ
in Malachi ii. 6, where Levi is eulogised in the words "he did
turn away many from iniquity." These words were applied specifi-
cally to Aaron (Aboth de R. Nathan, i. xii.), and Hillel already has
the saying : "Be of the disciples of Aaron, loving peace and
pursuing peace, loving mankind, and bringing them near to the
Torah" (Mishnah Aboth, i. 12). Here is the same spirit as "the light
of the law which was given to lighten every man " (Testament of Levi,
xiv.). This "bringing men near" applies to proselytism, but in
Rabbinical literature it is again and again used of active labour in
rescuing sinners. Nitai the Arbelite cautioned against association
with the wicked (Aboth, i. 7, on the relation between this and i. 12 see
Jewish commentaries). But this was not the only view held. Aaron,
we are told, would offer friendly greetings to the wicked (Johanan b.
Zakkai, we are told, Ber. 17 a, punctiliously greeted heathens in the
market-place), who would thus be shamed from their sin (Aboth de
R. Nathan, loc. cit.); he would go out on the roads at night, intercept
those who were about to transgress, and with soft, affectionate words
of intimate comradeship, would divert them from their intention
(Buber Tanhuma, Numbers, p. 10), and thus "all Israel loved Aaron,
men and women." To "bring another man near" to the Torah was
to create a soul (Aboth de R. Nathan, n. xxvi.). This ideal, pre-
Christian in Rabbinic literature, was also post-Christian. There is the
oft-cited case of R. Meir (to whom was due a first draft of the Mishnah).
Hard by his abode lived men who were violent criminals, and they so
troubled Meir that he prayed for their extinction. But his wife Beruria
checked him, and at her instigation he admitted that it was better to
pray for their conversion (Berachoth, 10 a). Meir, it will be remembered,
was noted for his persistent friendship to his heretic and sinful master
and friend, Elisha ben Abuya, for whose return to the fold he so
tenderly exerted himself. Even more to the present point is the
conduct of R. Ze'ira. In his neighbourhood were robbers and highway-
men, but Ze'ira showed them intimate friendship, so that they might
be brought to penitence, which indeed came about in their sorrow at
the Rabbi's death (Sanhedrin, 3 7 a). Pathetic, too, is the idea of
VII. PUBLICANS AND SINNERS 61
R. Joshua ben Levi that the Messiah would eventually be found at
the gates of Rome, among the sick poor, binding up their wounds
(Sanhedrin, 98).
And so the story might be continued. The Rabbis could see the
good in all men, and might exalt above those of spotless reputation one
engaged in what they considered unsavoury and demoralising occupa-
tions. Gazing over the crowd, Elijah picked out as assured of the
future life a jailor, who had cared for the morals of his prisoners
(Ta'anith, 22 a). On occasion of a drought in Judaea, people reported
to Abbahu that they knew a man whose prayers for rain were
infallible. His popular name was Pentekaka (lit. the man of Five Sins).
R. Abbahu interviewed him, inquired as to his means of livelihood,
whereupon Pentekaka said that his name corresponded to his profession.
" I am occupied with harlots, I clean the theatre, I carry the vessels to
the bath, I amuse the bathers with my jokes, and I play the flute."
But, asked the Rabbi: "Have you ever done a good thing in your
life?" Pentekaka answered: "Once I was sweeping out the theatre
and I saw a woman standing between the pillars, bitterly weeping.
I spoke to her and ascertained that her husband was a prisoner, and
she could only buy his freedom by sacrificing her chastity. So I sold
my bed and my pillow and all my possessions, and I gave the money
to her, bidding her go ransom her husband and not sell her honour to
strangers." Hearing such words from such a man, Abbahu exclaimed :
" Thou art the man fit to pray for us in our hour of trouble " (Talmud
Jer. Ta'anith, i. 2).
VIII. "GIVE UNTO CAESAR."
To Samuel of Nehardea (c. 165 c. 257 A.D.) belongs the honour of
formulating the principle which made it possible for Jews from the
early middle ages onwards to live under alien laws. Jeremiah had
admonished his exiled brothers : "Seek ye the peace of the city whither
I have caused you to be carried away captives, and pray unto the
Lord for it ; for in the peace thereof shall ye have peace." It grew
necessary to become more explicit, and the Rabbis proclaimed a
principle which was as influential with the synagogue as " Give unto
Caesar that which is Caesar's " became with the Church. '* The law of
the government is law " (dina dtmalchutha dina, T.B. Baba Qama 1 13 b ;
Baba Bathra 54 a ; Gittin 10 b ; Nedarim 28 a) said Samuel, and ever
since it has been a religious duty for the Jews to obey and accom-
modate themselves as far as possible to the laws of the country in
which they are settled or reside (cf. my remarks in Encyclopaedia
Britannica, ed. IT, vol. xv. p. 404). "To Jeremiah and Mar Samuel,"
says Graetz, "Judaism owes its possibility of existence in a foreign
country" (Geschichte der Juden, iv. 2, iii.).
What Mar Samuel, however, did was not to devise a new principle,
but to give that principle the precision of law. Very much in the
history of civilization has depended on the power of moralists to
concentrate a theory into an epigram ; the sayings of Jesus and
Samuel are apt illustrations. Long before Samuel, however, the same
attitude prevailed. At the period of the disastrous Bar Cochba insur-
rection, when Roman law and Roman administration were bitterly
resented, the Rabbinic teachers impressed on their brethren the
absolute duty of paying the taxes imposed by the Government.
According to the statement of Johanan ben Zakkai (Mechilta on
Exodus xix. i, ed. Friedmann, p. 61 b top) the Romans, after the
destruction of the Temple, imposed the enormous tax of fifteen shekels ;
and though the exact significance of this is doubtful, it may have been
viii. "GIVE UNTO CAESAR" 63
a tax on leases ; we know that the Roman imposts were very con-
siderable (cf. Biichler, The Economic Conditions of Judcea after the
destruction of the Second Temple, 1912, pp. 62 seq.). Yet it was held
obligatory to pay these taxes with the utmost scrupulosity, in so far as
they were lawfully imposed, and were not the whimsical exactions
of the publicans (T.B. Baba Qama 113 a; Nedarim 28 a ; Tosefta,
Nedarim iii. 4; Semah ii. 9 where evasion of taxation is denounced
as equivalent even to murder, idolatry, incest, and profanation of the
Sabbath).
Nor does the evidence extend only to the Hadrianic period. It
goes back even further. On the text in Ecclesiastes viii. 2 (" I counsel
thee, Keep the king's command, and that in regard of the oath of
God "), the Midrash (Tanhuma on Genesis viii. 1 6, Noah i o ; ed.
Buber, p. 33) comments thus : " The Holy One said unto Israel, I
adjure you that even though the (Roman) Government decrees against
you harsh decrees ye shall not rebel against it for anything that it
decrees, but keep the king's command. But if it decrees against you to
abandon the Torah and the commandments and deny God, then do not
obey it, but say unto it : I keep the king's laws only in those things
which are necessary for the government." The Midrash goes on to
cite the conduct of Daniel's three friends who assure Nebuchadnezzar :
" In so far as duties and taxes are concerned, in all that thou decreest
upon us, we will obey, and thou art our king, but to deny God we
have no need to answer thee in this matter... we will not serve thy
gods, nor worship the golden image which thou hast set up " (Daniel
viii. 1 6). The difficulty of this compromise was twofold. First, bad
government is incompatible with the Kingdom of God (Schechter, Some
Aspects of Rabbinic Theology, p. 106), and the Roman Government
was often deserving of inclusion in the category of bad government.
Secondly, the tendency of Roman emperors to assert their divine
status, and to found their authority on a theory somewhat approaching
that of divine right, made Roman rule in general obnoxious to Jewish
sentiment. Nevertheless, as Tacitus admits, the Jews were long
patient under the irritation ; they rebelled only when chronic irrita-
tion was transformed into specific provocation : " Duravit tamen
patientia Judseis usque ad Gessium Florum procuratorem ; sub eo
bellum ortum." That occurred in the spring of 66 A.D. Some of the
previous procurators had so far studied Jewish susceptibilities as to
strike (probably employing Jewish workmen) coins of special designs
for local circulation in Judaea. There were on these no figures of
64 viii. "GIVE UNTO CAESAR"
animate objects, but only ears of corn, palm-trees or branches (?), the
cornucopia, diota, covered vase, or wreath. In 35 A.D. Pontius Pilate
struck coins decorated simply with the laurel wreath and the lituus or
augur's wand (T. Reinach, Jewish Coins, ed. Hill, p. 41). At a much
later period we find a Rabbi (Nahum b. Simai) described as remarkable
for his " holiness " because " he never looked upon the form of a coin "
(Pesahim 104 a and parallels in Bacher, Agada der Paldstinensischen
Amorder in. 616); this probably refers, however, to coins on which
were figures of the emperors.
Very clearly belonging to the period of the Vespasian war is the
saying recorded in the Mishnah, Aboth iii. 2. The authority cited is
Haninah (Hananiah), the prefect of the priests, who was a con-
temporary of Johanan ben Zakkai, and like him a member of the
peace party. "Pray for the peace of the kingdom," said Haninah,
" since but for the fear thereof men would swallow one another alive."
This may allude specifically to public prayer on behalf of the ruler
(see Ezra vi. 10 ; Baruch i. n ; I. Mace. vii. 33 ; Philo, Leg. adCaium,
xxiii., xlv. ; Josephus, War u. x. 4; T.B. Yoma 69 a; I. Timothy
ii. i, 2, and cf. Schiirer ii. 24; Singer, Transactions of the Jewish
Historical Society of England iv. 103). It is interesting to add a
conjecture made by Dr Bacher. We have no record of the precise
liturgical phraseology of the prayer for the Government unless
Dr Bacher has discovered it in the A both de Rabbi Nathan (u.
ch. xxxi. p. 68, ed. Schechter). By a slight emendation of the text,
the words of the prayer would be : "May it (the Roman Government)
rule over us for all time" (D^TI ^ m ni^>1K> KnnE> Bacher, Agada
der Tannaiten, ed. 2, vol. i. p. 52).
But though thus prepared to obey Rome and abide by all its lawful
regulations, there was to be no compromise when Caesar infringed the
sphere which appertained to God. This distinction we have already
seen in the Midrash, but we find the same very clearly expressed in
the pages of history. Josephus records several instances of the
readiness of the Jews to suffer death rather than admit the images of
Caesar (e.g. War n. ix. 3). Most nearly illustrative of the subject
before us is the passage in which the historian describes what took place
when Caius Caesar (who succeeded Tiberius as emperor in 37 A.D.)
sent Petronius with an army to Jerusalem to place his statues in the
Temple ; he was to slay any who opposed this step, and to enslave the
rest of the nation (War n. x. i). Petronius marched from Antioch
southwards towards Judaea ; but when he reached Ptolemais in Galilee
VIIL "GIVE UNTO CAESAK" 65
he was met by a deputation of Jews. Prevailed upon by the multitude
of the supplicants, he summoned a meeting of all the men of note to
Tiberias, where he declared unto them the power of the Romans and
the threatenings of Caius, and also pronounced their petition un-
reasonable. " For as all the nations subject to Rome had placed the
images of the emperor in their several cities among the rest of their
gods, for them alone to oppose it was like the behaviour of rebels, and
was insulting to the emperor." Josephus then proceeds as follows, and
the passage may usefully be cited in full ( 4, 5) :
And when they insisted on their law, and the custom of their country, and how
it was not lawful for them to put even an image of God, much less of a man, in
any profane part of their country, much less in the Temple, Petronius replied,
" And am I not also bound to keep the law of my lord? For, if I transgress it and
spare you, I shall justly perish. And he that sent me, and not I, will war against
you; for I am under command as well as you." Thereupon the whole multitude
cried out that "they were ready to suffer for their Law." Petronius then tried
to quiet their noise, and said to them, "Will you then make war against the
Emperor ? " The Jews said that they offered sacrifices twice every day for the
emperor and the Roman people ; but if he would set up his statues, he must first
sacrifice the whole Jewish nation ; and they were ready to expose themselves to be
slain with their children and wives. At this Petronius felt both astonishment and
pity on account of their invincible regard to their religion, and their courage which
made them ready to die for it.
Petronius yielded, and incurred the censure of Caius, but the
latter's death in 41 intervened to save him from the consequences of
his complacency to the Jewish steadfastness towards their God, and
his own disobedience towards Caesar. Philo (Leg. ad Caium xxxii.,
xxx vi.) narrates the same circumstances at greater length ; but he,
too, records that the Jews willingly and even enthusiastically accepted
the sovereignty of Caius, in all matters except the proposed " innova-
tions in respect of our Temple;... the honour of the emperor is not
identical with dishonour to the ancient laws (of Judaism)." Caius
well represents the opposite case when he retorts (xliv.) : " Ye are
haters of God, in that ye deny me the appellation of a god," though
he was generous enough to attribute this blindness to the Jews as a
misfortune rather than as a fault: "These men do not appear to me
to be wicked so much as unfortunate and foolish, in not believing
that I have been endowed with the nature of God." This misfortune
and unwisdom the Jews never abandoned, and thus were always
protagonists in the refusal to give unto Caesar that which is God's.
IX. JEWISH DIVORCE IN THE FIRST CENTURY 1 .
Social conditions in Palestine at the beginning of the Christian era
were bewilderingly complex. Restricting our attention to the question
of marriage, we find at the one extreme a sect (the Essenes) which
advocated celibacy, and possibly at the other a sect (the Zadokites)
which forbade divorce, or at all events remarriage. Then there were
the aristocrats of the court circle who had adopted Roman ways. For
instance, Josephus records two instances in which women of the
Herodian house (Salome, 25 B.C., and Herodias, contemporary with
John the Baptist) divorced their husbands, and paralleled the excesses
denounced by Juvenal in his sixth satire (Mark x. 1 2 may be directed
against such licentiousness). The Pharisaic Judaism of the same period
regarded marriage as the ideal state, yet freely permitted divorce. If
the ideal were shattered it seemed to accord best with the interests of
morality to admit this, and afford both parties to the calamity a second
chance of lawful happiness. The marriage bond should be inviolable,
but must not be indissoluble.
The progress of law and custom in Jewry tended not to modify the
theoretical ease of divorce, but to increase its practical difficulties. The
Gospel view was that the Deuteronomic divorce was a concession to
human weakness, a lowering of the earlier standards of Genesis which
held marriage to be indissoluble. The Rabbinic reading of history was
different. The Pentateuch introduced the formality of the written
1 This Note was written at short notice, to comply with the urgent request of
the late Lord Gorell, Chairman of the Royal Commission on Divorce and Matrimonial
Causes. The Note was presented to the Commission, and the author was examined
on November 21, 1910. I received valuable help from Dr M. Berlin. For various
reasons it seems best to leave the Note without substantial change. Hence it is
impossible to allude to the interesting views of Prof. L. Blau published in 1911
1912. In his Juduche Ehescheidung und derjiidische Scheidebrief, pages 45 72 of
Part I are devoted to an exposition of the New Testament passages on Divorce, with
Rabbinic and other parallels.
IX. FIRST CENTURY DIVORCE 67
Letter of Divorce, and Rabbin ism regarded this as an advance in
civilization, not a retrogression. The Deuteronomic divorce was a
restriction of the earlier right or power of the husband to discard his
wife at will and with scant ceremony. Rabbinism contrasted the
decent formalities of the Mosaic Code with the arbitrary indelicacy of
primitive custom (Genesis Rabba ch. xviii.).
The Pentateuch, however, contemplates the husband as alone having
the right to effect a divorce. In the Babylonian Code of Hammurabi
the wife had some power of initiative, and when recently the Egyptian
papyri of the fifth century B.C. were discovered, it was thought that
these Aramaic documents showed the Jewish woman in possession of
the same status as man in regard to initiating divorce. Closer study,
however, shows that at most the woman of the papyri could claim a
divorce, she could not declare one. This condition remained unaltered
in the first Christian century. Josephus (Antiq. xv. viii. 7) distinctly
asserts : " With us it is lawful for the husband to do so (i.e. dissolve a
marriage), but a wife, if she departs from her husband, cannot marry
another, unless her former husband put her away." In two cases the
husband's right of divorce was abrogated by the Pentateuch (Deut.
xxii.), if he ravished a virgin or if he falsely accused his wife of ante-
nuptial incontinence. In the first case the man was compelled to wed
the woman in an indissoluble union, in the second case he could not
divorce his wife. In later Rabbinic law a divorce if pronounced was
technically valid ; the Biblical law, however, does not deal with such
a case, and the wife was immune from divorce. t But what was her
position 1 The option rested with her. She could compel her husband
to retain her, or she could accept a divorce. Philo declares (ii. 313 KOL
/AtWv re dTraAAaTTeo-flcu, this last word being Philonean for divorce)
that she could divorce him, but it is not probable that the law ever
agreed with Philo's view. At most the injured wife may have been
entitled to move the court to compel her husband to write her a Letter
of Divorce. The situation reminds one of Meredith's Diana of the
Crossways.
We are in possession of a clear piece of evidence as to the Jewish
progress in divorce law in the period preceding the Christian era. In
Matthew xix. 10 the disciples after hearing Jesus' declaration on the
indissolubility of marriage, object : "If the case of the man is so with
his wife, it is not expedient to marry." Here, the difficulty of divorce
is treated as a bar to wedlock. This is the man's point of view. What
of the woman's? Now in the first century B.C. it would seem that,
52
68 IX. FIRST CENTURY DIVORCE
from the woman's side, the facility of divorce was a bar. In face of
the ease with which a husband could whistle off his wife, women
refused to contract marriages, and men grew grey and celibate
(T. J. Kethuboth, end of ch. viii. ; T. B. Kethuboth 82b, Tosefta xii.)7
Thereupon the Pharisaic leader, Simon b. Shetah, the reputed brother of
Queen Alexandra, enacted that the wife's Kethubah or marriage settle-
ment was to be merged in the husband's estate, that he might use it as
capital, but that his entire fortune, even such property of his as had
passed into other hands, should be held liable for it. This effectively
checked hasty divorce (cf. 'Erubin 41 b), and indeed the rights of wives
under the Kethubah were throughout the ages a genuine safeguard to
their marital security. In respect to holding property and possessing
independent estate the Jewish wife was in a position far superior to
that of English wives before the enactment of recent legislation.
Another point of great importance was this. Jewish sentiment was
strongly opposed to the divorce of the wife of a man's youth, and men
almost invariably married young. The facilities for divorce seem mostly
to have been applied or taken advantage of in the case of a widower's
second marriage (a widower was expected to remarry). "What the
Lord hath joined, let no man put asunder" represented the spirit of the
Pharisaic practice in the age of Jesus, at all events with regard to a
man's first marriage. It is rather curious that while in the Gospel so
much use is made of the phrase of Genesis "one flesh" to prove marriage
indissoluble, no reference is made to another verse in the same context
"It is not good thatthe man should be alone" which obviously requires
marriage and not celibacy. It may be that Jesus, anticipating the near
approach of the Kingdom, was teaching an " interim " ethic, which
would have no relation to ordinary conditions of life (cf. the view that
Angels do not marry Enoch xv. 3 7, Mark xii. 25 and the later
Rabbinic maxim that in the world to come there is no procreation
(Berachoth lya)). But it is more likely that he was laying down a rule
of conduct only for his own immediate disciples, declaring that " all
men cannot receive this saying." That, however, a belief in the divinity
of the marriage tie was compatible with a belief that the tie could be
loosened, is shown by the course of Jewish opinion. The Rabbis held
with Jesus that marriages are made in heaven (see Jewish Quarterly
Review, n. 172), and several Old Testament phrases point to the same
roseate view. Of the marriage of Isaac and Rebecca it is written "the
thing proceedeth from the Lord" (Gen. xxiv. 50). "Houses and
Riches are the inheritance of fathers," says the author of Proverbs
IX. FIRST CENTURY DIVORCE 69
(xix. 14), "but a prudent wife is from the Lord." Again, " Fear not,"
said the Angel to Tobias (Tobit vi. 17), "for she was prepared for thee
from the beginning." The Pharisees fully accepted this amiable theory
of divine fatalism. "God," said the Rabbi, "sitteth in heaven
arranging marriages." Or it was more crudely put thus : "Forty days
before the child is formed a heavenly voice proclaims its mate"
(T. B. Moed Qaton i8b; Sota 2 a). In the Middle Ages, belief in the
divine arrangement of marriage affected the liturgy, and on the sabbath
following a wedding, the bridegroom proceeded to the synagogue with a
joyous retinue, and the congregation chanted the chapter of Genesis
(xxiv.) in which, as shown above, the patriarch's marriage was declared
as ordained by God. Naturally this belief in the pre-ordainment of
marriage must have strengthened the Jewish objection to divorce.
"For I hate divorce, saith the Lord" (Malachi ii. 16) was a verse much
honoured in Pharisaic thought, and Malachi's protest gave rise to the
pathetic saying : "The very altar sheds tears when a man divorces the
wife of his youth," and to the sterner paraphrase "He that putteth her
away is hated of the Lord" (T. B. Gittin 90. Of. Prov. v. 19 ; Eccles.
ix. 9 ; Ecclus. vii. 26, yet see also xxv. 26).
But though divorce is hateful, continuance of the marriage bond
may be more hateful still. Perfect human nature could do without
divorce, but it could also do without marriage. Adam and Eve, it has
been well said, went through no marriage ceremony. The formalities
of marriage are not less the result of human imperfection than is the
need of divorce. Were it not for the evil in human nature, said
the Rabbis (Gen. Rab. ix. ; Eccles. R. iii. u), a man would not marry
a wife not that the married state was evil, on the contrary, it was
held to be the highest moral condition but the passions which are
expressed in the marital relationship are also expressed in the lower
lusts. We may also perhaps read another idea into this Rabbinic
conception. X needed the marriage bond to limit his own lusts and
also to ward off Y. And just as, in this sense, man's evil side requires
a marriage contract, so in another sense his good side demands the
cancellation of the contract, if its continuance be degrading or in-
harmonious.
Hence, though the strongest moral objection was felt against
divorce, and though the vast majority of Jewish marriages were
terminated only by death, the Pharisaic law raised no "bar to divorce
by mutual consent of the parties, just as marriage, despite its sacred
associations, was itself a matter of mutual consent. It should be
70 IX. FIRST CENTURY DIVORCE
remembered that in the Jewish document of divorce no ground for
the act is defined, the husband simply declares his wife thenceforth
sni juris and free to re-marry. She could, and often did, re-marry
her husband, unless he had divorced her for unchastity, or unless she
had in the meantime contracted another marriage. In the time of
Jesus it was not necessary for a divorce by mutual consent to come
before a regular court or Beth Din of three Rabbis, as later became
the practice. The whole ceremony could, at the earlier period, be
gone through privately, in the presence of two witnesses. An expert
Rabbi was, however, probably required to ensure the proper drawing
up of the document, and the due fulfilment of the legal delivery to,
and acceptance by, the wife. Thus if Joseph of Nazareth and his
betrothed bride had mutually consented to a divorce, there is no
reason in Jewish law why he should not have " put her away privily "
(Matthew i. 1 9). There is little ground for thinking that such divorces
by mutual consent were either frequent or productive of social evils,
though it may be that the woman's assent was occasionally extorted
by harsh measures. But though the Rabbis could oppose no legal
bar to divorce by mutual consent, it was their duty to exhaust every
possible expedient of moral dissuasion. Aaron, in Hillel's phrase
(Aboth i. 12), was the type of the peace-maker, and this was tradi-
tionally explained (Aboth de R. Nathan, ch. xii.) to mean that his
life-work was, in part, the reconciliation of estranged husbands and
wives (see above, Note VII).
But the case was different when one of the parties to the divorce
was unwilling to assent, or when one party had something to gain by
treating the other party as unwilling. From the eleventh century it
has been customary in Jewish law to require that in all cases the
wife shall assent to the divorce, except where her misconduct or failure
could be shown to be sufficient cause why the marriage might be
forcibly dissolved by the husband. But this condition of the woman's
assent was not necessary at the beginning of the Christian era, when
neither Rabbinic sanction nor the wife's consent was obligatory. The
rule in the first century was (Yebamoth xiv. i) : "A woman may be
divorced with or without her will, but a man only with his will."
If, however, the wife contested the divorce, it is highly probable that
the husband had to specify his reasons and bring the matter before
a regularly constituted Beth Din. This was certainly the case if he
suspected her of adultery (Sota i. 3 4). The accusing husband took
his wife before the local Beth Din or court of three, and after a first
IX. FIRST CENTURY DIVORCE 71
hearing two Rabbis would conduct the accused to the Supreme Court
in Jerusalem, which alone could deal finally with such charges. If she
confessed, she forfeited her marriage settlement and was divorced;
otherwise the ordeal of the waters (Numbers v.) was applied. We
may well suppose that in other cases, especially such as involved a
stigma on the wife, the matter would be made a matter of public
inquiry if she so claimed. It is only thus that we can fully explain
the different views taken at the early period as to lawful grounds
of divorce. The schools of Hillel and Shammai differed materially
(Gittin, end) : the former gave the husband the legal right to divorce
his wife for any cause. Of. Matthew xix. 3, Josephus Antiq. iv. viii.
23 (" for any cause whatsoever "). Philo uses similar language (Spec.
Laws, Adultery, ch. v.). The school of Shammai limited the right to
the case in which the wife was unchaste. The " schools" or "houses"
of Hillel and Shammai belong to the first century. It is uncertain
whether this particular difference of opinion on divorce goes back to
Hillel and Shammai themselves, and thus to the very beginning of
the Christian era. It is barely possible that the teaching of Jesus
on the subject led to further discussion in the Pharisaic schools, and
that the rigid attitude of Jesus influenced the school of Shammai.
This, however, is altogether improbable, for the view of the latter
school is derived from Deuteronomy (xxiv. i) by a process which
closely accords with the usual exegetical methods of the Shammaites.
Matthew v. 32 (as the text now stands) with its \6yov Tropvua^ is
certainly derived from the school of Shammai, for the text of Deut.
xxiv. i reads -QI nny, and it was the school of Shammai who turned
the words round into niiy "Q1 (Gittin ix. 10), which corresponds
in order with the text of Matthew. Hillel's language : " even if she
spoiled his food," is of course figurative, and may point to indecent
conduct, a sense which similar metaphors sometimes bear. Hillel was
a teacher noted for his tender humaneness ; it was he who popularised
in Pharisaic circles the negative form of the Golden Rule before Jesus
stated it positively. Hence, it is not just to speak of his view on
divorce as "lax" or "low," even if (as no doubt later Rabbinic
authority assumed) Hillel used this forcible language to preserve as
inalienable the ancient norm that a husband possessed complete right
to divorce his wife for any cause. For it must be observed that his
" lax " and " low " view of divorce was also a more rigid and elevated
view as to the necessity of absolute harmony in the marriage state.
Still, his view (or its interpretation) did produce a condition of sub-
72 IX. FIRST CENTURY DIVORCE
jection in the woman's status, and left room for much arbitrariness
on the part of the husband. Yet 'Aqiba who went beyond Hillel in
maintaining the husband's arbitrary powers (" even if he find another
woman more beautiful"), was in fact no friend of divorce, for he
applied the severest rules in estimating the pecuniary rights of the
wife under the marriage settlement. "Sell the hair of your head to
pay it," said 'Aqiba (Nedarim ix. 5) to a would-be divorcer who com-
plained that the payment of the heavy demands of the settlement
would impoverish him. As D. Amram in his excellent book on the
subject of The Jewish Law of Divorce (Philadelphia, 1896) puts it,
neither Hillel nor 'Aqiba was making law, they were stating it,
"regardless of their personal views or opinions" (p. 37). It is
true, however, that their statement of the law helped to make and
perpetuate it for future times. The injurious effect was much miti-
gated, though never theoretically removed, by subsequent modifications.
We can trace the gradual incidence of restraining enactments and
customs. Already in the year 40 A.D. we find various reforms intro-
duced by Gamaliel, who ordained e.g. that the Get or divorce letter
must be subscribed by the witnesses, and withdrew from the husband
the right to cancel the Get unless the wife or her attorney were present
(Gittin iv. 2). Such cancellation was made before Gamaliel's reform ;
the husband would locally constitute a Beth-din of three Rabbis ad
hoc. Though, as stated above, the divorce itself needed no Court,
many questions (as to settlements etc.) arising out of the divorce
would have to be brought before the Beth-din.
There were, indeed, certain grounds on which husband or wife could
claim the help of the Court in effecting a divorce against the other's
will. In all such cases, where the wife was concerned as the moving
party, she could only demand that her husband should divorce her; the
divorce was always, from first to last, in Jewish law the husband's act.
The matter was not, however, always left to the parties themselves.
"Joseph being a righteous man, and not willing to expose her to
shame, determined to divorce her secretly." This implies that Joseph
had no option as to discarding his wife. Cf. Montefiore, Synoptic
Gospels, p. 454. This work contains an excellent analysis of the various
Gospel passages on divorce, see pages 235 242, 454, 508 510,
688692, 1000 i. To return, if the husband suspected his wife of
unchastity while betrothed to him, he was compelled, as a " righteous
man," to divorce her (betrothal was so binding that divorce was
necessary to free a betrothed couple). His only option was between
IX. FIRST CENTURY DIVORCE 73
divorcing his bride privately with her consent, or formulating a charge
of infidelity against her, thus subjecting her to public disgrace as well
as divorce. Divorce was not in itself a disgrace, seeing that it might
occur on grounds involving no moral stigma. The case was aggravated
by the circumstance that Mary was with child, until Joseph, in
Matthew's account, received the assurance that his whole suspicion was
erroneous. The wisdom books and the Rabbinic doctors agreed in
regarding adultery as peculiarly heinous when it resulted in the birth
of a child (Ecclus. xxiii. 23, Hagiga i. 7). The offence was a three-fold
sin : against God, against the husband, against the family (Hamburger,
Real-Encyclopadie des Judenthums I. 258). In Jewish law adultery
was the intercourse of a married woman with any man other than her
husband. Though his conduct was severely reprobated, and at all
events in later centuries gave his wife a right to claim a divorce, a man
was not regarded as guilty of adultery unless he had intercourse with a
married woman other than his wife. For though monogamy had
become the prevalent custom in Jewish life long before the Christian
era (cf. Jewish Encyclopedia, vin. p. 657), the man could legally marry
several wives, and sometimes did so. Thus an unmarried and un-
betrothed woman with whom a married man had intercourse might
become his wife; indeed such intercourse could be legally construed
into a marriage. By the Pentateuchal law the penalty for adultery
was death.
But this law can never have been frequently enforced. It needed
eye-witnesses (hence the "taken in the very act" of John viii. 4). More-
over, as Dr Biichler has pointed out, the husband would hesitate to charge
his wife, and the detected adulterer would offer heavy compensation to
save his own life which was forfeit. The husband could privately
divorce his wife, she naturally losing all her rights under the marriage
settlement. A charge of adultery would have to be public, and tried
before the central court. It is not probable that the death penalty for
adultery was inflicted at all in the age of Jesus. The Jewish courts had
lost the general power of capital punishment in the year 30 A.D. (T. J.
Sanh. 1 8 a, T.B. 41 a). The Mishnah cites a single case which would
fall within the age of Jesus, but it does so doubtfully (Sanh. vii. 2), and
Josephus' casual assertion that the penalty for adultery was death is
rather an antiquarian note than a record of experience (Apion ii. 25).
On the other hand it would seem that the ordeal of the bitter waters, as
applied in case of suspected adultery of the wife, was still prevalent,
for the Mishnah records (Sota ix. 9) that the ceremony was only
74 IX. FIRST CENTURY DIVORCE
abolished during the Roman invasion (circa 70 A.D.), though Queen
Helena of Adiabene a proselyte to Judaism in the first century A.D.
sought to restore the practice (Yoma iii. 10, Tosefta Yoma ii. 3). It is
interesting to note that 'Aqiba whose view on divorce was so "lax"
nobly said of the ordeal : " Only when the (accusing) husband is
himself free from guilt will the waters be an effective test of his wife's
guilt or innocence" (Sifre, Naso 21; Sota 47 b). With this may be
usefully compared the fine utterance (John viii. 7) : " He that is
without sin among you, let him first cast a stone at her" (Jewish
Encyclopedia, i. 217 ; Hastings, Encyclopaedia of Religion, I. 130, from
my article there I have taken some passages). The abolition of the
ordeal is attributed by the Mishnah to the great prevalence of adultery,
and it may be that, just as on the inroad of Hellenism some unsettlement
of native morals occurred in the towns and among the wealthy (this
being all that the attacks on harlotry and unchastity in the Wisdom
Literature implies), so in the disturbed conditions due to the Roman
regime a temporary laxity of morals intruded itself. The Rabbis held
adultery in the utmost detestation. Not all a man's other virtues could
save the adulterer from Gehenna (T. B. Sota 4 b). Unchastity drives
away from man the Divine Presence which dwells only in the chaste
soul. It is impossible, however, to attempt to collect here the mass of
Pharisaic maxims against such offences. In the year 135 A.D., at the
crisis of the disastrous revolt against Hadrian, a meeting was held at
Lydda. The assembly was attended by several famous Rabbis (includ-
ing 'Aqiba), and the question was discussed as to the extent of
conformity with Roman demands which might justifiably be made
rather than face the alternative of death. The result is a remarkable
testimony to Jewish abhorrence of unchastity. It was decided
(Sanh. 74 a) that every Jew must surrender his life rather than commit
any of the three offences : idolatry, murder, or gittui 'arayoth, a phrase
which includes both adultery and incest.
The penalty for proven adultery, when the capital punishment was
abolished, was mitigated into the divorce of the woman (the husband
having no option) ; the wife also lost all her rights under the marriage
contract, and was not permitted to marry her paramour (Sota v. i).
The husband could, nay must, divorce her on suspicion, but her settle-
ments would be intact. It would therefore be to his advantage
sometimes to prefer a public charge against her. The male adulterer
was scourged, but was not compelled to divorce his own wife unless
she insisted. In general, when the Mishnah speaks of "compelling" the
IX. FIRST CENTURY DIVORCE 75
husband to execute a Bill of Divorce, the Court could scourge, fine,
imprison, and excommunicate him, and had practically unlimited power
to force him to deliver the necessary document freeing his wife. By a
legal fiction which undeniably had moral justification, the act would
still be described as voluntary on the husband's part. But in case of
his determined contumely, there would be no redress, as the Court
could not of its own motion dissolve a marriage, though it could
pronounce a marriage ab initio void. The secular courts might be used
to enforce the desire of the Beth-din (Gittin ix. 8). But the Beth-din
could not be induced to return the compliment, and validate a divorce
pronounced in a Roman Court (Gittin i. 5). For the whole tenour of
Jewish divorce depended on the theory that divorce was the act only and
solely of the husband, and no Beth-din could validate a divorce which
was the act of any court, and not of the husband, in the prescribed
forms. Moreover, on matters affecting marriage and divorce the Jewish
courts would be most jealous of external interference. In modern
times, however, the London Beth-din would refuse to sanction or
validate a divorce which had not been previously effected in the civil
courts of the country.
Other consequences followed from the theory that divorce was the
willing act of the husband. The divorce of the insane husband of a
sane wife would be impossible (Yebamoth xiv. i), as he could not
execute the deed of divorce. Nor could the insane wife of a sane
husband be divorced by him, because she stood in all the greater need
of his protection. (If the insanity were proved to have existed before
marriage, the marriage could be pronounced initially void, for the
marriage of the insane was illegal.) It should here be pointed out
that though the sane husband could divorce his sane wife on a variety
of grounds, and in the first century could do so without the intermedia-
tion of a Court, he could not secure himself against the divorced wife's
claims for maintenance unless he satisfied the Court that the divorce
had been properly executed, and that the wife's just rights had been
satisfied. In that sense, the Courts would have a power to revise his
personal acts, even in the early period under review. Apart also from
legal duties, the husband was expected to show every possible con-
siderateness to his divorced wife. She was, of course, no longer under
his jurisdiction, she was sui juris, and her husband lost the usufruct
of her estate. This last fact was a constant preventive of arbitrary
divorce (T. B. Pesahim ii3b). But the husband was expected, as a
humane son of Israel, to save his divorced wife from penury. " It is
76 IX. FIRST CENTURY DIVORCE
related of Rabbi Jose the Galilean (about 100 A.D.), that after his
divorced wife had remarried and was reduced to poverty, he invited her
and her husband into his house and supported them, although when
she was his wife she had made his life miserable, and his conduct is the
subject of Rabbinical laudation. 'Do not withdraw from thy flesh,'
said Isaiah (Iviii. 7); this, Rabbi Jacob bar Aha interpreted to mean,
' Do not withdraw help from thy divorced wife'" (Amram, op. cit. p. 1 10).
If the divorced woman retained charge of infant children, the former
husband not only had to maintain her, but he was also required to pay
her for her services. But, in general, as to the custody of the children,
the regulations were extremely favourable to the wife, who was treated
with every conceivable generosity. These regulations, however, except
as concerned the infant up to the time of weaning, were not formulated
so early as the first century. It is clear that a husband was very
reluctant to divorce his wife if she were also the mother of his children.
Though it was held a duty to divorce an "evil woman" an incurable
scold and disturber of the domestic peace nevertheless if she were a
mother, the husband would waive his right and endure his fate as best
he might ('Erubin 41 b).
We have already seen that the insane husband was incompetent to
deliver a Bill of Divorce. In certain other cases of disease though
not of mere infirmity the wife could claim a divorce. If she became
deaf-mute after the marriage, he could divorce her; if he contracted the
same defects he could not divorce her (Yebamoth xiv.). If the husband
fell a victim to leprosy the wife could claim a divorce, and in the second
century the Courts could enforce a separation in such cases against the
will of the parties, unless the latter satisfied the authorities that there
would be no continuance of sexual intercourse. The wife could claim a
divorce in other cases of loathsome disease, as well as when the
husband engaged in unsavoury occupations which rendered cohabitation
unreasonably irksome (Kethuboth vii. 9). In those cases the wife
retained her settlements. The husband could divorce the wife with loss
of her settlements if she transgressed against the moral and ritual laws
of Judaism, and some Rabbis of the first century held that the same rule
applied if the wife made herself notorious by her indelicate conduct
in public. If he became impoverished and unable, or if he were un-
willing, to support her adequately^ if he denied to her conjugal rights,
she could by rules adopted at various times claim the right to her
freedom (Kethuboth v. 8 9), indeed such treatment on his part was
a breach of the contract made in the marriage deed. Similar rights
IX. FIRST CENTURY DIVORCE 77
accrued to the wife some of these concessions belong to a considerably
later period if he restricted her liberty, if he became an apostate, if
he committed a crime which compelled him to fly the country, if he
violently and persistently ill-treated her, if he refused marital rights,
and if he were openly licentious in his life. In case of desertion, the
wife could not obtain a divorce ; though, in order to presume his death,
the Court would waive some of its usual strictness as to the reception
of evidence. If the whereabouts of the husband were known, the
local Court would use every effort to compel him to return or grant a
divorce. The excellence of intercommunication between Jewish settle-
ments would enable the Court to trace him. But the Court could not
grant a divorce to the wife if the husband had merely vanished and
left no trace, unless they saw valid ground for presuming death. The
persecutions, to which the Jews were subjected, compelled many
men to leave home in search of a livelihood, and in the Middle Ages,
out of love and consideration for his wife, the husband would some-
times give her a conditional divorce which would become effective if
he failed to appear within a stated term. It is said that in ancient
times a Jewish soldier, on going to active service, delivered such a
divorce which would be valid if he died on the field. The effect would
be to save his widow from the levirate marriage, from which as a
divorcee she was free. In course of time the position of the woman
was continuously improved, generation after generation of Rabbinical
jurists endeavouring to secure to her an ever greater measure of justice
and generosity.
The wife's barrenness, after ten years' married life, was a ground
for divorce (Yebamoth 64 a); later on it was disputed whether , the
Court should leave the man to follow his own feeling in the matter, or
whether it should compel him to divorce his wife, or alternatively (in
countries where monogamy was not demanded by law) marry an addi-
tional wife. Philo gives us reason to think that at the earlier period
husbands were reluctant to make use of their power to divorce a
barren wife. But childless marriages were regarded as a failure, and
the point gave much trouble at various epochs. It was a religious duty
to beget offspring, this was the fundamental purpose of marriage. We
very rarely come across a celibate among the well-known Pharisees.
Ben-'Azzai (Tos. Yebamoth viii. 4, Sota 4 b etc., cf. J. E. n. 672) was a
rare exception. He belongs to the beginning of the second century,
and he remained unmarried though he denounced celibacy. When a
colleague remonstrated with him, pointing out the inconsistency between
78 IX. FIRST CENTURY DIVORCE
his conduct and his doctrine, Ben-'Azzai replied : " What shall I do ?
My soul clings in love to the Torah (Law); let others contribute to the
preservation of the race." But it was not believed that this prime
duty to society could be vicariously performed, and e^ery Jew was
expected to be a father. The act of sexual intercourse was consciously
elevated by this view from an animal function to a fulfilment of the
divine plan announced at the Creation.
From this brief summary it will be seen that the Jewish law of
divorce must be judged in relation to the general principles of social
and domestic ethics. Rules for marriage and divorce cannot be appre-
ciated apart from many other factors. Jewish teaching and training
were directed towards producing moral sobriety, continence, purity.
It did this by word and deed, by formulating moral maxims and
fostering moral habits. Society usually attacks the problem at the
wrong end; it penalises marital offences instead of making those
offences rare. The ancient Synagogue dealt with the youth and maid
in the formative period of their lives. The Jewish law of divorce
applied to a society of firm domestic solidarity, it was the law of a
society in which young marriages predominated, and the contracting
parties entered into a life-long wedlock straight from a pious and
virtuous home, a home in which harmony and happiness were the
rule, and the relations between husband, wife and children were
distinguished by a rarely equalled and never surpassed serenity and
reverence. As a saying (certainly not later than the first century)
runs (Yebamoth 62 b): "Our masters have taught, He who loves his
wife as himself, and honours her more than himself; who leads his
sons and daughters in the straight path, and marries them near their
time of maturity; to his house the words of Job apply (v. 24): Thou
shalt know that thy tent is in peace." With much of this ideal the
modern world has lost sympathy, but the Judaism of the first century
maintained it, and built on it a moral structure which stands high
among the manifold attempts to erect an effective discipline of life.
X. WIDOWS' HOUSES.
t
That in all ages, and not inconspicuously in our own, men are
tempted to make undue use of their influence over wealthy women in
the cause of religious institutions is a familiar fact. In the second
century, in Sepphoris, the women resented the duty of supporting
scholars (Baraitha in Pesahim 49 b). But, on the other hand, we have
the testimony of Jerome that Jewish women were not only among the
regular performers of this obligation, but were eulogised by him on
this very ground, "Ex quo apparet eum de aliis sanctis dixisse
mulieribus, quae juxta morem Judaicum magistris de sua substantia
ministrabant, sicut legimus ipsi quoque Domino factitatum " (Adversus
Jovinianum i. 277 ; cf. A. Biichler, Sepphoris, p. 75).
These last words of Jerome are a striking reminder of the unequal
measure with which the Pharisees and their opponents are judged, not
by Jerome but by more recent writers. The influence exercised by
the early preachers of the Gospel over women is well attested, and
held the reverse of blameworthy. When, then, Josephus complains
of the " great influence over women " which a certain Pharisaic faction
possessed (Antiq. xvn. ii. 4), it is scarcely just to endorse his con-
demnation, or to forget two points : (a) he distinctly speaks of a
faction only (/xopiov), carefully avoiding the word by which he usually
designates the main body of the Pharisees (atpeVeis) ; (6) his animosity
is directed against the political activity of this faction, who committed
what to Josephus was the height of iniquity, in that "when all the
rest of the people gave assurance by oath of their good-will to the
Emperor and to the King's government, these very men would not
swear, who were more than 6000 ; and when the King imposed a fine
upon them, Pheroras' wife paid the fine for them."
Moreover, it must be remembered that such charges were part of
the ordinary invective of controversy. In the Psalms of Solomon
(see particularly Ps. iv.) the Pharisees themselves make a very similar
80 x. WIDOWS' HOUSES
attack on the Sadducees. In the Assumption of Moses, again, the
Pharisaic author (vii. 6) assails either the zealots of his own order or
the priestly caste in the words that they are " devourers of the goods
of the poor," saying they do so out of mercy (misericordiam, according
to Charles the word means justice). Colani's contention that this last
phrase is to be explained by the decree of the Sanhedrin (Kethuboth
50 a) in the second century forbidding a man to give more than one-fifth
of his fortune and income to the poor is monstrous. The decree of
the Sanhedrin was due to the excessive generosity which led men
to impoverish themselves in the cause of charity, with perhaps
(as Dr Kohler ingeniously suggests) some intentional opposition to the
Essenic communism and to such ideas as Matthew xix. 21 (J. E. in.
p. 668). The Talmud gives the former reason, and in any event the
expression " devourers of the goods of the poor " cannot be explained
by any such incident. Dr Charles thinks the Sadducees are attacked ;
if so, one must not assume that the attack of their critics was just. The
poor no doubt often felt the pressure of the taxes imposed on them, and
there is a late Midrash (Shohar Tob on Ps. i., cf. Yalqut) in which a
biting satire is put into the mouth of Korah. He adduces the case of
a widow who is deprived of her crops and sheep by the many demands
made on her slender resources by the priests. Certainly the Pharisees
were themselves the most severe critics of the possible abuses of their
own system. When, however, M. Friedlander remarks (Die religiosen
Bewegungen inner halb des Judentums im Zeitalter Jesu, p. 112) that
the Pharisees themselves said quite as severe things as did Jesus about
certain abuses (" schlimmeres wahrlich hat auch Jesus nicht von diesen
Weltverderbern ausgesagt"), he misses the significance of this fact.
If the Pharisees were thus critical, then it is manifestly unjust to treat
the criticism as though it could apply against Pharisaism as a whole.
To justify the words " which devour widows' houses " as a descrip-
tion of average scribes, would require much more evidence than has
ever been adduced. "Widows were known there (in Jerusalem), it
appears, who had been reduced from comfort to beggary by giving up
their means to religious uses at the suggestion of scribes" (Menzies
on Mk xii. 38, p. 229). The text hardly requires us to make this
assumption. But then there comes the incident of the Widow's mites.
" She of her want did cast in all that she had, even all her living "
(Mk xii. 43). This sacrifice is eulogised, and justly. Yet the acceptance
of such a gift might be denounced by a hostile critic as a " devouring "
of the widow's substance. Jesus, however, praises it, just as the
x. WIDOWS' HOUSES 81
Pharisaic Scribe does in the story (cited by Schottgen). A priest who
had scorned a certain woman's handful of flour was rebuked in a vision
overnight: "Despise her not; it is as though she offered her life"
(Leviticus Rabba iii. 5). It need hardly be added that the Pharisees
attached much importance to the exiguous gifts of the poor (cf. the
passages adduced by Schottgen on Mark, p. 251 ; Baba Bathra loa;
Leviticus Rabba iii., where the poor's offering of two doves is preferred
to King Agrippa's thousand sacrifices "JDlp iy$ *?W P"ip; see also
Wiinsche, p. 402, he quotes : Numbers Rabba xiv., Mishnah, Menahoth
xiii. i; and add Pesahim u8a). On the other hand, Gould (Mark
xii. 40) suggests that "the devouring of widows' houses would be
under the forms of civil law, but in contravention of the Divine law of
love."
But the forms of civil law were by no means harsh on widows.
The prevalent custom in Jerusalem and Galilee was to allow a widow
to remain in her husband's house, and be maintained from his estate
during the days of her widowhood (Mishnah, Kethuboth iv. 12). In
Judaea (apart from Jerusalem) the widow might be compelled to
receive her settlement, and then leave the house. Such a rule might
have pressed hard in certain cases. Strong language is used in a
late passage in the Palestinian Talmud against those who help the
" orphans " to take this harsh course against " widows " (T. J. Sota
on iii. 4). But on the whole the widow was well protected by the
Jewish civil law (see L. N. Dembitz in the Jewish Encyclopedia,
xii. p. 514). The example of the widow of Zarephath was held up
for imitation (Cant. R. ii. 5, 3) and Jerome's praise would well apply
to such a case. But to "devour widows' houses" was no common
failing of those who based their lives as the Pharisees did on the
Scriptures which so often and so pathetically plead the widow's cause.
Moralists in all ages have had to repeat this urgent appeal, and there
was no doubt adequate ground for such a homily in the age of Jesus.
But the Pharisaic teachers were keenly alive to their duty in all
periods to take up the cause of the widow. And they expressed
themselves emphatically on the subject again and again; nowhere,
perhaps more forcibly than in their saying Exodus R. ch. xxx.
(y\>rh ^nj 1^3 jbnJn ^D), "He who robs the widow and orphans is
as though he robbed God himself."
XL THE CLEANSING OF THE TEMPLE.
From a not unreasonable point of view the dignity and worth of
the Temple in relation to national life must be considered as enhanced
and not diminished by the association of that life with the Temple
environment. The sacro-sanctity of the inner courts would be, as it
were, humanised by the secularisation of the more remote precincts.
To many a modern mind it is attractive rather than repellent to read
of the popular uses to which the Temple was sometimes devoted. The
famous celebration of the semi- religious function of the Water-Drawing,
during the Feast of Tabernacles, with its deep spiritual significance
allied to merry, carnival-like rites, is a case in point. Modern writers
are too apt to confuse Pharisaism with Puritanism ; more than half of
the contrasts imagined between Hellenism and Hebraism arise from
this same confusion. Josephus, moreover, records the holding of even
more pronouncedly secular assemblages within or close to the Temple
precincts (War i. xx ; n. i, xvi ; v. v). The tendency to treat the
modern Synagogue as a place formally restricted to purposes of worship
was a reaction which is happily breaking down, especially in America,
where so many of the so-called Jewish reforms are reversions to ancient
traditions.
" But indeed in those days nearly every priest must have been a
trader." With these words Dr G. A. Smith concludes his brilliant
account of the Temple Revenues, Properties and Finance in the first
century of the Christian era (Jerusalem, Vol. i. p. 366). But surely
the same might be said with equal validity of the governing bodies of
many a Church and University in our own times, without implying
that the financial side of these institutions was unduly prominent.
The question always is : what is the implication 1 There is little
ground for the supposition that the people were, in general, oppressed
by the Temple financial arrangements. The Temple, again, was made
a place of safe deposit for private money, but no trading was involved,
XI. THE CLEANSING OF THE TEMPLE
83
and the authorities who speak of these deposits in the Temple almost
explicitly state this. Thus the stores alluded to in II. Maces, iii. were,
as Dr Smith points out, " laid up for the relief of widows and fatherless
children," and in part belonged to Hyrcanus son of Tobias. "It was,"
writes the same authority (II. Maces, iii. 12), "altogether impossible
that [by confiscating this money] such wrongs should be done unto
them that had committed it to the holiness of the place, and to the
majesty and inviolable sanctity of the Temple, honoured over all the
world." The priests would clearly have no financial operations at all
in relation to such funds, while Josephus (War vi. v. 2) when he says
that in the Temple treasuries "the rich had built themselves store-
chambers there " refers to a time of stress, when the Temple would, as
a fortified place, be an obvious asylum. Again, here, however, the
language of Josephus does not suggest that the priests in any way
traded with the money. From the same historian's earlier account of
the Parthian raid on Jerusalem (War I. xiii. 9) it may be gathered
that private persons were not in normal times in the habit of using
the Temple treasury as the store-house of their property. It is scarcely
worth while citing the mass of facts available to show that sacred
edifices have in many ages been used as safe-deposits, without neces-
sarily incurring any suspicion of the taint of commercialism.
The presence in the Temple precincts of money-changers for a
full account of whose operations see S. Krauss, Talmudische Archa'ologie,
i piii II. 411 is generally conceded to have been an arrangement
designed for the advantage of the pilgrims. The Temple-tax of half a
shekel had to be paid in definite coinage. It could not be paid in
ingots, but only in stamped coins (T.B. Berachoth 47 b with reference
to Deut. xiv. 25 ; cf. Sifre ad loc.). It must not be paid in inferior
alloy but in high grained silver (T.B. Bechoroth 51 a). Again and again
we are informed that the only coins accepted were Tyrian (Mishnah,
Bechoroth viii. 7 ; Tosefta, Kethuboth xiii. 3, ed. Zuck. p. 275), which
indeed were so emphatically the legal tender in the Temple that they
were termed Jerusalemite as well as Tyrian. But it is not quite clear
which Tyrian coins were meant. T. Reinach points out that among the
conditions imposed on the vanquished Jews by Antiochus Sidetes was
the withdrawal of the right of coining silver, though the striking of
small bronze coins, intended for local circulation, was intermittently con-
tinued. This was in 134 B.C. But "very few years after the surrender
of Jerusalem, in 126 B.C., when the civil war was waging between the
sons of Demetrius II and the usurper Alexander Zebinas, the wealthy
62
84 XL THE CLEANSING OF THE TEMPLE
town of Tyre seems to have snatched from one of the pretenders to the
throne the practical acknowledgment of its independence and the right
to issue a silver coinage of its own. The Tyrian coinage, which lasted
for almost two centuries, consists mostly of shekels (staters), bearing
as types the head of the town God Heracles and the Ptolemaic eagle ;
their legend Tyre the holy and inviolable (Tvpov lepbv KCU aa-v\ov) seems
to be imitated from the Yerushalem Kedoshah of Simon's shekels. The
dates are reckoned from the new era of 126 B.C. These coins, notwith-
standing their heathen types and Greek lettering, were of so exact a
weight and so good an alloy that they enjoyed a large circulation in
Judzea, and were even officially adopted as sacred money, that is to say
the Rabbis decided that the annual head-tax of one [half-]shekel due
from every Israelite to the Temple treasury was to be paid in Tyrian
money." It is strange enough that while the bronze coins circulated
in Judaea should conform scrupulously to the tradition and represent
nothing but inanimate objects, the payment of Temple dues should not
only be accepted but required in coins containing figures on them.
Reinach meets this objection by the suggestion that "once thrown
into the Temple treasury, all gold and silver coins were melted down
and transformed into ingots" (T. Reinach, Jeivish Coins, ed. Hill,
1903, pp. 20 23). At all events, while the coins most current in
Syria were the Roman tetradrachms and denarii (such a silver denarius
is referred to in Matthew xxii. 15), the Temple demanded payment on
the Phoenician standard (cf. Krauss, op. cit., p. 405), and the money-
changer for this (and for other reasons) was therefore an actual
necessity.
In passing it may be remarked that there is no ground for supposing
that the ordinary business of money-changing went on in the Temple.
In the N.T. the word /coAAu/3io-rr/s is always used in describing the
scene of the cleansing of the Temple, and it must be interpreted to
mean the receiver of the qolbon (jIU^lp), or fee for changing other
currencies into Temple currency and exclusively for Temple use. When
Mark (xi. 16) adds the detail that Jesus "would not allow any one
to carry a vessel through the Temple," the meaning no doubt is
that he sided with those who ordained that the Temple must not be
made a public thoroughfare (T.B. Yebamoth 6 b). Others went further,
and forbade frivolous behaviour outside the Temple precincts and in
the neighbourhood of the Eastern Gate (Berachoth 54 a). Similar
rules were applied to the Synagogues (Megillah 27 28), and one may
cite the regulation in Cambridge against carrying trade parcels through
XI. THE CLEANSING OF THE TEMPLE 85
the College precincts. That Jesus is applying an established rule and
not innovating is confirmed by the fact that he cites old prophetic
texts (Isaiah Ivi. 7, Jer. vii. n) in support of his attitude.
Granting, then, that certain commercial operations were necessary
for the maintenance of the Temple or convenient for those who had
occasion to present themselves in its courts, there was nothing in such
circumstances inherently censurable. If there was a sort of market
within the Temple enclosure, it is impossible to assent to Dr Eders-
heim's easy conclusion : " It needs no comment to show how utterly
the Temple would be profaned by such traffic." On the contrary, it
needs much comment to show this. Equally exaggerated is Lightfoot's
characterisation of the money-changer's profit as "unholy gain."
Gould, in his note on Mark xi. 17, clearly sees that such attacks imply
not merely an invective against an illegitimate use of the Temple, but
a thorough-going antipathy to trade as such. Yet if the money-changer
were necessary his profits were not " unholy." The labourer is worthy
of his hire: Thus, there was considerable labour, and that of an
expert kind, involved in the examination of animals to pronounce
them perfect or blemished, and a fee was naturally charged (Mishnan,
Bechoroth iv. 5). These fees as well as the profits of the money-changers
were strictly limited by law and usage. Dr Edersheim seriously over-
estimated the gain. "If we compute the annual Temple-tribute at
about 75,000, the bankers' profits may have amounted to from ,8000
to 9000, an immense sum in the circumstances of the country." We
have, on the other hand, the clear statement that the profit was only
one in twenty-four or one in forty-eight (Tosefta, Sheqalim i. 8, ed.
Zuck. p. 174; Maimonides, Sheqalim iii. 7; Krauss, Talmudische
Archdologie n. 413). Even if we take the higher estimate, that of
Rabbi Meir, Edersheim has overrated the changer's earnings by three
to one.
Nor is it at all certain that this profit found its way regularly into
private pockets. The Babylonian Talmud (Menahoth io8a) has no
suggestion of the secular destination of the changer's gain. Maimonides
(loc. cit.) decides that the profit was used for the Temple purposes.
Here he was following the tradition of Meir. In the Jer. Talmud
there is indeed an opinion expressed that the money-changer himself
took the profit. But this opinion is only one among several, and very
probably refers to the provincial money-changers and not to those in
the Temple. From the fifteenth to the twenty-fifth of Adar the
money-changers set up their "tables" in every country place (Mishnah,
86 XI. THE CLEANSING OF THE TEMPLE
Sheqalim i. 3), and it is probable that the banker received the com-
mission of one in twenty-four for himself. Schwab, in his French
translation of the Palestinian Talmud (Yol. v. p. 268) inserts the words
"en province," which is a manifest impropriety, for though this may be
the sense, the words do not occur in the text, which runs as follows :
"To what use were the qolbons turned? R. Meir says, they were
added to the fund of the sheqalim ; R. Lazar says, they were employed
for free-will offerings nedabah ; R. Simeon of Shizur (Saijur) says,
they provided with them gold-plates and covering for the Holy of
Holies; Ben Azzai says, the bankers took them as their profit; and
some say they used them for the expense of keeping the roads in repair "
(T. J. Sheqalim, chapter i. last lines). The roads were put in order at
the beginning of Adar (Mishnah, Sheqalim i. i). This association of
the repair of the roads with Ben Azzai' s view may justify the conclu-
sion that he was referring to provincial and not Jerusalem transactions
(the scene of the money-changing was transferred to the Temple on
Adar the twenty-fifth ; Mishnah, Sheqalim i. 2). In the parallel passage
in the Tosefta, however, the words about the repair of the roads are
wanting. Nevertheless, the weight of evidence is in favour of the
verdict that the gains of the exchange were devoted to public and not
to private ends. When once the money had been paid over to the
Temple treasury, it was held unlawful to use it to gain profit even for
the Sanctuary (at least this was Aqiba's view, Mishnah, Sheqalim
iv. 3); but as the qolbons were paid before the money was actually
received by the Sanctuary, they would not be profit directly made by
the use of the sacred funds as capital.
We may conclude that besides the ordinary traders in money-
changing, there were also operators of a less commercial type. The
former would not have been permitted to carry on their trade in the
Temple precincts ; the latter were only authorised in the outer Court
of the Temple between the 25th of Adar and the ist of Nisan, an
interval of about one week (Mishnah, Sheqalim i. 3. Cf. D. Oppenheim,
Literaturblatt des Orients, Vol. x. 1849, P- 555)- As, in this case, the
profits were destined for public and sacred uses, and the operator
received no gain from the transactions, it would seem likely that the
money-changing for purposes of the Temple-tax was performed by
officials of the Temple, that is by the priests. This would ensure
that in normal circumstances the people would be fairly treated, and
it was only under the aristocratic regime of the Temple's last decades
that we hear of oppression. This occurred less with regard to the
XI. THE CLEANSING OF THE TEMPLE 87
money-changing than with regard to the prices of pigeons and so forth
for the sacrifices, the actual buying and selling of which moreover do
not seem to have been normally carried on within the Temple precincts
(cf. Oppenheim, op. cit. p. 556). When oppression occurred, the popular
defenders of the people in such cases were the Pharisaic leaders. We
find on record the action of various Rabbis which lowered the prices of
pigeons even to the point of modifying the law on the subject (Mishnah,
Kerithoth i. 7, where by reducing the number of pigeons to be brought
by women the price of the birds was lowered by Simeon ben Gamliel
from a gold denarius to half a silver denarius that is to one-fiftieth of
the original price). An earlier Rabbi (Baba ben Buta, contemporary
with Herod) actually brought in 3000 sheep so that offerers might have
animals for use. But Edersheim adds to the latter story a detail absent
from the source he quotes (T.J. Hagigah ii. 3). Baba ben Buta found
the Temple desolated as he termed it, but not because the grasping
priests had limited the supply to maintain a high price, but because it
was a festival and the ruling priests held that it was not lawful for
private offerings to be brought on a holy day. The question was one
at issue between the Schools of Hillel and Shammai, and Baba ben Buta,
though a disciple of the latter, in this detail followed the decision of
the former. But there is evidence enough that certain rapacious
priestly families were detested by the people (witness the case of the
House of Hanan) and that the Pharisees themselves denounced such
practices (T.B. Pesahim 57 a). While, then, it is impossible to agree
that the whole of " this traffic, money- changing, selling of doves, and
market for sheep and oxen was in itself, and from its attendant
circumstances, a terrible desecration" (Edersheim), there might well
have been occasions on which indignation such as that of Jesus would
be justified. But we must not magnify an exception into the rule.
The danger always lies in this tendency to confuse a system with
its abuses. This, as it seems to me, is an error made by many
commentators on the Gospels, who seek to expand the often-enough just
criticism of Jesus against abuses, into an unjust condemnation of the
whole Pharisaic system. It is fair enough for the anti-Nomists to
criticise and judge Pharisaism as a religion based on Law ; but there
is no justice in refusing to consider the legalistic point of view and its
possible merits. Still less is it fair to confuse legalism with externalism,
or to assume without close examination of each instance that the moral
abuses, which seem superficially inherent in a legalistic system, were
really the logical result of the system, or did actually occur in
88 XI. THE CLEANSING OF THE TEMPLE
Pharisaism as lived by those who believed and rejoiced in it.
(R. . Travers Herford's Pharisaism, its Aim and its Method, London
1912, appeared after the present volume was mostly in type. Other-
wise frequent reference would have been made to this brilliant and
successful attempt to do justice both to Jesus and to the Pharisees.)
The Cleansing of the Temple is a good case in point. And, therefore, I
venture to repeat here what I wrote at an earlier date, when pleading
for a revision of this tendency where the judgment on Pharisaism is
concerned (Jewish Quarterly Review, 1899, P- 641). " Externalism
needs the most careful watching, and ritual is always in need of
freshening under the inspiration of the ideas which lie behind it. But
Pharisaism was not ritualism. I, and many Jews with me, have no
resentment whatever against the general spirit of the criticism to
which the Law was subjected by Jesus, against his healthy onslaught
against externalism. When Jesus overturned the money-changers
and ejected the sellers of doves from the Temple he did a service to
Judaism But were the money-changers and the dove-sellers the
only people who visited the Temple ? And was everyone who bought
or sold a dove a mere formalist? Last Easter I was in Jerusalem,
and along the fagade of the Church of the Holy Sepulchre I saw
the stalls of the vendors of sacred relics, of painted beads and in-
scribed ribbons, of coloured candles, gilded crucifixes, and bottles
of Jordan water. There these Christians babbled and swayed and
bargained, a crowd of buyers and sellers in front of the Church sacred
to the memory of Jesus. Would, I thought, that Jesus were come
again to overthrow these false servants of his, even as he overthrew his
false brothers in Israel long ago. But I will also tell you what I did
not think. I did not think that the buying and selling of sacred relics
was the sole motive which brought thousands of pilgrims to Jerusalem ;
I did not say : Here is the whole of the Gospel, this is its inevitable
end, its sure outcome. I knew that there is more in Christianity than
this, that there are other Christians than these. Nay, as I turned
away, I thought that perhaps if I had the insight to track a dealer in
relics to his inmost soul, I might after all find there a heart warm with
the love of Christ."
It must finally be remembered that the payment of the Temple-tax
was a privilege as well as a burden. It was the typical illustration of
the democratic basis of Jewish life. The daily sacrifices being for all
Israel were paid for by all Israel. " All Israel were partners in this "
(Pesiqta Rabbathi x, ed. Friedmann, p. 33 b). An individual might
XI. THE CLEANSING OF THE TEMPLE 89
not claim the privilege to pay for the whole cost of the continual
offerings (see Friedmann's note ad loc.) : all Israel must share in the
burden and the privilege. In estimating the effect of the Temple dues
on the popular life this element must not be overlooked. It colours
the whole estimate we have to form of the system. There were
amenities as well as sacrifices involved in the sacrificial institution.
It was not founded on exaction nor corrupted by peculation. These
were the occasional abuses of a regime which, on the whole, secured
popular enthusiasm for a beloved tradition.
XII. THE PARABLES.
"The parable became a truth, proved upon the pulses of men."
These words, used by a modern writer in another connection, aptly
characterise the abiding significance of the New Testament Parables.
A vast amount of religious and literary genius has been directed,
throughout the ages, to the worthy object of extracting the fullest
meaning from the Parables attributed to Jesus. But far more effective
has been the process by which these Parables have been " proved upon
the pulses of men."
It is generally felt that Jesus was not the originator of the method
[_of teaching by Parables. Even Jiilicher, who advances so strenuous
a plea for the originality of the contents of the New Testament Parables,
does not claim of course in presence of the Old Testament Parables
cannot claim that the method was a new creation (Die Gleichnisreden
Jesu, i. 164). Bousset roundly asserts that, though as an exponent
of the Parabolic art Jesus "spoke" while the Rabbis "stammered,"
nevertheless "Jesus owed the vehicle on which he mainly relied in
his popular preaching the Parable to the Synagogue and the
Scribes " (Jesus, p. 30). And, again, " There can be no doubt that
he first learned such a manner of teaching in the Synagogue. All
that has come down to us in the way of Parables from Rabbinic
tradition later though they undoubtedly are bears so close a re-
semblance both in form and matter to the Parables of Jesus, that
no idea of accident can be entertained. And since any influence of
Jesus upon the later Jewish Rabbinism is out of the question, we can
only assume that Jesus caught the form of his Parabolic speech from
the Scribes in the Synagogue" (op. cit. p. 43). On both the points
raised in this last sentence Bousset is probably right, but he has
gone beyond the evidence in the vigour of his statement, for we know
very little as to the contemporary style of Synagogue homily. It is,
however, true that just in the case of ideas which affect the folk
XII. THE PARABLES 91
influence is most likely to be exercised without the consciousness of
imitation. Ziegler (Die Konigsgleichnisse des Midrasch, 1903, Intro-
duction) rightly maintains that many Parables must have been part
of the common fund (Gemeingut) of the people, and that Jesus may
have drawn upon and added to this common fund. Jesus had no need
to take his Parables from other Agadists, just as other Agadists had
no need to take their Parables from Jesus. But as Ziegler judiciously
sums up the matter, p. xxii : " It is indeed conceivable that Jesus
employed much that he had heard from his teachers ; it is also possible
that sundry Parables of Jesus became popular, lived on in the mouth
of the folk, and thence were taken over by later Agadists, without the
least inkling on their part as to the identity of their author, just as
to-day Heine is inadvertently quoted by the most pronounced Heine-
phobes yet it is out of the question to assert anything like a systematic
influence of one side on the other." There must have been a large
Jewish stock of fables and parables floating about long before they
were set down in writing (Fiebig, Altjudische Gleichnisse und die
Gleichnisse Jesu, 1904, 25), and it is possible that both the Tannaim
and Evangelists drew from the stock.
Close comparison of the Gospel Parables with the most similar of
the Rabbinic nearly always reveals dissimilarity amid the similarity.
Though in his earlier work just cited, Fiebig falls short of justice to
the Rabbinic Parables as a whole, I fully agree with a conclusion
which he reaches in his later work (Die Gleichnisreden Jesu, Tubingen,
Mohr 1912), which appeared after this Note was in type. Fiebig is
clearly right when he claims that the Gospel Parables are marked by "^
characteristic features which testify to an original and exalted '.,-.
personality in their authorship, or at least in their adaptation. Yet
the hand of the editor has been at work, and it is scarcely possible to
formulate canons of criticism by which the genuine Parables of Jesus
may be distinguished from the rest. It would be delightful could we
accept fully the view of the Rev. J. W. Hunkin (Journal of Theological
Studies, xvi. 381) that "the parables have been transmitted in the ]
Synoptic tradition very nearly in the form in which they were spoken
by Jesus." But without going this length, it is obvious that some of ;
the Synoptic Parables point to a strong personality. And the same
is true of the Rabbinic Parables. Amid the sameness one detects
individualities. Hillel, Aqiba, Meir, Joshua b. Levi, Abbahu, are to
a certain extent as distinct in their Parables and Similes as in their
doctrines, and if they drew on the common stock of their people's lore,
92 XII. THE PARABLES
reinforced as that stock was by accretions from the lores of other folk,
they made their borrowings, as their inventions were, personal by the
genius with which they applied them to living issues.
All authorities are agreed that there can have been no direct,
literary borrowing by the later Rabbis from the books of the New
Testament. Thus Prof. Burkitt suggests (J. T. S. xv. 618) that
Matthew vii. 24 7 is the ultimate source of the Rabbinic contrast
of two forms of building in Aboth de Rabbi Nathan xxiv. The parallel
is not close in detail, and an examination of the variant in the second
recension of the Aboth xxxv. renders it remotely possible that we have
here a confused reminiscence of some Philonean ideas on the Tower of
Babel (Mangey, I. 420). The Rabbis were, moreover, fond of comparing
the various aspects of the study and performance of Law to firm and
infirm structure such as a tree with many and few roots (Mishnah,
Aboth iii. 22). But if there were borrowing in the particular case
before us, Prof. Burkitt is clearly right in holding that "it was
probably second-hand, i.e. from one of the Minim," and that the
Midrash "put it down to Elisha ben Abuya [the heretic] to avoid
offence." Similarly, if it be the case that the Talmud (Me'ilah iyb)
borrowed from a Christian source the story of an- exorcism, the
borrowing must have been unconscious. (But see on this interesting
point the discussion in the Revue des fitudes Juives vii. 200, x. 60, 66,
xxxv. 285.)
Another instance of greater curiosity concerns the Parable of the
\ Prodigal Son. In the literary sense this is original to Luke. But
some of the phraseology seems traceable to Ahiqar, and the root idea
is Philonean (G. Friedlander, The Grace of God, 1910). Now, the
text of the Talmud must at one time have contained a passage
reminiscent of the Parable. For in a Genizah MS. (published by
L. Ginzberg in Gaonica, New York, 1909, ii. 377) Aha, the famous
eighth century Gaon, quotes Sanhedrin 99 a in a version no longer
fully extant in the Talmud texts. To illustrate the Pharisaic principle
that the penitent sinner stands on a higher level than the completely
righteous, Abbahu cites the parable of " a king who had two sons, one
of whom ordered his way well, while the other went out to depraved
living "
run nmrfc KV *in&o niDi "n nnK D^n w [fa vh vnc? *]W>
This looks like a reminiscence of Luke's Parable, and it may have been
removed from the Talmud text by scribes more cognisant than Abbahu
was of the source of the story. Dr Ginzberg, who recognised the
XII. THE PARABLES 93
similarity, takes another view. His words (op. cit. p. 351) are: "The
source for the parable... is not known to me. Obviously R. Aha must
have had it in his text of the Talmud.... In any event, it is the short,
original form of the New Testament parable of the prodigal son."
And here reference may be made to another instance. The Gospel I
Parable of the Sower is introduced by the medieval Jewish adapter of J
the Barlaam and Josaphat romance. Abraham b. Hisdai wrote his
Hebrew version (Ben Jia-melech we-hanazir) under the title "King's
Son and Nazirite," or as moderns prefer to render the Hebrew title
" Prince and Dervish," in the thirteenth century. The tenth chapter
contains the Parable of the Sower at great length. The main idea, ?
comparing the propagation of Wisdom to the Sower, must have occurred
in the original Indian of Barlaam (J. Jacobs, Barlaam and Josaphat,
1896, p. cxi). A well-known Indian parallel, moreover, is found in
the Sutta Nihata (cf. P. Carus, Gospel of Buddha 74); this is clearly
more primitive than the Gospel version. Yet Abraham b. Hisdai gives
us a form, the details of which are for the most part bodily derived
from the New Testament, a fact of which he was assuredly unaware.
The over-working of the Indian original of Barlaam by a Christian
redactor must have already occurred in the recension of the romance
used by the Hebrew translator as his base. (On the problem of the rela-
tion of the Hebrew to other versions of Barlaam see M. Steinschneider,
Die hebraeischen Uebersetzungen des Mittelalters, Berlin, 1893, 532.)
With regard to another suggestion of Rabbinic borrowing, the case is^
different. It has been argued that the beautiful Parable of the Blind
and Lame (see below) is not Rabbinic, but Indian. The Indian
parallels cannot, however, be the source of the Rabbinic Parable as it
now stands. In the Indian (E. Leumann, Die Avasyaka-Erzahlungen,
Leipzig, 1897, p. 19) a lame man gets on a blind man's back and
together they escape from a forest fire. This is not a source for the
Rabbinic Parable, which differs totally in idea. Nor can I be per-
suaded by Dr M. James (J.T.S. xv. 236) that the version of the
Parable (much closer to the Rabbinic than the Indian is) found in
Epiphanius (ed. Dindorf, n. 683) is older than the Rabbinic. The
Christian form seems to me derived from the latter. Finally I may
refer to the Parable of the Three Rings, made famous by Lessing in
his Nathan der Weise. There are many parallels to this, some using it
as a vindication of Christianity, others of Italian scepticism. In the
Hebrew Chronicle of Solomon ibn Verga, it is a pathetic plea for
tolerance by an oppressed faith, and M. Gaston Paris firmly maintains
94 XII. THE PARABLES
that if not originally Jewish, the Parable is presented in its original
form by the Hebrew Chronicler (Revue des fitudes Juives xi. 5).
Naturally, the preceding jottings to which others might be added
are not designed as a formal discussion of the problem of borrowing.
They may, however, serve as an indication of the vast amount of
research, literary and historical, yet remaining to be undertaken before
the problem can be seriously considered. One thing is clear; the
result cannot but be a triumph for humanism. That Buddha could be
made a hero for Christian and Jew is not the least of the episodes in
that triumph.
Free trade in good stories corresponds to the common experience
and common aspiration of mankind. We have, in the readiness of
men to adopt other men's superstitions, a sad comment on the
universality of the lower elements in human nature. But the adop-
tion by one and all from one and all of beautiful Parables is a mark
of the universality of the higher elements. It is of itself a beautiful
Parable " to preach the simple brotherhood of souls that seek the
highest good."
We must try to get closer to another aspect of the historical problem.
The Parable was used by Old Testament writers with perfection of art.
The Tannaim, from the latter part of the first Christian century onwards,
make a far more extensive use of the method. But, in between, the
later Biblical writers, the authors of the Pre-Christian Jewish Apoca-
lypses (with the possible exception of Enoch) and such a representative
Alexandrian as Philo have no parables. In one of his early works
(Markus-Studien, 1899, p. n), an able Jewish scholar, H. P. Chajes,
concludes that in the age of Jesus the Parable was an unusual device,
and that it had not yet won the place which it afterwards filled in the
Rabbinic method of popular instruction. He even suggests that this
is the original meaning of the Evangelists' discrimination between the
teaching of Jesus and that of the Scribes. " He taught as one having
authority " should read " he spoke in Parable." (Underlying the Greek
text cos ^ova-iav tx wv l & the Hebrew Ke-moshel /IPO? which Dr Chajes
would emend to be-mashal ?^9?.) Dr Chajes proceeds (p. 12): It
will easily be retorted, How could the mere use of Parables have made
so striking a sensation, seeing that the Masked (Parable) plays so
prominent a role among the Rabbis 1 Yes, among the Rabbis; but
it is extremely doubtful whether this was yet the case in the age of
Jesus. A real Agadic activity cannot be posited before the epoch of
Hillel, and no Parable can with certainty be assigned to that teacher.
XII. THE PARABLES 95
It was only at a later period, after the destruction of the Temple, that
the Parable attained high honour, as we already find it to be the case
with Johanan ben Zakkai, Joshua ben Hananya, and especially Meir
(cf. Mishnah, Sota, ix. 15 ; T. B. Sanhedrin, 38b, last lines).
This argument scarcely survives examination. One Rabbinic source
ascribes to Hillel (and, in some readings, also to his contemporary
Shammai) a mystic knowledge of the language of the hills, the trees,
the beasts and the demons, and a special predilection for parables or
fables (Safer 'im, xvi. 9). The authenticity of this ascription is doubted
by Bacher (Agada der Tannaiten, i. 10, notes 3 5). But the only
ground for this suspicion is the fact that the Talmud (T. B. Sukkah,
25 a) makes the same remark concerning Johanan ben Zakkai. Soferim
seerns to present the older tradition, for while it equally ascribes this
knowledge to Johanan, it also carries the statement back to Hillel,
whose disciple Johanan was. Weiss, the author of the History of
Jewish Tradition (in Hebrew) Dor dor vedorashav, i. 157, throws
no doubt on the trustworthiness of the passage in Soferim. That
HillePs thought sometimes ran in the direction indicated appears
also from the Mishnah (Aboth, iv. 8), for Hillel said : " The more
women, the more witchcraft" he may therefore have had an academic
interest in demonology as Soferim asserts. And it is otherwise quite
clear that at all events part of the statement in Soferim must be true,
for we have abundant evidence that Hillel was fond of Parabolical
forms of speech (cf. Weiss, op. cit. pp. i6oseq.). That Hillel was
interested in folk-lore is demonstrated by the anecdotes told of him
(T.B. Sabbath 31 a, Aboth de R. Nathan xv.). Again, in the last
reference, in his interview with a would-be proselyte, Hillel is recorded
to have compared the study of the details of the Temple service to the
etiquette at an earthly Court. This comes very near an actual
Parable. So, too, there is a compressed Parable in Hillel's striking
enunciation of the doctrine of retribution: "He saw a skull which
floated on the face of the water, and he said to it, Because thou
didst drown (others) they drowned thee, and in the end they that
drowned thee shall be drowned" (Mishnah, Aboth, ii. 7). Another
of Hillel's phrases : " He who serves himself with the tiara perishes "
(ib.) is a figurative condemnation of the self-seeker's appropriation
of the Crown of the Torah. Illustrating the covenant of love between
God and Israel Hillel said: "To the place that my heart loves my
feet carry me. If thou comest to My house, I will come to thine;
but if thou comest not to My house I will not come to thine"
96 XII. THE PARABLES
(Tosefta, Sukkah, iv. 3). There are several other such sayings
recorded of Hillel ; and frequent mention is made of his wide
acquaintance with popular lore as well as his readiness to enter
into familiar conversation with the common folk. All of this goes to
confirm the authenticity of the tradition reported in Soferim as cited
above. Besides this, there are quoted in Hillel's name two actual
Parables rudimentary, but bearing unmistakably the Parabolical
stamp. Bacher fully accepts the authenticity of these Parables
though they occur in a somewhat late Midrash (Leviticus ftabba,
Ixxxiv.). Chajes adduces no adequate ground for suspicion. The first
of the two Parables referred to is as follows : Hillel's disciples were
walking with him on a certain occasion, and when he departed from
their company they enquired "Whither goest thou?" He answered,
"I go to fulfil a religious duty." "What duty T' " To bathe in the
bath-house." " Is this, then, a duty?" "Ay," replied Hillel; "the
statues of kings which are set in theatres and circuses he who is
appointed concerning them cleanses and polishes them; he is sustained
for the purpose, and he grows great through intercourse with the great
ones of the kingdom. I, created in the image and likeness of God,
how much more must I keep my body clean and untainted." Ziegler
(op. cit. p. 17) agrees with Weiss and Bacher in holding this passage
a genuine saying. The authenticity is guaranteed (as Bacher argues)
on linguistic grounds, for whereas the preceding passage is in Hebrew,
the second Parable which immediately follows is in Aramaic, and this
very intermixture and interchange of Hebrew and Aramaic is charac-
teristic of several of Hillel's best authenticated utterances. The second
Parable is this : again Hillel is walking with his disciples (the parallel
to the journeys of Jesus in the company of his disciples may be noted);
he turns to part from them, and they ask his destination. "I go home,"
said Hillel, " to render loving service to a certain guest who sojourns in
my house." " Hast thou then a guest ever in thy house ? " " Is not
the unhappy soul a sojourner within the body ? To-day it is here, and
to morrow it is gone ! "
At this point a general remark may be interpolated. While
rendering these and other Rabbinic Parables, the translator feels
himself severely handicapped. Not only were the New Testament
Parables elaborated by the Evangelists far more than the Talmudic
were by the Rabbis, but the former have been rendered with inimitable
skill and felicity, while the latter have received no such accession of
charm. Even Herder's paraphrases of Midrashim are turgid when
XII. THE PARABLES 97
compared with the chaste simplicity of style and form under which the
New Testament Parables appear in the Vulgate, and even more con-
spicuously in Luther's Bible and the Anglican versions. These
versions are, from the point of view of literary beauty, actually im-
provements on the Greek, just as the Hebrew of the twenty-third
Psalm has gained an added grace in the incomparable English rendering
with which we are all familiar. No one has done as much for the gems
of Rabbinic fancy. They have remained from first to last rough jewels;
successive generations of artists have not provided increasingly be-
coming settings to enhance their splendour. But even so some modern
writers have been unfairly depreciatory of the Rabbinic Parables, for
while there is a considerable number of no great significance, there are
some which are closely parallel to those of the New Testament, and
some others which may be justly placed on the same high level. There
are no more beautiful Parables than that of the blind and the lame
(Sanhedrin, 91 a b, Mechilta, n!?KO ii.), which may be summarised
thus :
A human King had a beautiful garden in which were some fine early figs. He
set in it two watchmen, one lame and the other blind. Said the lame man to the
blind, "I see some fine figs, carry me on your shoulders and we will get the fruit
and eat it." After a time the owner of the garden came and asked after his missing
figs. The lame man protested that he could not walk, the blind that he could not
see. So the master put the lame man on the blind man's back and judged them
together. So God brings the soul and casts it in the body (after death) and judges,
them together.
It is difficult to understand why the excellence of such Parables
should be contested. Fiebig (p. 88) objects that it is very improbable
that a king should employ the lame and the blind as watchmen. One
wonders why not, seeing that in the East particularly the old and the
decrepit are much used for such sedentary work. It may be that the
difficult passage II. Samuel v. 6 implies the employment of the blind and
lame as sentinels of the citadel. Undoubtedly the idea of the watchmen
is necessary for the Rabbinic Parable which is not a mere adaptation
of the Parable which Dr James cites. In the Epiphanius parallel
(J.T.S. loc. cit.) the King is described as possessing among all his
subjects only two men unfit for military service, this is surely not less
improbable than the lame and blind watchmen. Besides, there are
many improbabilities in the New Testament Parables also (as e.g.
the refusal of a king's invitation to a banquet, Matt. xxii. 2 ; in
Luke xiv. 1 6 the banquet however is given not by a king but by " a
7
y XII. THE PAEABLES
certain man"). Such improbabilities are not defects in Parables at all.
We might have been spared some inept criticism of the New Testament
Parables, had due notice been taken of the wise Rabbinic maxim : Do
not apply your logic to a Midrash. Again, it is sometimes said that
the Rabbinic Parables fall below those of the New Testament in that
the latter deal with far greater subjects, Sin and Grace, Prayer, Mercy,
Love, the Kingdom of Heaven (Fiebig, p. 105). That in the enormous
mass of Rabbinic Parables many treat of trivialities in a trivial
fashion is true ; but simplicity must not be confused with insignificance.
There is a quality of homeliness about many of the Rabbinic Parables,
a quality inherited from the Bible, with its Ewe-lamb and its Song of
the Vineyard. It is this quality that distinguishes the Jewish from
the ordinary Eastern Parable; the former, far less than the latter,
merely illustrates a maxim. Many Oriental Parables are expanded
Proverbs, but the Rabbinic Parables cannot as a rule be compressed
into a Proverb. As to subject matter, very many of them are directed
to most of the subjects which Fiebig enumerates, and to other funda-
mental problems of life and death and the hereafter. Thus the Parable
quoted above of the lame and the blind expresses the unity of body
and soul, or rather the truth that a man is a single product of dust
and spirit. The persistence in later Jewish thought of the belief in
the bodily resurrection was in part, at least, due to the impossibility
of separating body and soul, even in the aspect of immortality.
The following summary from the excellent article by Dr J. Z.
Lauterbach (Jewish Encyclopedia, ix. 5 13 a) is a just though of
course incomplete statement of the subjects of the Rabbinic Parables :
In the Talmud and Midrash almost every religious idea, moral maxim, or
ethical requirement is accompanied by a Parable which illustrates it. Among the
religious and moral tenets which are thus explained may be mentioned the following :
the existence of God (Gen. E. xxxiv. i); his manner of retribution, and of punishing
sins both in this world and the next ('Ab. Zarah, 4 a, Yalq. Lev. 464, Sabb. 152 a);
his faithful governance ('Ab. Zarah, 55 a, Sanh. io8a); his impatience of injustice
(Suk. 30 a) ; his paternal leniency (Ex. E. xlvi. 6) and his relation to Israel
(t'6. xlvi. 4, Ber. 32 a) ; Israel's sufferings (Ber. 13 a) ; the folly of idolatry
('Ab. Zarah, 54 b 5 5 a); the Law as the guardian and faithful protector in life
(Sotah, 21 a); the sin of murder (Mechilta, "HIV, 8); the resurrection (Sabb. 91 a);
the value of benevolence (B. B., loa); the worth of a just man for his contem-
poraries (Meg. 1 5 a) ; the failure of popularity as a proof of intrinsic value
(Sotah, 40 a) ; the evil tendency of freedom from anxiety (Ber. 32 a) ; the
limitations of human knowledge and understanding (Sanh. 39 a); the advantage
frequently resulting from what seems to be evil (Niddah, 31 a) ; conversion
(Sabb. 153 a); purity of soul and its reward (ib. 152 b).
XII. THE PARABLES 99
This list could be much extended, but it suffices to demonstrate
that the depreciation of the Rabbinic Parables, on the ground of
triviality of motive, is a mere aberration of criticism. It can, therefore,
hardly be maintained with Fiebig (op. cit. p. 87), that "the manifold
situations of human life are only sparingly and pallidly depicted " in
the Rabbinic Parables. It is, on the other hand, a sound discrimina-
tion (p. 83) that there are in the Rabbinic literature a vast number of
royal Parables. Hillel and Johanan b. Zakkai present some examples
(Bacher, Agada der Tannaiten, I. 73, 81). Most of the royal Parables,
however, belong to the period later than the fall of Bethar in 135,
and they only begin to predominate with Domitian in the hands* of
Agadists like Meir and Simon b. Yohai (Ziegler, p. xxiii). By that
time the interest of Jewish moralists in good government as part of
the idea of the Kingdom of God (cf. Schechter, Aspects of Rabbinic
Theology, ch. vii.) led them to portray under royal metaphors the
relations of God to man, and they did this both by way of contrast
and similitude. Some of the oldest Parables in which the heroes are
kings, perhaps dealt in their original forms with ordinary men, and
kings was probably substituted for men in some of them (both Rabbinic
and Synoptic) by later redactors.
One point deserves close attention. It is not possible to assent to
Fiebig's characterisation that " in comparison with the Synoptic
Parables, it strikes one that the processes of Nature sowing and
harvest, growing, flowering and fruitage, were taken little account
of [in the Rabbinic Parables]." In the latter, besides many Parables
treating of trades, handicrafts, seafaring, school-life, domestic affairs,
there are many comparisons drawn from the fields, vineyards, streams,
flowers, trees, fruits, birds, beasts, and other natural objects. This
is perhaps more noticeable in those phases of the Agada which do
not assume the form of narrative Parables, but it is frequent in the
latter also, and the Rabbinic examples agree with the Synoptic in
treating of nature under cultivation rather than in a wild state.
With regard to the harvest, Schweitzer holds that the reference in
the New Testament Parables is eschatological, pointing at all events
to a definite note of time ; this particular harvest in the last year of
Jesus' life is to be the last harvest on earth, and the Kingdom is to
follow it immediately. In Joel iii. 13, Isaiah xvii. 5 n, as well as
in the Jewish Apocalypses (e.g. Baruch Ixx.), the harvest is synonymous
with the judgment. This is not altogether convincing, for it is curious
that the images of the sower and the mustard seed the harvest and
72
100 XII. THE PARABLES
the full-grown tree, processes of long maturation should express the
idea of a sudden consummation and nothing more.
The idea of the harvest in the Synoptics is probably a composite
one, the standing corn is regarded as food for the sickle, whether it be
the sickle of an angry Master or of the human reaper of the accumulated
reward of long drawn out endeavour. If the expression " the harvest
is large but the labourers are few " (Matt. ix. 37 38, Luke x. 2 ;
cf. John iv. 36) were the authentic exordium to the mandate to the
disciples in Q, we have here the harvest used in quite a different sense
from the Apocalyptic. Both these uses meet us in Rabbinic. In the first
place, with regard to the passage just cited, there is a Rabbinic parallel
nearer than is generally supposed, though so long ago as 1847 Zipser
suggested it (Literaturblatt des Orients, 1847, c0 ^ 75 2 )- I n ^ ne Mishnah,
Aboth ii. 19 (20), occurs a saying which in Dr Taylor's rendering runs
thus : " R. Tarphon said, ' The day is short, and the task is great, and
the workmen are sluggish, and the reward is much, and the Master of
the house is urgent. He said, It is not for thee to finish the work,
nor art thou free to desist therefrom ; if thou hast learned much Torah,
they give thee much reward ; and faithful is the Master of thy work,
who will pay thee the reward of thy work, and know that the recom-
pence of the reward of the righteous is for the time to come.'"
Dr Taylor sees in this Mishnah points of contact with the Parable
of the Vineyard in Matt, xx., "where the oiKoSecrTroTijs (Master of
the house) says to the labourers whom he finds unemployed, Tt oSSe
eo-Ty/Kcn-e o\.rjv rrjv ^/xepav apyot; ('Why stand ye here all day idle?')."
The first part of this Mishnah is usually taken to correspond to the
"ars longa vita brevis" of Hippocrates. But it is a very plausible
suggestion of Zipser's that the first clause of the Hebrew has been
wrongly punctuated. It is commonly read ">Xi? Bi*n ("the day is short"),
whereas the true reading should be IVi? Ei'5 ("to-day is harvest"
there is no need to emend to "VVi? as the Gezer Calendar Stone,
published in the Quarterly Statement of the P. E. F., Jan. 1909,
gives us several times over the spelling -|^p for "harvest"). This
is confirmed by another word in the saying, "Master of the House,"
for the Hebrew equivalent nun t>ID often means "landowner"
(cf. Dr A. Biichler, Sepphoris, p. 38, etc.) just as the oiKoSeo-TroV^s
of Matthew does (this equivalence of the Hebrew and Greek just
quoted was noted by Dr Taylor, and has been elaborated I think by
Dr Nestle). The whole of Tarphon's saying would thus have an
agricultural setting. It may be pointed out in passing that this is
XII. THE PARABLES 101
not the only parallel between sayings of Tarphon and the New
Testament. Compare the "mote" and "beam" of 'Arachin, 16 b
with Matt. vii. 3 (there seems no reason for doubting with Bacher,
Agada der Tannaiten, i. 35 1 n., the authenticity of this saying as one
of Tarphon's). Tarphon lived during the existence of the Temple
(T. J. Yoma, iii. 7, 38 d), and was thus a contemporary of the
Apostles. He was a strong opponent of the Jewish Christians (Sabbath,
n6a), and hence his name was used by Justin Martyr (whose
Tryphon = Tarphon) as a typical antagonist. It is impossible that
Tarphon would have taken his similes from Christian sayings, and
the parallels point unmistakably to the existence of a common and
ancient source. The whole Mishnah is more elaborate than most of
the passages in Aboth and we may conclude that Tarphon is not the
author of the opening clauses but only of their interpretation in terms
of studying the Law.
These opening clauses however, when juxtaposed with Matt. ix.
37 8, present under the figure of the harvest a very different idea
from the Judgment. It is the goal of effort rather than the starting
point of doom, the reward of life rather than the precursor of death.
There is nothing apocalyptic about this, nothing catastrophic. " The
king does not stand (in satisfaction) by his field when it is ploughed,
or when it is hoed, or when it is sown, but he stands by it when it
is full of corn for the granary," said R. Simon (Tanhuma Miqes on
Gen. xxviii. 13). On the other hand there are some Rabbinic passages
in which the harvest is a type of the Judgment in the sterner sense
(Leviticus Rabba, xviii. 3).
Several of the New Testament Parables are clearly inconsistent with
a firm belief in the immediate approach of the end; there is no
"interim morality" in the Parable of the Talents (Matt. xxv. 14 30,
Luke xix. 12 27, cf. Mark xiii. 34 37). It is improbable, however,
that the same Jesus who said "Be not therefore anxious for the
morrow" (Matt. vi. 34), and "Sell all thou hast" (ib. xix. 21), should
have cried " Well done, good and faithful servant " to those who
had traded with their capital. To the idea of this story we have
a Rabbinic parallel, but not in Parable form ; it is cited as an incident
(Debarim Rabba y in. 3), and in some particulars the moral is other
than in the New Testament. For, after all, the five and the two
talents were risked, and might have been lost in the trade. In the
Midrash incident this objection does not suggest itself. This is the
incident referred to : " R. Phineas ben Jair [second half of second
102 XII. THE PABABLES
century] lived in a certain city of the South [Lydda?], and certain
men went to support themselves there. They had in their possession
two seahs of barley, which they deposited with him. These they forgot
and left the place. And R. Phineas ben Jair went on sowing them
year by year ; he made a granary for them, and stored them. After
seven years these companions returned to claim their seahs. Immedi-
ately R. Phineas ben Jair recognised them, and he said to them, Come,
take your stores. Lo, from the faithfulness of flesh and blood thou
recognizest the faithfulness of the Holy One, blessed be He." (This
last clause reminds one of the "faithful servant.")
In Rabbinic parallels to several others of the Synoptic Parables
the inferiority is not always on the Rabbinic side as Jiilicher in
particular thinks. In the first place the parallels sometimes strike a
note which finds no exact echo in the Synoptic examples. It is strange
that Fiebig can cite (from Mechilta JBeshallah, ed. Friedmann, 29 b)
the following as he does (op. cit. p. 34) without noting that it is a
somewhat unique expression of the relation between God and man.
Rabbi Absolom the Elder says : A Parable. To what is the matter like ? To
a man who was angry with his son, and banished him from his home. His friend
went to beg him to restore his son to his house. The father replied : Thou askest
of me nothing except on behalf of my son? I am already reconciled with my son.
So the Omnipresent said unto him (Moses), "Wherefore criest thou unto me?"
(Exodus xiv. 15). Long ago have I become well disposed to him (Israel).
Here then we have the idea that the Father is reconciled to his
erring son even before the latter or any intercessor makes appeal, in
accordance with the text: "Before they call I will answer" (Isaiah Ixv. 24).
Compare also the similar idea in the Pesiqta Rabbathi ch. v. (ed.
Friedmann, p. 1 7 b) ; these expressions of the Father's love seem to go
even beyond the beautiful pathos of Luke xv. 20.
A King ordered the men of a certain district to build a palace. They built it.
Then they stood by the gate and proclaimed : Let the King come in ! But what did
the King do ? He entered by a wicket door, and sent a herald to announce : Shout
not, for I have already come to the palace. So, when the Tabernacle was erected,
Israel said : Let my Beloved come to his garden ! The Holy One sent and said
unto them : Why are ye anxious? Already have I come into my garden, my sister,
my bride.
So, too, the medieval poet Jehuda Halevi sang, though he was thinking
more of the divine omnipresence :
Longing I sought Thy presence,
Lord, with my whole heart did I call and pray;
And going out toward Thee,
I found Thee coming to me on the way.
XII. THE PARABLES 103
Another note of the Rabbinic Parables (which has I think no echo in
the Synoptics) is the idea of " Chastisements of love " (Serachoth 5 a)
which finds expression in many comparisons, among which perhaps the
following is the most characteristic (Exodus Rabba, xxi. 5). The
Midrash very pathetically puts it that God wishes Israel to cry to him,
he longs to hear Israel's voice raised in filial supplication. Just as he
chastises Israel to discipline him, so he tortures Israel to force from
him the prayer which Israel refuses to yield while free from racking
pain. The divine ear yearns for the human voice. This profound,
mystical thought, is expressed with both quaintness and tenderness in
the following Parable :
Why did God bring Israel into the extremity of danger at the Red Sea before
saving him ? Because he longed to hear Israel's prayer.. Said R. Joshua ben Levi,
To what is the matter like ? To a king who was once travelling on the way, and
a daughter of kings cried to him: "I pray thee, deliver me out of the hand of
these robbers ! " The king obeyed and rescued her. After a while he wished to
make her his wife ; he longed to hear her sweet accents again, but she was silent.
What did the king do ? He hired the robbers again to set upon the princess, to
cause her again to cry out, that he might hear her voice. So soon as the robbers
came upon her, she began to cry for the king. And he, hastening to her side,
said: "This is what I yearned for, to hear thy voice." Thus was it with Israel.
When they were in Egypt, enslaved, they began to cry out, and hang their eyes
on God, as it is written "And it came to pass... that the children of Israel sighed
because of their bondage... and they cried..." Then it immediately follows : "And
God looked upon the Children of Israel." He began to take them forth thence with
a strong hand and an outstretched arm. And God wished to hear their voice a
second time, but they were unwilling. What did God do ? He incited Pharaoh
to pursue after them, as it is said, "And he drew Pharaoh near." Immediately
the children of Israel cried unto the Lord. In that hour God said: "For this
I have been seeking, to hear your voice, as it is written in the Song of Songs, My
dove in the clefts of the rocks, let me hear thy voice; thy voice, the same voice
which I first heard in Egypt.
Again, the following is a gracious Parable, which, were one on the
look-out for Rabbinic foils to the Gospels, might be contrasted with
Matthew xxi. 9.
When R. Isaac parted from R. Nahman, the latter asked for a blessing. Said
R. Isaac : I will tell thee a Parable. A traveller was passing through a desert, and
he was hungry, faint, and thirsty. He found a tree, whose fruit was sweet, whose
shade was pleasant, and at whose foot there flowed a stream. He ate of the fruit,
drank of the water, and sat in the shade. On his departure he said : tree, tree,
how shall I bless thee ? If I say to thee, May thy fruit be sweet, lo thy fruit is sweet
already ; that thy shade shall be pleasant, lo it is pleasant now ; that a stream shall
water thee, lo this boon is thine at present. But I will say: May all the saplings
104 XII. THE PARABLES
planted from thee be like thyself ! So, thou, How shall I bless thee? With Torah?
Torah is thine. With wealth? Wealth is thine. With children? Children
are thine. But I say: God grant that thy offspring may be like thyself!
(Ta'anith 6 a).
Or, to turn to another idea, the following is an original note, at all
events there is no full Synoptic parallel. The citation of the passage
will serve also a secondary purpose ; it will again illustrate the frequent
Kabbinic habit of syncretising the Parable of idea with the application
of historical incident.
B. Hanina bar Idi said : Why are the words of the Torah (Scriptures) likened
unto water, as it is written (Isaiah Iv. i) Ho every one that thirsteth come ye to the
water ? To say unto thee : Just as water forsakes a high place and goes to a low
place, so the words of the Torah find a resting-place only in a man whose character
is lowly. R. Oshaya also said : Why are the words of the Torah likened to these
particular liquids, water, wine, and milk, for the text continues : Come ye, buy wine
and milk without money ? To say unto thee : Just as these three liquids are kept
only in the simplest of vessels, so the words of the Torah are only preserved in
a man of humble spirit. It is as once the Emperor's daughter jeeringly said to
B. Joshua b. Hananya : " Ho ! Glorious Wisdom in a foul vessel ! " He replied :
"Ho ! daughter of him who keeps wine in an earthen pitcher !" "In what sort of
vessel should wine be kept, then?" asked the princess. "Important people like
you should store their wine in pitchers of gold and silver." She persuaded the
Emperor to follow this course, but soon men came to him to report that the wine
had turned sour. "My daughter," said the Emperor, "who told you to suggest
this thing?" She replied that her adviser was B. Joshua b. Hananya. The latter
was called, and in answer to the Emperor's questions replied: "As she spake to me,
so spake I unto her." (Taanith, 7 a; Nedarim, 50 b).
Various Kabbinic parallels to New Testament Parables have been
detected by various scholars. One must here remark that the similarity
of idea must not be confused with identity of Parabolical treatment.
Philo has no true Parables, but several of his ideas are found later
on developed into that literary type. For instance, what became a
favourite Rabbinic Parable, the comparison of the creation of the
world to the planning of a palace (Genesis Rabba, i.), a comparison
associated by Bacher with the schools of Hillel and Shammai, is
already found fully developed in Philo (de opif. mundi, 4, 5).
Leaving the study of parallels, if the Rabbinic Parables are con-
sidered absolutely, without comparative reference to those of the New
Testament, it is clear that they must be allowed to rank high in
literature of the kind. The Parable took a very firm root in the
Jewish consciousness, though for some centuries it was not transplanted
from its native soil Palestine to Babylonia, and Rab (died 247)
XII. THE PARABLES 105
scarcely presents any instances of the Mashal (Bacher, Agada der
babylonischen Amoraer, 1878, p. 31). But the influence of the
Palestinian Midrash prevailed, and throughout the middle ages and
the modern epoch, Jewish homilies have been consistently illustrated
by Parables. Now, as of old, the Parable was the instrument for
popularising truths which in an abstract form were not so easily
apprehensible.
Professor Bacher elsewhere describes the Mashal (Parable) as " one
of the most important elements of the Agada." Agada must here be
understood in its widest signification : the exposition of Scripture and
the application of the precepts of the Law to the elucidation of principle
and the regulation of conduct. The utility and even necessity of the
Mashal for understanding the Torah are variously enunciated in a
series of fine similes in the Midrash, and the passage (Canticles
Rabba, i. i. 8 ; Genesis Rabba, xii. i ; Eccles. Rabba on ii. 1 1 ;
T. B. Erubin, 2 1 b footnote) may here be paraphrased in full :
^ R. Nahman said : A great palace had many doors, and whoever
entered within it strayed and lost his direction (for the return).
There came one of bright intelligence who [cf. Ariadne] took a clue
of rope and tied one end of it to the entrance, and went in and out
along the rope. Thus before Solomon arose no man could understand
the words of the Torah, but all found it intelligible after the rise of
this King." Further, said R. Nahman, "It is like a wild thicket
of reeds, into which no man could penetrate. But there came a clever
wight who seized a scythe and cut a path, through which all men could
come and go. Thus was it with Solomon." R. Jose said : " it is com-
parable to a great case full of fruits, but the case had no handles and
no one could move it. Then there came one who made handles, and
everyone could move it." R. Shila likened Solomon's service to that
of a man who provided a handle to a huge cask full of hot liquid.
R. Hanina put the same thought in these terms : "It was like a deep
well, full of water, and the water was cool, sweet and wholesome, but
no creature could reach it to drink. A certain one came and joined
rope to rope and cord to cord; he drew water from the well and drank.
Then, for the first time, all could draw and drink. Thus from word
to word, from Mashal to Mashal, Solomon reached the uttermost secret
of the Torah. And this he did by means of the Mashal." So, the
passage continues, " The Rabbis said, Let not the Mashal be light in
thine eyes, for by means of the Mashal a man Can stand in the words
of the Law, for it is comparable to a king who lost gold from his
106 XII. THE PARABLES
house, or a precious pearl, and found it by means of a clue worth
a Roman as." There is in all this a two-fold meaning. Solomon
added certain things to the Law, the Rabbis assigned to him a number
of takkanoth or new regulations which made the Law practically
usable; he also popularised the Law, making it accessible to the
masses by means of the Mashal. As the Midrash continues, Solomon
by means of the Mashal attained to a knowledge of legal minutiae ;
he also made the Law popular. "Rabbi Judan said, Whoever
speaks words of Law in public (among the many) is worthy that
the Holy Spirit should rest upon him, and this thou learnest from
Solomon."
In this analytical passage, the term Mashal is used in a very wide
sense, and includes all forms of applied morality. Parable thus
becomes part and parcel of the instrument for arriving at truth and
for making truth prevail. Truth, to Pharisee and Evangelist alike,
is the will of God, and the Parable was at its highest when seeking
to understand and to do that will. The Parables of Talmud and
Gospels are (so Zipser put it) derived from a common source, the
systematised teaching of Hillel and Shammai. Parables were not
merely an entertainment, they were not merely designed to interest
the people. They were the method by which the mysteries of pro-
vidence and the incidences of duty were posted and illustrated.
Sometimes these mysteries and incidences are beyond understanding \
and when then Mark (iv. r i) describes the Parable as actually employed ]
by Jesus to prevent men from understanding, the description is happily
characterised by Bousset when he calls it " preposterous," and dismisses
it as "the dogmatic pedantry of a later age." The same idea is found
in all the Synoptics and cannot be dismissed in this easy way. What
is " preposterous " is the supposition that Jesus taught in Parable in |
order that men might misunderstand. This is to mistake an Orientalj
process of thought by which consequences are often confused with
motives. (Cf. Skinner on Isaiah vi. 10.) The Parable has this danger
that it may imply more than it says, and may leave behind it more
puzzles than it solves. It is not an exact instrument ; it works without
precision. The consequence of a Parable may be misunderstanding,
or what is equivalent, partial understanding, and it is certain from
the language of the evangelists that the Parables ascribed to Jesus
were liable to this consequence. Hence, as it was improper to admit -
that Jesus used an imperfect form imperfectly, consequence was
translated into intention, and the misunderstanding was described as
XII. THE PARABLES 107
designed in order to prevent the Jews from turning and finding for-
giveness. Later on, when the eschatological element in the teaching
of Jesus was forced into greater prominence, the supposition that the
Parable was used in order to veil a Messianic secret may easily have
arisen. The latter, however, cannot be the original force of the
reference, for it is plain enough that many of the New Testament
Parables, different though they be to explain in all their details, are
absolutely simple inculcations of moral and religious truths, profound
but not mysterious.
XIII. DISEASE AND MIRACLE.
Rabbinic Judaism took over from the Old Testament a belief that
disease was a consequence of sin (Leviticus xxvi. and parallels in
Deuteronomy). This theory was especially held to explain general
epidemics, and also those afflictions the origin of which was at
once most obscure and their effects most dreaded such as leprosy.
It is not necessary to do more than recall the cases of Miriam, Joab,
Gehazi, and Job.
The Rabbinic sources contain many assertions as to the relation
between sin and disease. (Of. the valuable discussion in the Tosafoth
to Aboth iv. n.) "Measure for Measure" applied here as in other
aspects of Rabbinic theology (Mishnah, Aboth v. n 14). R. Ammi
(of the third century, but his view was shared by earlier authorities)
asserted sans phrase that there was no affliction without previous
sin (Sabbath, 55 a). R. Jonathan said : " Diseases (D^3) come for
seven sins : for slander, shedding blood, false oaths, unchastity,
arrogance, robbery, and envy" ('Erachin, 16 a). In particular leprosy
was the result of slander (Leviticus Rabba, xviii. 4). On the other
hand, "When Israel stood round Sinai and said, All that the Lord has
spoken we will do, there was among the people no one who was a
leper, or blind, or halt, or deaf," and so forth (ibid. ; Sif r6 i b, the sin
of the golden calf, like other acts of rebellion, caused leprosy and other
diseases, Pesiqta Rabbathi vii., ed. Friedmann p. 28). Thus obedience
prevented disease, just as disobedience produced it. This, to a large
extent, moralised the idea : it set up the moral life as the real
prophylactic. In general the principle enunciated in Exodus xv. 26
was adopted by the Rabbis, though it must be remembered that so
great an authority as R. Meir altogether disputed the theory as to the
connection between suffering and transgression. God's dealing with
men, he held, was an unfathomable mystery. Leprosy, again, like
XIII. DISEASE AND MIRACLE 109
other diseases might, in another view, merely be the beneficent earthly
penalty designed to save the sufferer from tribulations in thp future
(Lev. R. xvii.).
To exemplify the application of the " Measure for Measure " idea,
the case of blindness will suffice. Nahum of Gimzu (first century)
explained his blindness as the consequence of his inhumanity to a poor
sufferer (Ta'anith, 2 1 a). The man who accepted bribery and perverted
justice would not pass from the world unless he suffered the infliction
of physical blindness corresponding to his moral lapse (Mechilta, Mish-
patim, 20, p. i oo a, Sifre, on Deuteronomy, 144). The case of one
blind from birth was more difficult to fit into the theory, and in
John ix. i Jesus denies that such an affliction was due to sin at all.
It is there explained that the congenital blindness had been imposed
that it might be cured, so "that the works of God should be made
manifest in him." This explanation is identical with that of Eccle-
siasticus xxxviii, except that Sirach applies it to the doctor's art.
"The Lord hath given men skill, that he might be honoured in his
miraculous works. " Disease more particularly pestilence was ascribed
also to sins which were not punished by human tribunals. In general
it was thought that sin left its material impress, and the later mystics
put it that it disfigured the image of God (Schechter, Studies in
Judaism, n. 274).
Two points only must be further indicated; the legal position of
the leper in Rabbinic law is sufficiently indicated in the Jewish
Encyclopedia vm, 10 a. ("Leprosy was not considered contagious.")
The first point is that the moral stigma attaching to disease soon
took a more amiable form. As Dr Schechter well puts it (Studies in
Judaism, i. 269) : ' The only practical conclusion that the Rabbis drew
from such theories as identify suffering with sin was for the sufferer
himself, who otherwise might be inclined to blame Providence, or even
to blaspheme, but would now look upon his affliction as a reminder
from heaven that there is something wrong in his moral state. Thus
we read in tractate Berachoth (5 a) : " If a man sees that affliction
comes upon him, he ought to inquire into his actions, as it is said, Let
us search and try our ways, and turn again to the Lord (Lam. iii. 40).
This means to say that the sufferer will find that he has been guilty of
some offence." '
The second point is that though leprosy was regarded as the punish-
ment for the worst crimes, it was not thought lawful or right to leave
the leper to his fate. Sympathy with suffering was not diminished by
110 XIII. DISEASE AND MIRACLE
any theories as to the origin of the suffering. In Ecclesiasticus Rabba
(on ix. 7) is told the touching story of Abba Tahna. As the sun was
near its setting on a Friday afternoon, Abba Tahna was going hlpme
with all his worldly goods in a bag on his shoulders. At the cross-road
he saw a man smitten with leprosy. The latter entreated the Rabbi
in these terms : " My master, show me charity and carry me to the
city." The perplexed Rabbi said : " If 1 leave my goods, how shall
I sustain myself and my household ? and if I leave this leper I shall
commit a mortal sin." Abba Tahna conquered the suggestion of his
evil inclination, left his bag, and bore the leper into the town. In the
end he did not suffer for his action. But the whole passage is an
effective comment on Luke x. 30.
Demoniac "possession" as a cause of disease, and "exorcism" as
its cure, were well known to the Rabbis. But it is certain that these
beliefs and practices were uncommon in Palestine at the time of Jesus.
The easy assumption to the contrary has no foundation. Though the
Enoch and other apocalyptic literature has a developed demonology,
and Acts xxiii. 8 implies a Pharisaic angelology, there is a remarkable
infrequency of references to the subject in the Mishnah and the
Tannaite literature (L. Blau, Das altjiidische Zauberwesen, p. 23).
Quite early was the power attached to prayers for rain. The fact that
Onias (on whom see Jewish Encyclopedia ix. 410 and refs.) stood in a
ring while praying for rain has a " magical " look, but it is not clearly
a charm. There is nothing of the magician or spell-worker in the
picture of Onias drawn in Josephus (Antiq. xiv. 2, i). Hillel (p. 95
above) was a student of demon-lore, perhaps un<ier Parsic influence
he was by birth Babylonian. Compare the prayer cures of Haninah b.
Dosa (6rst century) he had magical leanings (see J.E. vi. 214), but
the female demon Agrat mentioned in his case was Persian. Persian
influence reached Palestine in the first century (Darrnesteter in Revue
des Etudes Juives I. 195) but became more pronounced after the
Palestinian schools were superseded by the Babylonian early in the
third century. Members of the Sanhedrin were expected to under-
stand magic in order to deal with causes in which the question arose
(Sanhedrin 17 a. See refs. in Taylor, Aboth v. 9). The same Mishnah
(v. 9) refers to demons, but this like Hagigah 16 a apparently belongs
to the late second century. It is in the Babylonian Talmud that we
find an appalling mass of demonology which, though it stands in rela-
tion to earlier beliefs, Biblical, Apocalyptic and Rabbinical cannot
properly be cited as applicable to the time of Jesus in the Holy Land
XIII. DISEASE AND MIRACLE HI
(Peiies on Bousset, p. 35. Bousset frankly admits the validity of
Perles' objection in the second edition of his Religion des Judentums,
p. 388, n. 4, but hardly corrects his general statements in accordance
with the admission). Probably, therefore, the Pharisees were amazed
at the attitude and actions of Jesus, so that it is intelligible that
Jesus was afterwards called a "magician" (Sabbath, 104 b), though
subsequent schools of Pharisaism would have been less amazed than his
contemporaries were. It may be, indeed, the fact that the Essenes
were (as Geiger supposes) "healers," in which case we should have a
further bond between Jesus and this sect. There was between the
years 150 and 450 a great increase in Jewish circles in the belief in
demons and their influence. (Cf. Conybeare, Jewish Quarterly Review,
ix. 87.) It is undeniable, however, that some cases of exorcism are
recorded earlier. But it is curious that they are all associated with
the Roman imperial family. Josephus, who makes indeed a general
assertion as to demoniac possession (Wars vn. vi. 3), only recites an
actual cure by exorcism performed in the presence of Vespasian
(Antiquities vm. ii. 5). So, too, the notorious instance of exorcism
reported of a second century Rabbi, Simon b. Yoliai, was not only
performed in the case of a Roman lady of the imperial family, but
actually occurred in Rome, if it be not indeed a mere reproduction
of a Christian story (see p. 92 above). Again, though the Jewish
exorcists (Acts xix. 13) were "strollers," yet the scene of their
exploits is not Judaea but Ephesus and the impression conveyed is
that they were playing with foreign fire. It does not seem, there-
fore, appropriate to the purpose of these Notes to enter at large into
the Rabbinic parallels to New Testament ideas on demonology. (See,
besides the literature already referred to, Kohler in Jewish Encyclopedia,
IV. 517 b.)
In the earlier period we find the physician held in high repute
(Ecclus. xxxviii. i seq.), though Sirach accepts the theory that disease
is connected with sin. The "confections" of the apothecary are
associated with prayer in effecting a cure. Moses prays for Miriam's
relief, and God is the " Healer." The prayer for such divine healing
found a place in the oldest part of the Synagogue liturgy, the eighteen
benedictions, the words used being derived in part from Jeremiah xvii.
14. This two-fold conception always finds expression in Jewish
thought. Prayers for the sick go side by side with the demand that
every community shall have its doctors (Sanhedrin, 1 7 ; Maimonides
112 XIII. DISEASE AND MIRACLE
Sanh. i. 10). Rabbinic "medicine" has very much of the "sympathetic"
and the folk-cure and the exorcist about it, but there is no ground
whatever for Bousset's assumption that the Rabbinic demonology arose
from any supposed surrender of the divine omnipotence, and the
yielding of part of his powers to demons and the like. The Rabbis
considered, in one sense, every recovery from sickness as a " miracle."
Said they: "Greater is the miracle that occurs when a sick person
escapes from a perilous disease than that which happened when
Hananiah, Mishael and Azariah escaped from the fiery furnace"
(Nedarim, 41 a).
XIV. POVERTY AND WEALTH.
The twelve were sent forth "two by two," just as was the rule
with the Jewish collectors of alms (T.B. B. Bathra 8 b) ; indeed
solitary travelling, especially at night, was altogether antipathetic to
Jewish feeling. According to all three synoptics (Mark vi. 7, Matt.
x. 10, Luke ix. 3) the disciples were to take nothing for their journey,
no provisions, no wallet, no money. Even so did the Essenes travel,
according to the report of Josephus ( War n. viii. 4) : " They carry
nothing at all with them when they travel." The twelve were to
accept hospitality wherever it was offered, and the Essenes " go (on
their journeys) into the houses of those whom they never knew before,"
the houses, however, belong to brother Essenes. The Essenes carried
weapons with them, while Matthew and Luke distinctly assert that
the twelve were not even to carry a staff. This seems an improbable
restriction, for the staff (pa/?8os) was a common necessary for the
traveller, serving at the same time as a help to walking and as a
weapon. The ordinary Jewish traveller carried a staff and a bag (see
Dictionaries s.v. ^Din)- Mark distinctly states that the twelve were
to carry a staff (el prj pd(38ov /xovov), and later on we find one or two of
the disciples in possession of weapons (Mk xiv. 47, Matt. xxvi. 51),
Luke (xxii. 38) reports that there were two swords. Luke seems to
feel the contradiction between the earlier commission and this, and so
inserts the passage (xxii. 35, 36) to explain the divergence.
The Essenes were " despisers of riches " (Josephus, loc. cit. 3) but
they were not worshippers of poverty. "Among them all there is no
appearance of abject poverty, or excess of riches," says Josephus.
Theirs was a rule of equality, a regime of simple sufficiency not of
common insufficiency. A life of such poverty was the natural corollary
of life in a society aiming at a holy life, and we find a similar rule
among the Therapeutae described by Philo; though the Therapeutae
were closer to the later Christian monastics than were the Essenes.
A. 8
114 XIV. POVERTY AND WEALTH
That the pursuit of certain ideals was incompatible with the desire to
amass material wealth is, however, a common thought of the Rabbis :
" This is the path to the Torah : A morsel with salt shalt thou eat,
thou shalt drink also water by measure, and shalt sleep upon the
ground, and live a life of trouble the while thou toilest in the Torah.
If thou doest this, happy shalt thou be and it shall be well with thee
(Ps. cxxviii. 2); happy shalt thou be in this world, and it shall be well
with thee in the world to come " (Mishnah, Aboth vi. 4).
But this implies no cult of poverty. Among the blessings prayed
for by Abba Areka were "wealth and honour" (Berachoth 16 b).
From time to time, ascetic movements have arisen in Judaism (cf.
Jewish Encyclopedia ii. 167), and the value of such movements cannot
be denied (cf. C. G. Montefiore Truth in Religion pp. 191 seq.). On
the whole, however, Pharisaic Judaism had, on the one hand, too full
a belief in calm joyousness as a fundamental and generally attainable
ideal of life, and on the other hand too acute and recurrent an ex-
perience of the actualities of destitution, for it to regard poverty as
in itself a good. (Cf. Note XVI below.) Even in the pursuit of the
Torah, there comes a point where poverty is a preventive rather than
a help. Eleazar ben 'Azariah, who succeeded the second Gamaliel as
President of the Sanhedrin, and was himself wealthy (Qiddushin 49 b),
summed the truth up in his epigram : " Without food, no Torah ;
without Torah, no food" (Aboth iii. 26). That destitution may "be
a bar to the ideal is an experience of many an idealist. After the
Bar Oochba war, there was so general an impoverishment in Palestine,
that the study of the Torah was intermitted. (Cf. the lurid picture
drawn by Dr A. Biichler in his essay on Sepphoris in the Second
and Third Centuries, pp. 70 seq.) " God weeps daily alike over the
man who could study Torah but omits to seize his opportunity, and
over the man who cannot study yet continues to do it " (T.B. Hagigah
5 b). In other ways, too, the Rabbis recognised that poverty was an
evil. " Poverty in the house of a man is more distressful than fifty
plagues" (T.B. Baba Bathra 116). The sufferings endured are so
intense that they save a man from seeing Gehinnom ('Erub. 41 b, cf.
Yebamoth 102 b). Poverty is an affliction equal in severity to all the
curses in Deuteronomy combined (Exod. Rabba xxxi.). The contrast
between the earthly lot of rich and poor is found in well-known
passages of the Wisdom literature. Very pregnant is the saying
attributed in the Talmud to Sirach, though the passage is not found in
any known text of the apocryphal book. It runs thus (Sank. 100 b) :
XIV. POVERTY AND WEALTH 115
"All the days of the poor are evil (Prov. xv. 5): Ben Sira said,
the nights also. The lowest roof is his roof, and on the highest hill
is his vineyard. The rain off (other) roofs (falls) on his roof, and the
soil from his vineyard on (other) vineyards" another illustration of
the truth that to him that hath shall be given, and from him that hath
not even his little shall be taken away. Poverty dogs the footsteps of
the poor, putting him at a constant disadvantage (T.B. Baba Qama 92 a).
Poverty even affects the personal appearance. "Beautiful are the
daughters of Israel, but poverty mars their face " (Nedarim 66 a).
But though an evil, poverty was not the consequence of sin, unless
that sin be the misuse of wealth (Leviticus R. xxxiv.). There is a
wheel revolving in the world, and wealth ill-spent ends in poverty
(Exod. Rabba xxxi.; T.B. Sabbath 151 b). But the poor though
deserving of human pity have no right to complain of the Divine justice.
As Philo says : " Poverty by itself claims compassion, in order to
correct its deficiencies, but when it comes to judgment... the judgment
of God is just" (Fragments, Mang. n. 678). In fact the Rabbinic
analysis goes deeper, and makes it necessary for us to qualify the
general statement that Poverty is an evil. "There is no destitution
but poverty of mind" (nynn K^N :y p Nedarim 41 a). Compare
with this the sarcastic allusion to "the poor man who hungers but
knows not whether he is hungry or not" (Megillah 16) this is the
real poverty, the lack of original insight, the absence of self-sufficiency
in character. Poverty, as we have seen, may be so crushing as to
destroy the victim's ideals. Far be it for an arm-chair moralist to
inveigh against those who listen not to a Moses because the iron of
misery has entered into their souls, so that they cannot hear for anguish
of spirit, and for cruel bondage. But the excuse cannot be accepted.
There was none so poor as Hillel, yet he worked for a half-dinar a day
and paid a moiety to the door-keeper for admission to the house of
study, sometimes braving the winter snow. Thus the cares of poverty
are no defence against the charge of neglecting the Torah. And,
continues the same Talrnudic passage (T.B. Yoma 25 b), there was
none so wealthy as R. Eleazar ben Harsom, yet he forsook his wealth,
and with a skin of flour spent his days in the house of study. The
cares of wealth are no defence. Man must rise superior to either. As
the Midrash puts it (Exod. R. xxxi.) : Happy is the man that can
endure his trial, for there is none whom the Holy One trieth not.
The rich God tries whether his hand be open to the poor, the poor He
tries whether he can calmly endure affliction. If the rich man sustain
82
116 XIV. POVERTY AND WEALTH
his trial, and worketh righteousness, lo, he eateth his money in this
world and the capital endureth for the world to come and God
delivereth him from Gehinnom. And if the poor man sustain his trial
and kick not against it, lo ! he receives a double portion in the world
to come. Then the Midrash proceeds to distinguish between the
wealth which doeth evil to its owner and the wealth that doeth good
to him, and so with the qualities of strength and wisdom. Suffering,
indeed, was the lot of rich and poor alike. A life of unbroken pros-
perity was the reverse of a boon. An old baraitha (of the school of
R. Ishmael) asserts that "he who has passed forty days without
adversity has already received his world in this life" ( l Erachin
1 6 b foot) ; one who was not afflicted would not belong to the category
of Israel at all (Hagiga 5 a). Here we read the note of experience.
It was Israel's lot so to suffer that it was forced to fall back on the
theory that only by " chastisements of love " (Berachoth 5 a) might he
obtain purification and atonement (Sit re 73 b). So, too, in another
sense, the difference between men's condition not an absolute differ-
ence, for wealth was accessible to all possessed of knowledge, i.e. virtue
(Sanhedrin 92 a on the basis of Proverbs xxiv. 4), while there was a
ladder in men's affairs up which the poor rise and the rich descend
(Pesiqta ed. Buber 12 a) or a wheel revolving to similar effect (Sabbath
151 b) was a means of atonement when sacrifices ceased (see quota-
tions p. 128 below).
There is no cult of poverty neither is there a cult of wealth. Both
are conditions of good and ill rather than good or ill themselves. Not
the possession of wealth but too absolute a devotion to its acquisition
and too ready a surrender to its temptations were feared. It was the
gold and silver showered on Israel by a bountiful God that provided
the material for the golden calf (Berachoth 32 a). Hillel held that
increase of property meant increase of anxiety (Aboth ii. 7). Yet Rabbi
Judah honoured the rich, and so did Aqiba (T.B. 'Erubin 86 a), for
the rich maintain the order of the world when they turn their
possessions to the service of their fellows : the rich support the poor,
and the poor support the world, says the Talmud (loc. cit.) a not
inept statement of the relations between capital and labour as under-
stood until the inroad of recent economic theories. Equality, whether
in the degree of wealth or poverty, was regarded as destructive of the
virtue of charity. If all men were equal, all rich or all poor, who
would perform the loving kindness of truth of Psalm Ixi. 1
(Tanhuma, Mishpatini ix. inmV p HDK1 1DH ms? VD^iy rCTK Dtf)
XIV. POVERTY AND WEALTH 117
Thus, there must be inequality. This theory, that the poor are
necessary to the rich, runs through the Jewish theory of alms-giving
and charity in all subsequent ages. Wealth becomes an evil when
it is made the instrument of oppression (Aboth de R. Nathan n. xxxi.),
or when the acquisition of it leads to the neglect of the Torah. The
poor are God's people (Exod. R. loc. cit.) and "poverty becomes Israel
as a red halter a white horse" (Hagiga gb) it sets off and augments
the beauty in each case. And it moreover acts as a restraint against
the abuses which luxury may induce. Extreme wealth is hard to bear
(Gittin 70 a), yet charity is its salt (Kethubotk 66 b), and is more
efficacious than any of the sacrifices (Succah 29 b). Yet, if wealth
often leads to a materialistic life, poverty may impel to unworthy
pursuits (Kiddushin 40 a). The wealthy man may win Paradise like
Monobazus, storing up wealth in heaven by generous use of his riches
on earth (T.B. Baba Bathra IT a). The poor man is equally able to
attain bliss. Most of the Rabbis were poor artizans, but some were
rich (Nedarim 50 a seq.). The wealthy among them scorned the idea
that wealth, as such, made up any part of the man's real account
(Pesahim 50 a).
For, " when Solomon built the Temple, he said to the Holy One in
his prayer : Master of the Universe, if a man pray to thee for wealth,
and thou knowest that it would be bad for him, give it not. But if
thou seest that the man would be comely in his wealth (wyi n&o),
grant wealth unto him " (Exodus Rabba xxxi. 5). To sum, again,
poverty and wealth are conditions not ends. Hence the test of wealth
is subjective, not objective. Who is rich] IntheMishnah(.46o^m. 3),
contentment is the definition of wealth. "Who is rich 1 ? he who is
contented with (literally, he who rejoices in) his lot; for it is said,
when thou eatest the labour of thine hands, happy art thou, and it
shall be well with thee (Ps. cxxviii. 2), happy art thou in this world
and it shall be well with thee in the world to come." It may be
difficult but it is not impossible for one and the same person to eat
at the two tables.
XV. THE CHILDREN.
The passages depicting Jesus' love for children are marked by
a singular tenderness and beauty. In several points there is contact
here with the stories of Elijah and Elisha. There is, however, a painful
contrast between the Synoptics (Mark x. 13 16 and parallels) and
the incident of Elisha and the bears (2 Kings ii. 23). But this is a
good illustration of the need to examine the judgment passed by the
Pharisees on certain Old Testament incidents. What did the Pharisees
make of Elisha's conduct 1 ? From the text (2 Kings xiii. 14), "Now
Elisha fell sick of the disease of which he died" the inference was
drawn that the prophet must previously have suffered from diseases of
which he did not die. " The Rabbis have taught (in a baraitha),
Elisha suffered three illnesses, one because he thrust Gehazi off with
both his hands, one because he incited the bears against the children, and
the one of which he died" (T.B. Sota 47 a, Baba Mezia 87 a).
Simplicity of faith, such as characterises the child's confidence in
its parent, is the motive of Psalm cxxxi. "Lord, my heart is not
haughty... Surely I have stilled and quieted my soul like a weaned child
with his mother." The weaned child in the Orient would be old
enough to run alone. Cf. I. Samuel i. 22. In 2 Mace. vii. 27 the mother
of the seven martyrs speaks of suckling her child for three years,
and in the Rabbinic period the average age for weaning was between
the second and third year (cf. Krauss Talnmdische Archaologie ii.
p. 9 and notes p. 436). Young pupils were termed sucklings
(Taanith 9 a). Hence the Psalmist's point of comparison is not the
helplessness of the child, nor its contentment in spite of the loss of
what once seemed indispensable ; but its natural readiness to return to
its mother despite the fact that it no longer needed her. This Psalm
(though the particular metaphor is differently explained) is thus the
model for man's attitude towards God (Midrash on the Psalm quoted).
David made it the guide of his life in all his vicissitudes (ibid.- } cf.
XV. THE CHILDREN 119
T.B. Sota 10 b). Just as only the man could enter the Kingdom who
sought it as a child (Mark x. 15), so he who makes himself small
(perhaps as a child pt3pn) in this world is made great (perhaps " grown
up " ^na) in the world to come, and he who holds himself as a slave
for the Torah here is made free hereafter (Baba Mezia 85 b). In the
Old Testament God's relation to Israel is compared to the relation
between a father and his young child. This relation was much
treasured in the Midrash (see Ycdqut on Jeremiah i. 5 and Hosea xi. 3
and parallels). God's nearness to the child is expressed also by the
thoughts (i) that the young is without sin (N^H DVD DtfD tibw iW p
Yoma 22 b, cf. Niddah 30 b, Low Lebensalter p. 65); and (2) that
the Shechinah is with the young. The whole passage which follows
has several other striking ideas which lead up to the most striking
of all : " Rabbi used to despatch R. Assi and R. Ammi to visit
the towns of Palestine in order to see that local affairs were well
ordered. Once they went to a place and asked to see its Guardians.
They were confronted with the Chiefs of the Soldiery. These,
said the Rabbis, are not the Guardians of the town, they are its
destroyers. Who, then, are the true Guardians? The teachers of
the children.... The nations asked, Can we prevail against Israel?
The answer was given, Not if you hear the voices of the children
babbling over their books in the Synagogues... See how deeply loved of
God the children are. The Sanhedrin was exiled, but the Shechinah
(Divine Presence) did not accompany its members into exile; the
Priests were exiled, but still the Shechinah remained behind. But
when the children were exiled, forth went the Shechinah with them.
For it is written (Lam. i. 5) : Her children are gone into captivity,
and immediately afterwards : And from the daughter of Zion all her
beauty is departed" (Echo, Rabba Introd. and I, 32).
The antiquity of the custom of blessing children by laying on of
hands is attested by Genesis xlviii. 14. The same passage (the very
words of verse 21 are used) was the source of the modern Jewish
custom of blessing the children especially in the home and on the
Sabbath eve. " Before the children can walk, they should be carried
on Sabbaths and holidays to the father and mother to be blessed ;
after they are able to walk they shall go of their own accord with
bowed body and shall incline their heads and receive the blessing."
This is from a book published in 1602 (Moses Henochs' Brautspiegel
ch. xliii.). Similarly the children are taken to the Rabbi, who places
his hand on the head of the children in the Synagogue and blesses
120 XV. THE CHILDREN
them, especially on Friday nights. It is not easy to say how old these
customs are. From Biblical times onwards the teacher regarded his
pupils as his children, and constantly called them so. (For the part
assigned to children in public worship see p. 4 above, and my Jewish
Life in the Middle Ages, pp. 31-2. Very beautifulis the passage in
Sota 30 b, in which is related how the infant on its mother's knee, and
the babe at the breast, no sooner saw the Shechinah at the Red Sea,
than the one raised its head, the other took its lips from the breast
and exclaimed : This is my God arid I will glorify him.) Such customs
as just described do not always find their way into literature (cf.
D. Philipson in Jewish Encyclopedia iii. p. 243), and they are often far
older than their earliest record. They suffice to show how fully in
accord with the Jewish spirit was Jesus' loving regard for the young.
In olden times, the Jewish child began to learn the Pentateuch with
the Book of Leviticus. Why ? Because the sacrifices are pure and
the children are pure. Said B. Assi, " Let the pure come and occupy
themselves with what is pure " (Leviticus Rabba vii.).
XVI. FASTING.
Philo did not represent Pharisaic teaching as to the relation
between body and soul ; he held that they formed a dualism, while
the Rabbinic view was that they constituted a unity. "Righteous-
ness," he says, " and every virtue love the soul, unrighteousness and
every vice the body" (i. 507; cf. Drummond, Philo-Judaeus i. 23).
Pharisaism, on the other hand, placed the seat of good and evil, virtue
and vice, equally in the heart (cf. Porter, op. cit. p. 52 above). But
on the subject of asceticism Philo and the Rabbis were at one. His
theory would naturally lead, on the contemplative side, to such
developments as the societies of the Essenes and Therapeutae, which
belong, just as the medieval and modern Hassidic asceticisms belong,
to Judaism quite as much as do any of its more normal institutions.
Yet, despite his admiration for these societies, Philo steered a sane
course between extremes, and so on the whole did Pharisaism. He,
like them, had no love for excesses in table luxury; he, like them,
thought that enjoyment was possible and laudable without excess.
Philo disapproved of the sumptuous Alexandrian banquets which took
toll of the world to supply rare dainties (i. 81), but, he adds, "Do not
turn to the opposite course and immediately pursue poverty and
abasement, and an austere and solitary life." And, as Drummond
(i. 24) summarises Philo's conclusion (on the basis of the passages
quoted and of i. 549 51), the philosopher counselled: "On the
contrary, show how wealth ought to be used for the benefit of others ;
accept posts of honour and distinction, and take advantage of your
position to share your glory with those who are worthy, to provide
safety for the good, and to improve the bad by admonition; and
instead of fleeing from the banquet-table exhibit there the virtue of
temperance." Cf. F. C. Conybeare, Philo about the Contemplative Life,
1895, p. 270. This became precisely the predominant Jewish view.
Maimonides (Eight Chapters iv., ed. Gorfinkle, pp. 62, 65) concedes
122 XVI. FASTING
that Jewish pietists at various periods deviated into extremes of
asceticism, but he diagnoses their conduct as a medicine against
disease, the medicine being noxious to the healthy. "The perfect
Law which leads to perfection recommends none of these things. It
rather aims at man's following the path of moderation " ; but in order
"that we should keep entirely from the extreme of the inordinate
indulgence of the passions, we should depart from the exact medium,
inclining somewhat towards self-denial, so that there may be firmly
rooted in our souls the disposition for moderation " (cf. Guide iii. 35).
Self-discipline is not self-torture, and man's right and duty to partici-
pate in all lawful happiness is illustrated in such remarks as that
of Abba Areka in the famous Talmudic passage : " On the day of
reckoning man will have to give account for every good which his eyes
beheld and which he did not enjoy" (T.J. Qiddushin, last lines).
In the first century we find, however, an unsettled condition of
opinion. Whether or not it belong to the original source (it is absent
from Mark), yet the outburst in Matt. xi. 18, Luke vii. 33 is an apt
summary of the conflict of views. John was addicted to fasting he
had a devil ! ; Jesus was not so ascetic, therefore he was a glutton and
a wine-bibber ! These' passages suggest also another contrast, that
presented by II. Samuel xii. 21 23, and Mark ii. 19, 20 (incidentally
it may be remarked that the custom of a bridal pair fasting on the
wedding-morn is only imperfectly traceable to a baraitha in T.J.
Bikkurim iii. 65 c).
II. Samuel xii. 2123. Mark ii. 19, 20.
Then said his servants unto him And John's disciples and the Phari-
[David], What thing is this that thou sees were fasting : and they come and
hast done? thou didst fast and weep say unto him, Why do John's disciples
for the child, while it was alive ; but and the disciples of the Pharisees fast,
when the child was dead, thou didst but thy disciples fast not? And Jesus
rise and eat bread. And he said, While said unto them, Can the sons of the
the child was yet alive, I fasted and bride-chamber fast, while the bride-
wept, for I said, Who knoweth whether groom is with them ? as long as they
the Lord will not be gracious unto me, have the bridegroom with them they
that the child may live ? But now he is cannot fast. But the days will come
dead, wherefore should I fast ? can I when the bridegroom shall be taken from
bring him back again ? I shall go to them, and then will they fast in that
him, but he shall not return to me. day.
These passages are interesting from another point of view. They
suggest (in David's saying) the addiction to fasting as a form of
XVI. FASTING 123
supplication, and (in the saying of Jesus) as a form of mourning.
Both of these ideas are abundantly illustrated by the Old and New
Testaments, and also by other evidence available from the beginning
of the Christian era. Thus R. Zadok fasted for forty years to ward
off the destruction of the Temple (T.B. Gittin 56 a). Fasting was
always thought one of the means of causing an alleviation of calamity
(T.J. Ta'anith ii. 65 b top ; cf. Mishnah, Aboth iv. n), but this, as we
shall see, was only admitted by the moralists with the condition that
such fasting be associated with true repentance. In time of drought
and other exceptional natural visitations public fasts were decreed
during Temple times (see Mishnah, Ta'anith passim ; the rule was not,
however, continued in Babylonia, T.B. Pesahim 54 b), just as was done
in the Maccabean age under the stress of political crises (I. Mace,
iii. 47 ; II. Mace. xiii. 12, cf. the Elephantine Papyrus ed. Sachau i. 15,
p. 7). Before starting on his journey from Babylon to Jerusalem,
a journey likely to be attended with danger, Ezra, thinking it un-
becoming to ask for a mounted guard, calls a fast, and this is efficacious
as protection (Ezra viii. 23). Such examples would naturally be long
imitated. When, at the beginning of the fourth century A.D., Zeira
was about to travel also from Babylon to Palestine, he fasted 100 days
(T.B. Baba Mezi'a 85 a. The number is no doubt exaggerated, the
Jerusalem Talmud, Ta'anith 66 a, speaks of Zeira's 300 fasts. Cf.
Bacher, Agada der Paldstinensischen Amorder iii. 6). It is unnecessary
to illustrate the prevalence of fasting as a mourning rite (cf. the fast
decreed on the death of R. Judah, T.B. Kethuboth 104 a) ; David's
action stands out from the normal idea. So, on the opposite side, does
Judith's ; with certain (rather numerous) exceptions, she fasted all the
days of her widowhood (Judith viii. 6. For the medieval Jewish
custom of fasting on the anniversary of a parent's death see Shulhan
Aruch, Yoreh Deah 402, 12, gloss).
Fasting as a penitential rite was, in the Rabbinic view, allied to
sacrifice. But this idea only came to the front after the destruction
of the Temple. The Talmud (T.B. Berachoth 17 a) records that
R. Shesheth (third century A.D.) on fast days was wont to pray:
"Master of the Universe, it is revealed before thee that while the
Temple stood, a man sinned and brought a sacrifice, of which only the
fat and . blood was offered, and this atoned for him ; and now I have
sat fasting and my fat and blood has been diminished. May it be thy
will that it may be accounted unto me as though I had offered it on
the altar, and do thou accept it from me with favour." According to
124 XVI. FASTING
some Mishnaic texts, at an earlier period, while the Temple was in
existence, the delegation (ma'amad] of Israelites who were appointed in
association with the priests officiating in Jerusalem, remained in their
cities and fasted four times a week during their sacrificial term
(Mishnah, Ta'anith iv. 3) ; but this passage is missing in the best
texts (including the Cambridge Mishnah, and the Munich codex, on
which see Rabbinovicz, Variae Lectiones, Ta'anith, p. 160) and cannot
therefore be relied upon. One may perceive a trace of the same idea
in the preference given to fasting over alms-giving as a means of
expiation ; alms-giving is a sacrifice of money, fasting of one's body
(T.B. Berachoth 32 b, top). Yet it must not be forgotten that according
to Mar Zutra the value of fasting lay in the accompanying alms-giving
(Berachoth 6 b). Far older and more continuous than the idea of
fasting as sacrifice is the association of fasting with initiation and the
reception of sacred messages. The Talmud (Sanhedrin 65 b) speaks of
the one who fasts in order that the spirit of purity may rest upon him
(cf. Exodus xxxiv. 28; Deut. ix. 9, 18 ; Daniel ix. 3). In early
Christianity this idea was more fully developed than in the Pharisaic
system, for there is no exact Rabbinic parallel to Acts xiii. 2, xiv. 23.
But from the Apocalypse of Baruch (v. 7, ix. 2) it is clear that in the
latter part of the first century fasting was the " usual preparation for
the reception of supernatural communications" (cf. Daniel ix. 3, and
several instances in IV. Esdras ; see Charles on the Baruch passages).
Jesus fasts for 40 days (Matt. iv. 2) as a preparation to his ministry.
In later centuries Jewish mystics practised fasting in hope of close
communion with God, in the third century already Joshua b. Levi
fasted much whereupon Elijah resumed his interrupted visits (see
refs. in Bacher, Agada der Paldstinensischen Amoraer i. 189). On the
other hand, though fasting might be regarded as a specific for the
preservation of the knowledge of the Torah in a pietist's progeny
(see Baba Mezi'a 85 a), nevertheless religious joy rather than a mood
of sadness was the pre-requisite for the reception of the Shechinah
(T. B. Pesahim 117 a), as also for entering on prayer (Berachoth 3 1 a).
This idea must be set against the assumption that Pharisaic fasting
was conducted in a dismal manner or with a sad countenance (on the
basis of Matt. vi. 17). In the Testament of Joseph (iii. 4), the patriarch
declares : "I fasted in those seven years, and I appeared unto the
Egyptians as one living delicately, for they that fast for God's sake
receive beauty of face" (cf. Daniel i. 15). The Day of Atonement was
a day of joy (Mishnah, Ta'anith iv. 8). How uncharacteristic of
XVI. FASTING 125
Pharisaic piety, moreover, is the public display of fasting, may be seen
from the categorical statement of the Code (Shulhan Aruch, 0. H.
565, 6) : " He who fasts and makes a display of himself to others, to
boast of his fasting, is punished for this." On occasions of public
fasts, naturally the fasting was public, for all the community assembled
at devotions in the public ways (Mishnah, Ta'anith ii. i); it was indeed
an offence for an individual to dissociate himself from the community
on such occasions, perhaps because he was not personally affected by
the calamity which had called forth the general fast (T.B. Ta'anith
ii a). But on private fasts it was the duty of the pietist to avoid
publicity. It is not easy to decide the extent to which private fasts
were developed at the beginning of the Christian era. In later times
they became very frequent ; against bad dreams fasting was declared
by Abba Areka as efficacious as fire is against flax (T.B. Sabbath 1 1 a).
Excessive private fasting was, however, discountenanced in the second
century by Jose ben Halatta, though apparently it was permitted by
the general opinion (Ta'anith 22 b). From a passage in the Psalms of
Solomon iii. 8, 9, it would seem that in the homes of pietists private
fasting was common : " The righteous man maketh inquisition con-
tinually in his own house to the end to put away iniquity ; with his
trespass offering he maketh atonement for that wherein he erreth
unwittingly, and with fasting he afflicteth his soul." But this may
refer to the Day of Atonement. The statement in Luke xviii. 12 has
been held to prove that the Pharisees fasted every Monday and
Thursday, but it is plausible to explain this as exceptional. "The
simplest view seems to be that Luke xviii. 12 (as well as Matthew vi.
1-6, Mark ii. ii, etc.) refers to the exceptional fasts during October
November, when severe pietists fasted on Mondays and Thursdays if
the rain failed. At the close of the period every one was required to
fast, but the Pharisee of Luke puts himself forward as a specially
strict observer of the rite, and such pietists (yehidim) fasted several
Mondays and Thursdays during the drought (T.B. Ta'anith 10 a and b).
Didache viii. i has the same autumn fasts in mind " (Biichler, Journal
of Theological Studies, x. 268. Similarly, the trumpet-blowing before
giving alms, Matthew vi. 2 etc., refers to the public fasts ; the Pharisees
were much opposed to public alms-giving and took various measures to
prevent the identity of the donor becoming' known to the recipient
Baba Bathra 10). The Monday and Thursday fasts became more
regular later on (Ta l anith 12 a), and it is possible that they go back
to the age of Luke. After the destruction of the Temple, private fasts
126 XVI. FASTING
became frequent, though the cases of those who fasted constantly must
have remained exceptional, as their cases are specifically cited (cf.
Hagiga 22 b; Nazir 52 b; Pesahim 68 b). And opinion was much
divided as to the laudability of the habit. Meir held that Adam was
a saint in that he fasted for many years and imposed other austerities
on himself ( l Erubin 18 b), while Mar Samuel declared the constant
faster a sinner (Ta'anith n a, foot). A student (talmid hacham) was
forbidden to fast overmuch as it rendered him physically unfit for
" the work of heaven " (ibid. 1 1 b, top). And even in the bitter
sorrow which followed immediately on the destruction of the Sanctuary
by Titus, Joshua b. Hananiah, a disciple of Johanan ben Zakkai,
opposed excessive asceticism, though actual fasting is not named
(Tosefta, Sotah, end ; T.B. Baba Bathra 60 b). It is also probable
that when Paul (II. Cor. xi. 2) refers to frequent fastings, he was
referring to that kind of self-denial which is so pathetically described
in the Mishnah (Meir vi. 4 quoted above p. 114).
On the most important aspect of fasting the Pharisaic record is
peculiarly clear, though they are habitually assailed on the very
subject. If there is one thing evident from the continuous record of
Judaism, it is the determined effort made by prophet and scribe to
prevent the fast becoming a merely external rite. The fifty-eighth
chapter of Isaiah remains, of course, the most spirited homily en-
forcing the true significance of fasting. But there are several powerful
reinforcements of the prophet's protest.
Ecclus. xxxiv. 25, 26. Tosefta, Ta'anith i. 8.
He that washeth himself after touching If a man keep the object of defile-
a deadbody, and toucheth it again, ment (sheres) in his hand, though he
What profit hath he in his washing ? bathe in the waters of Siloam and in all
the waters on earth he is not clean.
Even so a man fasting for his sins,
And going again, and doing the same ; Mishnah, Yoma viii. 9,
Who will listen to his prayer? He who says I will sin and repent,
And what profit hath he in his humilia- I will sin and repent, he hath no power
tion ? of repentance.
The passage quoted from the Tosefta also occurs in the Jerusalem
Talmud (Ta'anith ii. 65 b) in an interesting context. We have there
recorded a series of actual homilies spoken on fast days. Before citing
some of these, reference must be made to a more familiar instance.
The Mishnah (Ta'anith ii. i) ordains that on a fast after a continued
XVI. FASTING 127
drought, all having assembled with the Ark containing the Penta-
teuchal Scroll in the public thoroughfare, and having sprinkled
themselves (and the Ark) with ashes, the oldest present is to address
the assembly in these terms : " Our brethren : it is not said of the
men of Nineveh that he saw their sackcloth and their fast, but he saw
their acts, that they turned from their evil way (Jonah iii. 10), and in
the prophet (Joel ii. 13) it is said: Rend your heart and not your
garments." In the Jerusalem Talmud (loc. cit.\ besides the homily
referred to above, we have the address of R. Tanhum bar Illai, on the
text (II. Chron. xii. 6, 7) : "Then the princes of Israel and the king
humbled themselves, and they said, The Lord is righteous. And when
the Lord saw that they humbled themselves, the word of the, Lord
came to Shemaiah, saying, They have humbled themselves, I will not
destroy them." On which the Rabbi comments : " It is not written
here they fasted, but they humbled themselves, I will not destroy them"
Of R. Haggai the same passage tells us that he always cited on every
fast day the saying of R. Eliezer : " Three things annul the decree :
prayer, alms-giving and repentance, and all three are derived from the
same text (II. Chron. vii. 14) : 'If my people, which are called by my
name, shall humble themselves, and pray, and seek my face, and turn
from their wicked ways, then will I hear from heaven, and will forgive
their sin, and heal their lnd ' " (seek my face is defined to mean
alms-giving on the basis of Psalm xvii. 15). It is manifestly unjust
to charge with ritualism fasts on which such homilies were a regular
feature.
The main point was that neither fasting nor confessing sufficed
unless with it went a practical amendment of conduct (T.B. Ta'anith
1 6 a). No doubt alms-giving may degenerate into an external and
mechanical rite, but it was sought to so combine it with an inward
sense of sin and a conscientious aspiration towards amendment that
the danger of degeneration was lessened. It was an old theory, and
Tobit (xii. 8) already expresses it : " Good is prayer with fasting and
alms and righteousness." A fine turn was given to the idea when the
alms-giving was not regarded as a direct agent in turning away
the divine disfavour, but as an imitation of the divine nature.
R. Tanhuma (Genesis Rabbah xxxiii. 3) addressed his assembled
brethren on a fast day in these terms : " My children, fill yourselves
with compassion towards one another, and the Holy One blessed
be he will be full of compassion towards you." It must moreover
be remembered that, after the fall of the Temple, Johanan ben
128 XVI. FASTING
Zakkai comforted his mourning disciples with the saying that the loss
of the Sanctuary by removing the sacrifices had not deprived Israel of
the means of atonement.' Charity remained. And the word used by
Johanan for charity is not alms-giving but the bestowal of loving-
kindness (DHDH ni^DJ) and the Rabbi cites the text (Hosea vi. 6) :
I desire loving-kindness and not sacrifice (Aboth de R. Nathan,
ch. iv., ed. Schechter, p. n). It was the same Rabbi who before the
destruction of the Temple had said : " Just as the sin-offering atones
for Israel, so charity (npTV) atones for the Gentiles" (T.B. Baba
Bathra 10 b).
XVII. THE SABBATH.
In no other detail of the differences of the Gospels with the Pharisees
do the latter appear to more advantage than in their attitude towards
the Sabbath. As against his critics Jesus, indeed, sums up his position
in the reasonable epigram : " The Sabbath was made for man, not man
for the Sabbath " (Mark ii. 27), but the Pharisees would have done, nay,
did do, the same. In the higher sense, it is true, this principle cannot
be maintained. The Philonean conception of Sabbath was that of the
divine effortless activity (De Cherub, xxvi., i. 154), and man was most
closely imitating the divine exemplar when he made the approach to
such a state the ideal purpose of his being. So the Rabbis also taught.
The observance of the Sabbath constitutes a man the partner of God
in the creation of the world (T.B. Sabbath 1190); if he keep the
Sabbath man makes it (Mechilta on Exod. xxxi. 16, ed. Friedmann,
p. 104); by hallowing the Sabbath, Israel brings redemption to the
world (T.B. Sabbath 118); and by fulfilling the Sabbatical precepts,
man bears testimony to the divine ordering of the Universe (Mechilta
on Exod. xx. 17, ed. Fr., p. 70 b). In this higher sense then, man
was made for the Sabbath, the destined purpose of his being was the
establishment of harmony with the divine. God kept the Sabbath
before man kept it (Jubilees ii. iSseq.), and man was made that he
might fulfil on earth the custom of heaven.
But in its practical application to ordinary human life, the Gospel
rule is salutary. Life must be fitted to religion, not religion to life ;
but there can be neither religion nor life when the one is allowed to
crush out the other. And this the Rabbis felt. The commandments
were given that man might live by them (DrQ m Levit. xviii. 5), and
this text was the basic ground of the Rabbinic permission of many
acts which, in themselves, and apart from their necessity for the
preservation of human life, were more or less flagrant invasions of the
Sabbatical rest (T.B. Yoma 85 b). The parallel between the view of
A. 9
130 XVII. THE SABBATH
Jesus and that of the Pharisees is, however, still closer. For, as is
well known, a principle almost verbally identical with that of Mark
ii. 27 is found in the name variously of R. Simon b. Menasya (Mechilta
on Exod. xxxi. 13, ed. Fr., p. 103 b) and of R. Jonathan b. Joseph
(T.B. Yoma, loc. cit.). Both these authorities were Tannaim, the latter
belonging to the beginning, the former to the end of the second century.
The variation in assigned authorship suggests that the saying originated
with neither, but was an older tradition. For the principle that
the Sabbath law was in certain emergencies to be disregarded was
universally admitted (T.B. Yoma 85 a), the only dispute was as to the
precise Pentateuchal text by which this laxity might be justified.
Such discussions always point to the fact that a law is older than the
dispute as to its foundation. One Rabbi bases the principle on the
text (Leviticus xviii. 5) already cited ; another in the Talmud, Simon
b. Menasya on the text: "Wherefore the children of Israel shall
keep the Sabbath to observe the Sabbath throughout their generations "
(Exod. xxxi. 1 6), and the Rabbi argued that one may profane a
particular Sabbath to preserve a man for keeping many Sabbaths.
Then follows another suggested justification : "The Sabbath ; holy unto
you " (Exod. xxxi. 14) : unto you is the Sabbath given over, and ye are
not given over to the Sabbath " (Ftitth DniDD DHX W miDD niE> D^).
As I have previously contended (Cambridge Biblical Essays, p. 186),
the wording of the Hebrew saying is noteworthy. Given over is from
masar (= to deliver up). The maxim seems to go back to Mattathias.
War was prohibited on the Sabbath (Jubilees ii. 12) but the father of
the Maccabee, under the stress of practical necessity, established the
principle (i Mace. ii. 39) that self-defence was lawful on the Sabbath
day, for to hold otherwise was to "deliver up" man, life and soul, to
the Sabbath. In the age of Josephus, Jewish soldiers would not
march, bear arms, or forage on the Sabbath (Antiquities xiv. x. 12), just
as at an earlier period they would not continue the pursuit of a
defeated enemy late on a Friday afternoon (2 Mace. viii. 26). But
these acts were not necessary, in a primary sense, and therefore were
avoidable; self-defence fell into a different category, and Josephus
attests (Antiquities xn. vi. 2) that "this rule continues among us to
this day, that if there be necessity, we may fight on the Sabbath days."
The distinction, however, between offensive and defensive warfare was
not without its dangers (see Josephus Antiquities xiv. iv. i ; Wars i.
vii. i ; n. xi. 4.1, and the Judseans suffered from the distinction when
Pompey took advantage of it. Shammai held that though offensive
XVII. THE SABBATH 131
warfare might not be initiated, an offensive already in progress might
be continued (Sabbath 19 a). Thus though Shammai and his school
took a severer view than did Hillel and his followers, the former made
concessions to necessity.
The exact limits within which the early halacha permitted the
infringement of the Sabbath law are not easily denned, for no subject
is more intricate than the history of the principle of the subordination
(fPrn) of Sabbatarian rigidity. It has been maintained (e.g. by
F. Rosen thai in the Breslau Monatsschrift, 1894, pp. 97 seq.) that the
earlier law was the more lenient, and that custom became continuously
more severe. But this is not accurate. It is only necessary to compare
the prescriptions of the Book of Jubilees with the later halacha to see
that there was evolution in lenity as well as in severity. Compare,
for instance, the asceticism in the marital life of Jubilees (xlix. 8) with
the very opposite attitude in T.B. Kethuboth 82 b (and commentaries
on Nedarim viii. 6). The Essenes (Josephus, War n. viii. 9) avoided
other bodily necessities on the Sabbath, but such rigidity was quite
opposed to the Pharisaic view (Sabbatfi 81 a). Or again, the Book of
Jubilees is firm in its refusal to admit the presence of heathens at the
Sabbath meals of Jews : " the Creator of all things blessed it [the
Sabbath], but he did not sanctify all peoples and nations to keep
Sabbath thereon, but Israel alone : them alone he permitted to eat and
drink and to keep Sabbath thereon on the earth" (Jubilees ii. 31).
The later halacha radically modified this attitude, for not only might
meals be provided for heathens on the Sabbath (T.B. Beza 21 b), but
the very compiler of the Mishnah himself gave a banquet in honour of
Antoninus on the Sabbath (Genesis, Rabba xi. 4). It is even open
to question whether the halacha, as developed by Hillel, did not
introduce the important rule which permitted the bringing of the
paschal lamb on a Sabbath (rat? nnn HDD).
It is clear, then, that the later halacha permitted certain relaxations
of the Sabbath law. From this, however, it cannot be inferred that in
the time of Jesus there was such rigidity as would account for his
antagonism to the Pharisees. It may well be that greater severity
prevailed in Galilee and the North than in Judaea and the South (see
some references by J. Mann in the Jewish Review, iv. 526). In that
case, and if the dispute really occurred in Galilee, the controversy
between Jesus and his opponents was local, and has no relevancy to
the Sabbath law as established by the school of Hillel. For it is just
on the points in which the conflict occurred that the Pharisaic law
9-2
132 XVII. THE SABBATH
must already have reached its humane position in the first century at
latest. The controversies between the schools of Hillel and Shammai
are concerned with some details of Sabbath observance, but in no case
do these controversies touch the points raised by Jesus. The estab-
lished general rule was that the Sabbatical regulations might be, nay
must be, waived in order to save life, and this is throughout implied
in the Synoptic incidents. The Rabbinic phrase expressing this
general rule (Yoma 85 a mB> nnn fe?SJ mps) was derived from a
special case, that of removing a person from under a fallen mass of
debris (G?BJ mp2), whence the term came to apply, in general, to all
acts necessary for saving an endangered life (see dictionaries s.v. nipD).
The Mishnah treats the rule as well established even in case of doubt :
"Any case in which there is a possibility that life is in danger thrusts
aside the Sabbath law " (Mishnah, Yoma viii. 6, Tosefta, Sabbath xv. 16).
A generous inclusiveness marked the limits of this bare possibility.
No Sabbatical considerations would have prevented the actual prepara-
tion of food for those in danger of actual starvation. Ears of corn
might not be plucked and ground on the Sabbath under normal
circumstances, but so soon as the element of danger to life entered,
such and any other acts requisite for saving that life became freely
admissible (cf. the collation of the early Rabbinic laws in Maimonides,
Hilchoth Sabbath ch. ii.). "And such things" (says the Baraitha,
T.B. Yoma 84 b and Tosefta, Sabbath xvi. 12, of all active infringements
of the Sabbath law in cases of emergency) "are not done by heathens
but by the great men of Israel" (^NIC" ^113 V 'V) * these breaches
of the law were to be performed personally by the leading upholders of
the law. So, too, in the similar case of the Day of Atonement, the
Mishnah (Yoma viii. 5) allows a sick man to be fed on the fast at his
own desire, in the absence of doctors, or in their presence even if they
thought the patient's need not pressing, but in the case of the presence
of experts, the patient might be fed if they recognised the necessity.
The Talmud (T.B. Yoma 83 a) explains this to mean that whereas the
patient, who himself desired it, was on his own demand to be fed,
whether experts were present or not, he was to be fed, even against
his own inclination, if experts declared him in danger. Thus even
though the ministrations of the doctors involved them in a profanation
of the Sabbath (for the Day of Atonement was also a Sabbath) they
were required to compel the patient to accept those ministrations,
however unwelcome it might be to him.
On the other side the case is different with unnecessary interruption
XVII. THE SABBATH 133
of the Sabbath rest. Normally, food eaten on the Sabbath must be
provided on the Friday. On this rule the older and the later halacha
agree. Jubilees (ii. 29, 1. 9) already lays this down with emphasis :
"they shall not prepare thereon anything to be eaten or drunk." This
restriction might easily be derived from an expanded application of the
Pentateuchal law concerning the manna (Exod. xvi. 23, 25), and from
the direct prohibition against kindling fire on the Sabbath (Exod.
xxxv. 3). It is scarcely doubtful but that the prohibition of preparing
food on the Sabbath, involving as it must a variety of more or less
laborious operations, was essential to any real observance of the day of
rest. Even so, certain work, such as the removal of heavy boxes of
produce, might be performed on the Sabbath to make room for the
reception of wayfarers (Mishnah, Sabbath xviii. i), but whatever could
be done on Friday was to be done on that day (ibid. xix. i, specifically
of the circumcision rite, according to Aqiba). Friday is therefore
called in the Greek sources the day of preparation (irapao-KtvTJ), a title
authenticated by Josephus (Antiquities xvi. vi. 2) as well as by the
Synoptics (Mk xv. 42 ; Mt. xxvii. 62 ; Luke xxiii. 54 ; cf. John
xix. 14 with reference to the Passover). There is no exact Hebrew
or Aramaic term corresponding to this, but later on, at all events, the
technical word n33n (T.B. Besa 2 b) seems to show that irapao-Ktvij
must have been the paraphrase of some such older phrase. At all
events substantially the Greek word represents the fact. An important
element of this preparation was the provision of ample Sabbath meals
for needy wayfarers (Mishnah, Peak viii. 7). Such entertainment was
not to be accepted lightly, and those who refused to avail themselves
of this relief were praised (ibid. 9). On the other hand, one who, if
absolutely destitute, declined the food provided (not on Sabbath only)
was esteemed a self-murderer (D>D1 *1D1K> &*6G). Fasting, moreover,
was forbidden on the Sabbath, this was an old and continuously
observed rule (Jubilees 1. n, Judith viii. 6). It has been ingeniously
suggested (E. G. Hirsch in Jewish Encyclopedia x. 597) that Jesus
practically charges his critics with having neglected charity, in not
providing Sabbath meals for the needy. "Thus he answers their
charge with another. For the act of his disciples there was some
excuse; for their neglect to provide the Sabbath meals there was none."
But this view, arrestive as it is, hardly fits the language of the
Synoptics. The argument turns on the lawfulness or unlawfulness of
certain acts on the Sabbath. It cannot be, on the other hand, that
Jesus alleges that even the Galilean Pharisees would admit no abrogation
134 XVII. THE SABBATH
of the Sabbath law to meet a pressing necessity, for his whole conten-
tion assumes that certain abrogations were permitted. The incidental
question as to travelling on the Sabbath does not arise, for in the
Gospels this aspect is ignored, and we must suppose that the disciples
had not engaged on a long journey, for such a proceeding would
constitute an entire breach with the spirit of the Sabbath rest. If
the disciples were in imminent danger of starvation, then the Pharisees
must have admitted the lawfulness of their act under the pressure of
circumstances. But it is scarcely asserted in the Gospels that the
necessity was so absolute as this. The citation of the precedent of
David does not involve this. Though there are variations in detail in
the accounts of the Synoptics they all agree in the reference to David
(Mark ii. 25, Matt. xii. 3, Luke vi. 3). "When he (David) had need
and was an hungered " says Mark, and the other Gospels say much
the same thing : in i Sam. xxi. it is not specifically said that David's
young men were in a condition of starvation, for the context implies
haste rather than destitution as the ground for using the holy bread.
The Midrash (Yalqut ad loc.), however, clearly asserts that it was a
case of danger to life. (It may be remarked incidentally that the
Midrash supposes the David incident to have occurred on a Sabbath,
and this would make the Synoptic citation of the parallel more pointed.)
All things considered, it would seem that Jesus differed funda-
mentally from the Pharisees in that he asserted a general right to
abrogate the Sabbath law for man's ordinary convenience, while the
Rabbis limited the licence to cases of danger to life. The difference is
shown, too, in the citation of Temple analogies. The Pharisees thought
that \pork permitted in the Temple was to be specially avoided in
general life on the Sabbath (T.B. Sabbath 74 a), but Jesus cites the
Sabbath work of the Temple as a precedent for common use (Matt,
xii. 5). But the real difference lay in the limitation assigned by the
Pharisees, according to whom all labour, not pressing and postponable,
was forbidden on the Sabbath. That this is the true explanation is
confirmed by the cases of healing, and is indeed forcibly suggested in
Luke xiii. 14 : "There are six days in which men ought to work, in them
therefore come and be healed, and not on the day of the Sabbath."
And this argument of the ruler of the synagogue remains unanswered;
it is regrettable that the Synoptics do not in other cases present the
Pharisaic case so precisely. Pharisaism speaks with no uncertain
voice, and it is the voice of moderation and humanity. Every remedy
for saving life or relieving acute pain, such as those of child-birth
XVII. THE SABBATH 135
(Mishnah, Sabbath xviii. 3), the curing of snake-bites (Tosefta, Sabbath
xv. 14), the relief of various pains (T.B. Yoma 84), cooking for the
sick (Tosefta, ibid. 15), these and many other matters are detailed in
various parts of the old halacha (see the collation of these passages in
Maimonides, Hilchoth, Sabbath ch. ii.). It is interesting to note that
John vii. 22 reports Jesus as defending his general position from the
analogy of circumcision. Here we have yet another instance of the
Fourth Gospel's close acquaintance with Hebraic traditions, for the
most notable relaxation of the Sabbath law was just in cases of
circumcision (see Mishnah and Talmud, Sabbath ch. xix.). In Yoma 85 b
the very words of John vii. 23 are paralleled, and the saving of life
derived by an a fortiori argument from the rite of circumcision. Jesus,
however, traverses the Pharisaic position, in that he had no objection
to treat long-standing diseases, lingering maladies, and in general
cases where the treatment could be postponed without fear of dangerous
consequences. Jesus concedes, nay his argument is based on the
assertion that the Pharisees would permit the relief of an animal's
distress on the Sabbath indeed the principle was laid down in various
places (Tosefta, Sabbath xv., T.B. Sabbath I28b KTVm&n DW1 ^D njtt)-
But Jesus went further. No act of mercy, whether the need pressed
or not, was to be intermitted because of the Sabbath. This is an
intelligible position, but the Pharisaic position was as intelligible, and
it was consonant with the whole idea of the Sabbath rest. For there
are many categories of acts, clearly servile, and yet which might be
brought within the definition of the merciful, thus first invading, and
finally destroying, the day set aside for repose and communion with
God. The Pharisees permitted, nay required, the performance of all
necessary works of mercy, but refused to extend the licence too
indiscriminately, and never reconciled themselves to the theory that in
general the performance of a duty justified the infringement of a
prohibition. Whatever may be urged from other points of view
against the Rabbinic treatment of the Sabbath, and much may be so
urged, it is just on the subjects in dispute in the Gospels (cf. Orient
ix. 62) that their withers are entirely un wrung.
XVIII. THE PERSONAL USE OF THE TERM
"MESSIAH."
In the Hebrew Bible there is no indubitable instance of the use of
the term Messiah (Greek xP i(rT ^} as a personal description of the instru-
ment of the future redemption. There are several passages which tend
in that direction, but as Dalman remarks no single passage can be
made responsible for the use of the title. Dalman's discussion of the
whole subject is full, and, in the main, satisfactory (The Words of
Jesus, Edinburgh, Clark, 1902, pp. 268 ff., 289 ff.). The reader may
be referred to Dalman for much careful information on the Rabbinic
uses of the term Messiah. That, as applied to the future salvation,
the term is pre-Christian is shown by the Psalms of Solomon (between
70 and 40 B.C.), where however it has been doubted whether the
reading (xvii. 36) xpi<rro<> Kvpws is right or merely a mistranslation of
POT rV^D It should be mentioned that earlier Jewish critics have
altogether doubted the Jewish provenance of this passage ; Geiger held
that the Greek translator, Graetz (Geschichte der Juden, in. ed. 2,
p. 439) that the author, was a Christian, because of this very
phrase (Ryle and James, Psalms of the Pharisees, Cambridge, 1891,
pp. 141 143, notes). A similar remark applies to the use of the
phrase in Pss. of Solomon xviii. 6 8. But for this suspicion there
seems no sufficient ground, for in the passages cited (especially xvii. 36)
the Messiah is a scion of King David in contradistinction to the
Hasmonean kings. This falls well in line with the developed Pharisaic
tradition in which David becomes almost inseparably associated with
the Messiah. Almost, but not absolutely, for Aqiba recognised Bar
Cochba as Messiah, though there is no claim in the sources that he
was of Davidic descent. It is not possible to regard the non-Davidic
XVIII. PERSONAL USE OF "MESSIAH" 137
origin of the Messiah in any document as, of itself, evidence that the
document is Sadducean.
The simplest view seems to be that when a name was sought for
the king of salvation, the old phrase used of the royal dignitary
mn fWD Ar. "i KITTO " the Anointed of the Lord " was appropriated.
The transference would be helped by the Apocalyptic literature, and
it may be also by the existence of a military official the " Anointed of
War" (nDH^D IWD, Mishnah Sota viii. i). This office was probably
filled by Judas Maccabeus. As regards the mere name, the word
Messiah, with or without the article, is the common appellation in the
Babylonian Talmud for the personal Messiah. Dalman (op. cit. p. 293)
thinks that " the Babylonian custom of using IWD as a proper name
is incapable of being verified in regard to Palestine. It cannot,
therefore, be regarded as old, or as having had a determining influence
in Christian phraseology." This distinction, however, is one hard to
draw. What may be asserted is that the name Messiah does not
become common in Rabbinic usage till after the destruction of the
Temple. Its application to Jesus occurs at the moment when the
name began to be widely used, and the New Testament usage here, as
in many other points, is parallel to Rabbinic development and forms a
link in the chain. After the Bar Cochba war (135 A.D.) the name was
well established.
Assuming then that the older phrase-form was ntfV Pl'S^lD, it remains
to account for the dropping of the word " Lord." In Daniel ix. 25 6
the term is used absolutely, "an anointed one"; and in the Zadokite
Fragment (ed. Schechter) we find "his anointed," and also "an
anointed from Aaron Israel" (p. 20, 1. i). In another place the
text has "anointed of Aaron" (12, 1. i). Dalman (p. 291) urges
that "as the Tetragrammaton was not pronounced, and as there was
a reluctance to name God [a reluctance which Dalman thinks, p. 196,
was shared by Jesus], so here, as in other commonly used titles, the
name of God was omitted and only IWDH Aram. KrrtPD was said."
But though this explanation has cogency, it must be supported by
another consideration which Dalman omits. It rather seems that
it was a Hebrew tendency to omit the qualifying noun in titles,
whether the qualifying noun was the name of God or not. We
have an instance in Sirach. The Hebrew text of ch. xliv. is headed
D^iyn JTQN nap "Praise of the Fathers of the World," whereas
the later Greek translator abbreviates this into Trareptov vftvos, " Praise
of the Fathers." Then later again the term "Fathers'* was used
138 XVIII. PERSONAL USE OF
to mean the older Rabbis without any qualifying noun. We see
the same process in two famous titles which subsequently were much
used. These were Nagid (probably abbreviated from ^K Dy T3J, cf.
*?K Dtf -5> of i Mace. xiv. 28) and Gaon (abbreviated from 3pIT |1N3
Psalm xlvii. 5). At a far older period Nasi (fcOKO) seems an abbre-
viation of a longer expression. It may be noted in passing that the
term Nasi like Messiah was transferred from a political to a spiritual
function, and that at an earlier period than we can definitely trace
the same reference in the case of Messiah.
XIX. GOD'S FORGIVENESS.
Rabbinic Judaism rested its confidence in the divine forgiveness on
God's justice based on his knowledge of human nature, and on his
mercy based on his love. Divine pardon is the logical correlative of
human frailty. "He knoweth our frame" as the Rabbis translated
it " our yeser, our evil propensities " ** he remembereth that we are
dust " (Psalm ciii. 13). Hence, Repentance forestalled sin in the order
of creation ; the means of grace was premundane; the remedy preceded
the disease (Aboth, ii. 4). All moral basis for the world was lacking
until this pillar of Repentance was set firm in place (Pesahim, 54 a;
Pirke R. Miezer, iii. ; Genesis Rabba, i. 4. Cf. Schechter, Rabbinic
Theology j 128, 314). This idea of premundane grace was deftly
supported by the citation of two juxtaposed verses of Psalm xc. :
" Before the mountains were brought forth, or ever thou gavest birth
to the world... thou didst turn man back to dust, saying, Return, ye
children of man." Again, God desires man's reverence, and to this end
he forgives. "There is forgiveness with thee, that thou mayest be
feared." A human tribunal punishes in order to vindicate the majesty
of the Law, but God maintains his reverence by mercy, he as it were
coaxes man to virtue by generously overlooking vice, and by making
the sinner realise that he has not erred beyond the range of pardon.
For the father yearns for the return of his erring children : " Like as a
father pitieth his children, so the Lord pitieth them that fear him "
(Ps. ciii. 13), and as we have just seen, this fear is on its side won by
the mercy which is the response to fear. He has no desire for the death
of the sinner, but would have him return and live (Ezekiel xxxiii. n).
"Neither the national and individual experiences recorded in the Old
Testament, nor the words and general language used, seem to suggest
any fundamental difference in the idea of forgiveness from that which
we find in the New Testament.... Indeed so far as the relation between
the individual and God is concerned, there is nothing to indicate that
140 XIX. GOD'S FORGIVENESS
the forgiveness granted by God in the experience of his people before
the coming of Christ, was different in kind from that which Christ
proclaimed" (Bethune-Baker, Hastings' Dictionary, n. 56). This is
clearly true, unless it be the fact that Jesus claimed the function of
mediatorship between man and God in the matter of forgiveness. The
Old Testament especially in the Psalter assumes that man has direct
access to the Father, and Pharisaism more than accepted it confirmed
and emphasised this assumption. The prophet whether John the
Baptist or another might bring men to forgiveness ; he did not bring
forgiveness to men ; it was not his to bring. The mediatorial idea
suggested by the allegorising interpretation of scripture on the one
hand and by the inroad of angelology and the doctrine of ancestral
virtue with its mediatorial appeal on the other was not altogether
absent from later Rabbinic theology, but on the whole it is true to
assert that the principle was left intact that God and God alone is the
object of worship and the sole and immediate source of forgiveness. A
human potentate is reached through his ministers ; but the presence of
God is attainable without any such interpository etiquette. (This, for
instance, is the moral drawn in Jer. Berachoth, 13 a, from Joel ii. 32 :
Whosoever shall call on the name of the Lord shall be delivered. Cf.
Schechter, op. cit. p. 45).
Important as this aspect of the relation between God and erring
humanity may be theologically, it is not more important practically
than another phase of the problem as to the direct and inalienable
accessibility of the divine mercy. Churches and Creeds do tend to
raise barriers between man and God. They ought to join; they too
often seek to keep asunder. They write their cheerless Quicunque
vult over the threshold of heaven. Israel, on his side, is the peculiar
treasure of God, for whom the rest of the world is of lower concern ;
not entirely so (for see p. 149 below) but to a considerable extent. On
their side, of the rest of the world each group has its own key to the
Presence, and the only route thither is marked on its especial and
exclusive chart. As a corrective to this natural dogmatism, there
recurrently rises an equally natural but a far more gracious humanism.
It cannot be that any quality of human nature can disqualify man
from the father's love ; be that quality inborn or acquired sinfulness or
unbelief. Inhumanity itself cannot rob its unhappy possessor of the
rights of humanity. " The Lord is gracious and merciful ; slow to
anger and of great loving-kindness. The Lord is good to all ; and his
tender mercies are over all his works " (Ps. cxlv. 8, 9). " He dealeth
XIX. GOD'S FORGIVENESS 141
not with us after our sins, nor rewardeth us after our iniquities " (Ps.
ciii. 10). But it is superfluous to multiply texts; the real, sufficing,
ultimate text is inscribed on the tablets of every humane heart. As
Philo says (De opif. mund. 61), "God exerts his providence for the
benefit of the world. For it follows of necessity that the Creator must
always care for that which he has created, just as parents do for their
children."
Two reasons, however, produce some inevitable modifications of this
amiable conception. In the first place, religion is disciplinary. Tt
must, in the interests of morality, somehow take account of consequences
in order to affect antecedents ; it must make forgiveness in some measure
dependent on desert. And, secondly, human nature, because it is
imperfect, tends to find analogues to its own imperfections in the divine
nature. In 1779 Erskine, defending Lieutenant Bourne for challenging
to a duel his commanding officer Admiral Wallace, said : "There are
some injuries which even Christianity does not call upon a man to
forgive or to forget, because God, the author of Christianity, has not
made our natures capable of forgiving or forgetting them." Men go
further, and assimilating God to their own image, assert that there are
injuries which God neither forgives nor forgets. To what extent have
Judaisjn and Christianity followed a similar course in this curious
limitation of God's mercy? "Out of the depths have I cried unto
thee....O Israel, hope in the Lord; for with the Lord there is mercy,
and with him is plenteous redemption. And he shall redeem Israel
from all his iniquities" (Ps. cxxx.). There is no limitation here. Or
again : " But thou hast mercy on all men, because thou hast power to
do all things, and thou overlookest the sins of men to the end that they
may repent. For thou lovest all things that are, and abhorrest none
of the things which thou didst make; for never wouldst thou have
formed anything if thou didst hate it. And how would anything have
endured, except thou hadst willed it ? Or that which was not called
by thee, how would it have been preserved ? But thou sparest all
things, because they are thine, O Sovereign Lord, thou Lover of men's
souls" (Wisdom xi. 23 26). And similar ideas may be readily
enough found also in the Gospels. "Knock, and it shall be opened
unto you... and to him that knocketh it shall be opened" (Matthew
vii. 7). " He maketh his sun to rise on the evil and the good, and
sendeth rain on the just and the unjust" (Matthew v. 45). And
though it be difficult for certain men to enter into the Kingdom of
God, yet such things " are possible with God" (Luke xviii. 27). "I will
142 XIX. GOD'S FORGIVENESS
arise and go to my father," says the Prodigal Son, but " while he was
yet afar off, his father saw him, and was moved with compassion and
ran and fell on his neck and kissed him " (Luke xv. 18 20). In such
passages there is the fullest possible admission of the divine accessibility
to all men. Jesus indeed was animated by a strong, one may even
say a unique, sense of his own relation to and unbroken intercourse
with God. But this sense of nearness is weakened for all other men
when the intercourse with God is broken by the intrusion between
them and God of the person of Jesus. In this respect of the
universality of access the Pharisaic position varied, but it was in the
main, as no doubt the Gospel position was represented by such
thoughts as are enshrined in the following Parable :
A King's son went out into evil courses, and the King sent his guardian
(iraidaywyds) after him. "Keturn, my son," said he. But the son sent him back,
saying to his father: "How can I return, I am ashamed." His father sent again
saying : " My son, art thou indeed ashamed to return? Is it not to thy father that
thou returnest?" pTin HHN "p3K b^N s6. Deut. Rabba ii. 24, in the name of
K. Meir).
The Synoptists, not once or twice but often, dispute the general access
to God. The contrast of sheep and goats, of wheat and tares the
gnashing of teeth and weeping of the iniquitous as they are cat into
the fire while the righteous bask in the sunshine of God of narrow
and broad ways ; the declaration that those who refuse to receive
Jesus or his apostles are in a worse case than the men of Sodom and
Gomorrah ; the invariable intolerance and lack of sympathy when
addressing opponents, and the obvious expectation that they will be
excluded from the Kingdom these things make it hard to accept
current judgments as to the universality of all the Gospel teaching in
reference to the divine forgiveness.
Under the stress partly of dogmatic controversy, partly of psycho-
logical experience, certain sinners were generally declared outside the pale
of pardon. Philo, whose doctrine on the divine relation to man is, on the
whole, so tenderly humane, holds that those who blaspheme against the
Divine, and ascribe to God rather than themselves the origin of their
evil, can obtain no pardon (De prof. 16, Mang. i. 558). This is parallel
to, though less emphatic than, Mark iii. 29: "he that blasphemeth
against the Holy Spirit hath no forgiveness for ever." Similarly, there
are Rabbinic passages in which "the sin of the profanation of the
Name of God" is described as exempt from forgiveness (Aboth de R.
Nathan, 58b). So, too, the man who causes many to sin cannot repent
XIX. GOD'S FORGIVENESS 143
(Aboth, v. 1 8). But the inability was not absolute for, as some texts
of Yoma 87 a read, it is only said to be well-nigh (tDVDD) not entirely
out of the power to repent. And in such cases "death atones"
(Mechilta, 69 a, Yoma, 86 a. Comp. Schechter, Aspects of Rabbinic
Theology, pp. 328 seq.). Yet it is true that certain sinners are
" excluded from a portion in the world to come " (Mishnah, Sanhedrin,
x. [xi.] i), having "denied the root-truths of Judaism " and thus "gone
out of the general body of Israel." (Comp. Maimonides on the Mishnah
cited, and see Jewish Quarterly Review, xix. p. 57).
Such views, however, were theoretical metaphysics rather than
practical religious teaching. In its dogmatic precisions religion may
think of exclusions ; in its humane practice it thinks of inclusions.
"God holds no creature for unworthy, but opens the door to all at
every hour : he who would enter can enter " (Midrash on Ps. cxx).
This is the basic doctrine of all religion, including Pharisaism, and it
is repeated again and again in various terms in Rabbinic literature.
(For references see Montefiore in Jewish Quarterly Review, xvi.
229 seq.)
God owes it, as it were, to his own nature to forgive. " God, the
father of the rational intellect, cares for all who have been endowed
with reason, and takes thought even for those who live a culpable life,
both giving them opportunity for amendment, and at the same time
not transgressing his own merciful nature, which has goodness for its
attendant, and such kindness towards man as is worthy to pervade
the divinely ordered world " (Philo, de prov. Mangey, n. 634). But
this view is not new to Philo; it underlies the whole Biblical and
Rabbinic theory as to Providence (see E. G. Hirsch in Jewish Ency-
clopedia, x. 232 3). In the oldest liturgical prayer the "Eighteen"
Benedictions a prayer in essence pre-Maccabean in date, as all au-
thorities are now practically agreed God is the sustainer of the
whole world in all its natural and human relations, and immediately
after the expression of his omnipotence comes the appeal to him
as the God who " delights in repentance," who " is gracious and doth
abundantly forgive." This last phrase is from Isaiah lv., a chapter
which is a most gracious comparison of God's fertilising energy in
nature to his ever-ready love to the erring human soul. "Let the
wicked forsake his way and the unrighteous man his thoughts; and
let him return unto the Lord, and he will have mercy upon him ; and
to our God, for he will abundantly pardon." And this graciousness
is based on the very greatness of God. " As the heavens are higher
XIX. GODS FORGIVENESS
than the earth, so are my ways higher than your ways," and this
power is correlatively shown in the divine interest in the affairs of
men. No passages in scripture are more often cited in the Rabbinic
literature than this, unless it be Hosea's messages (ch. xiv), Ezekiel's
(ch. xviii.), Isaiah's noble utterances in xliii. 25, xliv. 22, and Daniel's
(ix. 9). "To the Lord our God belong mercies and forgivenesses."
Rabbinic exegesis had no doubt as to the categorical sense of Isaiah i.
1 8, "Though your sins be as scarlet, they shall be as white as enow,"
though the moderns mostly render this sentence interrogatively.
This last fact is of some importance. To render : "7/*your sins be
as scarlet, shall they be white as snow ? " may suit the context in
Isaiah better, but it is doubtful grammar. However, Rabbinic exegesis
does often throw much light on the point we are considering. Forgive-
ness was an inherent attribute of the divine nature, as Philo says and
as the Rabbis also maintain. But the texts on which the Rabbis base
their conclusion as to the divine mercy are statements also of the divine
retribution. In particular is this the case with the greatest text of all,
Exodus xxxiv. 6 7, "The Lord, the Lord, a God full of compassion
and gracious, slow to anger and plenteous in mercy and truth ; keeping
mercy for thousands, forgiving iniquity and transgression and sin :
and that will by no means clear (the guilty) ; visiting the iniquity of
the fathers upon the children, and upon the children's children upon
the third and the fourth generation." The difficulty might have been
met by the application of the principle that Morality and the Law are
the expression, not of God's mercy or any other quality, but simply of
the divine will. This idea is expressed in the passage (T. J. Berachoth
v. 3) which denounces the ascription of such laws as that of the bird's
nest to the mercy of God. It is the divine will that bids man show
kindness to the bird, and not the divine love. This idea of Will did
not, however, find much favour in Rabbinic theology, for it was
directed against Gnosticism and had but a temporary value and vogue.
That, on the contrary, the Law is an expression of Love was deep-
rooted and permanent in that theology. Man's mercy to man was a
reflection of God's mercy to man (see p. 166 below). God is the
" Merciful One," " the Loving One " (8Om), and the very same epithet
is transferred to the Law itself, which is often cited as "The Merciful"
(see dictionaries of Levy and Jastrow s.v.). Hence, the retributive
conclusion of the great pronouncement of God's mercy must be
explained in terms of mercy.
This mercy is sometimes expressed in terms of postponement. This
XIX. GOD'S FORGIVENESS 145
is not to be confounded with limitation. God's power and readiness
to forgive are absolute. But even in his attribute of judge he is not
harsh, and again and again postpones sentence. Philo's words (n. 634)
re-appear in the Midrash (Tanhuma Buber, Numbers i2b): "the
Merciful does not become the Tyrant (nfDN n^Jtt !?niltDn ftf)." But,
on his side, man must not behave as though God's patience is infinite.
God holds over thrice, but he strikes the fourth time ( Yoma 86 a on
basis of Amos ii. 6). This is not literal, and the stress must be laid
on the thrice of the forgiveness, not on the fourth of the punishment.
Yet though there is no limitation to God's forgiveness, there must be
a limit to man's taking advantage of it. (Aboth de It. Nathan xxxix).
How does Pharisaism reconcile the contradiction 1
This was done by calling into play the other note in the harmony
between God and man. If it be the nature of God to offer forgiveness,
it is the nature of man to need and to crave it. God's metier is pardon,
man's part is repentance. But though God's grace is in large measure
conditioned by man's desire for it, by his repentance, nevertheless God
makes repentance easy (Pesiqta xxv. i63b). "He who sins and
regrets his act is at once forgiven" (Hagiga 5 a). God would have men
seek him in reverence, therefore he forgives. He is long-suffering, and
does not requite offence with penalty. He holds it over, giving the
sinner a long respite. He visits the sin of the fathers on the children
if the children carry on the tradition of sin. He is merciful and so
he accepts the repentant. This is the meaning attached to the phrase
quoted above from Exodus. The Rabbis do not translate np^ tib np31
(Exod. xxxiv. 7) "he will by no means clear the guilty," but stop at the
emphatic infinite (np3l) " and he will altogether clear " the repentant,
though (np:^ &6) "he will not clear" the unrepentant, unless amend-
ment follows at least in a subsequent generation (Yoma 86 a).
But this very idea of postponement is practically identical with the
idea that no man is ultimately obdurate. Even the worst type of
sinner "he who makes others sin" is not regarded as in a hopeless
case, even he may come to repent (Yoma 87 a). This thought under-
lies the liturgy. It will be noticed, e.g., that the Confession of Sins on
the Day of Atonement a confession older according to Dr Rendel
Harris than the Didache includes offences of the most varied kind,
including breaches of the Decalogue and also those sins (" profanation of
the name" and so forth) which in the theoretic theology were pro-
nounced unpardonable. Yet after enumerating them the worshipper
adds : " For all these, O God of forgiveness, forgive us, pardon us, grant
10
146 XIX. GOD'S FOKGIVENESS
us remission." Now, as Philo put it, God's mercies are uncircumscribed,
but not so the faculties of the recipients " (Drummond, Philo Judaeus,
ii. 57). Because God is to man as a father, therefore he does not
forgive without discipline. God judges while he pities. u He not only
pities after he has judged, but he judges after he has pitied : for with
him pity is older than judgment " (Quod deus immut. M. i. 284). Yet
he also judges, in relation always to man's finitude. God works on
human nature, a nature which though imperfect is not impotent for
good. In a sense the Jewish doctrine is something like the synergism
of Erasmus, which as his opponent saw was radically opposed to the
Pauline theory of grace. Repentance and confession lead to grace,
says Philo (De excer. 8, 11. 435), and the Rabbis held the same view.
Suppose there is no repentance, is there grace? The Rabbis would
probably have answered that the supposition is a wild one, but that in
any case there is grace. One by one they rescued from the category of
the unforgivable the few individuals whom by name they had relegated
to the category. For God cannot divest himself of his attribute of
mercy. This is the meaning of God's prayer to himself that his grace
may overcome his wrath (Berachoth, 7 a, Moed Qaton, 16 b). This is
the meaning of Aqiba's saying (A both, iii. 20) that "the world is judged
by grace (lllDl), yet all is according to the amount of the work." The
antinomy is the ultimate doctrine of Pharisaism. Man's part in the
divine scheme of mercy must be real. He must turn and live. But the
world is nevertheless judged by grace. This does not mean that man
can or ought to escape the consequences of sin. Man must pay : but
God is a lenient creditor, and he himself provides the coin for the
remission of the debt. Man recognizes, too, that God has the right
to bring man back to himself by any means that he chooses. The
main thing is that man must take his part seriously. Sometimes man
cries to be turned back to God by mild means. There is a very human
note in the prayer of Rabah (Berachoth, 17) : "0 my God, before I
was formed I was nothing worth, and now that I have been formed I
am but as though I had not been formed. Dust am I in my life : how
much more so in my death. Behold I am before thee like a vessel
filled with shame and confusion. O may it be thy will, O Lord my
God and God of my fathers, that I may sin no more, and as to the sins
I have committed, purge them away in thine abounding compassion
though not by means of affliction and sore diseases." As Maimon puts
it in his Letter of Consolation (Jewish Quarterly Review, n. 68) :
" When a child is rebellious against us, we punish him in a gentle way,
XIX. GOD'S FORGIVENESS 147
giving him instruction, inflicting pain upon him, the effect of which,
however, will not be permanent, with a thong which gives pain,
but leaves no trace, and not with a whip which leaves a permanent
mark, or with a rod, which would make a mark for the time, but
cleaves not the flesh, as it is said, " If thou beatest him with a rod,
he shall not die " (Prov. xxiii. 13). This idea readily passes over into
the idea of " chastisements of love " (Ber. 5), which on the one hand
are not the stripes for sin but the stigmata of service, a means of
repentance. On the other hand, the prevalent idea is that God is the
father, who corrects "as a father chastises his son" (Deut. viii. 5), who
demands from his son genuine tokens of contrition and amendment,
but whose love goes out to those who are weakest and least able to
return. Philo on Genesis xxviii. 3 has a striking explanation of Isaac's
selection of Esau for the blessing. He determines, in the first instance,
to bless Esau not because he prefers him to Jacob, but because Esau
is in greater need of the blessing. Jacob can "of himself do things
well," but Esau is " impeded by his own character, and has no hope of
salvation but in the prayer of the father." Thus, the father forgives
just because the son does not deserve it. The Pharisaic position will
never be understood by those who fail to realise that it tried to hold
the balance between man's duty to strive to earn pardon, and his
inability to attain it without God's gracious gift of it. Perhaps the
point may be made clear by contrasting two Rabbinic parables. The
first is from Deut. JKabba ch. iii. :
A King's bride brings two gems as her dowry, and her husband gives her two
other gems. She loses her own gems, and the King takes back his two. When she
again finds her two gems, he restores his two. So Israel brought into the covenant
with God the gems of justice and righteousness (Gen. xviii. 19), inherited from
Abraham. God added two other gems, loving -kindness (Deut. vii. 12) and mercy
(xiii. 18). When Israel lost his gems (Amos vi. 12) God took away his (Jer. xvi. 15).
When Israel again finds his lost gems (Isaiah i. 27), God restores his gift (Isaiah
liv. 10) and the four jewels of justice, righteousness, loving-kindness and mercy
together form a crown for Israel (Hosea ii. 21).
Here we have the idea that God's mercy is a gem the possession of
which is conditioned by Israel's righteousness. It is surely noble
teaching, but it is not the whole truth. The other half is told in
another type of thought. If, in the famous saying of Antigonos of
Socho (Aboth i. 3), Israel must serve without hope of reward, then on
the other side God's gifts must be bestowable by him without condition.
Israel must work without pay; God must pay without work. On
102
148 XIX. GOD'S FORGIVENESS
the text : / will be gracious to whom I will be gracious, and I
will show mercy on whom I will show mercy (Exodus xxxiii. 19) the
Midrash has a remarkable passage. It occurs on this text in the
Rabba (ch. xlv. end), Tanhuma, and Yalqut ( 395). It proclaims that
though righteousness will receive its reward, God's grace extends in
full measure to those who have not deserved it. The idea is the same
in all versions, except that in the Yalqut the point is missed that the
last treasure was the largest. (This point is brought out in the prose
rendering of S. Singer in Lectures and Addresses, London 1908, p. 74.)
The following verse translation was made from the Yalqut (Alice Lucas,
Talmudic Legends, London 1908, p. 10) :
A legend tells, that when th' Almighty Lord
Proclaimed to Moses his eternal word,
He in a vision showed to him likewise
The treasures that lie stored in Paradise.
And at each one in turn the heavenly voice
Spake: "This the treasure is, that shall rejoice
His soul who freely giveth alms, and here
His portion is who dries the orphan's tear."
Thus one by one were all to him made known,
Until unnamed remained but one alone.
Then Moses said: "I pray thee, what is this?"
And answer made the Lord most High: "It is
The treasure of my mercy, freely given
To those who else were treasureless in heaven."
This idea, that the Father gives undeservedly, is strongly brought
out in the Philonean passage oft-alluded to, and now quoted as follows.
(For another version see Eusebius, Prep. Evangel, viii. 14, Mangey
u. 634, Aucher, de Providentia II. 53) :
Quaestiones in Genesim Genesis xxvii. 3
Aiiot? OVTWV vwv, TOV ntv dyaOov, TOV 198. Quippe quod duo sunt filii:
5 virairlov, TOV p.h virairiov ii\oyf]<Tiv unus bonus, alter sub causa (sc. crimine,
<f>r)<riv OVK Trei.8i) TOV o-irovSaLov irpoKpiveL culpa). Istum itaque, qui sub causa
TOVTOV, d\\' on eneivov olde 6V avrov est, benedicere ait, non quod plusquam
Ko.Topdovv 5vi>dfj,evoj>, TOVTOV S rots idiots bonum praeferat hunc, sed quia scit
aXtffKbfjLevov, fjt,tjde/jilav d tx VTa ilium per se solum posse recte rem per-
eXirlda, el (J.TJ ras evxa-s TOV ficere; istum vero ut a suis moribus
irarpos- wv fl /*TJ TVXOI, iravruv dv efy detentum impeditumque, spem salutis
KaKo8aifjiovt<rTaTos. habere in sola patris oratione: quam si
(J. Kendel Harris, Fragments of Philo n o n assequatur, prae omnibus miser erit.
Judaeus, 1886, p. 43.) (Aucher's Latin translation from
Armenian, p. 400.)
XIX. GOD'S FORGIVENESS 149
The Rabbis have another form of the same thought when they
pronounce the penitent sinner superior to the righteous ; the former
has overcome a weakness to which the latter is not susceptible. The
same thought underlies the Rabbinic discrimination between Jew and
Gentile in regard to God. Often there is strong particularism in
favour of Israel (Jewish Quarterly Review, xvi. 249 seq.), and Judaism
did, under the stress of the Roman persecution, regard the obdurately
unrepentant heathen as resting under the divine wrath, much as we
find it in the Apocalypses, and in the particularist passages of the
Synoptics. But the inherent universalism of Rabbinism reveals itself
not only in the beautiful hope for the heathen contained in the liturgy
(in the Alenu prayer), not only in such a saying as that the righteous
of all nations have a share in the world to come a saying which
Maimonides raised to the dignity of a Jewish dogma but the nations
are actually represented as finding repentance easier than Israel finds
it. (On the salvation of the heathen see M. Joseph, Judaism as Creed
and Life, ch. x, ed. 2, 1910, p. 116.) And most striking of all is the
use made of the story of Nineveh. The Book of Jonah is read on the
Day of Atonement, and it was also in earlier times the subject of a
discourse on fast days (Mishnah, Ta'anith, ii. i). Thus the accepted
repentance of a heathen nation was the model for the repentance of
Israel. " The Lord is good unto all ; and his tender mercies are
over all his works." This is a verse in the i45th Psalm which was
introduced thrice daily into the Rabbinic liturgy. Characteristically
enough, too, it was the recitation of this particular Psalm which, it
was held, opened the doors of paradise to men (Berachoth 4 b). And
it is an absolutely universalistic Psalm.
Some other aspects of the questions treated in this note will be
considered further in Note XX on "Man's forgiveness."
XX. MAN'S FORGIVENESS.
The Mishnaic tractate Aboth (Fathers) contains a collection of
maxims by the Tannaim, the teachers of Pharisaism from the century
before till the end of the second century after the beginning of the
Christian era. Among these maxims occurs one which is of unusual
character, for it consists merely of the citation of a passage of Scripture
without addition or comment. The maxim referred to is found in
ch. iv. ( 19 or 26) of the tractate: "Samuel ha-Qatan was wont to
say: Rejoice not when thine enemy falleth, and let not thine heart
be glad when he stumbleth." This citation from Proverbs xxiv. 17 is
remarkable. Samuel belonged to the end of the first century, and was
associated in esteem with Hillel (T. J. Sotah ix. 12) as one "worthy
of the Holy Spirit." The ingrained weakness of human nature, the
desire for revenge against an enemy, is thus pointedly attacked by a
great Pharisee, and in a manner as remarkable for its position as for its
form.
In the Old Testament inculcation of kindliness to man Pharisaism
found a firm basis for its own treatment of the subject. This doctrine
does not consist of a few stray texts; it is of the essence of Old
Testament religion. With regard to the special point before us, the
repression of rancour and vindictiveness, the Hebrew Bible is permeated
with example and admonition. No two nobler instances of forgiveness
are to be found in literature than the records of Joseph's conduct to
his brethren and of David's to Saul, culminating as the latter does
in the Dirge of " magnanimous forgiveness" with which to use
Sir G. A. Smith's phrase the second bo*ok of Samuel opens. And
the admonition finds expression in every part of the Bible. It is
found in the Law not in one but in many precepts. To the eternal
glory of the Old Testament, the great texts "love thy neighbour as
thyself/' "hate not thy brother in thy heart," "avenge not," "bear no
grudge," "love the stranger," are part of the Hebrew law of holiness.
There was little left for religion in subsequent ages except to draw
out the full consequences of these and similar injunctions. Nothing
XX. MAN'S FORGIVENESS 151
that has been added can compare in sheer originality and power to the
first formulation of these great principles. Theology, unhappily, has
been engaged in belittling the Old Testament contribution to the
gracious store, whittling away its words, or at best allowing to them
grudgingly the least that the grammatical words compel ! For
instance, in the note on Leviticus xix. 18, in the Cambridge Bible
for Schools (Leviticus Volume, 1914, p. 109), the editors are painfully
anxious that the young student should not over-rate the text before
him. And he is pointedly warned that the " stranger " of verse 34 is
only the " stranger who worshipped Israel's God." Did Israel, then,
worship Egypt's Gods? Yet the "stranger" is to be loved because
the Israelites " were strangers in the land of Egypt." Must, then, the
same word ger mean two different things within the compass of the
same Hebrew sentence 1 Whatever ger means in other contexts, and
in later ages, it is clear that in Leviticus xix. 34 it has a wide con-
notation. (On the whole question of the Rabbinic law on the stranger
see D. Hoffmann, Der Schulchan-Aruch und die Rabbinen uber das
Verhdltniss der Juden zu Andersgldubigen, Berlin, 1885.)
This, however, is a minor point. All honour to the great teachers
of later times who set themselves to read as much into the law of
brotherly love as they could. But the law is Hebraic. And it is not
the Pentateuch alone which contains it. The prophetical teaching is
saturated with the love of mercy. There is no need to quote.
Zechariah sums up what he regards as the message of the older
prophets : " Execute true judgment, and show mercy and compassion
every man to his brother : and oppress not the widow, nor the father-
less, the stranger, nor the poor; and let none of you imagine evil
against his brother in your heart" (Zech. vii. 9, 10; cf. viii. 16, 17
where there is added "love no false oath," with the glorious conclusion
" for all these things are things that I hate, saith the Lord ").
Similarly with the Wisdom literature of the Old Testament. Job,
Proverbs, Ecclesiasticus, have splendid sayings on the subject of
forgiveness. Again, there is no need to quote more than one passage.
I select this passage, partly for its intrinsic merit, partly for its
position in the Mishnah as already indicated, but mainly because it
became a fundamental principle of Pharisaism.
Eejoice not when thine enemy falleth,
And let not thine heart be glad when he is overthrown:
Lest the Lord see it, and it displease him,
And he turn away his wrath from him. (PBOV. xxiv. 17 18.)
152 xx. MAN'S FORGIVENESS
What does this mean? Ibn Ezra among the older, and Dr Charles
among the newer, commentators interpret the words to mean that
malicious joy defeats its own end. This would not be a low standard,
for many a bitter opponent has been restrained by the knowledge that
to press revenge too relentlessly rouses for the victim a sympathy
which would not otherwise be felt. But the great majority of
interpreters, ancient and modern, read the sentence differently.
C. H. Toy's explanation in the Proverbs volume of the International
Critical Commentary (p. 448) runs thus :
"The turn his anger from him (that is from the enemy) is not to be understood
as affirming that God will cease punishing a wicked man, because another man is
pleased at the punishment; the full force of the expression is 'turn from him to
thee,' and the stress is to be laid on the 'to thee.' 'Thou,' says the sage, 'wilt
then become the greater sinner, and Yahweh will be more concerned to punish thee
than to punish him.'"
The same view is taken in the Kautzsch Bible, where Kamphausen
(p. 808) renders the verses Proverbs xxiv. 17-18 thus: "Wenn dein
Feind fallt, so freue dich nicht, und wenn er hinsinkt, frohlocke nicht
dein Herz, dass nicht Jahwe es sehe und Missfallen empfinde und
seinen Zorn von jenem hinweg [auf dich] wende." On this insertion,
Wildeboer in Marti's Kurzer Hand-Commentar remarks : " Kamp-
hausen rightly inserts the words to thee." In the "Century" Bible
G. C. Martin takes the same view: "from him, i.e. 'lest the Lord turn
His anger from the wicked man to you.'" As will be seen later, the
Pharisaic theory consistently was that the unforgiving injured party
became the sinner through his implacability.
That the moderns are, however, supported by older exegetes is
clear. Thus, the most popular commentary on the Mishnah, that of
Obadiah of Bertinoro, has this remark on the passage already cited
(Aboth iv. 19 [26]): "1QK 1^J7D 3H57H: since it is not written ntn but
H^ni, the meaning is : He will transfer his anger from thine enemy
and will place it upon thee." The commentary on Aboth ascribed to
Rashi interprets similarly. So does Gersonides, and so again does the
popular Hebrew writer David Altschul in his commentary on Proverbs.
Accepting this meaning it is a noble saying, just as in the very same
chapter (Prov. xxiv. 29) is found that other noble verse: "Say not, I
will do so to him as he hath done to me ; I will render to the man
according to his work." This is the highest possible expression of
forgiveness as opposed to retaliation, unless the saying in Prov. xx. 22
be higher still : " Say not thou, I will recompense evil ; wait on the
XX. MAN'S FORGIVENESS 153
Lord and he shall save thee." Here, certainly, there is no reference
at all to revenge ; God does not avenge, he saves.
The doctrine read in, or into, Proverbs xxiv. 18 by most commen-
tators is confirmed by the opening of the great passage on forgiveness
to be found in Ecclesiasticus xxvii., xxviii. The passage is quoted
below ; here we are concerned with two introductory verses :
Wrath and anger, these also are abominations;
And a sinful man shall possess them.
He that taketh vengeance shall find vengeance from the Lord;
And He will surely make firm his sins. (EccLys. xxvii. 30, xxviii. i.)
This seems to mean that God exacts vengeance from the vengeful, just
as Prov. xxix. 18 teaches. At all events, the Pharisaic principle was
just that. The unforgiving man is the sinner (see quotations below).
And following on his elaboration of this principle, Maimonides (Laws
of Repentance ii. 10) adds :
It is prohibited for a man to be hard-hearted and refuse his forgiveness ; but he
shall be "hard to provoke and easy to pacify" (Aboth v. 14). When the sinner
seeks pardon, he must forgive with a perfect heart and a willing mind. Even
though one has oppressed him and sinned against him greatly, he shall not be
vengeful nor bear a grudge. For this is the way of the seed of Israel and those
whose heart is right. But the heathen, of uncircumcised heart, are not so, for they
retain their anger for ever. Therefore does the Scripture say of the Gibeonites
(2 Samuel xxi. 2), in that they pardoned not and proved relentless, " They were
not of the children of Israel."
No doubt it is a good thing for men to see themselves as others see them,
and the Pharisees have enjoyed the privilege without stint ! Is it not
well, too, for others sometimes to see men as they see themselves ? Let
the Pharisees enjoy this privilege too !
It is important to observe the reference made by Maimonides to
the incident of the Gibeonites' revenge. The claim of Maimonides
that forgiveness was a characteristic of Israel is made in the Talmud
also in reference to the Gibeonites (Yebamoth 7 9 a). Often it has
been urged that the presence of vindictive passages in the Psalter
must have weakened the appeal of the finer sentiments in other parts
of the Psalms and of the Scriptures generally. But the argument is
a fallacy. The New Testament teaching is not all on the same level
as the Sermon on the Mount, there are passages which express a vin-
dictive spirit. But Christians rightly treat such passages as negligible
in presence of the nobler sayings, which dominate and colour the
whole. So with the Jew and the Old Testament. He was impelled
154 xx.
invariably to interpret the lower in terms of the higher. The noblest
ideas dominated the rest. Never do we find in the Rabbinic literature
appeal made as precedents to those incidents at which the moral sense
boggled. What was disliked was explained away. " Eye for eye " was
never applied in practical Jewish law. Taken over theoretically from
the Code of Hammurabi, the lex talionis was not acted on in Israel.
No single instance of its application is on record. The unfavourable
reference to the law in Matthew v. 38 no more than the favourable
allusion to it in Philo (u. 329) implies that the law was extant as a
legal practice. The Talmud is emphatic that the retaliation was not
by mutilation of the offender but by the exactment of compensation by
fine. (Baba Qama 84 a, where only one authority argues for a literal
interpretation.) Perhaps the Dositheans were literalists in this respect,
but the phrase "eye for eye," with which so much play is made in
non-Jewish literature, was not familiar on Rabbinic lips. Some
writers do most erroneously confuse "eye for eye" (a principle of
human justice) with " measure for measure " (a theory of divine retri-
bution). The one is a truculent policy, the other a not ungracious
philosophy. The Pharisees who like the Synoptists adopted the theory
of "measure for measure," like them also rejected the principle of
"eye for eye." In fact the very objection to the lex talionis as
literally conceived was used to support the need of traditional inter-
pretation ; the law as written cannot be understood without the
Pharisaic mitigations (see the quotations from Saadiah in Ibn Ezra's
elaborate note on Exodus xxi. 24). Similarly with the imprecatory
Psalms. These could not mean what they seem to say, and why not?
Because they do not consist with the forgiving spirit of other parts
of the Scriptures. Thus Psalm xli. 1 1 reads " Raise me up that I may
requite them." This contradicts the humaner spirit of Psalm xxxv. 13,
vii. 5, and so David must have meant : " Raise me up that I may
requite them good for evil " (see the quotation from Saadiah in Qimhi's
note to Psalm xli.). This may be poor exegesis, but it is rich humanism.
There is another fact to remember. The imprecatory Psalms never
received a personal private interpretation.
Theologically we see the same phenomena. Anthropomorphisms are
brought into harmony with the developed spiritual conception of the
Godhead, by explaining them away, allegorising them. Economists
tell us that base coin drives out the genuine. But in Jewish history
we see the reverse process ; the genuine drives out the base. This
tendency is shown in the Bible itself. Contrast i Chron. xxviii. 29
XX. MAN'S FORGIVENESS 155
with i Kings ii. 1-12, whence it is seen that the author of the Book
of Chronicles entirely omits the passage assailed, thus revealing that
the feeling of the Chronicler was quite as tender and unvindictive as
that of any modern moralist. The example of Joseph so very deeply
impressed Jewish thought, that it is set up as an exemplar for God
himself ! Here is an oft-repeated idea ; it occurs in the Pesiqta
Rabbathi ed. Friedmann, p. 138 a, also in the Pesiqta d. R. Cahana, in
the Canticles Rabba on viii. i, and elsewhere :
Comfort ye, comfort ye my people, saith your God (Isaiah xl. i). This is what
the Scripture hath: that Thou wert as my brother (Cant. viii. i). What kind of
brother? . . . Such a brother as Joseph to his brethren. After all the evils they wrought
unto him Joseph said, Now therefore fear ye not: I will nourish you, and your little
ones. And he comforted them and spake to their heart (Genesis i. 21).... Israel said
unto God : Master of the World, come regard Joseph. After all the evils wrought
by his brothers he comforted them and spake to their heart; and we, on our part,
are conscious that we caused Thy house to be laid waste through our iniquities, we
slew thy prophets, and transgressed all the precepts of the Law, yet, that Thou
wert as a brother unto me ! Then the Lord answered : Verily, I will be unto you as
Joseph. He comforted his people and spake to their heart. So, as for you, Comfort
ye, comfort ye my people. Speak unto the heart of Jerusalem and say unto tier that
her warfare is accomplished, that her iniquity is pardoned.
The ideal traits of the Biblical heroes and saints were set up for
imitation, their faults never.
Like the Book of Proverbs, the Wisdom of Sirach (Ecclesiasticus)
inculcates a lofty ideal on the subject of forgiveness. It is clear that
the teaching is on the same line as that of the Synoptics : as is manifest
from the passages set out in parallel columns :
Forgive thy neighbour the hurt that he When ye stand praying, forgive, if ye
hath done unto thee ; have aught against any one;
So shall thy sins also be forgiven when That your father also which is in heaven
thou prayest. may forgive you your trespasses.
MARK xi. 25; MATT. vi. 14;
LUKE vi. 37.
One man eherisheth hatred against an- Forgive us our debts, as we also have
other, forgiven our debtors.
And doth he seek healing from the Lord? For if ye forgive men their trespasses,
He sheweth no mercy to a man like your heavenly father will also for-
himself, give you.
And doth he make supplication for his But if ye forgive not men their tres-
own sins? passes, neither will your father for-
Being flesh himself he nourisheth wrath : give your trespasses.
Who shall atone for his sins?
ECCLUS. xxviii. 35. MATT. vi. 12, 14, 15.
156 xx. MAN'S FORGIVENESS
Now this teaching of Jesus son of Sirach is absolutely identical with
that of Jesus of Nazareth. Dr Charles, who holds that the Testaments
of the Twelve Patriarchs belong to the second century B.C., cites from
these Testaments a view on forgiveness which he characterises as "no
less noble than that of the New Testament." I will repeat the quota-
tion made by Dr Charles from Test. Gad vi. i.
3. Love ye one another from the heart ; and if a man sin against thee, cast
forth the poison of hate and speak peaceably to him, and in thy soul hold not guile ;
and if he confess and repent, forgive him. 4. But if he deny it, do not get into a
passion with him, lest catching the poison from thee, he take to swearing, and so
thou sin doubly. 6. And though he deny it and yet have a sense of shame when
reproved, give over reproving him. For he who denieth may repent so as not again
to wrong thee: yea he may also honour and be at peace with thee. 7. But if he
be shameless and persist in his wrongdoing, even so forgive him from the heart,
and leave to God the avenging.
Thus the line of connected Jewish teaching is complete : Proverbs,
Sirach, Twelve Patriachs, Synoptics. Other links in the chain could be
indicated. Philo, with much else as elevated, has these sayings (cited
by C. G. Montefiore in his Florilegium Philonis in J. Q. R. vn.
543): "If you ask pardon for your sins, do you also forgive those who
have trespassed against you? For remission is granted for remission"
(Mang. ii. 670). "Pardon is wont to beget repentance" (n. 672 <rvy-
yi/w/x?7 fjierdvoLav TTC^VKC yevvav). "Behave to your servants as you pray
that God may behave to you. For as we hear them, so shall we be
heard ; and as we regard them, so shall we be regarded. Let us show
pity for pity, so that we may receive back like for like" (ibid.).
The teaching of Judaism on the subject of forgiveness is in fact the
brightest and strongest link in its golden chain. The doctrine was
adopted by medieval moralists who insist on it with extraordinary
frequency. And it was introduced into the authoritative Codes. As
Maimonides puts it in his Code (Laws of Repentance n. 9, ic): "The
man who does not pardon a wrong doing to him is the sinner ; it is
prohibited for a man to be vindictive (HTDX, lit. cruel, hard-hearted)
but he must forgive with a perfect heart and an eager soul." This is
the spirit in which the Jew approaches God with his supplication for
mercy on the great Day of Atonement. This is the teaching of
Pharisaism. To attribute any other doctrine to it is unhistorical.
There is no justification for representing as in a moral "backwater"
the humanitarian religion of Hillel, Johanan ben Zakkai, Nehunya ben
Haqana, Meir, and the rest of a long, continuous line of teachers in
XX. MAN'S FORGIVENESS 157
Jewry, who are organically connected with Sirach though they neither
begin nor end with him.
That there are "imprecations" in the Psalter, that the Pharisaic
literature shows some narrowness of sympathy where sectarians are
concerned, and that through its whole course, until the rise of the
liberal movement, Judaism has retained a "particularist" taint, these
facts must neither be ignored nor exaggerated. As to the Psalms, an
admirable treatment of the question may be found in an anonymous
little book (with Introduction by the Rev. Bernard Moultrie) entitled
The Use of the Psalms in the Christian Church with special reference to
the Psalms of Imprecation (St Leonards-on-Sea, 1908). The author
shows how Paul, in warning Christians against revenge (Rom. xii. 19, 20),
uses words borrowed from the Old Testament (Levit. xix. 18; Deut.
xxxii. 35; Prov. xxv. 21, 22). Job in the course of his spirited protest,
which contains the most perfect ideal of virtue ever formulated in
literature, exclaims (xxxi. 29, 30)
If I rejoiced at the destruction of him that hated me,
Or lifted up myself when evil found him ;
(Yea, I suffered not my mouth to sin,
By asking his life with a curse;)
As the author of the volume cited justly asserts (p. 63) : "The opposition
to revenge is so little peculiar to the New Testament, that the strongest
and most numerous passages against it are to be found in the Old."
The author goes on to show that, on the other hand, imprecations are
found in the New Testament. (He cites: Rev. vi. 15 17 ; Matt. xiii.
56, xxiii. 3336, xxiv. 50, 51, xxv. 41; Heb. x. 31, xii. 29 ; 2 Thess.
i. 6 I2 .) But these, like the "imprecations" of the Psalter, are all
' based on the theory : " Do not I loathe them, O Lord, that hate thee :
and am I not grieved with those that rise up against thee?" (Psalm
cxxxix. 21). If this theory be no longer tenable in modern times,
then those few whole Psalms, and single verses in other Psalms, which
are based on like theory, should be expunged from public worship without
casting a stone from the superior virtue heap at the former generations
of Maccabean Zealots or English Puritans who saw in the theory
nothing lowering or dangerous. The Synagogue has no need to
eliminate Psalms lix. and cix. (the chief of the imprecating Psalms)
because they are not used in regular Jewish public worship !
Of these two Psalms only this need be said. Of the imprecations
in Ps. cix. (619) it is almost certain that the "Psalmist quotes
the imprecations of his enemies in his complaint to God against them"
158 xx.
(W. Emery Barnes, Lex in Corde, 1910, p. 176). This view is disputed,
but there is much in its favour. Of Psalm lix. it is equally certain
that the imprecations are not directed against a personal enemy. It
may well be that the objects of animosity are the Samaritans, and
that the Psalm belongs to Nehemiah's age.
Mr Montefiore's lament that Jesus displayed animosity against the
Pharisees has been resented by critics of his volumes. His comment,
it has been said, is due to psychological misunderstanding. If this be
so, ought not the same principle to apply to the Pharisaic animosity
such as it was against sectarians 1 If Jesus might with propriety
assail the Pharisees with threats of dire retribution, the same measure
must be meted out to them, when they are the assailants of those
whom they thought wilfully blind to truth and open rebels against
righteousness. In no age have the sects loved one another over much,
and much as one may sigh at this display, among all creeds, of human
nature red in tooth and claw, it is happily true that the consequences
have not been entirely bad for the world. The prophet is almost
necessarily a denunciator, and the sect must fight if it would maintain
the cause. "The emulation of scholars increases wisdom" (B. Bathra,
2 1 a), and the same principle applies to sectarian differences. The
Pharisees of the age of Jesus were no doubt good fighters against
internal heresies, just as they were good fighters against the common
enemy, Rome. But there was more of this a century before and a
century after Jesus than in his actual age. For it is in fact found on
examination that the Jewish ill-feeling against the "nations" is cor-
related to the ill-feeling of the "nations" against Israel. The Maccabean
spirit of exclusiveness was roused by the Syrian plot against Judaism,
just as the later Pharisaic exclusiveness was roused by the Roman
assault on the religious life of Israel. And the same is true even of
the apocalypses, with their tale of doom. All of them must be placed
in their proper historical background if the picture is to be just. Un-
doubtedly, with the terrible experience of the Great War before our
eyes, with the recollection of much said and written and done burnt
into our minds, our world is better able to judge the past. And it
is not necessary to appeal to our own immediate experience of the hour.
One would not deduce the theory of brotherly love held by Dutch
Christendom from the language of Boers regarding English during the
South African War; one would not entirely gauge the condition of
Elizabethan Anglicanism in relation to the forgiving spirit by its
language or actions regarding Spanish Catholics. Nor would one be
XX. MAN'S FORGIVENESS 159
just to Puritanism if one read a complete theory of its attitude towards
the persecutors of the Church into Milton's fiery sonnet on the massacre
by the Piedmontese :
Avenge, O Lord, thy slaughtered saints, whose bones
Lie scattered on the Alpine mountains cold!
National, sectarian, animosities, even humanitarian indignations against
the cruel and the unrighteous, do indeed stand on a different plane to
personal vindictiveness, and men sometimes do well to be angry.
It is, however, not the case that the Pharisaic liturgy enshrines
any vindictiveness against Christianity. This denial is obviously true
of the first century, but it is also absolutely true of later centuries.
As a Jewish heresy, early Christianity was the subject of antipathy,
as an independent religion it was scarcely assailed at all. Paganism
was another matter ; against idolatry the Synagogue waged war, and
sometimes idolaters came in for their share of the attack, and were,
in moments of stress, regarded as outside the pale of the brotherhood
of man. But even then, it was internal heresy that was more bitterly
resented, and the deliberate sinner, the man of immoral and heretical
life within the fold, was far more the object of recrimination than any
one who stood outside. Here, again, we have a fact of human nature,
not of Pharisaic nature only, and it is a pity that the Pharisees are
made to bear the burden which should be put on the shoulders of man-
kind.
The Rabbinic sayings to the effect that it is^ permissible to " hate "
the wicked within the fold, have no reference to personal wrongs. The
offences which make "hatred" justifiable are invariably breaches of
morality or of the law of God which should not be condoned until the
offender had repented. The personal foe does not come into the
category. The same page of the Talmud (Pesahim 1 1 3 b) which
records the duty to show detestation of the adulterer records also that
beloved of God is he who forgives wrongs personal to himself. "I
believe it to be quite one of the crowning wickednesses of this age
that we have starved and chilled our faculty of indignation " (Ruskin,
Lectures on Art, 1870, p. 83; compare Sir J. Stephen, History of the
Criminal Law of England, 1883, Vol. I. p. 478). In the category of
those who were to be the object of this " indignation," were sometimes
included the heretic and the disloyal (Aboth de ft. Nathan xvi.). But
almost always the offences were indeed detestable (e.g. Ta'anith 7 b).
Beruriah, the wife of R. Meir, in an oft-quoted passage explained
160 xx.
Psalm civ. 35 as a prayer that sin not sinners should be made an end
of (Berachoth 10 a). It is not easy, in this tender fashion, to discriminate
between sin and sinners, but one ought never to lose sight of the
general Pharisaic repugnance against hatred. " Hatred of mankind "
(the term used is the widest possible rwon nN3K>) is one of the three
things (the other two are the "evil eye" and the "evil yeser" envy
and lust) which " put a man out of the world " (Mishnah, Aboth ii. 1 1
[15]). So that we have in a late Midrash the splendid generalisation
that : Whoever hates any man is as one who hates Him who spake
and the world was (Pesiq. Zut. on Numbers viii. seq.). This prohibition
applied to all men, even to Rome (see the strong rebuke in Eccles.
JRabba xi., on the text Deut. xxiii. 8). Even the command to remember
Amalek was explained by one Rabbi to mean : Remember your own
sins which led up to Amalek's assault :
A King owned a vineyard, round which he built a fence. He placed inside the
fence a savage dog. The King said : Should one come and break through the fence,
the dog will bite him. The King's own son came, and broke down the fence. The
dog bit him. Whenever the King wished to mention how his son had offended in
the matter of the vineyard, he said to him: Eemember what the dog did to you!
So, whenever God wishes to recall Israel's sin at Bephidim (Exod. xvii. 8), he says
unto them : Eemember what Amalek did to you ! (Pesiqta K. y iii. 27 a).
In passing, though the fact is of more than passing importance, let
note be taken of the quotation from the Pesiqta Zutarta. To hate man
is to hate God. We have the same thought underlying the preference
shown by Ben Azzai for Genesis ii. 4 as the "greatest commandment"
(cf. p. 20 above). R. Aqiba declared in favour of Leviticus xix. 18
"Love thy neighbour as thyself." But this is open to the objection
that if a man is himself in despicable state, he may despise his
neighbour (n*2n nt3JV TVT3nJ1 Win). Hence, says Ben Azzai, greater
is the text : " These are the generations of the heaven and the earth
when they were created " (Gen. ii. 4). As R. Tanhuma comments :
" If thou showest low regard for any man, remember whom thou art
despising : for the text says : In the image of God made he man."
Another aspect of the sectarian question is apt to be overlooked.
Sects, while their first inspiration is fresh and their numbers small,
have always been distinguished for the strength of brotherly love
within their own body. But when the membership transcends local
bounds, and the initial impulse is materialised into a systematised
organisation, that warmth of complete and unreserved fraternity is
necessarily apt to cool. It is superfluous to show how Christianity
XX. MAN'S FORGIVENESS 161
was compelled by its own success to become less a brotherhood than a
Church. Within Judaism we find at every epoch, from the period
before the Christian era down to the present time, the continuous
formation of new unions, which display intensity of brotherhood while
young and small, and which progress in the normal way towards
greater aloofness as the body grows older and bigger. Religion is kept
fresh by the outbreak of sectarianisms ; this is the great good accruing
from the creation of new sects. For these recurrent outbreaks of
sectarianism are also outbreaks of brotherliness within the new sect,
they are the renewals of the religious stream, the openings up of new
wells of the humane spirit which comes direct from God. And so we
find Hippolytus saying of the Essene : " He will observe righteousness
towards men and do injustice to none : he will not hate anyone who
has done him injustice, but will pray for his enemies" (Refutatio
Omnium Haeresium, ix. 18 28. Cf. Kohler, Jewish Encyclopedia,
v. 239; Josephus, War, n. viii. 6 7). So, passing across many centuries,
we have Luria (the mystic leader of Safed, 1534 1572) opening the
day with the invocation: "Lo! I hold myself ready to fulfil the divine
behest : Thou shalt love thy neighbour as thyself," and at night closing
the day with the declaration: "Lo! I pardon everyone who has angered,
or provoked me, or sinned against me, and I pray that no man what-
soever shall be punished because of me." (Cf. Steinthal, Zu Bibel und
Religionsphilosophie, p. 161.)
And the same sensitiveness is observable at normal periods. There
is, for instance, a whole series of more ancient personal prayers
preserved in the Jerusalem Talmud (Berachoth iv. 2). "May it be
thy will, O Lord my God and God of my fathers, that hatred and envy
of us enter not into the heart of man, nor hatred and envy of any man
enter into our heart." On the same page may be seen the student's
prayer. "May it be thy will, that I be not angered against my
fellows, nor they against me." Yet another prayer occurs in the same
context. " Bring us near to what thou lovest, keep us far from what
thou hatest." These beautiful petitions may be paralleled by that
of Mar Zutra, who every night on retiring to his couch said : "For-
giveness be to all who have troubled me" (py"j JND !?D^> rch H&?
Megillah 28 a).
Turning from the necessary distinction suggested above between
public, national, humane enmities and private, individual, inhuman
vindictiveness, we are arrested by an aspect of the subject which is an
important element in the Pharisaic doctrine of forgiveness. The
A. 11
162 xx. MAN'S FORGIVENESS
injured party must forgive, but what of the man who has done the
wrong ? Pharisaism did not reserve all its sympathy for the inflicter
of the wrong ; it had sympathy, too, with the sufferer of the wrong.
It said to the injurer : You, too, pray for God's mercy, but you must
not go to God red-handed. Before you ask God's forgiveness, seek the
forgiveness of your injured fellow-man. Not even the Day of Atone-
ment atones for wrongs done by man to man (Mishnah Yoma viii. 9).
The man who brought a sin-offering and remembered at the very altar
that he still held the stolen goods, was ordered to stop his sacrifice,
make restitution, and then come back to his sacrifice (Tosefta, Baba
Qama x. 18, p. 368. Of. Cambridge Biblical Essays, 1909, p. 189; see
also Philo de opif. chs. i and iv). And if the sin-offering prescribed in
the Pentateuch was thus of no avail unless practical atonement had
preceded, it is not surprising that we find the same declaration of the
futility of prayer to God unless it had been preceded by an appeal to
the injured neighbour. Undo the injury, beg your neighbour's forgive-
ness, realize the wickedness of wrong-doing, do not throw all the
burden of reconciliation on the person wronged. He must forgive, but
you must try to earn his forgiveness. It is not merely a piece of
French wit : Que messieurs les assassins commencent I The criminal
must not expect all the consideration, he must show some on his part
to the rights of society. The Pharisees softened punishment by their
theory that it was part, the main part of atonement : the prisoner
came out, not crushed by disgrace, but ennobled if chastened by the
sense that he had borne punishment to put himself right with the
outraged moral law. It then became the duty of society to forgive
on its part : to clean the slate, and forget the record. And so with
regard to wrongs which do not fall within the scope of the law at all.
Here, too, the perpetrator of the wrong must bear his share in the hard
labour of atonement. It is almost pathetic to read in Jewish moral
books how the offender must humble himself, must again and yet again
present himself before his offended brother, seeking pardon, refusing to
accept a rebuff. (Of. Yoma 87 b.)
The Synoptics, on the whole, imply the same view. The Gospel
exhortations to forgive take it for granted that, though the response
must be prompt and complete, it is response rather than initiative that
is contemplated. There are thus two elements : (a) approach by the
offender, (b) pardon by the offended. Some theologians who, without
foundation in fact, contrast the Pharisaic doctrine unfavourably with
the Gospel teaching, in their just admiration of (b), which the Pharisees
xx. MAN'S FORGIVENESS 163
fully shared with the Synoptists, ignore (a), which the Synoptists fully
shared with the Pharisees. The Gospel view is most clearly seen in
the effective Parable of Matthew xviii. 23 35. The defaulting debtor
is forgiven the debt after admitting it and praying for patience
(v. 26 7). The debtor then refuses a similar prayer by his debtor
(v. 29). In punishing this act and the Parable of forgiveness a little
loses its grace by making over vindictive the lord's resentment of
unforgiveness the lord says : " Thou wicked servant, I forgave thee
all that debt, because thou besoughtest me" (v. 32). This be it
remembered is the illustration of the injunction *' until seventy times
seven" (v. 22). Clearly the injured is expected to do his part in
seeking pardon from the injured (cf. Hermas, Mandate iv. 8, 9).
And if the injured party be dead ? Then at his grave must pardon
be asked : the living appealing to the dead (Maim. Teshuba, ii. 1 1 ;
Yoma 87 a). This terrible aspect of the case had great weight in
completing the practical Pharisaic mechanism of forgiveness. For
there are wrongs done by us over which we weep in vain. It is not
that our friend will not always forgive; sometimes he cannot. The
injury may have passed beyond him : it may have affected too many :
you may fail to catch up with all its ramifying consequences. Or he
may have died. It is the most heart-breaking experience, especially
in family dissensions. You are hard, you will not bend : then you
relent too late : the other side has hardened : or the other side has
passed from earth, and heart cannot find the way back to heart this
side of the grave.
It was this last consideration that impelled the Rabbis to pour all
the vials of their indignation on the man who increases the inherent
difficulty of reparation by his obduracy when asked to forgive. Such
a one, Maimonides on the basis of the Mishnah (Baba Qama viii. 7 etc.,
Bam. Rabba 19, Berachoth 12, Yalqut Samuel i. 115) pronounces
a sinner and a typical representative of the spirit of cruelty and hard
nature. Here the theory of measure for measure was applied, the
theory which finds so effective an expression in the Lord's prayer
("Forgive us our trespasses as we forgive them that trespass against
us"). "If a man offends his neighbour and says: I have sinned, the
neighbour is called sinner if he does not forgive" (see refs. just cited).
"So long as thou art forgiving to thy fellow there is One to forgive
thee ; but if thou art not pitiful to thy fellow, there is none to have
mercy on thee" (Buber Tanhuma Genesis, p. 104, cf. Sabbath, 151 b).
The Midrash also argues that Job and Abraham received signal
112
164 xx. MAN'S FORGIVENESS
instances of God's beneficence when they had prayed for the pardon of
others, and much more to the same effect. The injured party must
pray for the pardon of his injurer (Tosefta B. Qama ix. 29 ed. Zucker-
mandel, p. 366 top), otherwise, he himself will suffer (Midrash Jonah,
p. 102).
On the other side, of Nehunya ben Haqana it was said (Megilla 28 a)
that the curse of a comrade never went to bed with him. In response
to R. Aqiba he said: "I never stood on my rights" (to exact revenge
or even apology) : and so Rava said (loc. cit., and Rosh Hashana 17 a,
Yoma 23 a) : He who forgives (vnHD bjJ TiyDil) received forgiveness,
for in the Scripture (Micah vii. 18) the words "pardoneth iniquity"
are followed by the words "passeth by transgression, i.e. God pardoneth
the man who passes over wrongs." There is no self -righteousness
here, no aggravating sense of superior virtue. Those, of whom the
Rabbis speak, who humbled their spirit and heard their "reproach in
silence," were, the same sentence (Pesiqta Rabbathi, p. 159 a) continues,
also those who "attributed no virtues to themselves." To bear reproach
and answer no word was an oft-praised virtue (Sabbath 88 b, Sanh.
48 b 49 a). This noblest of all applications of the principle of measure
for measure which goes back to Psalm xviii. 25, 26 is found again and
again in the Rabbinic writings. It is not "incidental" to them, it is
permeative. " R. Judah says in the name of Rabban Gamliel : See,
the Scripture saith, And He will show thee mercy and have compassion
on thee and multiply thee ; this token shall be in thy hand, Whilst
thou art merciful, the Merciful will have mercy on thee" (Tosefta Baba
Qama ix. 30).
The principle of "measure for measure" (see Matthew vi. 14 15)
supplies the most efficient motive for forgiveness, but passing beyond
that, the Rabbis make the duty of forgiveness absolute. The un-
forgiving man was the denier of God (Yalqut on Judges viii. 24);
many private Rabbinic prayers breathe the most thorough feeling for
a state of mutual good-will between men (e.g. Berachoth in both the
Talmuds on iv. 2). In the future world there is to be no enmity
(Berachoth 1 7 a), which is the Rabbinic mode of setting up the same
ideal to be striven for on earth. The acme of the saintly disposition
is slowness to be enraged and quickness to be reconciled (Aboth v. n).
And although we do not find in the Rabbinic literature a parallel
to the striking paradox Love your enetnies, we do find the fine saying
(already quoted by Schottgen): "Who is mightiest of the mighty?
He who makes his enemy his friend" (iiniK 1KJ15? nt?W ^D, Aboth
XX. MAN'S FORGIVENESS 165
de B. Nathan, xxiii). This ancient saying received more than lip-
homage. Samuel ibn Nagrela was made Yizir of Habus, the Berber
king of Granada in 1027. Near the palace of Habus, says Graetz
(History of the Jews, E.T. in. viii.), there lived a Mussulman seller
of spices, who no sooner beheld the Jewish minister in the company
of the king, than he overwhelmed him with curses and reproaches.
Habus, indignant at such conduct, commanded Samuel to punish
this fanatic by cutting out his tongue. The Jewish Vizir however
knew how to silence him who cursed. He treated him generously,
and by his benefactions converted the curses into blessings. When
Habus again noticed the seller of spices, he was astonished at the
change, and questioned Samuel about it. The minister replied, "I
have torn out his angry tongue, and given him instead a kind one."
So, to return to the older period, the greatest crown of all was that
won by Moses when he entreated God, not on his own behalf, but to
forgive sinful Israel (Yalqut on Ps. xc. i).
A Prayer of Moses (Psalm xc. i). To what is the matter like? To three men,
who came to seek the royal amnesty. The first came and made obeisance. " What
seekest thou?" "Amnesty for my rebellion." His petition was granted. So with
the second. Then came the third. "What seekest thou ?"" For myself, nothing.
But such and such a Province is laid waste, and it is thine; command that it be
rebuilt." Said the King, "That is a great crown it is thine." So David and
Habakkuk pray on their own behalf (Psalm xvii. i, Habakkuk iii. i). When Moses
came, God asked him: "What seekest thou?" "Forgive the iniquity of this
people " (Numbers xiv. 19). God answered : " This is a great crown it is thine, in
that I change my will because of thee " (Yalqut, Psalms, 841).
Very fine too is the following expression given to the desire to convert
enemies into friends by the exhibition towards them of love :
If thine enemy be hungry, give him bread to eat,
And if he be thirsty, give him water to drink;
For thou shalt heap coals of fire upon his head,
And the Lord shall reward thee. (PEOV. xxv. 21 22.)
E. Hama b. Hanina said : Even though he has risen up early to slay thee, and
he come hungry and thirsty to thy house, give him food and drink. Why? Because
thou heapest coals of fire on his head and the Lord will make him at peace with thee
(v. 22). Bead not yeshalem, will repay, but yashlimenu, will make him at peace
with thee (Midrash ad loc. Cf. T. B. Megillah 15 b. See the passages as quoted
in Yalqut ha-Machiri, Proverbs, p. 58 b).
As another Babbi could claim, at the close of a long life (Megillah 28 a),
"I never went to bed with the curse of my fellow" (rMp nn!?y K 1 ?
hy nan).
166 xx. MAN'S FORGIVENESS
By a natural, and assuredly not dishonourable, stretch of moral
chauvinism, this very quality of forgiveness, which is so rashly denied
to the Pharisees, was by them treated as a special characteristic of
Israel (cf. p. 153 above). "He who is merciful towards all men (nin^n)
thereby shows himself of the seed of Abraham" (Besa 32 b. In all such
passages the context shows that pm merciful, used indeed in the
widest sense, is particularly employed in the meaning forgiving).
Carrying to the extreme the maxim " Be of the persecuted not of the
persecutors " (Baba Qama 93 b and elsewhere), the Rabbis even said
"He who is not persecuted does not belong to Israel" (Hagiga 5 a).
"Three gifts the Holy One bestowed on Israel : he made them for-
giving, chaste, and charitable" (Bam. Rabba viii., Yebamoth 7 9 a). Or
to sum up : " Ever shall a man bestow loving-kindness, even on one
who does evil unto him ; he shall not be vengeful nor bear a grudge.
This is the way of Israel" (Midrash le'olam, ch. vii.).
And why 1 Because Israel is the child of God, and must strive to
be like his Father. The great foundation of the forgiving spirit is not
to be sought in the principle of measure for measure. Its basis is the
Imitatio Dei, an idea which is very old, very frequent in the Pharisaic
literature, and included by Maimonides as one of the precepts of the
Pentateuch (Affirmative laws 8). Portia, in her sublime praise of
the quality of mercy, says:
But mercy is above this sceptred sway,
It is enthroned in the hearts of Kings,
It is an attribute to God himself;
And earthly power doth then show likest God's,
When mercy seasons Justice.
Her rebuke, cast at the Jew, almost reads like a quotation from the
Jew's own books. "As God is merciful and gracious, so be thou
merciful and gracious," is the Pharisaic commentary on "Ye shall be
holy, for I the Lord am holy" (Sifra 86b, and many other passages,
Mechilta 3 7 a, Sabbath i33b, and often. See Schechter, Some Aspects
of Rabbinic Theology, ch. xiii.). " The profession of the Holy One,
blessed be he, is charity and lovingkindness, and Abraham, who will
command his children and his household after him 'that they shall
keep the way of the Lord ' (Gen. xviii. 1 9), is told by God : ' Thou
hast chosen my profession, wherefore thou shalt also become like unto
me, an ancient of days ' " (Genesis Rabba, Iviii. 9, Schechter, p. 202).
Or to cite but one other passage (Sota, 14^), "Rabbi Hamab. R. Hanina
said : What means the Biblical command : Walk ye after the Lord
XX. MAN'S FORGIVENESS 167
your God '? (Deut. xiii. 4). Is it possible for a man to walk after the
Shechinah ? Is it not previously said : The Lord thy God is a con-
suming fire? (Deut. iv. 24). But the meaning is: to walk after the
attributes of the Holy One. As he clothed the naked Adam and
Eve in the Garden (Genesis iii.) so do thou clothe the naked ; as
the Holy One visited the sick (appearing unto Abraham when he
was ailing, Genesis xviii.), so do thou tend the sick ; as the Holy One
comforted the mourners (consoling Isaac after the demise of his
father, Genesis xxv.) so do thou comfort the mourners; as the Holy
One buried the dead (interring Moses in the valley, Deut. xxxiv.),
so do thou bury the dead. Observe the profundity, the ingenuity of
this Rabbinic exegesis : from first to last, from Adam's days in the
beginning to Moses' death in the end, from Genesis to Deuteronomy,
the law, according to the Rabbi, bids the Israelite Imitate God" (cf.
Jewish Addresses, 1904, pp. 41 51). Most frequently this Imitatio
Dei interprets itself as an admonition to mercy. God imparts of his
attribute of mercy to men that they may be merciful like himself
(Gen. R. xxxiii.). That the connection of the law of holiness with
the Imitatio Dei goes back to the beginning of the Christian era is
shown from Philo's saying : " Holiness consists in imitating the deeds
of God," just as "earthly virtue is an imitation and representation of
the heavenly virtue" (/xt'/u^/xa is used several times in this context), a
"warder-off of the diseases of the soul" (De alleg. legum, i. 14, Mangey,
I. 52). For God is the supreme archetype (see Drummond, Philo-
Judceus ii. 81), and as all virtue is a reflection of his moral nature, so
man becomes moral when he strives to liken his character to the
heavenly exemplar. So the "rewards of the virtuous, which fill the
soul with a transcendent joy" are, with Philo, the attainment to some
share in the nature of God (Drummond, n. 323). This extension of
the idea is Pharisaic as well as Philonean (Pesiqta R. xi. end). On
earth man is an appanage of God, cleaving to him in the desire to
imitate. But hereafter man becomes self -existent in his resemblance
to God (Dni DW Dn).
XXI. THE LIFE OF THE RESURRECTION.
The question as to the exact physical conditions of life after death
has often divided Jewish opinion. Maimonides (Hilch. Teshubah
viii. 2) unreservedly asserts: "In the world to come there are no
bodies, but only the souls of the righteous, without bodies like angels."
This view Maimonides based on Talmudic authority ; but some of his
critics protested against it and quoted such Rabbinic sayings as clearly
inculcate the view that at the resurrection the dead arose with the
same physical defects as in life (T.B. Sanhedrin 91 b), though these
were forthwith healed, that the dead arose clothed (Kethuboth 114 a).
So, in the Apocalypse of Baruch 1. 2 : " the earth will then assuredly
restore the dead, making no change in their form," though (li.) the
aspect of the resurrected saints would thereafter be transformed.
On the other hand, Maimonides rested his statement on the saying
(Berachoth 17 a): "In the world to come there is neither eating nor
drinking, no marital relations, no business affairs, no envy, hatred nor
quarrelling ; but the righteous sit with their garlands on their heads,
enjoying the splendid light of the Divine Presence (Shechinah) as it is
said : And they beheld God and they ate and drank (Exodus xxiv. n)."
This saying (parallel to Mark xii. 25) is cited by the Talmud in the
name of Abba Arika (Rab), who died in 247 A.D. But the main
ideas involved in his sentence are all much older, and are not incon-
sistent with the belief in the bodily resurrection. In the first century
the schools both of Hillel and Shammai believed in the restoration of
the material form (Genesis Rabba xiv., ed. Theodor, p. 129; Leviticus
R. xiv.). But it is certain that this bodily resurrection was only
regarded as one stage in the process of attaining to immortality, and
much ingenuity has been exercised (as by Nahmanides in Shaar
hagemul) in reconciling with one another the various Rabbinic state-
ments (including the famous parable of the lame and the blind, on
which see p .98 above).
XXI. THE LIFE OF THE EESURRECTION 169
The main metaphors in Rab's picture of the future life are (i) the
banquet, (2) the light, (3) the crown. "The righteous sit with garlands
on their heads, enjoying, etc." is a figure obviously derived from the
banquet (Low, Gesammelte Schriften iii. 417). It is a familiar figure
which the evidence shows goes back in Rabbinic literature to the first
century. In a famous passage of the Mishnah, Aqiba (Aboth iii. 16,
last words) speaks of the future life as a banquet, which is prepared for
all, wicked as well as righteous, for the sinner is to enjoy it when he has
paid the penalty for his evil life (this universal interpretation is clearly
derivable from the context, and the Bertinoro rightly so interprets).
Aqiba held that the judgment on the wicked in Gehinnom lasted only
twelve months (Mishnah, Eduyott ii. 10). The same figure is carried
out in the Mishnaic saying of R. Jacob (Aboth iv. 16). He compares
the earthly life to the TrpoOvpov (vestibule or outer door) and the future
world to the rptfcAivov (dining hall. The force of R. Jacob's comparison
is well brought out by L. Low, Gesammelte Schriften i. 127; cf . also
iii. 417). An amplification of the figure, belonging, however, to an
earlier date, is seen in the parable of the wise and foolish guests (T.B.
Sabbath 153 a. This parable is ascribed to Johanan b. Zakkai by the
Talmud, and to Judah the Patriarch in Eccles. Rabbah on ix. 8. The
former ascription is adopted by Bacher, Agada der Tannaiten, ed. 2,
vol. i. p. 36, and it is the more probable seeing that R. Meir knew it.
The figure is frequently found in the Rabbinic literature of later
centuries; cf. T.B. Pesahim H9b; Baba Bathra 74 b, where the
Leviathan appears as the main dish at the banquet).
Even older is the idea of the heavenly light which the righteous
were to enjoy. In Daniel xii. 2 the resurrected saints are to shine as
the brightness of the firmament, and in the Ethiopic Enoch (cviii. 12)
they are to be clad in raiments of light. The term light played a great
part in Jewish mystical terminology. The angels fed on the shining
light of the Shechinah (Numbers Rabba xxi 16) and the mystics made
much play with the thought. The figure of the crown is also an old
conception. Thus in Wisdom (v. 15 seq.) the righteous live for ever,
and they shall receive the royal robe (/?o<rt\eiov) and the diadem of
beauty (8ia8?//xa rov /caAAovs) from the Lord's hand. It is not clear
from the context whether this crowning of the righteous is regarded
as part of the protection on earth or whether it is a feature of the
life hereafter, but the two ideas lie near together. The crown may
imply the notion of victory, or possibly the exact thought is of
freedom. The phrase "with their crowns on their heads" occurs in
115
170 XXI. THE LIFE OF THE RESURRECTION
the Sifra (Behar Perek ii.) ed. Weiss, p. 106 d) in a context which
leads Weiss to make this suggestion (foot of the page cited) : the
freed slaves "ate and drank and rejoiced with their crowns on their
heads " between the first and tenth of Tishri in the Jubilee year. In
his book, The Immanence of God in Rabbinical Literature (p. 88),
Dr Abelson has a fine passage in which he summarises the view of
Nahmanides. In Exodus xvi. 25 the text says of the Manna : "to-day
ye shall not find it in the field," on which the Mechilta remarks : " Ye
shall not find it in this life, but ye shall find it in the life to come."
Dr Abelson thus reproduces Nahmanides' comment : " The worthy
Israelite will find his manna, i.e. his source of continued vitality, even
after death ; he will find it in that blessed union with the Shechinah
for which he has qualified himself in ascending stages of spiritual
saintliness. He will wear the crown upon his head. Does not the
prophet predict that ' in that day the Lord of Hosts shall be a Crown
of glory ' (Isaiah xxviii. 5) 1 There will be a complete merging of the
human life with the divine life."
INDEX I
Of Names and Subjects
Aaron as type 60, 70; in relation to
Messiah 137
Abba Areka 105, 114, 121, 125, 168
Abbahu 6r, 92
Abbott, E. A. on Bath-Qol 48 ; on Dove
5o
Abelson, J. on Shechinah 170
Ablution. See Baptism
Abodah 2
Abraham 166
Abraham b. Hisdai 93 .
Absolom, Rab'bi 102
Abudarham on baftara 5
Adam 39, 69, 126
Adeney, W. on John the Baptist 32
Adultery 70, 73
Agada 13, 91, 94, 99, 105
Agatharcides 3
Agrat no
Agrippa 81
Ahiqar 92
Alexander, Rabbi 52
Alexander Zebinas 83
Alexandra 68
Alexandria 3, 121
Allen, W- on leaven of the Pharisees 53
Almsgiving 113, 124, 127
Altschul, D. 152
Anialek 160
Am-haares 56
Ammi 108, 119
Amoraim, exegesis of 14
Amram, D. on Divorce 72
Angelology no, 140, 169
Antigonos of Socho 147
Antiochus Sidetes 83
Antoninus 131
Apocalypses, Jewish 16, 80, 94, 99, 124,
168
Aqiba 19, 29, 33, 35, 72, 74, 91, 116,
133' !36 !46 1 60, 164
Aquila 33
Asceticism 40, 114, 121
Assi 119, 120
Atonement 40, 124, 132, 145, 149, 156,
162
Augustine 22
Authority 13
Baba b. Buta 87
Bacher, W. 6, 14, 19, 50, 64, 105, 169
Banquet (Messianic) 169
Baptism ch. iii. and iv.
Bar Cochba 14, 62, 114, 136-7
Barlaam and Josaphat 93
Barnes, W. E. 31, 158
Bath Qol 47
Baya 54
Ben Azzai 20, 24, 77, 160
Ben Nanas 28
Ben Zoma 28, 49, 50
Benedictions 42, 55. See also Eighteen
Benedictions
Bernays, J. on Golden Rule 21
Bertinoro, 0. of 152, 169
Ber uriah 60, 159
Bethune-Baker, J. 140
Blasphemy 142
Blau, L. 4,. no
Blindness 109
Bread 51
Bousset, W. 42, 90, 91, in, 112
Buchler, A. 7, 9, 15, 54, 57, 63, 79,
100, 104
Buddha 94
Burkitt, F. C. 12, 31, 38, 92
Caesar, Give unto ch. viii.
Gaius 64
Capital punishment 73
Chajes, H. P. on Parables 13, 94
Charles, B. H. 124, 152, 156
Chastity 57, 74
Cheyne, T. K. 33, 34
Children ch. xv., 4, 77, 145
Circumcision 37
Coins 64, 83
Colani 80
Commandment, Greatest ch. ii., 160
Confession 145, 162
Conybeare, F. C. in, 121
Cosmology 48, 50
172
INDEX I
Creed 24
Crown (Messianic) 169
Cures, miraculous no
Dalman on Messiah 136, 137
Darmesteter, J. on Persian influence in
Palestine no
David 23, 117, 122, 123, 134, 136, 150,
i54 165
Decalogue 28
Dembitz, L. N. 81
Demetrius II 83
Demonology 95, no
Didache 21, 28, 40, 125
Disease and Sin 108 seq.
Divorce ch. ix., 15
Doctors ch. xiii., 132
Dove ch. v., 36
Dreams 125
Drought 61, 123
Drummond on Philo 121, 167
Edersheim 36, 85, 87
Eighteen Benedictions 9, 143
Elbogen, I. on the haftara 5
Elephantine papyrus on divorce 123
Eleazar b. Azariah 114
Eleazar b. Harsom 115
Eleazar b. Hyrqanos 56
Eleazar b. Jacob 37
Elijah 61, 124
Elisha 118
Elisha b. Abuya 56, 60, 92
Elmslie, W. 56
Enoch no
Epidemics 108 seq.
Epiphanius 93, 97
Eschatology 34, ch. xxi.
Essenes 3, 16, 29, 30, 32, 34, 43, 55,
66, in, 113, 121, 131, 147, 161
Eucharist 55
Eve 39, 69
Example, effects of 51, 55
Exorcism in
Externalism 126, 127
Eye for Eye 154
Ezra 10, 123
Fasts ch. xvi., 6, 40
Fatherhood of God 41, 119, 139, 140,
143, 146, 148
Festivals 3, 10, n, 12, 40, 51, 133
Fiebig on Parables 91, 97, 112
Fire, baptism by 44
Forgiveness chs. xix., xx., 60
Fourth Gospel 12, 135
Friday 133
Friedlander, G. on Prodigal Son 92
Friedlander, M. on Synagogue i, on
Pharisees 80
Gaon 138
Galilee 12, 15, 81, 131
Gamaliel 72, 164
Geiger, A. in, 136
Gehinnom 74, 114
Gerlach on Josephus 31
Gersonides 132
Gibbon and the Golden Eule 21
Gibeonites 153
Ginzberg, L. on Prodigal Son 92
Gnosis 50
God, see Fatherhood. Name of 45 ;
justice of 139, 140 ; mercy of ch. xix.,
144 ; as Friend 167 ; as Healer m
Golden Calf 108
Golden Eule 21 seq.
Gould, E. P. 18, 81, 85
Grace 146
Graetz, H. 23, 62, 136, 165
Giidemann, M. on Golden Rule 24, 27
Haber 54
Habus 165
Hadrian 63, 74
Haftara ch. i.
Haggai, Eabbi 127
Hama b. Haninah 165, 166
Hamburger, J. 21, 73
Hanan, house of 87
Hananiah 64
Haninah b. Dosa no
Harnack 31
Harvest 99 seq.
Hassidim 121
Hatred 159 seq.
Healing, miraculous ch. xiii. ; on Sab-
bath ch. xvii.
Heathen 56, 128, 140, 151
Helena of Adiabene 74
Hellenism 74
Henochs, Moses 119
Herod 55, 87
Herod Antipas 12 seq., 30
Herodias 31, 66
Herford, E. T. on Pharisees 88
Herzfeld 54
Hillel 14, 21, 28, 41, 70, 71, 87, 91,
94 95. 99. IIO > 1J 5 "6, 131, 132,
156, 168
Hippolytus 161
Hirsch, E. G. on Sabbath 133
Hiyya 52
Hoffmann, D. on heathen 151
Holiness, Law of 150
Holy Spirit 43, 49, 106, 142
Holzmann on the Dove 48
INDEX I
173
Humility 164
Hunkin, J. W. on Parables 91
Hyrcanus son of Tobias 83
Ibn Ezra 154
Image of God 18, 20
Images 65
Imitation of God 17, 127, 129, 166, 167
Indian Parables 93
Insanity 75
Inspiration 14
Ishmael, Rabbi 16, 116
Isocrates and Golden Rule 21
Israel 48, 166
Jacob, Rabbi 26, 76, 169
James, M. on ' Blind and Lame ' 93, 97
Jerome 79
Jerusalem i seq. 13, 37, 56, 64, 81 seq.,
ch. xi.
Jesus, in the Synagogue 4 seq. 13 ; and
the Kingdom 51 ; publicans and sin-
ners 58 ; on divorce 67 seq. ; and
Temple 84 ; healing 1 1 1 ; attitude to
children 117; and asceticism 122,
124; and Sabbath 129 seq. ; and for-
giveness 142, 158, 162
Johanan b. Zakkai 10, 55, 60, 62, 64,
95> 99 "6. 127, 156, 169
John the Baptist ch. iii., 13, 32, 35, 140
Jonah 149
Jonathan, Rabbi 108, 130
Jordan, River 33, 39, 106
Jose b. Halafta 76, 125
Joseph, as type 155
Joseph, M. on salvation of heathens 149
Joseph of Nazareth 70, 72
Josephus on Sects 16 ; John the Baptist
30; "testimony to Christ" 31; on
Images 64 ; on Herodian women 66 ;
on divorce 67, 73 ; on Pharisees 79 ;
on Temple 82, 83 ; on Onias no; on
exorcism in; on Essenes 113, 131,
161 ; on warfare on the Sabbath 130;
on Friday as Preparation 133
Joshua b. Hananiah 3, 50, 56, 95, 126
Joshua b. Levi 55, 61, 91, 103, 124
Jubilees, Book of 131, 133
Judah Halevi 102
Judah, Rabbi 25, 53, 69, 116, 123, 164
Judas Maccabeus 137
Judith 123
Jupiter 90
Justin Martyr i, 101
Juvenal 66
Kamphausen, A. 152
Kethubah 68
King, E. on the Triennial Cycle 10
Kingdom of God 51, 55, 63, 68, 99, 142
Klein, K. on the Fourth Gospel 12;
on Golden Rule 29
Kohler, G. on Didascalia 29 ; on John
the Baptist 31; on demonology in ;
on Essenes 161
Krauss 9, 31, 43, 84, 85, 117
Kuenen on the Great Synod 9
Lake, K. on Baptism 42
Lauterbach, J. Z. on Parables 98
Law, the 2, 23, 24, 67, 114
LeavenTch. vi.
Leprosy 41, 108, 109
Lessing and the Three Kings 93
Leszynsky, R. on the Sadducees 16
Letter and spirit 88
Leviathan 169
Lex talionis 154
Light and heavy precepts 26
Lightfoot 85, 169
Liturgy of Synagogue 9 seq., iir, 127,
Love, chastisements of 147 ; love of
enemies 160, 164
Low, L. 50, 169
Lucas, A. 148
Luria, Isaac 161
Lydda 27
Maamad 124
Maccabees i, 117, 123, 130, 157, 158
Machaerus 30
Magic no
Maimonides 24, 58, 59, 85, 121, 132,
143, 146, 149, 153. 156. l6 3> I<5 7> l68
Mann, J. 19, 27, 131, 170
Manna 11, 133, 168
Marriage 68 seq. 76
Martin, G. C. 152
Mary 72, 73
Mashal 105-6. See Parable
Master of the House 100
Mattathias 130
Meals 35 seq.
Measure for measure 108, 109, 154, 104
Medicine 112
Meir, Rabbi 60, 85, 86, 91, 95, 99, 108,
126, 142, 156, 159, 169
Mercy chs. xix. and xx.
Messiah ch. xviii., 42, 48, 49, 55> 61,
92, 107
Milton on Piedmontese 159
Minim 92
Miriam n
Monday, fasting on 125
Money-changers 82 seq.
Monobazos 116
Monogamy 73
174
INDEX I
Montefiore, C. G. 42, 72, 114, 143, 156,
158
Moses 23, 45, in, 115, 165
Musaph 3
Naber on John the Baptist 32
Nahman b. Isaac 23, 105
Nahmanides 168, 170
Nahum of Gimzu 109
Name of God 45, 142
Nash papyrus 28
Nasi 138
Nazareth 12
Nehuuya b. Haqana 156, 164
Neighbour, love of chs. ii. and xx.
Nineveh 127, 149
Nitai 60
Noachide laws 27
Noah 4 8
Nomikos 19
Obadiah of Bertinoro 152, 169
Olive 48
Onias and prayers for rain no
Ordeal 71
Paris, M. Gaston 97
Parables ch. xii.
Parables in New Testament
Prodigal Son n, 92, 142
The Vineyard 26
Good Samaritan 27
Leaven 51-3
Building on Eock 92
Sower 93
Banquet 97
Eoyal Parables 99
Talents 101
Lord and debtor 163
Parables, Kabbinic
Father and Son 26
Labourer and hire 26
Builder 92
Prodigal Son 92
Blind and Lame 93, 97
Floating Skull 95
Soul as Guest 96
King's Statue 96
Eoyal Parables 99
Harvest 100
God and Israel 102
Chastisements of Love 103
The blessed tree 103
Earth as Palace 104
Parables on the Parable 105
King and erring son 142
Israel as Bride 147
Treasures in Heaven 148
Park and dog 160
King and his Petitioners 161
Wise and foolish Guests 169
Parthian s 83
Passover 37, 51, 133
Paul, discourse in Synagogue 4, 8; on
leaven 51 ; on fasts 126; on forgive-
ness 157
Peace 55
Pentekaka 61
Perles, F. on demonology in
Perseus 51
Persian influence in Palestine no
Pesiqta 6
Petronius 64
Pharisees and reading of the Law 5 ;
prayer and study 9 ; in Galilee 1 2 ;
Pharisaic type of character 24 ; limits
of conformity to Eome 27 ; John the
Baptist 32; baptism, ch. iv. ; leaven
of 53 ; meals 55 ; divorce 66 seq. ;
moral discipline 78 ; Pharisaism and
Puritanism 82 ; angelology no; joy-
ousness n 4 ; treatment of children
117; fasting ch. xvi. ; externalism
125; Sabbath ch. xvii. ; God and
man 1 46 ; forgiveness chs. xix. and xx.
Philipson, D. on blessing the children
1 20
Philo on Greek Synagogues i, 3, 4 ;
Prodigal Son 1 1 ; allegorical discourse
14; sects 16; Golden Eule 21; light
and heavy precepts 25 ; leaven 51 ;
legation to Eome 65 ; divorce 67 ;
tower of Babel 92 ; absence of Parables
in 104; Therapeutae 113; Poverty
115; body and soul 121; Sabbath
129; heathen 141; blasphemy 142;
God as Father 143, 148; mercy 146;
Jacob and Esau 147 ; Eye for eye 154 ;
forgiveness 156; imitation of God
167
Phineas b. Jair 101
Plummer, A. on the preaching of Jesus
13
Pompey 130
Poor 81, ch. xiv.
Porter, on Yeser 52, 121
Portia 1 66
Prayer 9, 57, 61, 64, 104, 123, 127, 161,
164
Priests 3, 81, 82, 87
Prophets, readings from. See haftara
Proselytes 28, 36, 41
Providence 143
Psalms, imprecatory 157
Psalms of Solomon, 41, 79, 125, 136
Publicans and Sinners ch. vii.
Puritanism 157
INDEX I
175
Qimhi 154
Qolbon 84
Bab. See Abba Arika
Babah 146
Bain 61, no, 123
Bapoport on sermons 4
Bashi 49, J52
Beinach, T. on coins 64, 83
Bepentance 34, 39, 41, 42, 49, 58, 123,
127, 129, 142, 145, 149
Besurrection 98, 167
Betribution 144 seq.
Bevenues of Temple 82
Biches. See Wealth
Bighteousness and Grace 147
Bitualism 127
Bogers, C. F. on baptism 38
Borne 54, 62, 63, 64, 74
Bosenthal, F. 131
Boyal Parables 99
Buskin on Anger 159
Saadiah 154
Sabbath eh. xvii., i, 12
Sacrifices 3, 38, 44, 87, 88, 124, 128
Sadducees 5, 32, 80
Salome 66
Samuel of Nehardea 62, 126
Samuel ha-Qatan 150
Samuel ibn Nagrela 165
Sanhedrin 9, 13, 71 seq., 80, no
Saul 150
Schechter, S. on " I say unto you " 16 ;
Holy Spirit 43; dove 50; yeser 55;
Kingdom 63 ; on disease and sin 109 ;
repentance 139 ; imitation of God 106
Schools 119
Schottgen 45, 81
Schiirer 4, 30, 36, 54
Schwab, M. on money-changers 86
Schweitzer 99
Sepphoris 79
Sermons 4 seq. 105
Shammai 9, 15, 71, 87, 95, 131, 132
Shechinah 36, 49, 119, 120, 124, 167,
168, 169
Shekels 84
Shema 9, 28
Shesheth, Babbi 123
Shila, Babbi 105
Sibylline Oracles 39, 45
Sick, prayers for in
Simeon b. Gamaliel 87
Simeon b. Menasya 130
Simeon of Shizur 86
Simeon b. Shetah 68
Simeon b. Yohai 99, in
Simlai, Babbits
Sin 41, 42, 58, 59, 108, 119, 142
Singer, S. on the heavenly treasure 148
Sinners. See Publicans
Sirach on doctors 109; on poverty 115 ;
on forgiveness 153 seq.
Slander 108
Slaves 38 seq., 42, 45, 117
Smith, G. A. i, 9, 13, 83, 150
Solomon and Parables 105
Solomon Ibn Verga 93
Staff, traveller's 113
Steinschneider, M. 93
Steinthal 161
Stranger, laws regarding 151
Suffering 109
Sutta Nihata 93
Synagogue i seq., 9, 82, 84, 93, in, 149
Tabernacles u, 82
Table talk 55
Tacitus 63
Tahna, Babbi no
Taxes and Taxgatherers 54 seq., 63, 86
Taylor, C. 15, 18, 20, 28, 57, 100, 101
Tanhuma 127
Temple 2, 64, 89, 95, 123, 126, 134,
ch. xi.
Testaments of Twelve Patriarchs 28,
44, 60, 124, 156
Thackeray, H. St J. 31
Therapeutae 3, 113, 121
Thursday fasts 125
Tiberias 65
Titus 126
Tobit 21, 69, 127
Tongues, gift of 10
Torah. See Law
Toy, C. on influence of Synagogue i ;
on forgiveness 152
Two Ways 13
Tyrian coinage 83
Venetian er, L. on haftara 5
Vespasian 64, in
Vindictiveness 161
Voice. See Bath Qol
Water 40, 43
Water-drawing, Ceremony of 3, 82
War and the Sabbath 130
Wealth 68, ch. xiv.
Weapons 113
Weber, on Yeser 52
Weiss, I. on Hillel 95
Wellhausen on the Greatest Command-
ment 1 8
Widows ch. x.
Wiener on dietary laws 56
176 INDEX I
Wildeboer on forgiveness 152 Zadok, Rabbi 123
Will of God 144 Zadokites 38, 66, 137
Wisdom Literature 74 Zeira, Rabbi 55, 60, 123
Women 38, 67 seq. Ziegler on Parables 99
Wrath 149 Zipser on harvest 100
Wiinsche, A. on Authority 52 Zohar, the 51
Zunz, L. on Jewish sermons 4
Yeser 42, 52, 57, 69, 160, 161 Zutra, Mar 124
INDEX II
Of New Testament Passages
MATT]
EEW
XXI. 12
82 seq.
XXII. 2
i. 19
7<>
xxii. 15
84
1. 20
iii. 7
73
32
45
xxii. 21
xxii. 40
xxiii. 23
62 seq.
18, 24
IV. 2
V. 19, 20
124
2 5
xxm. 33
xxiv. 50
157
157
v. fs
v. 43-4 ~.
v :45
vi. 1-6
\l 4
"5
xxv. 14-30
xxv. 41
xxvi. 51
xxvii. 62
r 57
113
133
VI. 12
vi. 14, 15
155, 163
164
MARK
vi. 17
124
i. 1-8
34
vi. 34
IOI
i. 4, 5
3
vii. 3
IOI
i. 7
48
vu 7
141
1. 21
16
vn- 24-7
9*
1. 22
13 94
vii. 29
13, 94
11. II .
I2 5
IX. 14
ii. 18
32
ix. 35
13
n. 19-20
122
ix. 37-8
IOO, IOI
n. 25
134
X. 10
113
n. 27
129, I 3
XI. I
13
iii. 1-6
12
XL 13
35
111. 22
13. 16
xi. 18
122
in. 29
142
Xll. I
II
iv. 3
93
xu. 3. 5
134
IV. II -.
106
xiii. 1-9
93
iv. 29
99
30U. 30
99
v .i-4
12
xiii. 33
vi. 6
13
xiii. 56
157
vi. 7
"3
XV. I
13
vi. 39
ii
XVI. 12
53
vii. i
13
xviii. 3
118
vn. 4
33
xvm. 17
xviii. 23-5
::::::::::::::,
vin. 14-21
viii. 27
53
13
xix. 3
xix. 10
:::::::::::::$
X. I
X. 12
6*
XIX. 21
80, ioj
x. 13-16
118, 119
xx. 1-15
IOO
xi. 16-17
84, 85
XX. 10
26
xi. 25
155
xxi. 9
103
xii. 17
62 seq.
178
INDEX II
xii. 29 28
xii. 32 18
xii. 38 80
xiv. 40-43 80, 81
xiv. 47 113
xv. 42 133
73
.109
LUKE
111. 7
iii. 17 ,
iv. 17
iv. 32
v- 33
vi. 3....
vi. 37
vii. 30.
38
45
4, 7
'3,
3?
'34
155
32
94
vii. 33 ................................. 122
viii. i ..... ............................ 13
viii. 4-8 .............................. 93
ix. 3 ................................. 113
x. 2 .................................... 100
x. 26, 27 ........................... 19
x. 30 ................................. no
xi. 4 ......................... . ....... 168
x ". i ................................. 53
xiii. H .............................. 134
xiii. 22 .............................. 13
xiv. 16 .............................. 97
xv. ii ................................. 92
xv. 18-20 ........................... 142
xvi. 16 .............................. 35
xviii. ii .............................. 57
xviii. 12 .............................. 125
xviii. 27 .............................. 141
xx. 25 ................................. 62 seq.
xxii. 38 .............................. 113
xxiii. 54 .............................. 133
JOHN
ii. 13 .......................... ...... 82seq.
iv. 36 ................................. ioo
vi. 4 ................................. ii
vii. 23 ................................. 135
vii. 37 ................................. ii
viii. 4 ..........................
ix. i .........................
xviii. 20 .............................. 12
xi*. 4 ................................. 133
xix. 45 .............................. 82 seq.
ACTS
xiii. 2 ................................. 124
xiii. 15 .............................. 4, 8
xiv. 23 .............................. 124
xviii. 25 .............................. 32
x. 3 ................................. 3
xxiii. 8 .............................. no
EOMANS
xii. 19-20 ........................... 157
xiii. 10 ... ........................ 21
1 CORINTHIANS
v. 6 .................................... 51
xv. 33 ................................. 51
2 COEINTHIANS
xi. 2 ................................. 126
GALATIANS
v- 9 .................................... 5i
2 THESSALONIANS
i. 6-12 .............................. 157
i TIMOTHY
ii. 1-2 ................................. (
HEBKEWS
x. 31
xn. 29
'57
157
REVELATION
15, i7 157
CAMBRIDGE: PBINTED BY j. B. PEACE, M.A., AT THE UNIVERSITY PRESS
SELECTION FROM THE GENERAL CATALOGUE
OF BOOKS PUBLISHED BY
THE CAMBRIDGE UNIVERSITY PRESS
The Gospels as Historical Documents. By VINCENT HENRY
ST ANTON, D.D., Fellow of Trinity College, Ely Professor of Divinity.
Demy 8vo. Part I : The early use of the Gospels. 7s 6d net. Part II :
The Synoptic Gospels. 10s net. To be completed in four parts.
The Fourfold Gospel. By EDWIN A. ABBOTT, F.B.A. Demy
8vo. Diatessarica, Part X.
Section I. Introduction. 2s 6d net.
Section II. The Beginning. 12s 6d net.
Section III. The Proclamation of the New Kingdom. 12s 6d net.
Section IV. The Law of the New Kingdom. 12s 6d net.
The Gospel History of our Lord and Saviour Jesus Christ,
in a Connected Narrative in the words of the Revised Version. Arranged
by the Rev. C. C. JAMES, M.A. Third edition. Crown 8vo. 3s 6d net.
A Harmony of the Gospels in the words of the Revised
Version, with copious references, tables, etc. Arranged by the Rev.
C. C. JAMES, M.A. Third impression. Crown 8vo. 5s net.
The Witness of Hermas to the Four Gospels. By
C. TAYLOR, D.D. Fcap 4to. 7s 6d net.
The Parables of the Gospels in the Light of Modern
Criticism. Hulsean Prize Essay, 1912. By L. E. BROWNE, M.A,
Lecturer at St Augustine's College, Canterbury. Crown 8vo. 2s 6d net.
How Luke was written. (Considerations affecting the Two-
Document Theory with special reference to the Phenomena of Order
in the Non-Marcan Matter common to Matthew and Luke.) By
E. W. LUMMIS, M.A., B.D. Crown 8vo. 4s 6d net.
The Origin of the Prologue to St John's Gospel. By
RENDEL HARRIS. Demy 8vo. 4s net.
The Fourth Gospel and some recent German Criticism.
By HENRY LATIMER JACKSON, D.D. Crown 8vo. 3s 6d net.
Disarrangements in the Fourth Gospel. By F. WARBURTON
LEWIS, B.A. Crown 8vo. 2s net.
The Date of St Paul's Epistle to the Galatians. 'By
DOUGLASS ROUND, M.A. Crown 8vo. 2s net.
Destination, Date, and Authorship of the Epistle to the
Hebrews. By H. H. B. AYLES, B.D. Crown 8vo. 5s net.
A Discussion of the General Epistle of St James. By
R. ST JOHN PARRY, D.D. Demy 8vo. 5s net.
Studies in the Second Epistle of St Peter. By E. ILIFF ROB-
SON, B.D., Sixth Form Master of Felsted School. Crown 8vo. 2s 6d net.
[P. T. O.
SELECTION FROM THE GENERAL CATALOGUE
"The Son of Man/' or Contributions to the Study of
Thoughts of Jesus. By EDWIN A. ABBOTT, F.B.A., Hon. Fellow i
St John's College, Cambridge. Demy 8vo. 16s 6d net.
Light on the Gospel from an Ancient Poet. By EDWIN
ABBOTT. Demy 8vo. 12s 6d net.
Notes on the Translation of the New Testament, being
Otium Norvicense (Pars Tertia). By FREDERICK FIELD, M.A., LI
Reprinted with additions by the Author. Demy 8vo. 5s net.
The New Testament in the original Greek, according to
text followed in the Authorised Version, together with the Vari
adopted in the Revised Version. Edited by F. H. A. SCRIVENER, M.J
D.C.L., LL.D. Crown 8vo. Cloth. 3s 6d uet. Printed on India
5s net. Fcap 8vo. Cloth. 3s net. Printed on India paper, 4s 6d
(Both editions may also be had in leather bindings.)
The New Testament, Greek and English, being
Authorised Version arranged in parallel columns with the Revi
Version and with the original Greek, according to the text followed
the Authorised Version with the variations adopted in the Revised V
sion. Edited by F. H. A. SCRIVENER, M.A., D.C.L., LL.D. Demy
Cloth, 9s net. Leather, Persian grained, 21s net. Turkey Morocco,
net. The Revised Version is the Joint Property of the Universities of Ca e
bridge and Oxford.
An Atlas of Textual Criticism. Being an attempt to sh'
the mutual relationship of the authorities for the text of the New Tet
ment up to about 1000 A.D. By EDWARD ARDRON HDTTON, M.A., Vic
of St Michael's, Hargrave. Crown 8vo. 5s net.
On the Origin of the Ferrar-Group, a Lecture on the gen
logical relations of the New Testament MSS., delivered at M
College, Oxford, Nov. 6, 1893, by J. RENDEL HARRIS, M.A. Demy 8
Is 6d net.
Further Researches into the History of the Ferrar-Gn
By J. RENDEL HARRIS, M.A. Demy 4to. 10s 6d net.
The Bible of To-day. By the Rev. ALBAN BLAKISTON, M.
Large crown 8vo. 3s net.
A Companion to Biblical Studies. Being a Revised and
written Edition of The Cambridge Companion to the Bible. Edited
W. EMERY BARNES, D.D., Fellow of Peterhouse, Hulsean Professor
Divinity. With 8 illustrations and 10 maps. Royal 8vo. 15s net. ]
The Cambridge Bible for Schools and Colleges. Gem
Editors : A. F. KIRKPATRICK, D.D., and R. ST JOHN PARRY,
With introductions, notes, and maps. Extra fcap 8vo. A comply
of the volumes with prices will be sent on application.
Cambridge University Press
Fetter Lane, London : C. F. Clay, Manager
.leJdl'l
286