Intermediaries in Jewish Theology: Memra, Shekinah, Metatron

STOP

Early Journal Content on JSTOR, Free to Anyone in the World

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in
the world by JSTOR.

Known as the Early Journal Content, this set of works include research articles, news, letters, and other
writings published in more than 200 of the oldest leading academic journals. The works date from the
mid-seventeenth to the early twentieth centuries.

We encourage people to read and share the Early Journal Content openly and to tell others that this
resource exists. People may post this content online or redistribute in any way for non-commercial
purposes.

Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early-
journal-content .

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people
discover, use, and build upon a wide range of content through a powerful research and teaching
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please
contact support@jstor.org.

INTERMEDIARIES IN JEWISH THEOLOGY

MEMRA, SHEKINAH, METATRON

GEORGE FOOT MOORE

Harvard University

I. Memra, Shekinah

The Christian interpretation of the Old Testament was early
set upon finding in it a figure corresponding to the Son, or the
Word (Logos), in the New Testament, a divine being, inter-
mediary between God the Father and the world in creation,
revelation, and redemption. For Christian theology, with its
philosophical presumptions, a God who visibly and audibly
manifested himself to men in human form and action was
necessarily such a being; the Supreme God, in his supramun-
dane exaltation or his metaphysical transcendence, could not
be imagined thus immediately to intervene in mundane affairs.
In this assumption and to a considerable extent in their par-
ticular interpretations the Fathers had a precursor in the Jew-
ish theologian Philo. One of the chief ends of their apologetic
was to demonstrate to Jews — or against them — first, that
their own Scriptures made the existence of such a being un-
deniable; and second, that — incarnate, crucified, risen, en-
throned at the right hand of God, presently to come in judg-
ment — he was no other than the Messiah whom the Jews
had rejected and the Lord whom the Christians worshipped as
Saviour.

From apologetic this passed into the tradition of both exege-
sis and dogmatics, and was to Christian consciousness so self-
evident that no other understanding of the Old Testament
seemed possible. Accordingly, when argument with Jews was
revived in the thirteenth century it was assumed that ancient
and unprejudiced Jewish students of the Scriptures must have
understood them in the same way, however their successors,,
in the exigencies of controversy, might dissimulate the fact.

41

42 HARVARD THEOLOGICAL REVIEW

On this presumption Christian scholars searched the earlier
Jewish literature, the Targums, 1 Talmuds, and Midrash, for
Christian doctrine, or at least adumbrations of it. Such re-
search continued in the succeeding centuries down to the
eighteenth; a vast mass of testimony was uncritically accu-
mulated, and conclusions drawn which obtained general assent
and continue to be accepted in some quarters to the present
time. In the Memra of the Targums, the Word (Logos) was
recognized, so to speak, in his own name and character; the
Skekinah was sometimes taken for the Second Person of the
Trinity, sometimes for the Third; after cabalistic studies came
into vogue, the mysterious Metatron joined the ranks of the
intermediaries. 2

As was pointed out in a former article in this Review, 3 the
material that was diligently collected to prove that Jewish
theology made a place for a being (or beings) of divine nature
through whose mediation the ends of the Supreme God were
effectuated in the world of nature and of men as they were in
Christian theology by the Son and Spirit has more recently
been appropriated to prove that Jewish theology, unlike Chris-
tian, interposed intermediaries between God and the world,
rendered necessary by its 'transcendent' idea of God, of which
error, conversely, the invention of such intermediaries is the
proof. Christian investigation and discussion of the terms
Memra and Shekinah have thus in all stages been inspired and
directed by a theological motive, and the results come around
in a circle to the theological prepossessions from which they
set out.

Jewish discussion of the subject has generally approached it
as a phase of the problem of the anthropomorphisms of Scrip-
ture. 4 Maimonides, in particular, who combated the notion that
God had body or form not only as irrational but as a deadly
heresy, and expended much ingenuity at the very beginning
of his Moreh in interpreting the seemingly anthropomorphic
expressions of Scripture as metaphors or otherwise rendering
them innocuous, claimed the authority of Onkelos for this
principle and procedure. 'Onkelos, the proselyte, perfectly
versed in Hebrew and Aramaic, takes all pains to remove the

INTERMEDIARIES IN JEWISH THEOLOGY 43

ascription of corporeity (to God), and whenever the Scripture
employs an expression that suggests corporeity, he interprets
it according to its (true) meaning.' 'Onkelos avoids the ascrip-
tion of corporeity (to God), and everything that might in the
remotest way suggest it.' 6 Such, indeed, according to Maimon-
ides, must be the endeavor of every intelligent man.

Maimonides' own Arab-Aristotelian metaphysic prescribed
to him the idea of God as simple Unity in so rigorous a sense
as to exclude not only all likeness to man, bodily or mental,
but all attributes, whether defined as essential, accessory, or
relative, and led him to regard the ascription of any attributes
as only a subtler form of the anthropomorphism which at-
tributed to him organs or actions. Of the latter, motion was
peculiarly objectionable, since it put God in space; and rest,
because it implied motion as its opposite. Onkelos seemed to
him to share this objection, for he regularly paraphrases pas-
sages in which God is said to go or come, to ascend or descend,
etc., sometimes by the introduction of memra, sometimes of
yekara, most frequently of shekinta. In so doing, he believed
that Onkelos had given the true meaning, whilst the letter of
Scripture was levelled to the apprehension of the common man.
The Glory or the Presence of God, as he conceived it, was not a
reverent circumlocution for God, but a created light by which
God's invisible presence was manifest to men; and similarly
the voice, or the word, of God was a created sound. 6 In thus
describing them Maimonides excludes personality and par-
ticipation in the divine nature. His Memra and Shekinah may
be called intermediary agencies, not intermediate beings, if
there be any profit in labelling them at all. His contemporary,
R. Moses ben Nahman, in his commentary on Gen. 46, 4 (a
verse with which Maimonides had to wrestle as an apparent
exception in Onkelos), and on Exod. 20, 16, contests the ade-
quacy of the principle Maimonides ascribes to Onkelos to ac-
count for the phenomena, as well as the validity of his explana-
tion. How the critic himself conceived the Shekinah and the
rest is not made clear. None of these writers subjected the
usage of Onkelos to a comprehensive analysis; the discussion
turned chiefly about a few striking verses in the Pentateuch.

44 HARVARD THEOLOGICAL REVIEW

A sounder method of investigation is adopted by modern
Jewish scholars who have dealt with the question. First among
these stands Samuel David Luzzatto, with his ~)i amx (Philoxe-
nus), published in 1830. Luzzatto minutely analyses the
changes Onkelos makes in his translation, and classifies them
by the reasons for them. More than this, he put investigation
on the right track by laying down at the outset the proposi-
tion, 'The Targum was not made for scholars, but for the un-
lettered masses ' (p. 1) — a proposition which was, he was well
aware, as revolutionary as it is sound. Luzzatto's own discus-
sion of memra, yekara, and shekinta is brief, and he fortifies
himself with the authority of a long quotation from Isaac
Arama, who treats them as respectful circumlocutions. A
special investigation, largely occupied with these particular
terms, was made by Siegmund Maybaum, 'Die Anthropomor-
phien und Anthropopathien bei Onkelos und den spateren
Targumim (Breslau, 1870), which is a comprehensive — for
memra in Onkelos an exhaustive — mustering and classification
of the relevant passages by the side of the Hebrew original
with explanatory comment. The most thorough investigation
of the whole subject is that of M. Ginsburger, 'Die Anthro-
pomorphismen in den Thargumim,' in the Jahrbiicher fur pro-
testantische Theologie, XVII (1891), pp. 262-280 and 430-
458, which contains by far the most comprehensive collection
of examples, and is of especial importance for its full presenta-
tion of the usage of the Palestinian Targums, which his prede-
cessors had adduced only casually, devoting their attention
almost solely to Onkelos. In the second part of his study the
usage of the Targums on the Hagiographa is also set forth.

A re-examination of the subject from a philological point of
view is the purpose of the present article, in which no attempt
is made to record the very extensive literature or the history
of interpretation. 7 For a complete understanding of the motive
of the translators in using these particular terms it would be
necessary to consider them in the larger connection of the whole
usage of the Targums in their substitutions and paraphrases,
but this is much too extensive a subject to be entered upon
here. The English reader may profitably acquaint himself with

INTERMEDIARIES IN JEWISH THEOLOGY 45

Luzzatto's exhibition of this usage as summarized by E.
Deutsch in Smith's Dictionary of the Bible, American Edi-
tion, Vol. IV, pp. 3404-3407.

It may not be superfluous to correct at the outset any notion
that Onkelos and the others in their versions systematically,
if not altogether consistently, eliminate or neutralize the an-
thropomorphisms of the original. Assertions of this sort are
indeed still found in books that the layman naturally takes
for authorities. 8 The same erroneous impression may be ac-
quired from works which deal methodically with the procedure
of the translators in this matter. The attention of the author
and the reader is there concentrated on the cases and occasions
in which expressions are modified that, at least in the ver-
nacular, sounded irreverent or undignified, and circumlocu-
tions introduced where a literal rendering literally understood
might fortify the common man's imagination of a God who
behaved too much like himself. The complement of this one-
sided impression, namely the limited range of such paraphrases
and the wide extent in which the Targums leave the anthro-
pomorphisms of the original untouched, can only come by
continuous reading of the Targums, and only in the same way
can the peculiarities of the several Targums be learned. Any
one, however, who will take the trouble to read the Targum
of Onkelos on the story of the Garden of Eden in Gen. 2-3, or
on God's visit to Abraham in Gen. 18, will be disabused of the
notion that the translator shrinks from a literal rendering of
even the most palpable anthropomorphisms. On the other
hand, circumlocutions and buffer-words are introduced with a
good deal of consistency in places that seem to us much more
harmless.

This is strikingly true of the uses of memra, generally trans-
lated 'word,' and frequently printed with a question-begging
capital, 'the Word.' To dispel misunderstandings at the out-
set we may begin by showing when and how memra is not used.
First, then, 'the memra of the Lord' in the Targums is not
employed as the Aramaic equivalent of 'the word of the Lord'
(nw in) in the Hebrew Scriptures. The Hebrew dabar, in all
senses and uses, is customarily rendered in the Targums by

46 HARVARD THEOLOGICAL REVIEW

pitgama. The 'word of the Lord,' or 'of God/ is pitgama de-Y.
(e.g. Gen. 15, 1), not memrade-Y.; 9 and similarly in'my word,'
'thy word,' 'his word,' when the pronouns refer to God. The
word of the Lord to a prophet is pitgam nebu'a, a word of
prophecy, e.g. Hosea 1, 2, 'the word of prophecy from before Y.
which was with Hosea.' See also 1 Kings 12, 22; Jer. 1, 2, 4, 11,
13; 2, 1, etc. It is idle to multiply examples of a uniform
usage. It holds in the cases which seem to approach most
nearly to a personification of the 'word of the Lord,' such as
Isa. 40, 8, ' The grass withereth, the flower f adeth, but the word
of our God abideth forever'; Targum, 'The wicked man dies
and his plans perish, but the word of our God (pitgama de-
elahana) abideth forever; Isa. 55, 11 (pitgam tubi); Jer. 23, 29
(hoi pitgamai, cf. 5, 14 pitgame nebuati). The Targum on the
Psalms is too late to be taken in evidence here, but it may be
observed that, although memra occurs frequently in it, when it
comes to translate Psalm 33, 6, 'By the word of the Lord
(LXX r<J> \by<$ tov Kvpiov) were the heavens made, and all the
host of them by the breath of his mouth,' it renders 'word'
not by memra, but by the common milla. 10 It may be added
that memra is not employed, as seems sometimes to be imagined,
as a standing circumlocution for 'God said,' or 'God spoke';
the Targums have no scruples about translating these phrases
literally. 11

Thus, wherever the 'word of the Lord' is the medium or in-
strumentality of revelation, or of communication to men, in
Greek X670S or prjua, the term employed for this medium in
the Targums is not memra, but pitgama, or (seldom), as in the
example cited above from Psalm 33, 6, milla.

Further, where the creative activity of God is spoken of in
the Scriptures, the Targums do not represent this activity as
mediated by his memra. Isa. 45, IS is an apparent exception
of the kind which in the proper sense of the dictum probat
regulam. See also Deut. 33, 27, in a midrashic interpretation
of the difficult words translated in our Bibles, 'underneath are
the everlasting arms,' where Onkelos has, 'By his word (memra,
fiat) the world was created.' The full importance of these ob-
servations will appear in the sequel, when we come to consider

INTERMEDIARIES IN JEWISH THEOLOGY 47

the supposed relation of memra in the Targums to the Logos
in Philo. 12

With so much by way of introduction, we may turn to an
examination of the meaning and use of memra.

Memar (definite, memra) is the Aramaic counterpart of the
late Hebrew ma'amar, from amar, 'say.' The nouns signify
something that is said, dictum, in the widest sense of the term.
If what is said has authority behind it, it acquires from the
context the connotation of command ('edict'). So the 'ten
words ' (debarim) of Exod. 34, 28, become for us the Ten Com-
mandments; the world was created by ten commands (ma-
'amerim, 'fiats,' in Gen. 1), Aboth 5, 1, cf. Megillah 21b. In
this sense memar is used in the Targums to interpret the Hebrew
peh ('mouth') in idiomatic phrases, e.g. Gen. 45, 21, njna 'B i>y,
Targ. 'al memra de-Par' oh; the English versions, 'according
to the commandment (edict) of Pharaoh.' The accounts of the
tabernacle were rendered 'by order of Moses' ('al memra de-
Mosheh, Exod. 38, 21); 'at the order of Aaron ('al memar
Aharon) and his sons shall be all the service of the Gershonites,'
(Num. 4, 27). 'Moses . . . died there in the land of Moab
by the command of God ('al memra de-Y., Deut. 34, 5). Memra
is used to render other expressions which imply command;
for example, in Gen. 41, 44, Pharaoh says to Joseph, 'I am
Pharaoh, and without thee ("■pjfai) shall no man lift up hand
or foot in all the land of Egypt.' Onkelos renders, 'without
thy command' ("pO'DD na) shall no man raise his hand to lay
hold of a weapon,' etc.

When men disobey the command (literally 'mouth') of God
(pi Yahweh), or refuse obedience to it, 13 the Targum renders
by memra; e.g. Deut. 1, 26, 'Ye refused to go up, and re-
belled against the command (memra) of God.' u Similarly,
when they are said to transgress the commandment of God; 1S
e.g. Num. 14, 41, 'Ye transgress the commandment of God';
Onkelos, 'Ye transgress the decree of the edict of God' (gezerat
memra de-Y.) ie So also with other verbs. 17 Num. 11, 20,
'Ye spurned the Lord who is in the midst of you'; 'Ye spurned
the word (memra) of Y. whose presence (shekinta) abode among
you.' Num. 21, 5, 'The people spoke against God and against

48 HARVARD THEOLOGICAL REVIEW

Moses'; 'The people murmured against the word (memra) of
Y., and contended against Moses.' Deut. 32, 51, 'Because ye
proved false to me in the midst of the Israelites '; 'Because ye
proved false to my word' (memri) .

On the other hand, to hearken to God, or to his voice, is in
the Targums regularly to receive (implying, 'obey') the com-
mand (memra) of God; e.g. Lev. 26, 14, 'If ye do not hearken
unto me,' etc.; Onkelos renders, 'If ye do not receive my com-
mand' (memri); Deut. 28, 15, 'If ye do not hearken to the
voice (qol) of the Lord your God, by observing and doing all
his commandments'; Onkelos, 'If you do not receive the com-
mand' (memra), etc. As the latter example shows, the Hebrew
qol, 'voice,' when it implies a command, is rendered by memra;
when this implication is not present, it is interpreted qal memra;
e.g. Gen. 3, 8, Adam and Eve 'heard the voice of the Lord';
Onkelos, 'heard the sound of the word (qal memra) of Y. (the
sound of Y. speaking), who was walking in the garden' (cf. 23,
10). So also Deut. 5, 21, 'We have heard his voice out of
the midst of the fire'; Onkelos, qal memreh; and likewise in
verses 25 and 26; cf. Deut. 4, 36, 'From heaven he made thee
hear his voice'; Onkelos again, qal memreh.

In the phrase last quoted memra is not understood to imply
command, and this is the case in a large number of passages
to a consideration of which we now proceed. Notice may be
directed first to places where the Bible narrates that God came
to some one and spoke to him. Thus in Gen. 20, 3, ' God came
to Abimelech in a dream of the night, and said to him'; Onke-
los renders, 'A word (memar) from before Y. came to Abime-
lech in a dream of the night, and said to him.' Precisely so to
Laban (Gen. 31, 24), and to Balaam (Num. 22, 9, with no
mention of a dream). The paraphrase is natural; and there
is additional reason for it in the fact that the recipients of
these visits of God are not Israelites. Still stronger reason for
paraphrase is given in Num. 23, 3, where Balaam bids Balak
stand by his sacrifice, 'and I will go; perhaps the Lord will
come to meet me,' and verse 4, 'The Lord met Balaam, and
he said to Him, I have prepared the seven altars,' etc. Onkelos
renders, 'Perhaps an oracle from before Y. (memar min qadam

INTERMEDIARIES IN JEWISH THEOLOGY 49

Y.) will come to meet me, and the word (pitgama) that he shall
show me, I will disclose to thee'; and in verse 4, 'And an
oracle (memar) from before Y. met Balaam,' etc. In connection
with this, verse 5 must be noted: 'Y. put the word (pitgama)
in Balaam's mouth.'

Similar caution is evident, however, where God says that
he will meet with the Israelites at the Tabernacle on stated oc-
casions. Exod. 25, 22f ., Onkelos, 'I will cause my word (memri,
oracle) to meet thee there, and I will speak with thee from
above the place of atonement (kapporeth) , from between the
two cherubs,' etc.; Exod. 29, 42, 43, Onkelos, 'I will cause my
oracle (memri) to meet with you there, to speak with thee there;
and I will cause my oracle to meet with the Israelites, and it
(the Tabernacle) shall be sanctified by my glory.' In Exod.
19, 17, 'Moses brought the people out of the camp to meet
God'; Onkelos, 'towards the oracle (memra) of God.' In these
cases the paraphrase is natural, since in the first two the text
and in the last the context make the revealing of the will of
God the object of the meeting; but in both the motive for
paraphrasing at all is plainly to avoid the imagination of a
meeting between men and God in propria persona. For this
there was explicit warrant in Deut. 4, 12: ' Y. spoke to you out
of the midst of the fire; ye heard the sound of words (qal
pitgamin), but a form ye did not see, only the sound (voice).'
In Exod. 3, 18, where the Hebrew is, 'Say unto him (Pharaoh),
The Lord, the God of the Hebrews, met us,' Onkelos has, ' ap-
peared to us' (cf. vs. 16); see also 5, 3. 18 Note further Exod. 4,
12 (cf. 15), God says to Moses 'I will be with thy mouth, and
teach thee what thou shalt speak'; Onkelos, 'my word (memri)
shall be with thy mouth,' etc.

Here may perhaps most appropriately be introduced the
scene between Jacob and Laban, Gen. 31, 49 f.: 'The Lord
be on the lookout between me and thee when we are out of
one another's sight. . . . God is witness between me and
thee'; Onkelos in both verses, 'the word (memra) of Y.'

Natural paraphrase is to be seen also in such cases as Gen.
15, 6, Abraham 'believed in (put confidence in) God, and it
was reckoned to him for righteousness'; Onkelos, 'He be-

50 HARVARD THEOLOGICAL REVIEW

lieved in the oracle (memrd) of Y.,' namely, the promise con-
tained in verses 1-5. Exod. 14, 31, When the Israelites saw
the great work the Lord did on the Egyptians, they feared the
Lord, 'and believed in the Lord and in his servant Moses';
Onkelos, 'in the oracle (memra) of Y. and in the prophecy of
Moses his servant'; see also Num. 20, 12; Deut. 1, 32.

Where the Hebrew is literally, 'God said in his heart (mind),'
that is, said to himself, Onkelos renders ' said in (or, by) his word '
(bememreh; dixit in dicto suo) ; see Gen. 8, 21; cf. 6, 6. In the
same way Onkelos interprets the enigmatic 'and God knew'
of Exod. 2, 25, ' God said in his word to deliver them,' i.e. con-
ceived the purpose. When God swears by himself, as in Gen.
22, 16, Onkelos has, 'by my word (bememri) I have estab-
lished.' l9 So in Exod. 6, 8, 'I will bring you into the land which
I lifted up my hand (swore) to give to Abraham,' etc.; cf.
Num. 14, 30. The same formula is used when a man adjures
another by God. Places in which it is said that God 'repented'
(was sorry, changed his mind) are treated in various ways, ac-
cording to the context. An instance of the use of memra is
Gen. 6, 6, 'God repented that he had made man'; Onkelos,
'Y. turned in his word (memreh, thought, we should say) that
he had made man.' Correspondingly in vs. 7. So also 1 Sam.
15, 11 and 35; Zech. 8, 14, Targum, 'my word (thought,
memri) did not turn.'

Passages in which it is said that God will fight for the Israel-
ites are paraphrased; e.g. Deut. 3, 22, 'For the Lord your God,
he it is that fighteth for you'; Onkelos, 'For Y. your God,
his word (memreh) fights for you'; cf. Deut. 1, 30. An inter-
esting class of passages which seem to fall into the same category
are those in which God promises to be with some one, or it is
said that he was with some one. Thus in Exod. 3, 12, God says
to Moses, 'I will be with thee'; Onkelos, 'My word (memri)
will be in thy support.' 20 So in Gen. 21, 20, 'God will be with
the lad' (Ishmael); Onkelos, 'the memra of Y. will be in the
support of the lad.' 21 In such passages memra is probably the
effective word which gives victory or protection with no need
of such personal intervention as the phraseology of the original
suggests. So also in punishment, e.g. Deut. 18, 19, Onkelos,

INTERMEDIARIES IN JEWISH THEOLOGY 51

'the man who does not receive (obey) my word (pitgami)
which he (the prophet) shall speak in my name, my word
(memri, Heb. 'I') will demand satisfaction of him.' Compare
also Deut. 4, 24, 'The Lord our God is a devouring fire'; ' Y.
our God, his word (memreh) is a devouring fire'; cf. Deut. 9, 3.

Cognate in a measure to these are passages in which memra
is put for the protecting 'hand' of God. Thus Exod. 33, 22,
God says to Moses, 'I will cover my hand over thee till I have
passed by'; Onkelos, 'I will extend protection by my word
over thee.' The command of God, his expressed will, suffices
for protection. 22 So also Num. 11, 23, 'Is the Lord's hand be-
come short?' so that he is unable to provide food for the vast
host of Israelites in the desert, as Moses in the preceding
speech seems to imply; Onkelos, 'Is the word (memra, fiat) of
God restrained,' hindered from effecting his purpose? With
this compare the rendering of the same figure in Targum Isa.
50, 21, 'Is my might (geburathi) shrunken?' See also 59, 1.

Finally, attention should be directed to the introduction of
memra when God speaks of a covenant between himself and
men. Thus Gen. 9, 12, 'This is the sign of the covenant which
I make between Me and you'; the Targum, 'between my word
(ben memri) and you' (cf. vss. 13, 15, 16, 17); 17, 7, 'I estab-
lish my covenant between my word and thee' (cf. 17, 10); see
also Exod. 31, 13, 17; Lev. 26, 46. Here 'the word' seems to
serve only the purpose of a buffer, to avoid the impression that
God enters into a covenant with men, so to speak, on equal
terms. In so far as the promise or the requirement that is the
subject of the covenant is expressed in the context, it is a not
inappropriate buffer.

A different explanation is given by Maybaum, who regards
memri in these cases, and in many of those adduced above
under other heads, as equivalent to a reflexive pronoun, 'my-
self.' That memar was used in this way, especially in the late
Targums on the Hagiographa, was remarked long ago by Bux-
torf, who cites from a haggadic amplification in Targ. Ruth 3,
8: 'Paltiel bar Laish (2 Sam. 3, 15) was a pious man, who
stuck a sword between himself (ben memreh) and MichaF;
and adds 'Sic de Deo saepissime.' Similarly the Targum on

52 HARVARD THEOLOGICAL REVIEW

Job 7, 8, 'Thine eyes are upon me {memri, my person), and I
am gone.' Other examples are cited by Lightfoot, Horae
Hebraicae, on John 1,1. Maybaum quotes from the so-called
Targum Jonathan ben Uzziel on Gen. 9, 17: 'This is the sign
of the covenant which I have established between me and all
flesh that is upon the earth'; Jonathan, ben memri uben memar
kol bisra, 'between myself (my person, we might say) and the
person of every man.' The one known manuscript of this Tar-
gum 23 agrees with the printed editions in reading thus; but
the expression is unparalleled, so that Ginsburger's suspicion of
a blunder by a copyist does not seem an excess of scepticism.
In any case an isolated phrase in this Targum is no key to the
usage of the older Targums with which we are here concerned.
Undoubtedly, if we had to translate idiomatically many of the
passages in which memra is used, we should say 'myself, him-
self,' and the like; but inasmuch as the whole motive of the
paraphrase is to avoid bringing God 'himself into such im-
mediate relation to the act or circumstance, it can hardly be
supposed that the translator deliberately introduced a word
which would be understood by his hearers to emphasize the
relation. If he did not like to say simply that God did so and
so, he would be still less inclined to say that God himself did it.
We have now surveyed the various uses of memra in the
Targums on the Pentateuch and the Prophets. Instances might
be multiplied under almost all the heads specified, but no class
of cases has been passed over. Most of the uses of the word
are easily explicable in their contexts in the light of the ends
and methods of the synagogue interpretation. If analogy, or
some subtlety of interpretation that escapes us, has sometimes
introduced it on less obvious occasions, these are exceptions
which need cause us neither surprise not perplexity. The
inquiry must set out from the common and plain uses; and
our conclusions must be drawn from them, not from the resid-
uum, if there be such, of unexplained occurrences. Proceed-
ing in this way we find that God's memra 2i has sometimes the
connotation of command — we might in imitation of the ety-
mology say 'edict' — the expression of his will which is an
effective force in nature and providence; sometimes it might

INTERMEDIARIES IN JEWISH THEOLOGY 53

best be translated ' oracle,' the revelation of his will or purpose
(not, however, a specific word of prophecy); sometimes it is
the resolution of a metaphor for God's power, his protection,
and the like. In many instances it is clearly introduced as a
verbal buffer — one of many such in the Targums 26 — to
keep God from seeming to come to too close quarters with men
and things; but it is always a buffer-word, not a buffer-idea;
still less a buffer-person.

This would come out still more plainly if it were possible
here to direct special attention to the singular phraseology of
many of the passages in which memra comes in as a euphemism
or as what I have called a buffer. In the context the translator
habitually keeps as close as possible to the original, without
adapting it to the new situation he has created by the intro-
duction of his memra, and the result is often awkward and
unidiomatic. It may be surmised that, as in the case of similar
euphemisms and buffers introduced into the Hebrew text itself
or the masoretic punctuation, the Targumists intentionally
left the matter so that readers or hearers educated in the Scrip-
tures would recognize the original expression or meaning
through the veil cast over it. Such phenomena cannot, how-
ever, be exhibited in translation — in the translations above
they are indeed frequently effaced in the interest of intelligible
English — nor would they, even with explanation, show what
they are except to readers familiar with both unsophisticated
Aramaic diction and idiom and with the peculiarities of the
translation-Aramaic of the Targums. To such readers, how-
ever, these phenomena must be among the most convincing evi-
dence of the real character and motive of the memra passages.

The sum of the whole matter is that nowhere in these Tar-
gums is memra a 'being' of any kind or in any sense, whether
conceived personally as an angel employed in communication
with men, or as a philosophically impersonal created potency,
as in Maimonides' theory; or God himself in certain modes
of self-manifestation, which has been thought to be the opinion
of R. Moses ben Nahman. The appearance of personality
which in many places attaches to the memra is due solely to
the fact that the phrase 'the memra of Y.,' or, with pronouns

54 HARVARD THEOLOGICAL REVIEW

referring to God, My, Thy, His, memar, is a circumlocution for
'God,' 'the Lord,' or the like, introduced out of motives of
reverence precisely where God is personally active in the af-
fairs of men; and the personal character of this activity neces-
sarily adheres to the periphrasis. The very question whether
the memra is personal or impersonal implies, from the philo-
logical point of view, a misunderstanding of the whole phenome-
non; and every answer to a false question is by that very fact
false.

These conclusions are strongly confirmed by the fact that
memra is found only in the Targums; not in such Aramaic
texts as are preserved in the Midrashim, nor in the voluminous
Aramaic parts of the Talmuds, nor, so far as I am aware, in the
Zohar. In other words, it is a phenomenon of translation, not
a creature of speculation. 28

The error is magnified to immensity when memra is connected
with the Logos of Philo, whether it be supposed, as by Gf roerer,
that the Palestinian mystical theology represented in the
Targums (!) borrowed its intermediary being, Memra, from
the Logos of Alexandrian 'theosophy,' or, contrariwise, that
the Logos was derived and developed by the Alexandrians from
the Palestinian Memra. The former theory involves a com-
plete misunderstanding of what the Targums are and what
they were made for, as well as a misinterpretation of the memra
in them; the latter, besides a similar misinterpretation of
memra in the Targums, involves a fundamental misunderstand-
ing of what the Logos is in Philo, and what it is for.

It has been pointed out above (page 45 f.) that in the Targums
memra is not the term employed where the 'word of the Lord'
is the medium or instrumentality of revelation, and that it is
not the creative word in the cosmogony of Genesis or reminis-
cences of it. 27 It is needless to add that is not the divine reason
in the universe, nor the reason akin to the divine that is in
every man. Since these things are exactly what the Logos is
and does in Philo, the only tertium comparationis that would
seem to be left is that the Greek X670S is often properly under-
stood and translated ' word,' and that memra also is commonly
so translated.

INTERMEDIARIES IN JEWISH THEOLOGY 55

It is an error of equal dimensions, when, by association with
the Christian doctrine of the Logos and by abuse of a technical
term of Christian theology, the Memra is described as 'an
hypostasis.' For the modern reader 'hypostasis' has no use
or meaning except that which it acquired in the controversies
of the third and fourth centuries over the ontological relation
of the Logos-Son to the Father; and to employ this term, with
its denotation and all its trinitarian connotations, of the sup-
posed personal, or quasi-personal, 'Memra' of the Targums, is
by implication to attribute to the rabbis corresponding meta-
physical speculations on the nature of the Godhead. But of
speculation on that subject there is no trace either in the exo-
teric teaching of Judaism or in anything we know of its eso-
teric, theosophic, adventures into the divine mysteries.

Another paraphrastic expression upon which for our present
purpose it is unnecessary to dwell is yekara, 'glory, majesty'. 28
One example out of many must suffice. In Exod. 24, 10, Moses
and his companions, with the seventy elders of Israel, 'saw
the God of Israel'; Onkelos, 'saw the glory (yekar) of the God
of Israel.' The same interpretative periphrasis is used in Exod.
16, 17; Isa. 6, 1. Similarly, Gen. 17, 22, God ascended from
Abraham; 'The glory of God ascended.' Exod. 20, 17, God
has come to prove you; 'The glory of Y. has appeared to
you.' 29

Shekinah is another such word, properly Hebrew, but used
in the Aramaic of the Targums as a borrowed word with Ara-
maic endings. The large part it has played in Christian dis-
cussion renders a brief statement of the usage necessary. Its
origin and primitive significance are best seen where it para-
phrases the verb (shakan; 'dwell, reside, abide') from which
it is derived. Thus in Exod. 25, 8 God says, 'Let them make
me a sanctuary that I may dwell among them'; the Targum
has it, 'I will cause my presence (shekinti) to abide (or reside)
among them.' Exod. 34, 6, 'And the Lord passed before him';
Targum, 'The Lord caused his presence (shekinteh) to pass
before him.' In general, when God is said to be in a place or
among his people, Onkelos makes it 'his presence' there; in

56 HARVARD THEOLOGICAL REVIEW

Deut. 12, 5; 11, 21, it is 'his presence' not 'his name' that God
causes to abide in the place he chooses for a sanctuary. Con-
trariwise, when he leaves a place, he 'causes his presence to
ascend' (to heaven, and depart from men); Hos, 5, 6; etc.
Deut. 32, 19.

While memra, as has been observed above, is found only in
the Targums, shekinah is very common in the Talmud and
Midrash also. 'Often it is a mere metonymy for 'God,' as when
R. Jose ben Halafta says: 'Never did the Presence (shekinah)
descend to earth, nor did Moses and Elijah ascend to heaven;
for it is written the heavens are the Lord's heavens, and the
earth he has given to the children of men' (Psalm 115, 16). 30
Inasmuch as the same Rabbi elsewhere says that God is not
in any place, perhaps this is his meaning here. It has also been
suggested that the words are meant by inference to contradict
Christian teaching on the incarnation and the resurrection
(Bacher, Agada der Tannaiten, II, 185). In the parallel in the
Mekilta, kabod, 'glory,' is used where the Talmud has shekinah.
The doctrine of R. Jose seemed so paradoxical in the face of
such explicit texts as Exod. 19, 20; Zech. 14, 4, on the one hand,
and Exod. 19, 3; 2 Kings 2, 11, on the other, that some in-
genuity had to be exercised to save it.

Where the omnipresence of God is asserted, the word used is
'the Presence' (shekinah). The Lord was revealed in the thorn
bush to teach that there is no place on earth void of the Pres-
ence; it is the Presence which, like the sea flooding the cave,
filled the tabernacle with its radiance, while the world outside
was no less full of it. Successive sins of mankind beginning with
Adam caused the Presence to be taken up from earth and from
one heaven to another to the seventh and most remote; a
succession of righteous men from Abraham to Moses brought
it down again, stage by stage, to earth once more. 31 In a later
work ten descents of the Presence to the world are enumerated,
from the first in the Garden of Eden to the last, still future, in
the days of Gog and Magog; the Scripture proofs alleged are all
verses in which God or (the Lord) comes down to earth (Gen.
11, 5, etc.), or is upon the earth, as in Gen. 3, 8; Zech. 14, 4. 32

In a special sense God dwelt in the tabernacle and later in

INTERMEDIARIES IN JEWISH THEOLOGY 57

the temple. When he took up his abode in them a cloud en-
veloped the tabernacle, or filled the temple, and thus veiled the
glory of the Lord, too deadly bright for mortal eyes, which
filled them. 33 The association of the presence of God with a
manifestation of his glory and of the latter with light led to the
conception of the Presence (shekinah) as light.

All worship demands a praesens numen, and however men
may entertain the idea of the omnipresence of God, they find
it difficult to realize his specific presence in the particular place
where they gather for religious service without some aid to
faith or imagination. This is the origin and meaning of the
teaching that wherever ten men (the quorum of the synagogue)
are met for prayer, there is the Presence. 34 How many 'Pres-
ences' are there then? a caviller asked. R. Gamaliel (II)
answered by asking a slave, How does the sun get into that
man's house? The sun shines, he replied, on all the world. If
the sun, one of the millions of suns that are before the blessed
God, shines on all the earth, how much more the Presence of
God! (Sanhedrin 39a.)

R. Isaac, a pupil of Johanan and a favorite homilist of the
third century, says: 'Whenever Israelites prolong their stay
in the synagogues and schools, God makes his Presence stay
with them.' 36 The following is also handed down from Isaac:
'Whence do we learn that God is found in the synagogue
building?' Because it is said,'God standeth in the congregation
of God' (Psalm 82, 1). And whence that when ten are praying
together the Presence is with them? Because it is said, 'God
standeth in the congregation of God' (ibid.). 36 And whence
that when three are sitting as judges the Presence is with them?
Because it is written, 'In the midst of the judges (elohim) he
will judge' (Psalm 82, lb). And whence that when two are
sitting and studying the Law the Presence is with them? Be-
cause it is written, 'Then those who fear the Lord spoke one
to the other, and the Lord hearkened and heard,' etc. (Mai. 3,
16). And whence that even when one is sitting and studying
the Law the Presence is with him? Because it is written, 'In
every place where I cause mention to be made of my name, I
will come unto thee and bless thee' (Exod. 20, 21). 37

58 HARVARD THEOLOGICAL REVIEW

In all these cases the Presence (shekinah) is not something
that takes the place of God, but a more reverent way of say-
ing 'God.' Similarly Christians speak of God's being present
in their religious assemblies or of the presence of the Holy
Spirit, without intending any difference of meaning, notwith-
standing the personality of the Holy Spirit, and indeed with-
out reflection at all. This use of the phrase 'the Holy Spirit,'
ultimately derived from the Old Testament, was, it should
be remembered, long established in Christian speech and
literature before the dawn of hypostatic speculations.

In Jewish literature, also, 'the Holy Spirit' frequently oc-
curs in connections in which 'the Presence' is elsewhere em-
ployed, without any apparent difference of meaning; but the
fact that the words are within a certain range interchangeable
is far from warranting the inference that the shekinah and the
ruk ha-qodesh were identified in thought. Thus it is said in
the Tanhuma (ed. Buber, Shemoth 10, f. 3a) that until the
temple was destroyed the shekinah was placed in the temple
('The Lord is in his holy temple,' Psalm 11, 4); after the de-
struction of the temple, the shekinah ascended to heaven (' The
Lord, in heaven is his throne,' ibid.). With this compare
Koheleth Rabbah on Eccl. 12,7 (end) : 'When Jeremiah saw that
Jerusalem was destroyed, and the temple burned, and Israel
gone into exile, and the Holy Spirit taken up,' etc. The inter-
change is especially frequent in reference to persons to whom
the Spirit or the Presence comes, or on whom it rests. A good
example is Tos. Sotah 13, 3 compared with Bab. Sotah 48b;
Sanhedrin 11a. In the former the voice from heaven declares
that one of the company is worthy to have the Holy Spirit
rest upon him; the Talmud has 'the shekinah.' On the other
hand, revelation, or inspiration, the chief function of the Holy
Spirit in Judaism, is, so far as I know, never attributed to the
Presence (shekinah). Among the five things which were in the
first temple but were lacking in the second, Yoma 21b includes
both the Shekinah and the Holy Spirit. This list is evidently
padded and confused. What seems to be the soundest form of
the tradition counts the five things : the fire (that was kindled
from heaven), the ark, the priestly oracle (Urim and Thum-

INTERMEDIARIES IN JEWISH THEOLOGY 59

mim), the anointing oil, and the Holy Spirit (the spirit of
prophecy). So Jer. Taanith ii, 1, f. 65a; Jer. Makkoth ii, 7,
f. 32a; ef. Jer. Horaioth iii, 2, f. 47c end. The 'Shekinah' in
Bab. Yoma I. c. is intrusive, perhaps a doublet to the Holy
Spirit. It does not seem to be found in any of the parallels
in the Midrashim (Shir ha-Shirim Rabbah 8 (on verse 9);
Tanhuma ed. Buber, Behealotka 11 (f. 25b, top); Bemidbar
Rabbah 15, 10, etc.

Like memra, shekinah acquires what semblance of personality
it has solely by being a circumlocution for God in contexts
where personal states or actions are attributed to him.

60 HARVARD THEOLOGICAL REVIEW

NOTES. — I. MEMRA

1. The Targums were generally supposed to represent a traditional exe-
gesis older than the Christian era.

2. For a single example, M. Kahler, ' Christologie, Schriftlehre,' Pro-
testantische Beal-Encyclopaedie, 3 ed. iv, 7: EigentUmlich ist dem nach-
kanonischen Judentum die Umsetzung der anschaulichen AusdrUcke fur das
Walten Gottes in der Welt, namentlich auch seines offenbarenden Wirkens,
in gewissermassen selbststandige Werkzeuge Gottes; das schiipferische und
offenbarende Wort wird im Memra hypostasiert, die Gnadengegenwart
Gottes bei seinem Volk in der Schechina; dazu kommt bei den Rabbinen
noch der Metatron; alle diese Mittelwesen gleichen den Engeln und sind,
wie auch der Geist Gottes, geschaffen.

3. 'Christian Writers on Judaism,' Vol. XIV (1921), pp. 227, 233, and
elsewhere.

4. See L. Ginzberg, 'Anthropomorphism,' Jewish Encyclopedia I, 621-
625, and the literature there noted.

5. Moreh Nebukim, Part i, cc. 27-28.

6. Nor created ad hoc. All miraculous events that occur at a given mo-
ment of time seemingly at variance with the order of nature were really con-
stituted part of that order at the creation of the world. 'The Eight Chapters,*
c. 8 (ed. J. I. Gorfinkle, New York, 1912, p. 46, and ibid, translation, pp. 90 f„
with the references there given in a note); Maimonides, Commentary on
Mishnah, Aboth 5, 6; Moreh, Part ii, c. 29. Cf. Munk, Le Guide des Egares,
I, p. 296 n.

7. For brevity and simplicity I have restricted myself to examples from
Onkelos and the Targums on the Prophets, which had an authority not con-
ceded to the rest. Whatever peculiarities the Palestinian Targums present,
Ginsburger's investigation proves that in them also there is no personifica-
tion of memra or shekinta, to say nothing of 'hypostasis.' In the translitera-
tion of Hebrew and Aramaic I have not marked the quantity of the vowels.
Readers who know the language do not need this assistance any more than
in Latin; those who do not will be none the wiser for it.

8. 'In the Targums anthropomorphic expressions are put aside altogether.'
Oesterley and Box, Religion and Worship of the Synagogue, p. 153.

9. In translating from the Targums, I employ 'Y.' where they have the
customary abbreviation for the name.

10. Milla sometimes stands in the Palestinian Targums where Onkelos
has pitgama. The variation has no significance.

11. 'The memra of Y. said,' and the like, occurs only in Palestinian Tar-
gums, and apparently with especial frequency in the Fragmentary Targum.
See Ginsburger, p. 267f.

12. See below, p. 54.

13. niSV 'B J1K iTID. English versions often, 'rebel against the command-
ment of the Lord.'

14. See also 1 Sam. 12, 14, 15; 1 Kings 13, 21, 26, etc.

15. nw 'B m naj/.

INTERMEDIARIES IN JEWISH THEOLOGY 61

16. Num. 22, 18; 24, 13; 1 Sam. 15, 24, etc. For the expression cf. the
Targum on Isa. 40, 5; 58, 14, 'for by the edict (memra) of Y. it is thus de-
creed' (gezir ken). See also Num. 14, 35.

17. dkd ,'n nm ,>y».

18. Compare the shifts of the Greek and Latin versions in Exod. 3, 18 and
5, 3. They translate Sip, 'call.'

19. From motives of reverence Onkelos uses this verb for the oath of God;
when men swear he employs the usual Aramaic verb.

20. "PJJD2, with a buffer preposition.

21. See also Gen. 21, 22, 23; 26, 28; 28, 20; 31, 5, 42; 39, 21, 23, etc.

22. A more drastic figure is similarly paraphrased in Ezek. 16, 8, 'I
spread my skirt over thee'; Targum, 'I extended protection by my word
(memri) over thee.'

23. Edited by M. Ginsburger, Pseud o- Jonathan . . . nach der Lon-
doner Handschrift (Brit. Mus. add. 27,031). Berlin, 1903.

24. It is to be observed that memra does not occur without a genitive —
'the word of the Lord,' 'my word,' etc., or a circumlocution for the genitive,
'a memar from before the Lord.' 'The Memra,' 'the Word,' is not found in
the Targums, notwithstanding all that is written about it by authors who
have not read them.

25. The commonest — and in many phrases awkwardest — of these is mp,
' before, in front of.' For examples see Ginsburger, pp. 278-280, or the Lexicons.

26. For this reason alone the attempt to elucidate memra by the dibbur
of the Midrash is out of place, even if the usage of dibbur were not misstated.

27. Consequently, the theory that derives the Logos- Word of John 1,
1-5 straight from the Palestinian memra is fallacious.

28. Yekar is elsewhere the ordinary translation of the Hebrew kabod, in
Greek 56£a.

29. For other examples, see Maybaum, p. 49 f , Ginsburger, p. 277 f. In
similar cases Onkelos sometimes has memra, sometimes sheldnta.

30. Bar. Sukkah 5a (top); cf. Mekilta on Exod. 19, 20 (ed. Friedmann f.
65b).

31. Bereshit Rabbah 19, 7 and parallels.

32. Aboth de-R. Nathan 34, 5.

33. Exod. 29, 34 f., 1 Kings 8, 10 f., cf. Isa. 6, 1-4.

34. Sanhedrin 39a.

35. Pesikta ed. Buber, Shemini Asereth, f. 193a-b; Pesikta Rabbathi ed.
Friedmann (Supplement), f. 202b. For the exegetical derivation see the
editors' notes, and Bacher, Agada der palast. Amoraer II, 220 f . n. To the
same homilist Song of Songs 2, 8 f. suggests God's springing from synagogue
to synagogue and from school to school to bless the Israelites (Pesikta Rab-
bathi, f. 72a; less complete text, Pesikta ed. Buber f. 48b).

36. f>N mjD 3W DTI^N. In the first deduction *?$ n*IJ> is taken in the
sense of 7K '"IJTID in Psalm 74, 8; the second takes my as 'congregation,'
which consists of at least ten men (general rule based on Num. 14, 27). See
Bacher, Agada der palast. Amoraer II, 221.

37. Berakot 6a. On Exod. 20, 21 cf. Onkelos, 'In every place where I
make my presence (shekinti) to rest, thither will I send my blessing unto
thee and will bless thee.'

62 HARVARD THEOLOGICAL REVIEW

II. Metatron

In the foregoing there is nothing novel either in the facts or
the conclusions, and the only reason for working over the ground
again and presenting the results here is that the scholars whom
it most concerns to know about the subject almost universally
ignore the previous investigations, and are content to take their
facts and opinions directly or indirectly from Gfroerer and
Weber. In the case of Metatron, on the other hand, there ap-
peared to be room for a new philological and historical study
of the whole problem, such as will occupy the rest of this
article.

Christian attention was first directed to Metatron by caba-
listic studies, and it was from the Cabala and commentators
who interpreted the Old Testament in the spirit and sense of
the Cabala that Christian theologians got the notions about
him with which subsequent investigation has generally set out.
Metatron was for them an angel of the highest order, or a
mysterious being of higher than angelic rank, who was in a
peculiar sense a mediator and intercessor with God. Hermann
Witsius (d. 1708) was tempted to surmise that even the name
Metatron itself might be a deflected form of the Latin media-
tor, 'nam qui Mediatoris sunt, ea huic Angelo adtribuere
solent.' x If, instead of starting with cabalistic mysteries, or
mystifications, the investigation begins at the other end, there
will be a better prospect of finding out who or what manner
of thing Metatron was.

The oldest occurrence of the word is in Sifre on Deut. 32,
49 (§ 338), that is, in a Palestinian work the final redaction of
which falls early in the third century, but which in this part
is a Midrash of the school of Ishmael three quarters of a century
earlier. 2 Moses is bidden to ascend Mount Nebo in the land of
Moab opposite Jericho, 'and see the land of Canaan, which I
am going to give the Israelites as a possession.' On this R.
Eliezer comments: 'With his finger he (God) was a mefatron
to Moses 3 and showed him the whole land of Israel; so far the
boundaries of Ephraim; so far the boundaries of Manasseh.' 4
According to R. Joshua, Moses saw it for himself; God gave

INTERMEDIARIES IN JEWISH THEOLOGY 63

him such powerful eyesight that he saw from one end of the
world to the other. 5 The word meiatron was explained by R.
Moses ben Nahman and Eshtori Parhi as 'one who shows the
way,' a guide, and a corresponding gloss has found its way into
the text of Sifre. 6

Another occurrence in a Palestinian Midrash is in Bereshith
Rabbah 5, 4 (on Gen. 1, 9): 'R. Levi said, Some interpreters
interpret with Ben Azzai and Ben Zoma, 7 that the voice of
God was made a metatron 8 over the waters, according to the
words, 'The voice of the Lord was over the waters' (Psalm 29,
3) . The question, as appears from the preceding context, was
how the waters found their way into the ocean when God
gathered them together in one place; the answer, The voice
of the Lord guided them. The interpretation of Ben Azzai is
cited (independently of Bereshith Rabbah) by R. Berechiah
in Midrash Tehillim on Psalm 93, 3 (§5, end); 'The voice of
God was a metator before them.' The Aruk 9 quotes from
Midrash Yelammedenu on Deut. 2, 31 (Behold I have begun
to deliver Sihon and his land before thee) : 'If that gives thee
concern, I am thy metator. Do not wonder at these words;
am I not hereafter going to be made a metator before an un-
circumcised man, Cyrus, as it is written, 'I will go before thee,
and make the crooked places straight,' etc. (Isa. 45, 2) ; I am
going to go before a woman, before Deborah and Barak, as it is
said, 'Is not the Lord gone out before thee' (Judges 4, 14).

Besides this passage, which is not preserved in our recensions
of the Midrash Tanhuma, the Aruk cites in this sense, from the
same source at the end of the Parashah Ki Tissa (on Exod. 34,
27), where, in answer to the intercession of Moses for the people
after the sin of the golden calf, God recounts his ill-requited
goodness to Israel: 'And not only that, but in the desert I
go before them as metator — 'The Lord goeth before them by
day' (Exod. 13, 21) — levelling down for them the heights and
levelling up the depressions' (cf. Isa. 40, 3f.). 10 In the same
sense, and with the same Scripture reference, we find in a later
Midrash on Exod. 23, 20, 'Behold, I send an angel,' etc.: 11
'God said to Israel, When you were worthy of it, I myself was
made a messenger (shalih) for you, as I did for you in the

64 HARVARD THEOLOGICAL REVIEW

desert, as it is said, The Lord went before them by day
(Exod. 13, 21); but now that ye are not worthy, I turn you
over to a messenger (shalih), as it is said, Behold, I send an
angel,' etc. (Exod. 23, 20.). 12 At the plea of Moses (Exod. 33,
12 ff.), however, the captain (sar, cf. Josh. 5, 4, and below,
p. 65) did not actually assume authority over them till the
death of Moses. Here God going before Israel in the desert is
called shalih, precisely as in the passage first quoted from the
Tanhuma (Yelammedenu) he is called metator; the two words
are equivalent in sense. A third example given in the Aruk,
also from Yelammedenu, is from the Parashah Balak (on
Num. 22, 36: Balak heard that Balaam was come), ' Showing
that they had sent metatorin (plur.) before him.' 13 From these
passages R. Nathan gathers that the idea in metator is, 'pre-
ceding, going on before.' 14 The substitution in our texts of the
Tanhuma on Num. I. c. of sheluhim (lit., persons sent on a
mission or with a message, the Hebrew word represented
in the New Testament by airburokos) is a correct interpreta-
tion from the context. In all the passages thus far cited
metatron or metator — the forms interchange in parallels and
variants — is an appellative; and except in the last it is God
himself (or his finger or his voice) that is the metatron or
metator. In all the context requires some such general sense
as 'one who leads or shows the way, one who goes in advance.' 16
In the Babylonian Talmud Metatron is an angel. The pas-
sages in which he appears are few, and it will not take us too
far to examine them all. In the first of these (Sanhedrin 38b)
R. Nahman (ben Isaac) narrates a controversy between R. Idi
(probably a Palestinian teacher of that name in the latter part
of the fourth century) and a heretic (min), as an example of the
right way to answer such cavils. The heretic quoted Exod.
24, 1, 'And to Moses he said, Ascend unto the Lord,' etc.
Why not, Ascend unto Me? The Rabbi replied: It means
Metatron, 16 whose name is like the name of his master, 17 as the
Scripture says, 'for My name is in him.' 'If that is so, you
should worship him.' 18 'It is written, Do not exchange me for
him.' 'What does it mean then by the words, 'He will not
pardon your transgression?' 19 'In solemn truth! we did not

INTERMEDIARIES IN JEWISH THEOLOGY 65

accept him even as a precursor, 20 for it is written, 'And he
(Moses) said to Him, If Thy presence ("ps) go not (with
us), lead us not up hence' (Exod. 33, 15). Metatron is here
identified with the angel whom God proposed to send before
the Israelites to watch over and protect them in the desert and
lead them to the place God had prepared for them (Exod. 23,
1-4; 32, 34), but whose offices Moses declined — unless God
personally accompanied the expedition, he was unwilling to
set out on it. That Moses did thus refuse to set out under the
conduct of an angel is deduced in the Tanhuma from the same
texts. The same angel was later sent to Joshua (Josh. 5, 13 ff .) ;
he announces himself as the captain of the Lord's host (ibid,
vs. 14), and says: 'Twice have I come to bring Israel into its
inheritance. It was I who came in the days of Moses thy
master, and he rejected me, and was not willing that I should
go; now I am come again.' 21 Substantially the same is
repeated in later compilations; 22 see also Bereshith Rabbah
97,3.

From this survey of the usage of metator and metatron we
may proceed to the question of etymology. That tibbd is
nothing but the Latin word metator written in Hebrew letters
was recognized long ago by both Jewish and Christian scholars,
and metatron was rightly taken to be only another form of the
same word. 23 Thus R. Moses ben Nahman (d. ca. 1270), in
his latest and greatest work, the commentary on the Penta-
teuch, on Exod. 12, 12, identifies the envoy (shalih) from God
to accomplish all that God did in the land of Canaan, with
'the great angel who on that account (sc. as being an envoy)
is called metatron; for the meaning of the latter word is 'one
who shows the way,' as we read in Sifre, (etc. adducing the
passages from Sifre and Yelammedenu quoted above), and
so in many places. And I have heard that 'messenger' in
Latin is metator.' 24 The same derivation is given by Elias
Levita in his glossary entitled 'Tisbe' (1542): 'I have heard
from the cardinal, my pupil, 25 that metator in Latin is a mes-
senger (shalih), and this is perhaps the explanation.' More
exactly, Benjamin Mussafia, in a supplementary note to the
article in the Aruk (ed. Amsterdam, 1655), from his own knowl-

66 HARVARD THEOLOGICAL REVIEW

edge of Latin, writes: 'Metator in Latin is an officer who goes
in advance of an army to select for the soldiers a halting place
and quarters for the night.' Similarly another learned lexi-
cographer, David Cohen de Lara, in his 'Ir David,' 26 a glos-
sary of the foreign words found in rabbinical writings. He
defines metator in Hebrew and Latin as a military 'quarter-
master' (its meaning in Roman law), and adds the Spanish
equivalent, 'aposentador.'

This was also the common opinion of the learned among
Christians in the seventeenth and eighteenth centuries. The
most complete exposition of this, with an almost exhaustive
array of the evidence, is given by Danz, where also other
theories are discussed. 27

Danz presented at length the use and meaning of metator in
Latin. 29 Thus Vegetius gives the military definition : 'Metator
dicitur, qui praecedens locum castris deligit.' Another writer
on the military art describes metator es as 'Antegestores, qui
ante agmen eunt, et loca castris metandis idonea deligunt, et
viam, qua duci exercitus commode possit, vident.' From
Christian authors 30 are quoted Optatus, Contra Donatistas,
iii: Cum ante venturos milites metatores, ut fieri adsolet,
mitterentur. Cyprian, Epist. 6, 4, exhorts his readers to fol-
low the courageous example of the presbyter and confessor
Rogatianus, qui . . . primum hospitium in carcere prae-
paravit et metator quodammodo vester nunc quoque vos
antecedit. See also Cyprian, Epist. 22, 1 : Nam tu Deo volente
ipsum anguem majorem metatorem Antichristi . . . deter-
ruisiti. Ambrose, Exameron v, 10: Quis imperator piscibus
praeceptum dedit, quis doctor hanc tribuit disciplinam, qui
metatores itinera disponunt, qui duces iter dirigunt, ut nullius
desit occur sus? 31

Metator thus had an evolution closely parallel to the Eng-
lish 'harbinger'; S2 'One sent on before to purvey lodgings for
an army, a royal train, etc.; ... a pioneer who prepares the
way. One that goes before and announces the approach of
some one; a forerunner.' 33 And if 'harbinger' were not in
modern English so bookish and so predominatingly figurative,
it would be the best rendering for the Hebrew metator, me\atron.

INTERMEDIARIES IN JEWISH THEOLOGY 67

The latest authors to deal with the subject at length, Oester-
ley and Box, 'cannot agree with the writer on this subject in
the Jewish Encyclopedia [Ludwig Blau] when he says that
'the derivation from the Latin metator ( = 'guide') is doubtless
correct.' They have two objections: First, it is Elisha ben
Abuyah who first refers to Metatron under this name; the
belief regarding Metatron must consequently have been much
earlier than his time (first half of the second century); so
early a date makes it improbable that the word is derived from
the Latin, for Roman influence upon Jewish literature is not
likely to have been strong enough to lead Jewish teachers to
adopt a Latin word; a Latin derivation is all the less likely
because the word first occurs in the Babylonian Talmud.
Second, in Latin metator means 'divider,' or 'measurer,' 34 not
'guide.' 'It would be difficult to point to any instance of the
Latin word being used in this sense.' As regards the latter
point it is sufficient to refer to the 'instances' quoted above,
which are not ' difficult to point to,' inasmuch as since the end
of the seventeenth century they, with others, e.g. from the
codes and the civil lawyers, have stood in the dictionary to
which a scholar would first go with such a question, and are all
cited in full by Danz. The first objection is equally baseless.
That ' it is Elisha ben Abuyah who first refers to Metatron under
this name,' is a complication of errors. 36 The story about the
origin of Elisha's heresy will be discussed later. Suffice it here
to say that in that story Elisha does not 'refer to Metatron
under this name' at all. That the occurrence of the word in
a Babylonian story of uncertain date about Elisha is proof that
the word and idea were current in the age and environment of
the hero of the story 36 is a kind of inference that might have
curious results: for example, the translated Enoch is called
Metatron, therefore the word and the idea are older than the
flood. And finally, the probability or improbability of Latin
words having found their way into rabbinical Hebrew or the
vernacular Aramaic is not to be decided by what the authors
deem antecedently probable, but by reading the literature;
in default of which the special glossaries to words borrowed
from Greek and Latin might profitably be consulted — Krauss,

68 HARVARD THEOLOGICAL REVIEW

for example. In the particular case before us it should be ob-
served that from the time of the Roman occupation of Syria
Latin military terms — and such metator is — for which there
was no exact equivalent in Hebrew or Aramaic were adopted
with especial frequency. The authors must have forgotten
that in the Gospels not only Jesus but the poor demoniac use
the Latin legio.

The derivation from the Latin metator did not yield a sense
that seemed adequate to the rank and functions of Metatron
in mystical and cabalistic writings; and, assuming that the
name must have been coined to express his exalted station in
that literature, scholars sought for etymologies corresponding
to their interpretation of the figure. 37

In cabalistic vein R. Bahya ben Asher (Behai) in his com-
mentary on Exod. 23, 21, 38 finds in the word Metatron two
meanings, 'lord' and 'messenger,' deriving the former from
the rabbinical (really Latin) matrona 3 * and the latter from
the Greek, in which a messenger is called mentator (sic!). 40 For
good measure, he offers a third derivation for a piece of the
word: matrat stands in the Targum for Hebrew shemirah
('keeping, protection'), 'and because he is the keeper of the
world he is called the keeper of Israel.' By inverting the pro-
cess, he deduces circularly from the etymology that Metatron
is the lord of all beings of lower rank, for all the host above
and below are in his authority and under his power; and he
is an envoy (messenger) of Him who is over him, and higher
than he is He who gave him dominion over the universe and
appointed him lord of His house and manager of all His posses-
sions (cf. Gen. 24, 2). The Dta in the middle of his name are
numerically 18, and thus equivalent to v\ ('living'). It is no
wonder that Christian scholars found in Bahya's Metatron all
they were looking for.

The etymology which in recent times has enjoyed the most
approbation derives the name from ixtra and Opbvos. The
merit — whatever it is — of being the inventor of this pun is
frequently attributed to some modern who has repeated it
without due credit to his predecessors. 41 The first to propose
it, so far as I know, was J. H. Maius, Professor at Giessen, in

INTERMEDIARIES IN JEWISH THEOLOGY 69

his Synopsis Theologiae Judaicae (1698), p. 72. He submits
it to the judgment of the learned as a modest conjecture,
'an non commodius longe ac vulgo fit (sc. from metator) ex
Graecis vocibus pera et Opovov deduci queat, ut innuatur
Angelus crwdpovos Dei, seu ejusdem throni, majestatis et
gloriae cum Deo Patre participi,' etc. Hengstenberg 42 cites
Maius, and, of more recent authors, Joh. Fried. Meyer, 'Blat-
ter fur hohere Wahrheit' (Sammlung iv (1822), p. 168). 43
Hengstenberg rejects the derivation for the very good reason
that per&Opovos is not even a Greek word. Gfroerer, however,
into whose Alexandrian theosophy in Palestine such a divine
assessor fitted as well as into the old orthodoxy, and who was
not deterred by philological scruples, accepted the etymology;
Metatron is the being who is nerd, top dpbvov Oeov. Inasmuch
as the rabbis adopted into their vocabulary both irapedpos and
avyicadeSpos, and if these did not satisfy them, could as easily
have borrowed awdpovos, it is not clear why they should have
taken the trouble to invent peradpovos. Modern authors who
maintain this derivation are bound to attempt some explana-
tion of the second { in metatron — for Greek theta in that age
we should expect Hebrew tau.

The last named difficulty is escaped by another etymological
figment; metatron is perb+rvpavvos (peraripavvov) , a factitious
word which is defined, 'one who stands next in rank to the
ruler.' 44 It is a further objection — if any other is needed —
that in Hebrew the borrowed words rlpavvos, rvpavvia seem
to be uniformly spelled 'Tti.

Another etymology, about the priority in which there seems
to be some rivalry, discovers in jnoD'O (pronounce, Mittron)
the name of the god Mithra. Its most notable advocate was
Alexander Kohut. 45 Hamburger (Real-Encyclopaedie fur
Bibel und Talmud, II, 781) enumerates several predecessors,
beginning with Fried. Nork, 46 'Brahminen und Rabbinen,'
1836. The honor seems to belong, however, to a Christian
scholar, Heinrich Ed. Schmieder, who propounded the theory
in an excursus to his 'Nova Interpretatio loci Paulini Galat.
iii. 19-20' (1826; pp. 41-48). 47 Schmieder briefly recites and
despatches the older attempts on the word as he found them

70 HARVARD THEOLOGICAL REVIEW

in Danz, as well as Meyer's neraBpovos, and then proposes
his own solution: Mittron or Mettron is Mithras. He tries
to show that the Jews were capable of disguising the name in so
remarkable a manner — which, in view of other achievements,
can not be unqualifiedly denied — but is chiefly moved by the
remarkable agreement he finds between the character and func-
tions of Mithras and the Jewish 'Mittron,' dwelling particu-
larly, as might be expected, on Mithras as /woItijs, 48 'mediator,'
on which aspect of his nature Creuzer had expatiated. Schmie-
der then develops with considerable ingenuity the theory that
these Persian doctrines were introduced and cultivated among
the Jews by the Essenes, comparing the teachings and obser-
vances of the latter as described by Philo and Josephus with
accounts of Persian customs.

Retracing our steps from this excursion into the vagaries of
etymology, two passages in the Babylonian Talmud remain to
be examined. One of these is the story about Elisha ben
Abuyah to which allusion has already been made. A second cen-
tury tradition (baraita) preserved in both Talmuds, 49 tells, in a
few obscure words, of four eminent teachers of theosophical lean-
ings in the generation before Hadrian — Ben Azzai, Ben Zoma,
'Aher' (Elisha ben Abuyah), and Akiba — who 'entered Para-
dise,' 60 and of the disastrous effect what they saw there had
on the first three: one of them gazed and died; another gazed
and was stricken with madness; Aher gazed and 'cut down the
plants'; Akiba alone came off unscathed. The cryptic phrase,
'cut down the plants,' is explained in the Jerusalem Talmud
as meaning that he visited the schools and persuaded youths
to abandon the study of the law, and take to such trades as
those of builders, carpenters, hunters, tailors. The 'plants'
are thus the young scholars in the rabbinical seminaries. In
the Babylonian Talmud a different story is told: 'Aher saw
(in Paradise) Metatron, to whom permission had been given
to be seated while he recorded the good deserts of Israelites.
Whereupon Aher exclaimed, We have been taught that in
heaven no one (except God) sits .... Can it be that there
are two supreme powers!' Metatron was chastised with sixty
lashes of fire 51 for having given occasion for such an error by

INTERMEDIARIES IN JEWISH THEOLOGY 71

not rising; he was directed to erase from the record the good
deeds of Aher; and a heavenly voice was heard saying, 'Re-
turn (repent), ye apostate children (Jer. 3, 14) — except
Aher!' 62

The remaining mention of Metatron in the Babylonian
Talmud is in Abodah Zarah 3b, according to which among
God's regular occupations in heaven is to give religious in-
struction for three hours — the last quarter of the day — to
children whose early death has deprived them of opportunity
to study the Torah on earth. 53 If it be asked, Who teaches
them in the earlier part of the day? it may be answered, Meta-
tron. In both these places Metatron has a special office in
heaven in relation to Israelites, the recorder of their good works,
the teacher of children who died in infancy. These are offices
that might be performed by any angel appointed to them, and
there is nothing in either context to suggest that Metatron was
a being of a different nature or one who stood in a peculiar
relation to God; nor that he was in any sense a 'mediator'
between God and Israel. An examination of these passages —
the only places where the name occurs — shows that the be-
ginnings of the Metatron mythology, if so it should be called,
in the Babylonian Talmud are extremely modest. 64

A higher rank and larger functions are attributed to Meta-
tron by recent writers on the subject on the ground of certain
titles which are said to be given him. Thus Weber: 'InHullin
60a and Yebamot 16b he bears the name D^iyn 1K>, Prince
of the World; he represents God's sovereignty (Herrscherstel-
lung) in the world.' Oesterley and Box, confidingly following
Weber, refer to 'two passages, one from the Babylonian and
one from the Jerusalem Talmud, in which Metatron bears the
title of Prince of the World; a title which more probably
implies that he is the representative of God in the world.' 65
As I pointed out in a former article (Vol. XIV, p. 237), Weber
apparently fell into his error by a careless reading of Levy,
Chaldalsches Wbrterbuch II, 231, who refers to the Tosafot
on the two places cited, where is found, however, not an identi-
fication of the sar ha-'olam with Metatron, but a discussion of
a difficulty which RabbenuTam (d. 1171) discovered in certain

72 HARVARD THEOLOGICAL REVIEW

inconsistent expressions about Metatron in two mediaeval
synagogue hymns. The question is, how could Metatron, ac-
cording to one of them the translated Enoch who from flesh
was turned to fire, be the sar ha-'olam with whom the other
seems to identify him, when according to Hullin I.e. the sar
ha-'olam sounded the praises of God at the creation (Psalm
104, 31), generations before Enoch was born? The outcome
does not concern us; the important thing is that if there had
been any Talmudic authority for the identification of the sar
ha-'olam with Metatron there would have been no room for
the discussion. It may be observed also that in Sanhedrin 94a,
sar ha-'olam is interpreted by Rashi simply as an angel to
whom the whole world is committed, without any hint of
identification. The Responsum quoted below (note 73) iden-
tifies the angel of Gen. 48, 16 and Isa. 63, 9 with Metatron
'whom the rabbis call the prince of the presence,' but makes
no mention of the greater title, 'prince of the world.'

Sar ha-panim, a frequent title of Metatron in post-Talmudic
literature, is most simply explained as a breviloquence, the
prince, or chief, of the class who are called ' angels of the pres-
ence,' that is those who have immediate access to God's pres-
ence, like the principal ministers of a monarch who custom-
arily attend in his presence. Cabalistic speculation, however,
made it 'the prince who is the Presence' and taught that this
is meant in Exod. 33, 14, 'My Presence shall go with thee'; so
also in 'the angel of His Presence delivered them' — the angel
who is His Presence. 57

Metator, Metatron, was, as has been shown, originally an
appellative, in meaning and use corresponding closely to the
English 'harbinger.' In the three places where it is found in
the Babylonian Talmud it is the name of an angel. It is, how-
ever, a name of unusual type, 58 for the names of angels are
generally compounds containing the word el, 'God,' after the
pattern of the biblical Gabriel and Michael. It is therefore a
pertinent question whether the angel who is designated in the
Midrash by the appellative and named in the Talmud by the
appropriated appellative — if it be the same angel — is one
who is otherwise known to us under a proper name.

INTERMEDIARIES IN JEWISH THEOLOGY 73

We have seen that R. Idi, 59 in his answer to the heretic, de-
clares with emphasis that so far from worshipping Metatron, the
Israelites would not accept him even as a 'precursor' when God
offered to send him before them to guard them on their way
and guide them to the place prepared for them (Exod. 23, 20).
It was this angel, who after the death of Moses appeared to
Joshua (Josh. 5, 13 ff.). 60 As 'the captain of the Lord's host,'
he was identified by mediaeval commentators 61 with Michael
who is 'one of the chief princes,' 'your prince' (Dan. 10, 13, 21),
'the great prince' (ibid. 12, 1), the champion of the Jews. 62 Ac-
cording to others, it was Michael who led the Israelites through
the desert.

At an earlier point in the same discussion Rabbi Idi answers
the question, Who is the subject in the words, 'Unto Moses
he said, Ascend unto the Lord'? by saying, 'The speaker is
Metatron, whose name is like the name of his master.' The
same exegetical difficulty is discovered by the Palestinian
Targum on Exod. 24, 1, and resolved: 'Unto Moses, Michael,
the prince of wisdom, said, on the seventh day of the month,
Go up before (into the presence of) the Lord.' 63 At the death of
Moses, when God is lamenting his loss, asking who now will
intercede for Israel when they sin, Metatron came and fell on
his face, and said, Lord of the World, in his life Moses was Thine
and in his death he is Thine.' 64 In the parallel narrative in
Midrash Mishle, 'Michael came and prostrated himself before
God, and said, Lord of the World, in his life he was Thine,
and in his death he is Thine.' 65

Metatron appears in a somewhat similar r61e in one of the
collections of proems, or introductions, for the use of synagogue
preachers which are prefixed to the old Palestinian Midrash on
Lamentations. God was mourning over the destruction of the
temple; he had no longer a dwelling place on earth. 66 'In that
hour Metatron came and fell upon his face and said, 'Lord of
the World, I will weep, but Thou shalt not weep.' God replied,
'If thou do not let me weep now I will enter into a place into
which thou art not permitted to enter, and will weep (there),
as it is written, 'If ye will not hear it, My soul shall weep in
secret for your pride' (Jer. 13, 17).' 67

74 HARVARD THEOLOGICAL REVIEW

Even in late mystical texts the same functions are attributed
to Metatron and to Michael. The most striking example of
this is the presenting offerings in the celestial sanctuary. In
the Talmud this is the office of Michael — 'Michael the great
prince stands and offers upon the altar.' 68 In one recension of
the Seder Gan Eden, a mediaeval work, we read that in the
highest heaven (Araboth) the great prince Michael stands,
with an altar before him, and offers upon the altar the souls of
the righteous. 69 In another mediaeval Midrash this office is
performed by Metatron. 'R. Simeon said, In the hour when
God bade Israel to erect the tabernacle he made a sign to the
ministering angels that they should make a tabernacle, and at
the time when it was erected below, it was erected above. 70
This is the tabernacle of the Youth whose name is Metatron
in which he offers the souls of the righteous to atone for Israel
in the days of their exile.' 71

In various places, particularly in Palestinian sources, we
thus find the name Michael where parallel passages have
Metatron. In explanation of this fact the following hypothesis
may be advanced. The word metator, or metatron, as an appel-
lative, meaning one who leads the way, was first used of God
himself, particularly in reference to the migration of Israel
from Egypt to Canaan, or of the angel whom he commissioned
to guard and guide them in the way and bring them to their
destination — whether it was thought that he actually led
them through the desert, or that at Moses' petition his com-
mission was suspended, so that he did not assume his leader-
ship until after the death of Moses, when he announced him-
self to Joshua before Jericho as the captain of the Lord's host.
If it was asked who this angel leader was, the inevitable answer
would be Michael, the captain ('prince') whom God had ap-
pointed over his people, their champion and protector. 72

All the offices and function of the angel Metatron in the older
sources, and even in the Babylonian Talmud, are such as might
naturally be ascribed to the guardian angel of the Jews. 73 The
Metatron whom Elisha ben Abuyah saw in heaven was sitting
(by special privilege) and recording the good deserts of Israeli-
tes, an appropriate occupation for their special patron; and

INTERMEDIARIES IN JEWISH THEOLOGY 75

no other angel could more appropriately share with God him-
self the task of instructing in religion the little souls whose
early death had deprived them of human teachers. So again,
when God would weep over the destruction of the temple, the
words, 'I will weep, but do not thou weep,' have a fitness and
a force in the mouth of the angel to whom Israel was committed
as his peculiar charge; while they lack this point altogether if
supposed to be spoken by some mythical associate divinity.
Similarly, when God, bewailing the death of Moses, asks who
now will stand between the people and His righteous indigna-
tion, it is most fitly the angelic advocate of Israel who reminds
Him that Moses is still His; that is, I take it, that Moses may
intercede for Israel in heaven as he had done on earth. Even in
a late apocalyptic book, the Sefer Zerubbabel, Metatron says:
'I am the angel who led Abraham in all the land of Canaan;
it was I who ransomed Isaac and wrestled with Jacob at the
ford of Jabbok; and I that led Israel in the wilderness forty
years in the name of the Lord; I who appeared to Joshua at
Gilgal. I am he whose name is like the name of his master, and
his name is in me.' 74 These are things that are regularly as-
cribed to Michael, and in the text itself the name Michael
slips in for the angel who makes the revelation to Zerubbabel
and is otherwise consistently called Metatron. 75 Metator was
a foreign word; the precise technical meaning which so well
suited the earlier contexts in which we find it used easily passed
in ordinary apprehension into the vaguer sense of 'one who
leads, or shows, the way,' which is current in Jewish commen-
tators and lexicographers. In Babylonia especially, where
Roman military terms were not familiar as they were in Pales-
tine, it was naturally taken for a proper name instead of an
appellative, the proper name of an angel; though even there
the association with Michael was not wholly forgotten. 76

Detached from the original connection with the exodus,
Metatron became more exclusively a celestial figure, such as
he is in the story of Elisha ben Abuyah's fall, or when he is the
teacher of children in the school of heaven. This is as far as
the Babylonian Talmud goes. As sar ha-panim he is the chief
of the 'angels of the presence' who have immediate access to

76 HARVARD THEOLOGICAL REVIEW

God. In the succeeding Gaonic period, however, there was a
notable revival of curiosity about the mysteries of heaven and
earth such as at a much earlier time gave origin to apocalypses
like Enoch, with a corresponding recurrence to apocalyptic
tours through the heavens under angelic conduct; and, as in
the earlier time, with its mystery of creation and its chariot
speculation, a theosophic motive enters into these adventures
— it is one of the main sources of the older Cabala.

In this literature, which, characteristically enough, produced
new Enoch books and descriptions by the mystical travellers
of the heavens that rise one above another to the highest and
the throne of God itself, with the angelic hierarchy that in-
habits them, Metatron has a prominent place. It does not
fall within the scope of our present investigation to accompany
Metatron through this literature, much less to pursue him into
the later Cabala. On one point, however, a word must be said.
It has been argued above that the angel Metatron was origi-
nally Michael. In the writings now under consideration Meta-
tron is commonly the translated Enoch. 77 What led to this
identification can only be conjectured. The most probable
hypothesis is perhaps that it originated in the occupation of
Metatron in the story of Elisha ben Abuyah, where he appears
as a recording angel; 78 for a similar function is attributed to
Enoch in the Book of Jubilees 4, 23: 'He was taken away from
among men, and we conducted him into the Garden of Eden
(Paradise) with majesty and honor, and behold he writes there
the judgment and the sentence upon the world, and all the
evil deeds of mankind.' 79 So also Enoch (in heaven) is ad-
dressed, 'Enoch, thou scribe of righteousness' (Enoch 12, 4;
15, 1). To the same office the Palestinian Targum on Gen. 5,
24 probably refers: 'Enoch worshipped in truth before the
Lord, and he was not found among the inhabitants of the
earth, for he was rapt away and ascended to the firmament by
the word of God, and his name was called Mitatron, the great
scribe.' 80

Inasmuch as flesh and blood have no place in that upper
world, God transformed Enoch's body into the fiery matter of
which angels are made. 81 He prepared him a throne just like

INTERMEDIARIES IN JEWISH THEOLOGY 77

His own, covered with similar tapestry, and caused to be pro-
claimed that He had appointed him prince and chief of all the
princes of His kingdom and all the celestial beings except the
eight honored and revered princes who are called Ihvh, by His
own eternal name. Every angel who has anything to communi-
cate with God is to come to Metatron and speak to him, and
whatever he bids them in God's name they shall observe and
do, because he is the prince of wisdom and of intelligence, which
minister to him and instruct him in heavenly and earthly
wisdom, the wisdom of this world and the mystery of the world
to come. He is also appointed over all of the temple on high,
and over all the storehouses of life in the highest heavens.
Metatron-Enoch has seventy names corresponding to the
seventy languages of earth, but God himself calls him 'Youth'

This name, Youth, is explained by Metatron himself: it was
given him by the ministering angels over whom he was set,
because he was a junior among them and a youth (in comparison
with the angels of the original creation). 82 It is this late assump-
tion into the angelic ranks that is urged in the Tosafot
against the identification of Metatron-Enoch with the Tal-
mudic sar ha-'olam. 83 If a modern conjecture, at variance with
the explanation above put into the mouth of Metatron, is
legitimate, the suggestion may be ventured that 'Enoch, the
Youth,' has no more recondite origin than a purely verbal
association with Prov. 22, 6 (on 'B i>y -iju!> -pn). It is easy
to see how convenient these lucubrations were to Christian
scholars in search of a Jewish counterpart to the Second Person
of the Trinity. By way of the equivalence of the uncreated
Metatron and Shekinah which the cabalists offered them, they
found what they required; and the more concrete Metatron
answered their purpose even better than the (etymologically)
abstract Shekinah. 84

By a similar equation they found the Messiah in Metatron.
The Jews recognized Metatron in Jacob's angel 'redeemer'
(Gen. 48, 16), the 'angel of His presence' who saved God's
people (Isa. 63, 9), the 'angel of the covenant' who was sud-
denly to come to his temple (Mai. 3, 1), and, in general, in the

78 HARVARD THELOOGICAL REVIEW

Angel of the Lord, whose personality so often seems to merge
into that of the Lord himself. Christians identified this angel
with Christ in his office under the old dispensation — the
Messiah designate, we might say. Thus, by the axiom that
things which are equal to the same thing are equal to each
other, it was proved that Metatron was the Messiah.

The Christian authors who maintain this deceive themselves
by a fallacy of equivocation. There is no evidence that the
Jews associated either the Angel of the Lord or Metatron with
the Messiah. Hengstenberg, indeed, avers that 'the identity of
the Angel of Jehovah, or Metatron, with the Messiah was
recognized even by the later Jews'; but, besides the fact that
in the New Testament this 'is assumed to be universally ac-
cepted,' he adduces only a single passage from the Zohar
(through Sommer, p. 35), 86 where there is nothing whatever
about the Messiah, but in which he finds a prediction of the
incarnation of Metatron in a mother's womb 86 — certainly a
singular mark by which to recognize a Jewish reference to the
Messiah!

To summarize the results of this wide-ranging and in part
intricate investigation: 1. Metator (metatron) is originally an
appellative, the Latin metator, borrowed and first used in its
proper, almost technical, sense, an officer who goes in advance
of an army to choose and mark out the site of a camp, and to
explore and indicate the route by which the halting place is to
be reached. Israel's metator in the desert was God himself, or
an angel assigned and commissioned by him to this task; this
office was most naturally filled by Michael, the champion of
the Jews. 2. In two passages in the Babylonian Talmud Meta-
tron is the proper name of an angel whose office in heaven indi-
cates a peculiar relation to Israel and interest in them; and in
this stage the same offices, notably the ministry at the celestial
altar with the offering of the souls of the righteous, and the
same predicates, are assigned in different sources, now to
Michael, now to Metatron. 3. In the revived apocalyptic and
cabalistic literature of the Gaonic period and after, the trans-
lated Enoch becomes Metatron; his earthly body is transmuted
into fire, and he takes his place among the angels, over whom

INTERMEDIARIES IN JEWISH THEOLOGY 79

he is advanced to the first rank and supreme rule, thus taking
the place held in the older angelology by Michael. Theosophic
speculation seizes upon this angelic mythology, and elevates
Metatron to a still higher eminence, until as we have seen, he
— more properly, it — is identified with the Shekinah; it is an
emanation, not a creature; and, as the 'middle column,'
unites the four worlds that are superimposed in stages (the
worlds of emanations, of creative ideas, of creative formations,
and of creative matter), 87 etc.

In all this, from the metaphor in which he begins to the meta-
physical myth in which he ends, whatever else Metatron may
be or do, whether he is an individual created angel or an emana-
tion from the Absolute, he is neither in function, nor in essence
an 'intermediary,' or 'mediator,' in the sense in which that
word is generally understood and in which it is intended by
those who write about him in that category. As if the cabalis-
tic myths were not fantastic enough, Christian theologians
have added to them their own, at first to claim him for their
Christology, latterly to discredit Judaism with him.

80 HARVARD THEOLOGICAL REVIEW

NOTES — II. METATRON

1. Miscellanea Sacra, lib. i, c. 17, § 7. For a modern instance, see p. 70,
and below, n. 71.

2. See D. Hoffmann, Zur Einleitung in die halachischen Midraschim,
1887.

3. Another reading is, 'The finger of God was made a metatron,' etc. So
quoted by R. Moses ben Nahman on Exod. 12, 12; Eshtori Parhi, Kaftor
u-Perah, f. 49b (ed. Berlin, 1852, f. 34b).

4. Compare R. Akiba (Sifre on Num. 27, 12, § 136), 'God showed Moses all
the divisions (lit., compartments) of the land of Israel like a table laid out.'

5. The Rabbis named, Eliezer ben Hyrcanus and his most frequent op-
ponent Joshua ben Hananiah, were disciples of Johanan ben Zakkai and
flourished in the generation after the fall of Jerusalem. In Sifre on Num. 27,
12 (§ 136) it is the latter interpretation that is ascribed to Eliezer. See Bacher,
Agada der Tannaiten, I, 2 ed. 148.

6. Compare also Yalkut in loc. (§ 949). See the explanation in Pesikta
Zutarta cited in Friedmann's note on Sifre, Deut. loc. cit.

7. Contemporaries of Akiba, in the first third of the second century. See
below, p. 70.

8. J11I3DD. Variants (Theodor, p. 34) : IlKJtOD, "llD^D, al. In the editio
princeps and those that follow it: R. Levi said, Some interpreters interpret
with Ben Azzai and Ben Zoma, The voice of God was made a metatron to
Moses in the hour when He said to him, Go up to Mount Abarim (Deut. 32,
49 — it guided him to the unknown spot where he was to die and be buried).
The voice of God was made a mefafron over the waters, etc.

9. The first great lexicon of the language of the Talmud and Midrash.
compiled by Nathan ben Jehiel of Rome (d. 1106). This lexicon contains
many quotations from works no longer extant, and many readings in extant
works representing an earlier and often better text than our editions.

10. Tanhuma, Ki Tissa, loc.

11. On this passage, which is often referred to in the following pages,
there is a learned monograph by Joh. Jac. Cramer, Custos Israelis, seu Dis-
sertatio philologico-theologica in Exodi cap. xxiii. v. 20, 21, 22, 23, qua
Angelum Israelis tutorem unigenitum Dei filiuin, et verum aeternumque
Deum esse solide, etiain ex Hebraeorum consensu demonstratur. 1705. Cf.
also his Dissert, in Exod. xxxiii. 1-6; 12 ff., xxxiv. 5-10.

12. Shemoth Rabbah on Exod. 23, 20 (edit. Wilna, 1884, c. 32, 2, f. 60b).

13. So Aruk, edit. prin. (see Kohut, s. v.); later editions, metator (sing.).
Tanhuma, Balak § 10: 'Showing that he (Balak) had sent messengers
(sheluhim) to apprise him' (of Balaam's approach); ed. Buber, § 14, 'mes-
sengers to Balak to apprise him.'

14. He suggests a possible etymology in a different sense, from Aram. "1133,
'keep, protect,' equivalent to Hebrew ~SQW (cf. Exod. 23, 20).

15. See also Yalkut, Gen. § 44 (from Midrash Abkir): 'God despatched
Metatron as a messenger (shalih) to Shemhazai,' one of the angels who fell.
It is possible that shalih is a gloss. Cf. Jellinek, Bet ha-Midrash IV, 127 f.,
and the Book of Enoch, 12, 4 ff.

INTERMEDIARIES IN JEWISH THEOLOGY 81

16. The words are ambiguous: Was it the Lord who bade Moses ascend
to Metatron, or Metatron who bade him ascend to the Lord? The cabal ists
were divided on this question (Recanati, quoted by Danz, pp. 735 f .) ; modern
scholars equally. Bacher takes the former alternative; Kohut the latter.

17. See below, n. 72.

18. I.e. do not exchange (confound) me with him; taking 13 IDJl ?N 'do not
rebel against him' as u 'JTOn ^>N (cf. Jer. 2, 11). The same explanation is
given in Shemoth Rabbah 32, 4. where a commentator correctly remarks,
'Do not imagine that he is God.'

19. The heretic's argument is that 'he will not' implies that he has the
power to do so, a power that is a divine prerogative.

20. XpOlllQ, a Persian word found elsewhere in the Talmud in the sense
of 'precursor, courier, messenger,' corresponding thus in general sense with
metator, metatron, in the Palestinian sources discussed above. R. Hananel
(on Sanh. loc. interprets TTl ('scout').

21. Tanhuma, Mishpajim, § 18 init.; ed. Buber § 10.

22. See especially Agadath Bereshith 32 (ed. Buber, p. 64 f.).

23. The ending on is appended to many foreign words where it does not
properly belong. See S. Krauss, Griechische und Lateinische Lehnworter, I,
192. With the doublet compare "YirJD and jniroD (awrryopos) .

24. 'The language of Javan,' might be Greek or Latin. These Rabbis seem
to make no difference. The words of Ramban are translated in full by Danz,
pp. 723 f .

25. Egidio of Viterbo, general of the Augustinian order, in whose house
in Rome Elias lived for thirteen years (1513-27), as the cardinal's teacher,
especially in the Cabala.

26. Subtitle (Latin): Sive de convenientia vocabulorum rabbinicorum
cum Graecis et quibusdam aliis linguis Europaeis (1638).

27. J. A. Danz, Schekina cum piis cohabitans ad Joh. xiv, 23, Progr.
iii-iv. (In Meuschen, Novum Testamentum ex Talmude illustratum (1736),
pp. 721 ff.) Cf. also Joh. Jac. Cramer, Dissert, in Exod. xxiii, 20-23, pp.
103 ff. (1705).

29. Largely drawn — with due credit — from Du Cange, Glossarium ad
scriptores mediae et infimae Latinitatis (1678).

30. These are the examples cited by Du Cange and from him by Danz.
What I have done is to take the text from modern critical editions, and,
where it seemed to conduce to the understanding of the passage, to quote a
larger context.

31. Peter Chrysologus calls John the Baptist, Metator Domini.

32. Related to German 'Herberge,' originally, 'army quarters.'

33. Murray, New English Dictionary, s. ».

34. The authors evidently imagine that metator is derived from metiril

35. The Jewish Encyclopedia (VIII, 519 — Ludwig Blau), which is
alleged as authority for the statement, gives no excuse for the misunderstand-
ing, nor does Hagigah 15a (not '15b'), which the authors cite but have evi-
dently not read.

36. That the word is really as old as the beginning of the second century
appears from the Palestinian sources quoted above. Why if, as the authors
tell us, Elisha ben Abuyah (a Palestinian) used the word, a Latin derivation

82 HARVARD THEOLOGICAL REVIEW

is more improbable because the story as told in the Babylonian Talmud is
inscrutable.

37. The propensity of dilettanti for displaying their ingenuity and their
acquaintance with foreign dictionaries in combinations which justify the
ancient jibe, 'Etymology is a science in which vowels count for nothing and
consonants for very little,' is to be taken into account.

38. In ed. Amsterdam 1726, which I have used, f. 114a.

39. Matrona, Matronita, is a contribution from the Cabala. By a simi-
lar etymological path Levi ben Gerson (d. 1344), on Prov. 1, 8, 'Forsake not
the teaching of thy mother,' identifies metatron with the sekel ha-po'el (active
intellect) of his Aristotelian philosophy.

40. I.e. Latin metator. The n is Bahya's own addition to the confusion.
Bamban, whom he is copying, has correctly metator. Buxtorf, who quotes
the passage at length in his lexicon, makes the impossible conjecture that the
author was thinking of firjvvrcop (poetic for nr)wH]s), which he renders by
nuncivs; in reality it means an informer (delator). Gfroerer charges this
etymological juggling to Bahya. Danz guessed mandatarius!

41. Oesterley and Box express themselves as if this explanation originated
with Weber.

42. Christologie, 2d ed. III. 2, p. 79.

43. 'Mitatron ist namlich der Mitthroner Gottes, 6 fikroxos tov dpovov,
6 ovvQpovos, oder der Herr, der zur Rechten des Herrn sitzt (Ps. ex. 1), der
Sohn, der mit dem Vater auf seinem Throne sitzt (Apoc. iii, 21).'

44. Oesterley and Box, who, after Weber, give their readers the choice
of Metathronos and Metatyrranos, remark: ' We cannot, however, follow
Weber when he speaks of the analogy of the Crown Prince,' etc. This
'analogy' is not Weber's; it is a gratuitous ineptitude of Schnedermann in
the second edition, who to Weber's 'der nachste nach dem Herrscher' adds
' (gleichsam dem Kronprinz).'

45. 'Ueber die jiidische Angelologie und Damonologie' (in Abhandlungen
fur die Kunde des Morgenlandes, IV. 3 (1866), and separately, pp. 36-42.
Also in his Aruch Completum, V (1889), 119 f.

46. Pseudonym of Selig Korn.

47. An anniversary publication of Schul-Pforta, in which Schmieder was
a professor. To this rather rare pamphlet I was directed by a reference in
Hengstenberg.

48. Plutarch, De Iside et Osiride, 46. Many others have been captivated
by the word utairqs, ignoring Plutarch's own interpretation, a middle
thing between God and the Devil, or between light and darkness, or putting
conjectures of their own in its place.

49. Tos. Hagigah 2, 3; Hagigah 14b-15; Jer. Hagigah, ii, l(f. 77b).

50. In a rapture (cf . Paul's account of such an experience, 2 Cor. 12, 1-4 :
he was 'rapt into Paradise and heard unutterable things, which it is not
permitted a man to speak'). Such experiences are meant to be understood
literally.

51. Since the bodies of angels are constituted of fire, this is the natural
form of castigation.

52. On this passage see L. Ginzberg in the Jewish Encyclopedia, V, 138 f.;
also Tosafot on Hag. 14b.

INTERMEDIARIES IN JEWISH THEOLOGY 83

53. Deduced from Isa. 28, 9.

54. In the Jerusalem Talmud there is none of it.

55. Op. cit. p. 172. And on the next page (with the same references,
Hullin 60a; Yebamot 16b): 'His function of representing God is perhaps
seen most distinctly in the title that is given him of the 'Prince of the World'
(sar ha-'olam), which shows that he was thought of as the ruler of the world.'
That one of these passages is in the Jerusalem Talmud is an original discovery.
All that is said about the sar ha-'olam in the places cited is that he uttered
certain verses of Scripture on certain occasions. In Hullin, when God said,
'after its kind,' of the trees (Gen. 1, 11), the sar ha-'olam said, 'May the glory
of the Lord be forever; let the Lord rejoice in his works' (Psalm 104, 31).
What prompted him to this ejaculation was that the grasses and herbs, notwith-
standing that God did not say of them 'after its kind,' argue a fortiori that
he could not mean them to be all mixed up, and accordingly appeared by
species like the rest (vs. 12). In Yebamot the words of Psalm 37, 25, 'I
have been young and now am old (yet have I not seen the righteous forsaken,'
etc.), were uttered by the sar ha-'olam. upon occasion of the calamities de-
scribed in Lam. 5, 11, 'They have ravished the women in Zion, the maidens
in the cities of Judah.' In neither place does the name Metatron occur, nor
anything that remotely suggests him. Neither has the sar ha-'olam here any
symptom of a ruler of the world, or 'a kind of demiurge' (Levy). Rashi's
laconic gloss in 'an angel.'

56. Glosses by mediaeval French rabbis after Rashi. See on Yebamot
16b (pIDS); Hullin 60a (same catchword).

57. Thus, for example, Bahya ben Asher on Exod. 23, 21; cf. on Exod.
33, 14.

58. Sandalphon (owd5eX<£os) is similar, but occurs only in one place in
the Talmud (Hagigah 13b); elsewhere only in late writings, and never in
Palestinian literature.

59. Above, p. 64 f .

60. Above, p. 65.

61. So Rashi (taking 'the Lord's host' as Israel), and others.

62. The Hebrew word translated 'captain' and 'prince' is the same (sar).

63. Cf . the same Targum on Deut. 34, 6, where Metatron and three others
are the four 'princes of wisdom.'

64. Tanhuma, Weethanan §6 (ed. Buber, f. 7a). So also GrUnhut, Liqqu-
tim V, 105a (from Yelammedenu) .

65. MidrashMishleonl4,34(ed.Buber,f.39b). A manuscript in the library
of the Jewish Theological Seminary in New York has here Metatron instead
of Michael (Jewish Encyclopedia, X, 231).

66. Cf . Berakot 3a, below.

67. Ekah Rabbati, Proem 24 (ed. Wilna f. 6d, top). On the secret place
(nvoriipvov) which God has, whither to retire and weep, see Hagigah 5b (on
Jer. I.e.).

68. Hagigah 12b; Zebahim 62a; Menahot 110a. The formulation is
consistent. See Lueken, Michael, pp. 30 f .

69. Jellinek, Bet ha-Midrash, HI, 137.

70. So far, Tanhuma, Naso, 18.

84 HARVARD THEOLOGICAL REVIEW

71. Bemidbar Rabbah, 12, 12 (ed. Wilna 1884, f. 49a), cf. Rev. 5, 8; 8, 3
(Psalm 141, 2). Oesterley and Box (p. 175) assert that in this place 'the
term 'Mediator' is directly applied to Metatron, and, what is more signifi-
cant, he is represented as the reconciler between God and the Chosen People.'
The whole passage is quoted above in a literal translation. There is no word
in the context, far and wide, which could remotely suggest 'Mediator,' to
say nothing about being 'directly applied' to Metatron. (See this Review,
XIV, 249). Weber, from whom the authors have their (blind) reference
(without acknowledgment) renders the passage correctly. In so late a Mid-
rash it is probable that the 'Youth whose name is Metatron' is meant to be
the translated Enoch, of whom more is to be said hereafter; but the attribu-
tion of Michael's function to Metatron is independent of this identification.

72. It is possible that the meaning of the mysterious words 'for my name
is in him,' was found in the name Michael. The words are usually paraphrased
'whose name is like the name of his master.' Etymological midrash found in
^tO'D a compound of fl3D3 'D (Exod. 15, 11) 'Who is like Thee?' and ^SO fK
(Deut. 33, 26) 'There is none like God.' 'My name is in him,' would then
be the name 7X, 'God,' not, as the cabalists imagined, the proper name illiT,
so that they even call him J1t3pn rwv 'Jahveh Minor' (or 'Junior'), e.g.
Jellinek, Bet ha-Midrash V, 175. Hence a Moslem polemic speaks of 'the
little God' whom the Jews call Mitatrun (Mas'udi, quoted by Schreiner,
Zeitschrift der deutschen Morgenlandischen Gesellschaft, XLII, 598). The
Moslem puts into the mouth of Metatron the lamentation which in Ekah
Rabbathi, Proem. 24 (cf. Berakot 3a) is uttered by God.

73. In a Response of the Gaons Sherira and Hai (dated 992), it is said:
Jacob said, 'The angel who redeemed me,' etc. (Gen. 48, 16), and Isaiah,
'The angel of His presence saved them.' This is the Prince of the Presence
(sar ha-panim) of whom our Rabbis speak, Metatron. And we see how the
Merciful (God) extols him, and makes known to Moses his honor and his
greatness, as it is written, 'Behold I send,' etc., 'Beware of him,' etc. — A.
Harkavy, Responsen der Geonim (1887), No. 173, p. 189.

74. Jellinek, Bet ha-Midrash II, 55.

75. L.c. p. 56, 1. 11 from below. In another manuscript also at an earlier
point (p. 55, below). The inconsistency is doubtless to be attributed to copy-
ists; but testifies to the persistent survival of the identification.

76. The intercourse between Palestine and Babylonia and the study of
Palestinian Midrash in Babylonia sufficiently explain this.

77. An angel who stands with Metatron in the highest rank, and some-
times takes precedence of him, is Sandalphon (<rvva.8e\<t>os, sc. of Metatron),
whom later cabalists identify with the translated Elijah.

78. The angelic scribe, Ezek. 9, 2ff.; Enoch 89, 61 ff.; 90, 14, 22. In
these places in Enoch probably Michael, as the guardian angel of Israel.

79. Not quite the same, for Metatron writes down the good deserts of
Israelites; Enoch the judgment and sentence upon the world and all the evil
deeds of mankind. It is the generation before the flood.

80. In the Midrash Elle Ezkere (Jellinek, Bet ha-Midrash II, 66) : ' Meta-
tron the great scribe ' (or secretary) writes down and seals the decree of God
against Edom (Rome).

INTERMEDIARIES IN JEWISH THEOLOGY 85

81. With the assumption of Enoch and his ascent through the heavens
compare the Ascension of Isaiah 7-11.

82. Jellinek, Bet ha-Midrash, V, 172.

83. See above, p. 72.

84. A summary of the doctrine of Metatron in the Zohar and appendixes
with references to the sources may be found in Eshel Abraham by Abraham
Rovigo (Fiirth, 1701), f. 9 c. See also Danz, p. 735 (b) f. He is the first of
the creations of God; to him God gave dominion over all his hosts; servant
of God, the senior of his house, ruling over all; his name is like his master's,
he is created in his image and likeness. He is a priest of the Most High; he
takes the souls of the righteous up (to heaven) every night. In more dis-
tinctively cabalistic conception Metatron unites (or connects) the stages
(pJTl) of the four-story universe, from top to bottom and bottom to top.
He is the 'middle column' (sn'JJSDNT SIIDy) reaching up and down to
both extremities, like the 'middle bar' in the tabernacle, which passed
through from end to end (Exod. 26, 28). It is in Metatron that the Lord is
revealed in his Shekinah, e.g. to Ezekiel. Metatron, who is called sar ha~
panim, is as it were a vesture enveloping Metatron who is called Shaddai;
the Lord and his Shekinah are in the midst of the latter. Later cabalists
found the distinction between two Metatrons intimated in the spelling of
the word: pIDD'O is the Shekinah; JTiDDO is the angel of the Shekinah, an
envoy or minister (Tikkune Zohar, and Cordovero, Pardes Rimmonim,
quoted by Danz, 735 (b), 736 (c); more fully, Sommer, Theologia Soharica,
pp. 36 f.). The Eshel Abraham is innocent of this refinement. In the Yalkut
Rubeni the doctrine is even more explicitly stated. The author finds in the
Zohar two Metatrons, Major and Minor; the latter is a created being, a mes-
senger; the Great Metatron is an emanation (Danz, 737 (d)). It is perhaps
not superfluous to eall particular attention in the foregoing to the phrase
'the middle column,' with its context and explanation, since, especially in
the Latin columna medietatis, it is exposed to the misunderstanding that
some kind of an intermediary, or 'mediatorial,' place and function is attrib-
uted to Metatron.

85. The servant of God, the senior of his house, etc. is Metatron, of whom
we have said: (Gen. 24, 2).

86. Sommer: 'Futurus sit ut conjugatur corpori in utero materno/
Similia legimus de Christo Psalm lx, 7-9. — Sommer has apparently forgotten
that tru/xa Si ko.tt)Ptl<su> /uoi is LXX (and N. T.), but not in the Hebrew
text. If the language of the Zohar is rightly understood, it would only be one
of many evidences of the influence of Christian ideas, or a desire to match
them, in late developments of the Cabala.

87. Asilut, Beriyah, Yesira, 'Asiyah. Ginsberg, 'Cabala,' Jewish En-
cyclopedia, IH, 475.