Skip to main content

Full text of "Essays in Biblical Greek"

See other formats

υ » ^ χ 





Oxford University Press Warehouse 
Amen Corner, E.G. 



Biblical Greek 







[^// rights reservedly 




The present work consists of the substance of the 
Lectures delivered by the writer during his terms of office 
as Grinfield Lecturer on the Septuagint. It is designed 
not so much to furnish a complete answer to the questions 
which it raises as to point out to students of sacred litera- 
ture some of the rich fields which have not yet been 
adequately explored, and to offer suggestions for their 
exploration. It is almost entirely tentative in its character : 
and the writer has abstained from a discussion of the views 
which have been already advanced on some of the subjects 
of which it treats, because he thinks that in Biblical philo- 
logy even more than in other subjects it is desirable for 
a student in the present generation to investigate the facts 
for himself, uninfluenced by the bias which necessarily 
arises from the study of existing opinions. 

Those portions of the work which depend on the 
apparatus criticus of Holmes and Parsons must especially 
be regarded as provisional (see pp. 131, 132). The writer 
shares the gratification which all Biblical students feel at 
the prospect of a new critical edition of the Septuagint 
being undertaken by members of the great school of Cam- 
bridge scholars which has already done work of exceptional 
importance in the criticism of the New Testament : and he 
looks forward to the time when it will be possible to study 


the Greek text of the Old Testament with the same confi- 
dence in the data of criticism which is possessed by students 
of the New Testament. But instead of suspending all 
critical study until that time arrives, he thinks that the 
forming of provisional inferences, even upon imperfect data, 
will tend to accelerate its arrival. 

It is proper to add that in his references both to the 
Hebrew and to the Syriac version, the writer has had the 
advantage of the assistance of some distinguished Oxford 
friends : but he refrains from mentioning their names, 
because he is too grateful for their help to wish to throw 
upon them any part of the responsibility for his short- 

PuRLEiGH Rectory, 
September 19, 1888. 





Differences between Classical and Biblical Greek arising from the 
acts — 

(i) that they belong to different periods in the history of the 

language 3-8 

(2) that they were spoken not only in different countries but by 

different races 9-1 1 

Materials for the special study of Biblical Greek furnished by the 
Septuagint — 

i. in itself, in that it supplies a basis for induction as to the 

meaning (a) of new words, (<5) of familiar words . , . 11-14 
ii. in its relation to the Hebrew, in that 

(i) it gives glosses and paraphrases . . . . . . 14-Ϊ7 

(2) it changes the metaphors 17-20 

(3) it varies its renderings 20-23 

iii. in its relation to the other versions of the Hebrew, which are 

valuable not only in themselves as adding to the vocabulary, 

but also because they correct the Septuagint . . . 24-26 

(i) sometimes substituting a literal translation for a gloss . 26-27 

(2) sometimes substituting a gloss for a literal translation . 27 

(3) sometimes interchanging translations with it . . . 28-29 
Application of the foregoing method to a small group of words 30-3 2 

iv. in the variations and recensions of its MSS 32-33 

General summary of results 33-35 



'Ayyapivfiv (pp. 37-38), άναηινωσκΐΐν (pp. 38-39), άποστοματίζαν 
(pp. 39-40), άρ€τή (pp. 40-41), Ύλωσσόκομον (pp. 42-43), δεισιδαίμων, 
δεισιδαιμονία (pp. 43-45), διάβολο$, διαβάΚλω (pp. 45-47), διαθήκη 
(pp. 47-48), δίκαΐ05, δικαιοσύνη (pp. 49-5^)5 ^τοιμάζειν, ετοιμασία, 
ίτοιμο$ (pp. 5ΐ-55)> Θρησκεία (pp. 55-57)> μνστήριον (pp. 57-62), 
οΙκονόμο$ (pp. 62-63), ομοθυμαδόν (pp. 63-64), παραβολή, παροιμία 
(pp. 64-71), πειράζειν, πειρασμοί (pp. 7ΐ-73)> ττ^νη^, πραυ$, πτωχοί, 
ταπεινοί (pp. 73-77)> τΌνηρόί, πονηρία (pp. 77-82), παράκλητοί (pp. 
82-83), "ττίστίί (pp. 83-88), νπόστασίί (pp. 88-89), σνκοφαντεΐν (pp. 
89-91), ΰπόκρισίί, νποκριτήί (pp. 9^-93) 3^-93 





General principles on which such words should be treated . . 94~9^ 
i. Psychological Terms in the Septiiagint and Hexapla . . 96 

Application to (i) napbia, (2) πν(νμα, (3) ψνχή, (4) διάνοια, of the 
methods of investigation by noting 

(i) uniformities or differences of translation, i.e. (a) of what 
Hebrew words the Greek words are the translations, 
{d) by what Greek words the same Hebrew words are 
rendered in the Hexapla, {c) by what other Greek words 
the same Hebrew words are rendered in the LXX . . 98-103 

(2) the combinations and interchanges of the several Greek 

words in the same or similar passages, viz. (a) καρδία 
and Ίτν^υμα, {b) καρδία and ψυχή, {c) ΊΠ/^νμχι and ψυχτ], 
(d) καρδία and διάνοια ....... 103-104 

(3) the similarity or variety of the predicates of the several 

words 104-108 

ii. Psychological Terms in Philo 109 

(i) σώ/χα and ^υχί7 no 

(2) σώμα, σαρξ HO 

(3) ψ^χί in general 112-115 

(4) The lower manifestations of \i'vxJ7 11 5-1 20 

(5) The higher manifestations of ^υχΐ7 120-123 

(6) ψυχικοί 124 

(7) vovs 125-126 

(8) Ίτν^υμα 126-129 

General results 129-130 



The materials for the textual criticism of the Septuagint consist of 

(a) Greek MSS., (<^) Versions, (ίτ) Quotations 131-134 

Three recensions of the text existed in the time of Jerome : to one 
or other of them it is probable that the majority of existing MSS. 
belong : the question proposed is whether it is possible to go behind 
those recensions and ascertain the text or texts which preceded them 135-137 

The answer is to be found in the examination of quotations from 
the Septuagint in writings of the first two centuries A. D. : those 
writings may be dealt with by two methods, viz. 

(i) the quotations of a single passage may be compared with 

the other data for the criticism of the passage, 
(2) all the quotations from either a single book, or the whole 
of the Old Testament, made by a given writer, may be 
gathered together and compared 138-139 



Examples of the application of the first method to quotations 

of passages from Genesis and Exodus ..... 140-172 
, Examples of the application of the second method to quota- 
tions from (a) the Psalms, {b) Isaiah, in 

(i) Philo 172-174 

(2) Clement of Rome I75-I79 

(3) Barnabas 180-186 

(4) Justin Martyr 186-202 



The antecedent probability that collections of excerpts from the 
Old Testament would be in existence among the Greek-speaking 
Jews of the dispersion is supported by the existence of composite 
quotations 203-204 

Examination of such quotations in (i) Clement of Rome, (2) Bar- 
nabas, (3) Justin Martyr . . 204-214 



The existing LXX text of Job is the text as amplified by Origen : 
the earlier text is indicated in some MSS. and versions, and can 

consequently be recovered 215-217 

The question proposed is to account for the wide divergencies 
between the earlier and the amplified text. 

(i) Some of them are probably due to an unintelligent correc- 
tion of the earlier text 217-219 

(2) Some of them are probably due to a desire to bring the 

book into harmony with current Greek thought . . 219-220 
But neither of these answers would cover more than a small pro- 
portion of the passages to be accounted for : two other hypotheses 
are possible — 

(i) that the existing Hebrew text of the book is the original 
text, and that it was more or less arbitrarily curtailed by 
the Greek translator, 
(2) that the existing Hebrew text is itself the expansion of an 
originally shorter text, and that the original LXX text 
corresponded to the original Hebrew . . . . 220 

The remainder of the essay is a detailed examination of the second, 
third, and fourth groups of speeches, with the result of showing that 
the second hypothesis adequately accounts for the differences between 
the earlier and the amplified form 221-245 




The special difficulties of the textual criticism of the book . . 246 

(i) Short account of the Greek MSS. and of the inferences 
which may be drawn from their agreements and differences 
in regard to {a) forms of words, {b) inflexions, {c) use of 
the paroemiastic future, (i/) omission or insertion of the 
article, (e) syntactical usages 247-253 

(2) Short account of the Latin and Syriac versions, and indi- 

cation of the method of ascertaining their relation to 

each other 254-258 

(3) Examination of some important instances of variation . 258-281 
Some provisional results 281-282 

Index of Biblical Passages 283-293 


There is a remarkable difference between the amount of 
attention which has been given to the language of the Old 
Testament and that which has been given to the language 
of the New Testament. To the language of the Old 
Testament scholars not only of eminence but of genius 
have consecrated a lifelong devotion. The apparatus of 
study is extensive. There are trustworthy dictionaries and 
concordances. There are commentaries in which the 
question of the meaning of the words is kept distinct from 
that of their theological bearings. There are so many 
grammars as to make it difficult for a beginner to choose 
between them. In our own University the study is en- 
couraged not only by the munificent endowment of the 
Regius Professorship, which enables at least one good 
scholar to devote his whole time to his subject, but also 
by College lectureships and by several forms of rewards 
for students. 

The language of the New Testament, on the other hand, 
has not yet attracted the special attention of any consider- 
able scholar. There is no good lexicon. There is no 
philological commentary. There is no adequate grammar. 
In our own University there is no professor of it, but only 
a small endowment for a terminal lecture, and four small 

The reason of this comparative neglect of a study which 
should properly precede and underlie all other branches of 
19^ Β 

2 /.,\f jj/jl -'ί/ΐ jaHjTHE'V.AtUE AND USE 

theological study, seems to me mainly to lie in the assump- 
tion which has been persistently made, that the language 
of the New Testament is identical with the language which 
was spoken in Athens in the days of Pericles or Plato, and 
which has left us the great monuments of Greek classical 
literature. In almost every lexicon, grammar, and com- 
mentary the words and idioms of the New Testament are 
explained, not indeed exclusively, but chiefly, by a reference 
to the words and idioms of Attic historians and philoso- 
phers. The degree of a man's knowledge of the latter is 
commonly taken as the degree of his right to pronounce 
upon the former ; and almost any average scholar who can 
construe Thucydides is supposed to be thereby qualified to 
criticise a translation of the Gospels. 

It would be idle to attempt to deny that the resemblances 
between Attic Greek and the language of the New Testa- 
ment are both close and numerous : that the two languages 
are in fact only the same language spoken under different 
conditions of time and place, and by different races. But 
at the same time there has been, and still is, an altogether 
inadequate appreciation of their points of difference : and, 
as a result of this inadequate appreciation, those points of 
difference have not been methodically and exhaustively 
studied. Such a methodical and exhaustive study lies 
before the coming generation of scholars : it is impossible 
now, and it would under any circumstances be impossible 
for a single scholar. It requires an apparatus which does 
not yet exist, and which can only be gathered together by 
co-operation : it requires a discussion of some of its canons 
of investigation by persons not only of various acquirements 
but also of various habits of mind ; it requires also, at least 
for its more difficult questions, a maturity of judgment which 
is the slow growth of time. All that can be here attempted 
is a brief description of the points to which attention must 
primarily be directed, of the chief means which exist for 


their investigation, and of the main principles upon which 
such an investigation should proceed. 

The differences between the language of Athens in the 
fourth century before Christ and the language of the New 
Testament may be roughly described as differences of time 
and differences of country. 

I. Many differences were the natural result of the lapse 
of time. For Greek was a living language, and a living 
language is always in movement. It was kept in motion 
partly by causes external to itself, and partly by the causes 
which are always at work in the speech of all civilized 

The more important of the former group of causes were 
the rise of new ideas, philosophical and theological, the new 
social circumstances, the new political combinations, the 
changes in the arts of life, and the greater facilities of 
intercourse with foreign nations. 

Causes of the latter kind were stronger in their operation 
than the attempt which was made by the literary class to 
give to ancient models of style and expression a factitious 
permanence. By the operation of an inevitable law some 
terms had come to have a more general, and others a more 
special, application : metaphors had lost their original 
vividness : intensive words had a weakened force, and 
required to be strengthened : new verbs had been formed 
from substantives, and new substantives from verbs : com- 
pound words had gathered a meaning of their own which 
could not be resolved into the meaning of their separate 
parts : and the peculiar meaning which had come to attach 
itself to one member of a group of conjugates had passed to 
other members. 

In a large number of cases the operation of these causes 
which are due to the lapse of time, forms a sufficient ex- 
planation of the differences between Classical and Biblical 

Β 3 


Greek. The inference that this was the case is corroborated 
by the fact that in many cases the differences are not 
peculiar to Biblical Greek, but common to it and to all 
contemporary Greek. 

The following are examples of the operation of these 

ahuvarelv has lost its active sense 'to be unable to . . .' and 
acquired the neuter sense *to be impossible': e.g. LXX, Gen. 

1 8. 14 μτι ά8ννατησ(ΐ πάρα τω 0e<5 ρήμα ; S. Matt. 1 8. 20 ovbev άδυνα- 
τησα νμίν. Aquil. Jer. 32. 1 7 οίκ άδννατησ€ΐ από σον παρ ρημα^•=.ΣΧΧ. 
ου μη άποκρνβτ} άπο σον ονθίν. 

ακαταστασία : the political circumstances of Greece and the East 
after the death of Alexander had developed the idea of political 
instability, and with it the word ακαταστασία, Polyb. i. 70. i, 
S. Luke 21. 9, which implied more than mere unsettledness : for 
it is used by Symm. Ezek. 12. 19 as a translation of nj^'H 'dread' 
or 'anxious care,' and it is coupled by Clem. R. 3. 2 with 

δίωγ /ios. 

εΓΓροιτη had borrowed from a new metaphorical use of eVrpc- 
π^σθαι the meaning of ' shame,' i Cor. 6. 5 : cf. το Ιντρ^πτικόν Epict. 

I. 5• 3, 9• 

€ΤΓΐσκιάζ€ΐν had come to be used not only of a cloud which over- 
shadows, and so obscures, but also of a light which dazzles by its 

brightness, Exod. 40. 29 (35) • • • on Ιπ^σκίαζ^ν eV αντψ η νΐφίλη 

και 8όξη9 κνρίον €ν€πλησθη ή σκηνή : the curreut use of the word in this 
sense is shown by e. g. Philo, De Mundi Opif. i. p. 2, where the 
beauties of the Mosaic account of the Creation are spoken of as 
Tais μαρμαρυγαΐς τα^ των (ντνγχανόντων ψνχας εττισκιά^οντα : id. Quod 
OftiniS probus liber , ii. p. 446 St' άσθίνίίαν τον κατά ψ^νχην όμματος ο 
ταΐς μαρμαρυγαΐς πίφνκ^ν επισκιάζ€σθαί. 

ίπιτιμία had given up the meaning in which it is used by the 
Attic orators, ' possession of full political rights,' and acquired the 
meaning of the Attic (πιτίμησις or ίπιτίμιον, 'punishment,' or 
'penalty': Wisd. 3. lo; 2 Cor. 3. 6. 

ίργάζεσθαι had added to its meaning of manual labour, in which 
in the LXX. it translates ^?y, e. g. Exod. 20. 9, the meaning of 
moral practice, in which in the LXX. it translates bv^ especially in 
the Psalms, e. g. 5. 6 ; 6. 9 ; 13 (14). 4 ; in the N. T. e. g. S. Matt. 
7. 23; Rom. a. 10. 


ζωοποιεΐμ has lost its meaning ^ to produce live offspring ' (e, g. 
Arist. H. A. 5, 27. 3), and has acquired the meaning *to preserve 
alive,' e.g. Judges 21. 14 ras -^νναικα^ as ίζωοποίησαν άπο των θυγα• 
τίρων Ίαβίΐς Γαλαάδ (cf. Barnab. 6 πρώτον το παώίον μίΚιτι ΐΐτα γάλακτι 
ζωοποΐ€Ϊται), or * tO quicken,' e. g. 2 Kings 5. 7 ό θώς €γω του θανα- 
τώσαι και ζωοποιησαι ... ; S. John 5• 21 οΰτωί και ό vios ovs OeXei 
ζωοποκΐ. Rom. 4• Ι7 • • • ^^°^ ''"°^ ζωοποιοϋντος τους veKpovs. So 
also ζωογοκεΐΐ', which in later non-Biblical Greek has the meaning 
'to produce live offspring,' as Pallas was produced from Zeus, 
Lucian, Dial. Dear, 8, is used in Biblical Greek in the same senses 

as ζωοποΐ€Ϊν, e.g. Judges 8. 19 el (ζωογονηκητβ αυτούς, ουκ αν άπίκτΐΐνα 
νμάί. Ι Sam. 2. 6 κύριος θάνατοι καϊ ζωογονεί. S. Luke Ι7• 33 ^* 

&ν άτΓο\€στ) αυτήν ζωογονησβι αυτήν. Both words are in the LXX. 
translations of Π^^Π pi\ and kzp/i. (There is a good instance of the 
way in which most of the Fathers interpret specially Hellenistic 
phrases by the light of Classical Greek in St. Augustine's interpre- 
tation of the word, Quaest. super Levit. lib. iii. c. 38, ' Non enim 
quae vivificant, i. e. vivere faciunt, sed quae vivos foetus gignunt, 
i. e. non ova sed pullos, dicuntur ζωο-^ονουντα)' 

Ketpia, which was used properly of the cord of a bedstead, e. g. 
Aristoph. Av. 816, had come to be used of bedclothes, LXX. 
Prov. 7. 16 (where Aquila and Theodotion have πίριστρώμασι): 
hence, in S. John 11. 44, it is used of the swathings of a corpse. 

KTiais had come to have the meaning of κτίσμα, i. e. like creaiio, 
it was used not of the act of creating, but of the thing created : 
Judith 9. 12 βασίλβΰ πάσης κτίσεως σον. Wisd. 1 6. 24 η yap κτίσις 
σοι τω ποιησαντι υπηρετούσα. Rom. 8. 20 ttj yap ματαιότητι ή κτίσις 

λικμα»' had expanded its meaning of separating grain from chaff 
into the wider meaning of scattering as chaff is scattered by the 
wind, e. g. LXX. Is. 41. 15, 16 άλοησεις ορη κα\ XenTvveis βουνούς καϊ 
ως χνουν θησβις καϊ Χικμησΐΐς : hence it and διασπείρειν are used inter- 
changeably as translations of πη{ ' to scatter,' both in the LXX. 
and in the other translations of the Hexapla, e.g. Ps. 43 (44). 12, 

UKX.. 8ΐ€σπ€ΐρας, Symm. ίλίκμησας, Jer. 1 5- 7? LXX. ^ιασπβρώ, Aquil. 

Symm. λικμησω. Hencc it came to be used as the nearest meta- 
phorical expression for annihilation: in Dan. 2. 44 Theodotion 
uses λικμησ€ΐ to corrcct the LXX. αφανίσω as the translation of ^DH 
aph. from t]!iD ' to put an end to.' Hence the antithesis between 
σννβ\ασθησ(ται and λικμησει in S. Luke 20. 1 8. 


ττάροικο? had lost its meaning of ' neighbour ' and had come to 
mean ' sojourner/ so that a clear distinction existed between 
napoiKclv and κατοικάν, e.g. LXX. Gen. 36. 44 (37. l) κατωκα de 
'Ιακώβ €V TTj yfj ov παρωκησ^ν 6 πατήρ αντοϋ^ iv γη Χαναάν, cf. Philo J^e 
COnfus. ling. i. p. 416 .. . κατωκησαρ o)s ev ττατρ'ώι, ουχ ω: inl ^€νης 

πράκτωρ seems to have added to its Attic meaning ' tax-gatherer' 
the meaning 'jailer' : since in an Egyptian inscription in the Corp. 
Inscr. Graec. No. 4957. 15 ττρακτόρβιον is used in the sense of 
a prison, els τ6 πρακτόρ^ιον κώ. els tcis oKXas φνλακά5. Hence τω 
πράκτορι in S. Luke 12. 58 is equivalent to τω Ιπηρίττι in S. Matt. 

5• 25. 

ττροβιβάζβιΐ' had acquired the special meaning * to teach/ or 

* to teach diligently* : it occurs in LXX. Deut. 6. 7 πpoβιβάσeιs αντα 

Tovs viovs σου, where it is the translation of ]^ψ pi. ' to sharpen ' 

sc. the mind, and hence * to inculcate.' Hence S. Matt. 14. 8 17 de 

προβιβασθείσα νπο τηs μητρο5 avrrjs. 

συνοχή had acquired from the common use of σννίχεσβαί the new 
meaning of 'distress': S. Luke 21. 25 συνοχή Ιθνων iv απορία. In 
Ps. 118 (119). 143 Aquila uses it as the translation of ρΊ^9=^ΧΧ. 


υποζυγιοκ had narrowed its general meaning of ' beast of burden ' 
to the special meaning of ' ass ' : it is the common translation in 
the LXX. of "^iisn. Hence its use in S. Matt. 21.5; 2 Pet. 2. 16. 

It vi^ill be seen from these instances, which might be 
largely multiplied, that in certain respects the ordinary 
changes vi^hich the lapse of time causes in the use of v^^ords 
are sufficient to account for the differences between 
Classical and Biblical Greek. There are certain parts of 
both the LXX. and the New Testament in which no other 
explanation is necessary : so far as these parts are con- 
cerned the two works may be treated as monuments of 
post-Classical Greek, and the uses of words may be 
compared with similar uses in contemporary secular 
writers. It is probably this fact which has led many 
persons to overrate the extent to which those writers may 
be used to throw light upon Biblical Greek in general. 


But the application of it without discrimination to all 
parts of the Greek Bible ignores the primary fact that 
neither the Septuagint nor the Greek Testament is a single 
book by a single writer. Each is a collection of books 
which vary largely in respect not only of literary style, but 
also of philological character. A proposition which may 
be true of one book in the collection is not necessarily true 
of another : and side by side with the passages for whose 
philological peculiarities contemporary Greek furnishes an 
adequate explanation, is a largely preponderating number 
of passages in which an altogether different explanation 
must be sought. 

Before seeking for such an explanation, it will be ad- 
visable to establish the fact of the existence of differences ; 
and this will be best done not by showing that different 
words are used, for this may almost always be argued to be 
a question only of literary style, but by showing that the 
same words are used in different parts of the New Testa- 
ment in different senses — the one sense common to earlier 
or contemporary Greek, the other peculiar to Biblical 
Greek. The following few instances will probably be 
sufficient for the purpose. 

άγαθοτΓοιεΐι/ (i) is used in i Pet. 2. 15, 20 in its proper sense of 
doing what is morally good in contrast to doing what is morally 
evil: so Sext. Empir. 10. 70, 2 Clem. Rom. 10. 2. But (2) it is 
used in the LXX. Num. 10. 32, Jud. 17. 13 (Cod. A. and Lagarde's 
text, but Cod. B. and the Sixtine text άγαθυνεΐ), Zeph. i. 12 as the 
translation of 2p^ L•'. in the sense of benefiting and as opposed to 
doing harm. So in the Synoptic Gospels, S. Luke 6. 9, 35 ; 
S. Mark 3. 4 (Codd. A Β C L, but Codd. ^<D άγαθον ποιήσαι which is 
found in the same sense, and as a translation of PPS in Prov. 11. 17, 
where Symmachus has evepyeTel) : and in Codd. DEL, etc. Acts 
14. 17, where Codd. ίίΑΒ C have the otherwise unknown (except 
to later ecclesiastical writers) άγαθονργών. 

βλασφημεί»' and its conjugates (i) have in Rom. 3. 8, i Cor. 10. 
30, I Pet. 4. 4, and elsewhere, the meaning which they have both 


in the Attic orators and in contemporary Greek, of slander or 
defamation of character. 

But (2) in the Gospels they have the special sense of treating 
with scorn or contumely the name of God, as in the LXX., where 
(a) β\ασφημ€ίν translates ^l? pi. 2 Kings 19. 6, 22; in Num. 15. 
30, Is. 37. 23 the same word is translated by παροξύνειν, but in the 
latter passage the other translators of the Hexapla revert to βλασ- 
φημί'Ίν; (δ) βλάσφημων translates f^^ hithpo. in Isa. 52. 5, and its 
derivative n^SJ in Ezek. 35. 12 ; {c) βλάσφημος translates ]}^ "^T?.?? 
* he blesses iniquity' {i.e. an idol) in Is. 66. 3. 

διαλογισμό? (i) is used in S. Luke 9. 46, Phil. 2. 14, and probably 
Rom. 14. I, in the ordinary late Greek sense of discussion or dis- 
pute; but (2) it is used elsewhere in the Gospels, S. Matt. 15. 19 = 
S. Mark 7. 21 ; S. Luke 5. 22 (=S. Matt. 9. 4 4νθνμησ€ΐς) ; 6. 8 of 
thoughts or cogitations in general. This is its meaning in the 
LXX., where it is used both of the thoughts or counsels of God, 
e.g. Ps. 39 (40). 6; 91 (92). 5, and of the (wicked) thoughts or 
counsels of men, e.g. Ps. 55 (56). 6; Is. 59. 7. In all these 
instances it is the translation of Π3^ΠΌ or Γΐ3^πο. 

6'π•ιγι>'ώσκ€ΐι/, Ιπίγι/ωσίξ (ι) are used in S. Luke i. 4 in the 
Pauline Epistles, e.g. Rom. 3. 20; i Cor. 13. 12; Eph. 4. 13; and 
in Heb. 10. 26 ; 2 Pet. i. 2. 8 ; 2. 20, in the sense of knowing fully, 
which is a common sense in later Greek, and became ultimately the 
dominant sense, so that in the second century Justin Martyr, Tryph. 

3, defines philosophy as 4τη<ττημη τον ovtos καί του άΚηθοΰί (πίγνωσις ', 

and still later, in Const. Apost. 7. 39, it was the second of the 
three stages of perfect knowledge, γνώσις, €πίγνωσις, πληροφορία. 

But (2) in the Synoptic Gospels (πιγίνώσκ^ιν is used in the sense 
of recognizing or being conscious of: e.g. S. Matt. 7. 16; 17. 12 ; 
S. Mark 5. 30; S. Luke 24. 16. 

This variety may perhaps be partly explained by the 
hypothesis that some books reflect to a greater extent the 
literary language of the time, and others the popular 
language. But such an explanation covers only a small 
proportion of the facts. Even if it be allowed that what is 
peculiar to Biblical Greek reflects rather a popular than 
a literary use of words, the nature of that popular use 
requires a further investigation: and hence we pass to a 
different series of causes. 


11. Biblical Greek belongs not only to a later period of 
the history of the language than Classical Greek, but also 
to a different country. The physical and social conditions 
were different. This is shown by the change in the general 
cast of the metaphors. The Attic metaphors of the law- 
courts, the gymnasia, and the sea are almost altogether 
absent, except so far as they had indelibly impressed them- 
selves on certain words, and probably, in those words, lost 
their special reference through frequency of familiar usage. 
Their place is taken by metaphors which arose from the 
conditions of Syrian life and from the drift of Syrian ideas. 

For example, whereas in Athens and Rome the bustling 
activity of the streets gave rise to the conception of life 
as a quick movement to and fro, αναστρίφ^σθαι, ανάστροφη^ 
versari, conversation the constant intercourse on foot be- 
tween village and village, and the difficulties of travel on 
the stony tracks over the hills, gave rise in Syria to a group 
of metaphors in which life is conceived as a journey, and 
the difficulties of life as the common obstacles of a Syrian 
traveller. The conduct of life is the manner of walking, 
or the walking along a particular road, e.g. ζΐτορζνθησ-αν νψηλω 
rpa\rj\(o, ζττορξύθη kv όδω του irarpos αντου. A change in 
conduct is the turning of the direction of travel, βττύστρίφεσ^αι. 
The hindrances to right conduct are the stones over which 
a traveller might stumble, or the traps or tanks into which 
he might fall in the darkness, aKavbaka, προσκόμματα, Traytdes, 
βόθυνοι. The troubles of life are the burdens which the 
peasants carried on their backs, φορτία. Again, the com- 
mon employments of Syrian farmers gave rise to the 
frequent metaphors of sowing and reaping, of sifting the 
grain and gathering it into the barn, aireipeLv, Θ€ρίζ€ΐν, 
σινίάζβίν, συνάγ€ίν : the threshing of wheat furnished a 
metaphor for a devastating conquest, and the scattering of 
the chaff by the wind for utter annihilation, άλοαν, λίκμάν. 
The pastoral life provided metaphors for both civil and 


moral government : sheep astray {τΐλανώμ^νοι) upon the hills, 
or fallen bruised down the rocky ravines [ΙσκνλμΙνοι και 
Ιριμμίνοι) furnished an apt symbol of a people which had 
wandered away from God. The simple ministries of an 
Eastern household {Ιιακονάν, διακονία), the grinding of corn in 
the handmill, the leavening of bread, the earthen lamp on its 
lampstand which lit up the cottage room ; the custom of 
giving of presents in return for presents [αντατιο^ώόναι^ 
avTaiTOboaLs) ; the money-lending which, then as now, filled 
a large place in the rural economy of Eastern lands 
(bav€L(€Lv, όφζίλη, όφζίλημα, όφξίλ^τη^) ; the payment of 
daily wages (μυσθό^) ; the hoarding of money out of the 
reach alike of the robber and the tax-gatherer (Θησαυροί, 
Θησαυρίζζΐν) ; the numerous local courts with their judges 
and witnesses (κριτψ, μαρτυρεί, μαρτύρων, μαρτυρία) ; the 
capricious favouritism of Oriental potentates (ττροσωττοληψία)^ 
all furnished metaphors which were not only expanded into 
apologues or parables, but also impressed themselves upon 
the common use of words. 

But these changes in the cast and colour of metaphors, 
though they arise out of and indicate social circumstances 
to which Classical literature is for the most part a stranger, 
are intelligible without special study. They explain them- 
selves. They might have taken place with a purely Greek 
population. The difficulty of Biblical Greek really begins 
when we remember that it was Greek as spoken not merely 
in a foreign country and under new circumstances, but also 
by an alien race. The disputed question of the extent to 
which it was so spoken does not affect the literary monu- 
ments with which we have to deal. Whether those 
monuments appealed immediately to a narrower or a 
wider circle of readers, they undoubtedly reflect current 
usage. They afford clear internal evidence that their 
writers, in most cases, were men whose thoughts were 
cast in a Semitic and not in a Hellenic mould. They 


were not only foreigners talking a language which was not 
their own, as an Englishman talks French : they were also 
men of one race speaking the language of another, as 
a Hindoo Mussulman talks English. This affected the 
language chiefly in that the race who thus spoke it had 
a different inheritance of religious and moral ideas from the 
race to which it properly belonged. The conceptions of 
God and goodness, the religious sanction and the moral 
ideal, were very different in men whose traditions came 
down from Moses and the prophets, from what they had 
been in men whose gods lived upon Olympus, and whose 
Pentateuch was the Iliad. The attitude of such men 
towards human life, towards nature, and towards God was 
so different that though Greek words were used they were 
the symbols of quite other than Greek ideas. For every 
race has its own mass and combinations of ideas; and when 
one race adopts the language of another, it cannot, from the 
very nature of the human mind, adopt with it the ideas of 
which that language is the expression. It takes the words 
but it cannot take their connotation : and it has ideas of its 
own for which it only finds in foreign phrases a rough and 
partial covering. 

Biblical Greek is thus a language which stands by itself. 
What we have to find out in studying it is what meaning 
certain Greek words conveyed to a Semitic mind. Any 
induction as to such meaning must be gathered in the first 
instance from the materials which Biblical Greek itself 
affords. This may be taken as an axiom. It is too 
obvious to require demonstration. It is the application 
to these particular philological phenomena of the universal 
law of inductive reasoning. But at the same time it has 
been so generally neglected that in a not inconsiderable 
number of cases the meaning of New Testament words has 
to be ascertained afresh : nor does it seem probable that 


the existing confusion will be cleared up until Biblical 
Greek is treated as a newly discovered dialect would be 
treated, and the meaning of all its words ascertained by 
a series of new inferences from the facts which lie nearest 
to them. It will probably be found that in a majority of 
cases the meaning which will result from such a new induc- 
tion will not differ widely from that which has been 
generally accepted : it will probably also be found that 
in a majority of cases in which a new meaning is demon- 
strable, the new meaning links itself to a classical use. But 
it will also be found, on the one hand, that new and 
important shades of meaning attach themselves to words 
which retain for the most part their classical use : and, on 
the other hand, that some familiar words have in the sphere 
of Biblical Greek a meaning which is almost peculiar to 
that sphere. 

For the purposes of such an induction the materials 
which lie nearest at hand are those which are contained in 
the Septuagint, including in that term the extra-canonical 
books which, though they probably had Semitic originals, 
exist for us only in a Greek form. 

A. Even if the Septuagint were only a Greek book, the 
facts that it is more cognate in character to the New Testa- 
ment than any other book, that much of it is proximate in 
time, and that it is of sufficient extent to afford a fair basis 
for comparison, would give it a unique value in New Testa- 
ment exegesis. 

(i) This value consists partly in the fact that it adds to 
the vocabulary of the language. It is a contemporary 
Greek book with new words, and many words which are 
found in the New Testament are found for the first time in 
the Septuagint : — 

(a) Some of these words are expressions of specially Jewish 

ideas or usages ; άκροβυστία, oiXiayeiv, άναθψατίζΐΐν, άπ^ρίτμητος, αττο- 


dtKOTOvv, €νωΒΙα, (φημ^ρία, ματαιότης, πατριάρχης, π€ριτομη, τιροσηλυτος, 
ττρωτοτόκια, ραντισμός. 

(δ) Some of them are legitimately formed, but new compounds 

from existing elements : ακρογωνιαίος, aWoyevrjs, €κμνκτηρίζ(ΐν, €μ- 
τταικτης, ίν^υναμουν, Ινωτίζ^σθαι, (πισκοπη, fiboKia, ήττημα, κατακανχασθαι, 
κατακΚηρονομα,ν, καταννσσ(ΐν, κατοικητηριον, κανχησις, κΚν8ωνίζ€σθαι, 
κραταιονν, μί-γάλωσυνη, ορθρίζίΐν, irayihcvdv, παραζηλονρ, π€ποΙθησις, 
•ηΚηροφυρύν, σητόβρωτος, σκαν8α\ίζ€ΐν, σκάνδαλον, σκληροκαρΒΙα, σκληρο' 
τράχηλος, στυγνάζαν, νπακοη, νστίρημα, φωστηρ. 

('ζ) The other and more important element in the value 
of the Septuagint viewed simply as a Greek book is that it 
affords a basis for an induction as to the meaning not of 
new but of familiar words. Very few lexicographers or 
commentators have gone seriously astray with new words. 
But the meaning of familiar words has been frequently 
taken for granted, when the fact of their constant occurrence 
in the Septuagint in the same connexion and with predi- 
cates of a particular kind, afford a strong presumption that 
their connotation was not the same as it had been in 
Classical Greek. 

Instances of such words will be found among those which are 
examined in detail below, e. g. διάβολος, πονηρός. 

These characteristics attach not only to the Septuagint 
proper, but also to the deutero-canonical books, or 
' Apocrypha.' Those books have a singular value in re- 
gard to the syntax of the New Testament, which is 
beyond the range of the present subject. Some of them 
have also a special value in regard to some of the more 
abstract or philosophical terms of the New Testament, of 
which more will be said below. But they have also a 
value in the two respects which have been just mentioned : 

(i) They supply early instances of New Testament 
words : 

€KWi/€ta, Acts 26. 7, is first found in 2 Mace. 14. 38 : it is also 
found in Judith 4. 9. Its earliest use elsewhere is Cic. AU. 10. 
7. I. 


ίξίσχύειν, Eph. 3. 18, is first found, and with the same con- 
struction as in the N, T., in Sirach. 7. 6. Its earliest use else- 
where is Strabo 788 (but with ώστε). 

καταλαλιά, 2 Cor. 12. 20, I Pet. 2. i, is first found in Wisd. i. 11. 
Its earhest uses elsewhere are Clem. Rom. 30. 35 ; Barnab. 20. 

KTiats, Rom. 8. 19 sqq., etc., in the sense of things created and not 
of the act of creation, is first found in Wisd. 5. 18 ; 16. 24; 19. 6. 

σκαΐ'δαλίζειΐ', Matt. 5. 29, and freq., is first found in Sir. 9. 5. 

ύπογραμμός, I Pet. 2. 21, is first found in 2 Mace. 2. 28: its 
earliest use elsewhere is Clem. Rom. 5. 

φυλακίζειμ, Acts 22. 19, is first found in Wisd. 18. 4 : its earliest 
use elsewhere is Clem. Rom. 45. 

χαριτοΟκ, Luke i. 28, Eph. i. 6, is first found in Sir. 18. 17. 

{2) They also supply instances of the use of familiar 
words in senses which are not found in earlier Greek, but 
which suggest or confirm inferences which are drawn from 
their use in the New Testament. 

An instance of this will be found below in the meaning of 
πονηροί, which results from its use in Sirach. 

B. But that which gives the Septuagint proper a value in 
regard to Biblical philology which attaches neither to the 
Apocrypha nor to any other book, is the fact that it is 
a translation of which we possess the original. For the 
meaning of the great majority of its words and phrases we 
are not left solely to the inferences which may be made by 
comparing one passage with another in either the Septua- 
gint itself or other monuments of Hellenistic Greek. We 
can refer to the passages of which they are translations, 
and in most cases frame inductions as to their meaning 
which are as certain as an^^ philological induction can be. 
It is a true paradox that while, historically as well as 
philologically, the Greek is a translation of the Hebrew, 
philologically, though not historically, the Hebrew may be 
regarded as a translation of the Greek. This apparent 
paradox may be illustrated by the analogous case of the 
Gothic translation of the Gospels : historically as well as 


philologically that translation is, as it professes to be, 
a rendering of the Greek into the Moeso-Gothic of the 
fourth century A. D. ; but since all other monuments of 
Moeso-Gothic have perished, the Greek of the Gospels 
becomes for philological purposes, that is to say, for the 
understanding of Moeso-Gothic words, a key to, or trans- 
lation of, the Gothic. 

But that which makes the possession of this key to its 
meaning of singular value in the case of the Septuagint, is 
the fact that to a considerable extent it is not a literal 
translation but a Targum or paraphrase. For the tendency 
of almost all students of an ancient book is to lay 
too great a stress upon the meaning of single words, to 
draw too subtle distinctions between synonyms, to press 
unduly the force of metaphors, and to estimate the 
weight of compound words in current use by weighing 
separately the elements of which they are compounded. 
Whereas in the ordinary speech of men, and with all but 
a narrow, however admirable, school of writers in a literary 
age, distinctions between synonyms tend to fade away, the 
original force of metaphors becomes so weakened by 
familiarity as to be rarely present to the mind of the 
speaker, and compound words acquire a meaning of their 
own which cannot be resolved into the separate meanings 
of their component parts. But the fact that the Septuagint 
does not, in a large proportion of cases, follow the Hebrew 
as a modern translation would do, but gives a free and 
varying rendering, enables us to check this common 
tendency of students both by showing us not only in 
another language, but also in another form, the precise 
extent of meaning which a word or a sentence was intended 
to cover, and also by showing us how many different 
Greek words express the shades of meaning of a single 
Hebrew word, and conversely how many different Hebrew 
words explain to us the meaning of a single Greek word. 


These special characteristics of the Septuagint may- 
be grouped under three heads: (i) it gives glosses and 
paraphrases instead of literal and word for word ren- 
derings : (2) it does not adhere to the metaphors of the 
Hebrew, but sometimes adds to them and sometimes 
subtracts from them : (3) it varies its renderings of 
particular words and phrases. Of each of these charac- 
teristics the following examples are given by way of 

I. Glosses and paraphrases : 

(a) Sometimes designations of purely Jewish customs are glossed : 
e.g. njK' || 'the son of the year/ Num. 7. 15, etc., i.e. a male of 
the first year which was required in certain sacrifices, is rendered by 
(αμνός) έΐΊαύσιοξ I ί^^ΊΏΠ ^Ό 'bitter waters,' Num. 5. 18, etc., is 
rendered by τό ύδωρ του «λεγμου ; ITp. the ' separation ' or ' conse- 
cration ' of the Nazarite, Num. 6. 4, and even Ί'ίρ t^'i<"i ' the head 
of his separation,' ib. v. 9, are rendered simply by ευχή; ΠίΠ''3 Π*•"» 
*a savour of quietness,' Lev. i. 9, etc., is rendered by οσμή 

(δ) Sometimes ordinary Hebraisms are glossed : e.g. "15^ 1^ 'the 
son of the foreigner/ Ex. 12. 43, etc., is rendered simply by άλλο- 
γ€ρης; t^pvX 'things of nought,' Lev. 19. 4, etc., is rendered by 
«δωλα ; 'li?^ ' to visit ' (used of God), is rendered in Jeremiah and 
several of the minor prophets by cKdiKelv : ^''.^^ψ ?"]]{ ' of uncircum- 
cised lips,' Ex. 6. 12, is rendered by αλογός ίίμι. 

{c) More commonly, an interpreting word, or paraphrase, is sub- 
stituted for a literal rendering : similar examples to the following 
can be found in almost every book. Gen. 12. 9, etc., ^JJ 'the 
South' is interpreted by η έρημος•. Gen. 27. 16 ^PPJ!} 'the smooth- 
ness,' sc. of Jacob's neck, is interpreted by τά -γυμνά : Gen. 50. 3 
D''pjn 'the embalming' is rendered by the more familiar της ταφής, 
* the burial,' and in the following verse, Γ))2 the ' house ' of Pharaoh 
is interpreted by tovs Βυνάστας, ' the mighty men ' of Pharaoh : Num. 
31. 5 1"ip»*1 'were handed over,' sc. to Moses, = eξηpLθμησav, *were 
counted out': i Sam. 6. 10 ^"^Ψ^^, 'the men' is interpreted by ot 
αλλόφυλοι, 'the Philistines': Job 2. 8 ^£)ΧΠ ηίΠΞΐ 'among the 
ashes' is interpreted by im της κοπριάς, 'on the midden': Job 31. 


32 ΠΊ^ρ < to the way' (possibly reading nnk^ < to a traveller') 
is inteφΓeted by παντί (λθόντι: in Ps. 3. 4 ; 118 (119). 114 1^9 
'a shield' (used of God) is interpreted by άι/τίλτ^πτωρ : in Ps. 17 
(18). 3; 18 (19). 15; 77 (78). 35; 93 (94). 22 η^ϊ 'a rock' is 
interpreted by βοηθός, and in Ps. 117 (118). 6 the same Greek 
word is added as a paraphrase of the personal pronoun v, κύριος 
€μοΙ βοηθ05 : in Ps. 15 (16). 9 ''1^23 'my glory' is inteφreted by 
ή γλωσσά μου : in Ps. 38 (39). 2 ΟΊΟΠΰ ' a bridle' is interpreted by 
φνλακψ: in Ps. 33 (34). 11 Q^T^ri * young lions' is interpreted 
by πλούσιοι : in Ps. 126 (127). 5 iriB^i< « a quiver ' is interpreted by 

την ^πιθνμιαν. 

[d) In some cases instead of the interpretation of a single word 
by its supposed equivalent, there is a paraphrase or free translation 
of a clause: for example, Ex. 24. 11 'upon the nobles of the 
children of Israel he laid not his hand' : LXX. των (πιλότων του 
Ίσραηλ ου δΐ€φωνησ€ν oiide (ϊς, ' of the chosen men of Israel not one 
perished': i Sam. 6. 4 ' What shall be the trespass-offering which 
we shall return to him': LXX. τί τό της βασάνου άποδώσομ^ν αύτίί ; 
' what is the [offering for] the plague that we shall render to it' (sc. 
to the ark) : i Kings 21 (20). 39 ' if by any means he be missing' 
(Ij^Si nipk.) : LXX. eav δε €κπηδων (κττηδηση, ' if escaping he escape ' : 
Ps. 22 (23). 4 'through the valley {^^^^) of the shadow of death': 
LXX. €v μίσω σκιάς θανάτου : Ps. 34 (35). 14 Ί bowed down heavily 
as one that mourneth for his mother ' (DNI ''?^?) : LXX. ως πενθών και 

σκυθρωπάζων ούτως €ταπ€ΐνονμην : Ps. 43 (44)• 20 ' that thou shouldeSt 

have sore broken us in the place of jackals' (Q''|iJn) : LXX. 6τι 

€ταττ(ίνωσας ημάς iv τόττω κακώσεως : Is. 60. 1 9 'neither for brightness 

shall the moon give light unto thee': LXX. ovSe ανατολή σελήνης 
φωτΐ€ϊ σου [Cod. Α. σοι] την νύκτα, ' neither shall the rising of the 
moon give light to thy night ' (or ' give light for thee at night '). 

2. Metaphors : 

(a) Sometimes there is a change of metaphor, e. g. in Amos 
5. 24 |ri^i< 7Π: 'a mighty,' or 'perennial stream,' is rendered by 
χείμαρρους άβατος, 'an impassable torrent' : Micah 3. 2 2ΠΚ ' to love ' 
is rendered by ζητείν, ' to seek.' 

{δ) Sometimes a metaphor is dropped : e. g. Is. 6. 6 ' then /ew 
(p)y;i) one of the seraphim unto me,' LXX. άπβστάλη προς μέ iv των 
2€ραφίμ: Ps. 5. i3j ^^d elsewhere, ΠΟΠ «to fly for refuge' is ren- 
dered by ikniUw: Job 13. 27 ΠίΠΊΝ « ways ' is rendered epya, ' deeds.' 



{c) Sometimes a metaphor appears to be added, i. e. the Greek 
word contains a metaphor where the corresponding Hebrew word 
is neutral : e.g. Jer. 5. 17 ^Ψ^ po. *to destroy' is rendered by 
αλοαι/, ' to thresh ' : Ezek. 2 1 . 1 1 ^']Π * to kill ' is rendered by άπο- 
κΐντύν, and Num. 22. 29 by Uk^vtuv, 'to pierce through' (so as to 
kill) : Deut. 7. 20 "i3S hiph. ' to destroy' is rendered by βκτρίβίσθαι, 
'to be rubbed out' : ]^^ 'to dwell' is frequently rendered by κατα- 
σκηνονν, ' to dwell in a tent.' 

These tendencies both to the glossing and paraphrasing 
of the Hebrew, and to the changing or apparent adding of 
metaphors, will be best seen by analysing the translations 
of some typical word. The following is such an analysis 
of the translations of ]n^ ' to give.' 

(a) In the following cases there is a paraphrase. 

/os. 14. 12 'Give me this mountain/ LXX. αΐτουμαί ae τ6 ορός 


Deut. 21. 8 'Lay not innocent blood unto My people of Israel's 

charge,' LXX. Iva μη γίνηται αίμα άναίτιον iv τω λαω σου ^Ισραήλ. 

Esther 3• 1 1 'The silver is given to thee,' LXX. το μίν αργύρων 

Ezek. 45. 8 ' They shall give the land to the house of Israel 
according to their tribes,' LXX. τψ γην κατακΚηρονομησουσιν οίκος 
^Ισραήλ κατά φυλάς αυτών. 

(β) In the following cases a local colouring is given to 
the translation, so that the translation of the verb must be 
taken in its relation to the translation of the whole passage. 

Gen. 20. 6 'therefore suffered I thee not to touch her,' evcKu 
τούτου ού< άψήκά σβ ayj /ασθαι αύτης. 

Gen. 38. 28 'the one put out his hand,' 6 ef? ττροεξήμβγκε τψ 

Gen. 39. 20 'Joseph's master ... put him into the prison,' 

^ΐ'εβαλεΐ' αύτον (Ις το οχύρωμα. 

Gen. 41. 41 Ί have set thee over all the land of Egypt,' καθ- 
ιστημι σ? σήμερον eVi ττάσ-η yrj Αίγνπτον. 

Gen. 43• 23 * the man . . . gave them water and they washed 
their feet,' 'ην'€γκ€>' ίδωρ νί-ψ-αι τους ττοδαί αύτων. 


Exodus 3. 19 Ί am sure that the king of Egypt will not let you 

go, olha oTi ου ττροήσ6ται νμας Φαραώ. 

Exodus 7• 4 Ί will lay my hand upon Egypt/ επιβάλω την χΰρά 

μου eV A ly νπτον. 

Exodus 18. 25 'Moses . . . made them heads over the people, 

rulers of thousands . . ./ έττοί'(]σεν αυτούς in αυτών χιΚιάρχονς. 

Exodus 21. 19 'he shall pay for the loss of his time,' τψ apyeias 
αυτόν άΐΓθτίσ€ΐ. 

Exodus 27. 5 'thou shalt put it under the ledge of the altar 

beneath,' ύττοθήσεις αυτούς (sc. τους ΒακτυΧίους) υπό την Ισχάραν τοΰ 
θυσιαστηρίου κάτωθεν. 

Exodus 30. 19 'thou shalt put water therein,' €κχ€€Ϊ9 ets αύτον 


Lev. 2. 15 ' thou shalt put oil upon it/ εττιχεεΐς eV α\ηην 

Lev. 19. 14 ' Thou shalt not . . . put a stumbhng block before 
the blind/ anivavTi τνφΧον ου ττροσθήσεις σκάνδαΧον. 

Deui. 15-17 'Thou shalt take an aul and thrust it through his 
ear unto the door,' Χτγ^η το οπητιον κα\ τρυπήσεις τ6 ώτίον αυτού προς 
την βύραν. 

2 Sam. 18. 9 'he was taken up between the heaven and the 

earth/ εκρεμάσθη άνα μβσον του ουρανού κα\ άνα μίσον της yης. 

2 Kings 16. 14 ' . . . and put it on the north side of the altar,' 

ε$ειξεΐ' αύτο ein μηρον τον θυσιαστηρίου. 

1 Chron. 16. 4 'he appointed certain of the Levites to minister,' 
έταξε . . . ΐκ τών Α^υιτών X€lτoυpyoυvτaς. 

2 Chron. 16. 10 ' . . . and put him in the stocks,' παρε'θετο αντον 

ft? φυΧακην. 

Esth. I. 20 'all the wives shall give to their husbands honour/ 
τιασαι ax γυναίκας περιθήσουσι τιμήν τοϊς άνδράσιν ίαντών. 

Job 2. 4 ' all that a man hath will he give for his life,' οσα υπάρχει 
άνθρώπω vnep της ψνχης αύτου εκτίσει. 

/od 9. 18 'He will not suffer me to take my breath/ ούκ. έα γάρ 

μ€ άναν€νσαι. 

Job 35• 10 'who giveth songs in the night,' ό κατατάσσωμ φυΧακάς 

Job 36.3' Fo^ truly my words are not false,' βργοις de μου BUaia 

«ρώ eV* αληθείας. 

Prov. 10. 10 'He that winketh with the eye causeth sorrow,' 6 
Ivveicuv οφθαΧμοΊς μ€τα 8όΧου συΐ'άγει άν^ράσι Χνπας. 

C 2 


Prov, 21. 26 'but the righteous giveth and spareth not/ 6 bk 

8i<aios eXea και olKTeipci άφ€ΐδως. 

Is. ^. 4 Ί will give children to be their princes/ εττιστήσω veavl- 
σκους άρχοντας αυτών. 

Is. 43. 9 * let them bring forth their witnesses/ άγαγετωσακ tovs 

μάρτυρας αυτών. 

Jer. 44 (37). 15 'the princes . . . put him in prison in the 

house of Jonathan,' απ^στ^ιΚαν αυτόν 6if την οΐκίαν ^Ιωνάθαν. 

Ezek. 14. 8 Ί will set my face against that man/ στηριώ το 
πρόσωπον μου (πι τον ανθρωπον ζκύνον. 

3. Variations of rendering. 

(a) In a comparatively small number of cases a single 
Greek word corresponds to a single Hebrew word, with 
such accidental exceptions as may be accounted for by 
a variation in the text : it is legitimate to infer that, in 
such cases, there was in the minds of the translators, and 
since the translators were not all of one time or locality, 
presumably in current usage, an absolute identity of mean- 
ing between the Hebrew and the Greek: e.g. hovkos=• 
Ί2ν (or ifni>}. 

{b) In certain cases in which a single Greek word stands 
for two or more different Hebrew words, the absence of 
distinction of rendering may be accounted for by the para- 
phrastic character of the whole translation, and will not 
of itself give trustworthy inferences as to the identity in 
each case of the meaning of the Greek and the Hebrew 

e.g. €Ϊδωλοι/, €Ϊδωλα Stands for (i) Q^Ti /ξ? 'gods/ (2) Ci^Wn; 

'things of nought' ( = τά μάταια Zach. II. 17, βδίλύγματα Is. 2. 8, 20, 

χειροποίητα Lev. 26. I, Is. 2. 1 8, ctc), (3) D^^X 'terebinth-trees/ (4) 
n^D3 'high-places' (more commonly=Ta υ^ηΧά), (5) Ογ^τ^ 'Baalim,' 
(6) ryh'h^ 'idol-blocks/ (7) D^^?n ' vanities/ (8) n^J^n 'sun-pillars,' 
(9) D^BVy 'idols/ (10) D7''p3 'graven images' (also=ra γλυπτά), 
(ii) Dpy 'images' (also=ei/cwv), (12) Υψ^ 'abomination/ (13) 
D''Q"[)il 'teraphim.' 

It is clear that in the majority of these cases ά'δωλα is a para- 


phrastic or generic term, and not the exact equivalent of the 

(c) In certain cases a single Hebrew word is represented 
by two or more Greek words, not in single but in repeated 
instances, and not in different but in the same books or 
group of books ; it is reasonable to infer in such cases, 
unless a close examination of each instance reveals a 
marked difference of usage, that in the minds of the 
translators the Greek words were practically synonymous : 

e.g. in Psalm 36 (37) νψΐ occurs 13 times : in vv. 10, 12, 14, 
17, 18, 20, 21, 32, 40 it is rendered by αμαρτωλές, in vv. 28, 35, 
38 by άσεβης: it is difficult to account for this except by the 
hypothesis that the two words were regarded as identical in 

(d) In certain cases in which a single Hebrew word is 
repeatedly represented by two or more Greek words, the 
variation exists only, or almost only, in different books, 
and may therefore be mainly attributed to a difference in 
the time or place of translation, or in the person of the 
translator : but at the same time such a repeated render- 
ing of a single Hebrew word by two or more Greek words 
argues a close similarity of meaning between the Greek 
words which are so used : 

e. g. in Genesis, Exodus, Leviticus, Numbers ?Πζ is translated 
by συναγωγή ; in Deuteronomy and the following books to Nehe- 
miah inclusive (56 times in all), with only the exception of Deut. 5. 
22, it is translated by εκκλησία. 

In Exodus, Deuteronomy, Joshua, Judges, but elsewhere only 
2 Sam. 15. 8, ^^V is generally translated by λατρεύειμ : in Numbers 
by λειτουργάω : in Genesis, the historical books, and the prophets by 

In Exodus, Leviticus, and Numbers nmip is ordinarily, and fre- 
quently, translated by θυσία : in Genesis (except 4. 3, 5) by h&pov : 
in other books, e. g. Isaiah, by both words. 

It is reasonable in these cases to infer a close similarity of mean- 
ing between συναγωγή and €κκλησία; XarpeveiVy \ζΐτουργύν, and 

^ovKevuv ; and bmpov and θυσία, respectively. 


(e) But in many cases it is found that a single Hebrew 
word is represented by two or more different Greek words 
not only in various books of the Septuagint but sometimes 
also in the same book, and with sufficient frequency to 
preclude the hypothesis of accidental coincidence. It is 
also found that another Hebrew word, of similar meaning, 
is represented, under the same conditions, by the same two 
or more Greek words as the preceding. Consequently each 
of a small group of Hebrew words is represented by one or 
other of a corresponding group of Greek words, and, con- 
versely, each of the small group of Greek words stands for 
one or other of a small group of Hebrew words. It is 
reasonable to infer in such cases that the Greek words so 
used are practically synonymous : i. e. that whatever dis- 
tinctions may have been drawn between them by the 
literary class, they were used indifferently in current speech. 
For example, 

7W is rendered in Isaiah by (i) εξαιρεί^ c. 60. 16, (2) λυτρουμ 
c. 35. 9 : 41• 14 : 43• i• Μ - 44- 22, 23, 24 : 52. 3 : 62. i2 : 63. 9, 
(3) ρυεσθαι c. 44. 6 : 47. 4 : 48. 17, 20 : 51. lo : 52. 9 : 54. 5, 8 : 
59. 20: 63. 16. 

νψΐ hiph. is rendered by (i) εξαιρεΐι/ Jer. 49 {42). 11,(2) ρυεσθαι Is. 
5. 29: 36. 14, 15, 18, 19,20: 37. II, 12: 38. 6: 50. 2, (3) σώζειι/ 
Is. 19. 20: 25. 9: 30. 15: 33. 22: 35.4: 37. 20, 35: 43. 3, II, 
12 : 45. 17, 20, 22 : 46. 7 : 49. 25 : 59. i : 60. 16 : 63. 9. 

i^^D pi. is rendered by (i) έξαιρεΐμ 2 Sam. 19. 5, 9, i Kings 112, 
(2) ρυεσθαι Ps. 40 (41). 2: 88 (89). 49 : 106 (107). 20 : 114(116). 
4 : 123 (124). 7, (3) σώζεις I Sam. 19. 11, 12 : 27. i, i Kings 18. 
40: 19. 17 : 21 (20). 20, 2 Kings 19. 37. 

y^} hiph. is rendered in Isaiah by (i) εξαιρεϊμ c. 31. 5 : 42. 22 : 43. 
13: 44. 17, 20: 47. 14: 57. 13, (2)ρύ6σθαι C.44. 6: 47.4: 48. 17, 
2o: 49.7,26: 51.10: 52.9: 54. 5» 8: 59-20: 63.16,(3) 
σώζειν c. 19. 20 : 20. 6. 

ΠΙΩ is rendered by (i) λυτροΟμ Ps. 24 (25). 22: 25 (26). 11: 30 
(31)• 6 : 33 (34). 23 : 43 (44)• 27 : 48 (49)• 8, 16 : 54 (55). 19 : 
70(71). 23: 77(78). 42: 118 (119). 134: 129 (130). 8, (2)pu'- 
εσθαι Job 5. 20 : 6. 23, Ps. 68 (69). 19, (3) σώί^ει»' Job 33. 28. 


a?Bp{. is rendered by (i) εξαιρεϊμ Ps. 36 (37). 40; 70 (71). 2 : 81 
(82). 4, (2) λυτροΰΐ' Ps. 31 (32). 7, (3) ρύεσθαι Ps. i6 (17). 13 : 17 
(18). 44, 49 : 21 (22). 5, 9 : 30 (31)• 2 : 36 (37)• 4© : 42 (43). i : 
70 (71). 4: 90 (91). 14, (4) σώζειι/ (for the derivatives tD'-pB, 
riD^S) Is. 10. 20: 37. 32 : 45. 20: 66. 19: so also άι/ασώζει»/ Jer. 
51 (44). 14, etc., Βιασώζειΐ' Job 21. 10, etc. 

Conversely, εξαιρεΐι/ is used to translate (i) ^^l Is. 60. 16, (2) 
νψΐ hi. Jer. 49 (42). 1 1, (3) tOp^ 2 Sam. 19. 5, 9, i Kings i. 12, Ezek. 
33• 5» (4) ''-'t twelve times in the Pentateuch, thirty-three times in 
the historical books, thirty-two times in the poetical books, (5) t3pa 
pi. 2 Sam. 22. 2, Ps. 36 (37). 40 : 70 (71). 2 : 81 (82). 4. 

λυτρουμ is used to translate (i) 7δ^2 twenty times in Exodus and 
Leviticus, twenty-four times in the poetical books, (2) ΠΊΏ fifteen 
times in the Pentateuch, seven times in the historical books, nine- 
teen times in the poetical books, (3) u2^ pi. Ps. 31 (32). 7. 

ρύεσθαι is used to translate (i) W\ Gen. 48. 16 and twelve times 
in Isaiah, (2) W^ hiph. Ex. 2. 17 : 14. 30, Jos. 22. 22, Is. 49. 26 : 
63. 5, Ezek. 37. 23, (3) t229 P^' Job 22. 30, and in the above- 
mentioned five passages of the Psalms, (4) /V^ Exod. 2. 19 : 5. 23 ί 
6. 6 : 12. 27, fourteen times in the historical books, sixty times in 
the poetical books, (5) ΠΊΒ Job 5. 20 : 6. 23, Ps. 68 (69). 19, Hos. 
13. 14, (6) ώδ pi. 2 Sam. 22. 44, and in the above-mentioned 
ten passages of the Psalms. 

σώζειΐ' is used to translate (i) W hiph. Deut. 33. 29, fifty-six times 
in the historical books, nearly a hundred times in the poetical books, 
(2) ΰ^>η^2'. Gen. 19. 17, 22, ten times in the historical books, twenty- 
seven times in the poetical books, (3) ^V^ Gen. 32. 30, eight times 
in the historical books, fourteen times in the poetical books, (4) 
rriQ Job 33. 28, (5) i37Q or one of its derivatives. Gen. 32. 8, 
2 Chron. 20. 24, Neh. i. 2, Is. 10. 20 : 37. 32 : 45. 20: 66. 19, 
Jer. 51 (44)• 28. 

It is reasonable to infer that, in their Hellenistic use, the Greek 
words which are thus used interchangeably for the same Hebrew 
words did not differ, at least materially, from each other in mean- 
ing, and that no substantial argument can be founded upon the 
meaning of any one of them unless that meaning be common to it 
with the other members of the group. 

III. There is a further circumstance in relation to the 


Septuagint which requires to be taken into account to 
a much greater extent than has usually been done. It is 
that in addition to the Septuagint we possess fragments of 
other translations of the Hebrew, those of Aquila, Symma- 
chus, Theodotion, and of two anonymous translators, who 
are generally referred to as the Fifth and Sixth. 

Part of the value of these translations lies in the fact that 
they belong to the period when the right interpretation of 
the Old Testament had become a matter of controversy 
between Jews and Christians : but very little is positively 
known about their authors or their approximate dates. 

Accounts of Aquila are given by Irenaeus 3. 21. i (=Eus. H.E. 
5. 8. 10), Origen Epist. ad African. 2 (i. p. 13), Eusebius Dem. Ev. 
7. I. 32, Epiphanius de Mens, et pond. 14, Jerome Ep. 57 ad Pam- 
mach. (i. p. 314), Cata. 54 (ii. p. 879), Praef. in lib. Job (ix. p. iioo), 
Comm. in Jes. 8. 11 (iv. p. 122), Comm. in Abac. Ill (vi. p. 656), 
and in the Jerusalem Talmud Megilla i. 11, p. 71, Kiddush. i. i, 
p. 59. Accounts of Symmachus are given by Eusebius H. E. 6. 
17, Dem. Ev. l.c.^ Jerome, and Epiphanius II. cc. Accounts of 
Theodotion are given by Irenaeus and Epiphanius //. cc, Jerome 
//. cc, and Praef in Dan. (v. p. 619). 

But these accounts vary widely, and, especially those of Epipha- 
nius, appear to be in a large degree conjectural. 

In regard to their dates, Aquila is placed by the Talmud //. cc 
in the time of R. Akiba, R. Eliezer, and R. Joshua, i.e. early in the 
second century a. d. : but it has been inferred from the fact of his 
being mentioned by Irenaeus and not by Justin Martyr that he 
flourished in the interval between those two writers. The date of 
Symmachus may be inferred from the fact that he is not men- 
tioned by Irenaeus to have been near the end of the second cen- 
tury, a view which is in harmony with the account of Eusebius 
H. E. 6. 17, which places him a generation before the time of 
Origen. The date of Theodotion is more uncertain than that of 
the other two : he certainly lived before the time of Irenaeus, and, 
if the view be correct that his translation is quoted in Hermas, he 
may even have preceded Aquila. 

But the chief part of their value lies in the con- 


trlbutions which they make to the vocabulary of Biblical 
Greek. Some words which are found in the New Testa- 
ment are not found elsewhere within the range of Biblical 
Greek except in these translations. 

άτΓοκαραΒοκία, Rom. 8. 19, Phil. i. 20 (most Codd.), is interpreted 
by the verb άποκαραΒοκ^ϊν, which is used by Aquila in Ps. 36 (37). 7 
as the translation of ί'ί?ΊΠΓΐΠ (hithpa. of h^r\), for which the LXX. 
iKfTfvaov and Symm. Uerfve are less accurate renderings. The 
reading of Codd. FG. in Phil. i. 20, καραδοκία, is known only from 
its use by Aquila in Prov. 10. 28 as the translation of ^.<'D*'^ 
* expectation,' = Symm. νπομόνη, Theod, προσδοκία. 

ί-γκακΐΐν, in the sense of ' to be weary or faint,' is first found out- 
side the N. T. as Symmachus's translation of ^^i?i5 in Gen. 27. 46, 
= LXX. προσώχθικα, Aquil. (σίκχανα, Ε. V. Ί am weary of my life 
because of the daughters of Heth.' 

έμβριμασθαι, Matt. 9. 30, Mark I. 43: 14. 5, John 11, 33, 38, 
which in Classical Greek is found only in Aesch. Septem c. Theb. 
461, of the snorting of horses in their harness, is best explained by 
its use (i) as the translation of ^V\ 'to be angry' in Aquil. Ps. 7. 

12 €μβριμώμ€Ρος = 1^Χ.Κ. όργην (ττάγων, Alius άπ€ΐλούμ€νος '. SO (μβρί- 

μησις=ίίΐ6 derivative W_ in Aquil. Symm. Ps. 37 (38). 4=LXX. 
opyr\s'. in Theod. Is. 30. 2 7 = LXX. opy^y : and in Theod. Symm. 
Ezek. 21. 31 (36) = LXX. ορ^ψ, Aquil. άπίΐλψ : (2) as the trans- 
lation of "iy2 'to rebuke,' in Symm. Is. 17. 13 Ιμβριμησ^ται αντω=. 

LXX. άποσκορακια, αυτόν, Aquil. (πιτιμησ^ι ev αντω '. SO (μβρίμησις 

translates the derivative nnya in Symm. Ps. 75 (76). 7=LXX. Aquil. 


Ιΐ'θύμησίξ, Matt. 9. 4 : 12. 25, Heb. 4. 12 finds its only parallel 
in the sense of 'thoughts,' or 'cogitations,' in Symm. Job 21. 27 
(in the same collocation with εννοιών as in Hebrews 4. 12, Clem. 
Rom. 21. 9), where it translates ΓΐΐΠ^ΠΏ, which, like ^νθυμησις in 
S. Matthew, is used of malicious thoughts (e. g. Esth. 8. 3, 5). 

εττίβλημα, in the sense of a ' patch,' Matt. 9. 16 ( = Mark 2. 21, 
Luke 5. 36), is found only in Symm. Jos. 9. 11 (5). 

καταφερ€σθαι, the expressive word which is used for ' dropping 
fast asleep' in Acts 20. 9, finds its only parallel in this sense in 
Biblical Greek (elsewhere, Arist. De Gen. Anim. 5. i, p. 'j'jga) in 
Aquil. Ps. 75 (76). 7, where it translates D'^")^=LXX. (νύσταξαν. 

θεομάχοξ. Acts 5. 39, occurs elsewhere in Biblical Greek only in 


Symm. Job 26. 5 [=Theod. ytyai/res•), Prov. 9. 18 (=LXX. γτ/γβ^^Γ?, 
Theod. yiyavres), Prov. 21. 16 ( = LXX. Ύΐγάρτων) : in each case it 
translates ^^^^'\. 

opoBeaia, Acts 17. 26, is not found elsewhere, but the verb 
opoOeTelv (many MSS. όρωθίτβϊν) is found in Aquil. Deut. 19. 14, 
Zach. 9. 2, and in Symm. Exod. 19. 12. 

σττλαγχί'ίζβσθαι, which is found 12 times in the Synoptic Gospels 
(not elsewhere in the N. T.) in the sense ' to feel compassion,' is 
found as the translation of ΏΓίρρπ in Symm. i. Sam. 23. 21, ΙστίΚα-^- 
χνίσθητ€-=υΧΧ, €7Τον€σατ€, Theod. Ι^άσασθ^ (which is the LXX. 
translation of the same verb in Ex. 2. 6). The compound liri- 
σπλαγχνίζίσθαι is found in Symm. Deut. 13.8 (9). as the translation of 
the same verb, ^ LXX. ουκ ^πιποθησεις €π αντω. The active σπλαγ- 

χνίζειν occurs in 2 Mace. 6. 8, but in the sense of the Classical 
an\ayxvev€iv=to eat the entrails of an animal after a sacrifice 
(Aristoph. Av. 984). 

Another element in the value of these translations consists 
in the corrections which they make in the LXX. rendering, 
sometimes substituting a literal translation for a gloss, and 
sometimes a gloss for a literal translation. 

(i) Sometimes a gloss or paraphrase of the LXX. is 
replaced by a literal or nearly literal rendering : this is the 
case chiefly, though not exclusively, with Aquila: for 

Gen. 24. 67 /'Πδ^ 'tent*: LXX. (as frequently) υΐκος, Aquil. 


Ex. 6. 12 ^]^^Ψ ^"^V. ' uncircumcised in Hps': LXX. aXoyos ίίμι, 

Aquil, άκρόβυστοί xetXeac. 

Ex. 21. 6 t^^C^i^.v• ''? 'to the gods' (sc. probably the judges): 

LXX. π/30? το κριτηριον τον θΐου, Aquil. Symm. προς τους Oeovs. 

Lev. 4. 2, 22: 5• 15 "^ίί?^? 'through error': LXX. ακουσίως, 
Aquil. Symm. iv άγνοια. 

Lev. 26. 13 n^*p^*ip 'standing upright': LXX. μ^τα τταρρησίας, 

Alius άνισταμίνονς. 

Num. 21. 25 n^nis 75•?^ ^ 2ina. in all its daughters' (i.e. dependent 

villages) : LXX. κα\ iv πάσαις ταΐς συγκυρούσαις αυτί), Aquil. Symm. 
Theod. θυγατράσίν αυτής. 

Num. 23. 21 ^^ξ> nynn 'the shout of a king': LXX. τα ei /δο^α 


αρχόντων, Aquil. αλαλαγμοί βασιλβως, Symm. σημασία, Theod. σαλ- 

Deut. ΙΟ. 1 6 D5^ni) n^ny nt< 'the foreskin of your heart': 

LXX. τψ σκληροκαρ8ίαν νμών, Aquil. άκροβνστίαν καρδίας. 

Deut. 32. ΙΟ li^i^y^^ ' found him' : LXX. αύτάρκησ€Ρ αυτόν, Aquil. 
Theod. ηνρ€ν αυτόν. 

Job I. 6 : 2. I Q'''7''5i'7 ""^,? 'sons of God': LXX. οί άγγελοι του 

Oeox), Alius oi υΙο\ θίοϋ. 

Fs. 15 (16). 9 """l^i^l 'my glory': LXX. ή -γλωσσά μου, Aquil. 
Symm. Theod. δόξα μου. 

Fs. 30 (31). II ^^ψν 'have waxed old': LXX. 4ταράχθησαν, 

Aquil. ηυχμώθη, Symm. ίυρωτίασαν. 

Fs. 31 (32). 6 δ<^9 ^5f? 'in a time of finding': LXX. iv καφω 

€υθ4τ<ύ, Aquil. ds καιρόν (υρίσ^ως αυτού. 

^•f- 34 (35)• 15 ^^ψ? ''Vr'V? ' in my halting they rejoice ' : LXX. 

κατ €μου (υφράνθησαν, Aquil, iv σκασμω μου ηυψράνθησαν, Symm, 
σκάζοντας de μου ηυφραίνοντο. 

Fs. 40 (41)• 3 ''"*?1^ ^?ί.? ' unto the soul (i. e. will) of his enemies ' : 

LXX. els χείρας βχθροΰ αϋτοΰ, Aquil. iv ψυχ^ βχθρού, Symm. els ψυχας 

(2) Sometimes, on the other hand, a literal rendering of 
the LXX. is replaced by a gloss or paraphrase in one or the 
other translation : this is the case chiefly, though not ex- 
clusively, with Symmachus : e.g. 

fudges 8.21 D^JΊΠb'Π-n^? ' the little moons ' (ornaments) : LXX. 

τους μηνίσκους, Symm. τα κόσμια. 

I Sam. 20. 30 J^T]V. 'uncovering': LXX. άποκαλύψ^ως, Symm. 


I Sam. 2 2. 8 ""^Τζ^'Π^ Π?!ΐ 'uncovering the ear': LXX. άποκα- 
λνπτων το ώτίον, Alius φανερον ποίίϊ. 
Job 1 . 1 6 ί^.?^^^^ ' devoured ' : LXX. κατίφαγ^ν, Symm. aneKTeivev. 
Fs. 21 (22). 17 i2"'?^3 ' dogs' : LXX. Kvves, Symm. θηραταί. 
PS' 37 (38)• 4 ""ί^ί^ϊ^Π ^:5)Ώ ' from the face of my sins' : LXX. από 

προσώπου των αμαρτιών μου, Symm. δια τας αμαρτίας μου. 

Fs. 40 (41). 9 ^^Ρ^ ^'OS^'i^b 'will not add to rise up': LXX. 

ού προσθησ€ΐ του άναστήναι, Symm. ουκετι άναστησ^ται. 

(3) But the chief contribution which these translations 
make to Biblical philology is that they enable us to correct 


or corroborate the inferences which are drawn from the 
relation of the Septuagint to the Hebrew, by supplying us 
with a number of new and analogous data for determining 
the meaning of words. It is found in a large number of 
instances that the word which one or other of the trans- 
lators substitutes for the LXX. word is itself used in other 
passages of the LXX. as the translation of the same 
Hebrew word : it is also found that, conversely, the LXX. 
word is used elsewhere by the other translators for the 
same Hebrew word. The inference to be drawn in such 
cases is that the words which are so interchanged are 
practically synonymous. 

Gen, 8. 13 •^JpD, LXX. στί-γψ, Aquil. Symm. κάλυμμα, which is 
the LXX. rendering of the same word in Num. 8. 10, 11, 12, 25. 

Gen, 24. 61 i^^VJ, LXX. άβραι, Aquil. παώίσκαι, which is the LXX. 
rendering of the same word in Ruth. 4. 12, Amos 2. 7: Symm. 
κοράσια, which is the LXX. rendering of the same word in Ruth 2. 
8, ef al. 

Ex. 2. 22 "*!!, LXX. πάροικος, Aquil. προσηλντος, which is much 
the more frequent translation of the same word in the LXX. 

Ex. 3. 16 "".^iPr^^, LXX. την yepovaiav, Aquil. τους πρεσβυτέρους, 

which is the ordinary translation of the same word in the LXX. 
outside the Pentateuch. 

Ex. 23. 16 ^I?i<n, LXX. συντίΚΐ'ιης, Aquil. συλλογή?, Symm. σνγ- 
κομώης : the word occurs elsewhere only in Ex. 34. 22, where the 
LXX. renders it by συναγωγής. (The use of συντΐΚίΐα in the sense of 
harvest is noteworthy in its bearing upon S. Matt. 13. 39.) 

Lev. 2. 6 C]'']jlQj LXX. κλάσματα, Aquil. Symm. Theod. ψωμούς: 
but in Judges 19. 5 the MSS. of the LXX. vary between ψωμω 
and κλάσμητι as the translation of the same word. 

Zev. 3. 9 n»''pri, LXX. αμωμον, AquU. reXeiav, which IS the LXX. 
rendering of the same word in Ex. 12. 5 ^/ al. Symm. όλόκληρον, 
which is the LXX. rendering in Lev. 23. 15. 

Zev. 6. 2 (5. 22) ρψν, LXX. η^ίκησε τι, Aquil. Symm. Theod. Ισυ- 
κοφάντησε, which is the LXX. rendering of the same word in Job 
35• 9, etc. 

Num. 25. 4 ^Ϊ^Ίϊ^, LXX. παρα^ειγμάτισον, Aquil. άνάττηξον, Symm. 


Deut. 7. 2 ΰ'ΊΠί!! ί3"ΊΠΠ, LXX. άφανισμω άφαΐΊίΓ?, Aqutl. Symm, 
Theod. άναθζματίσζΐς, which IS the rendering of the LXX. in Deut. 

13. 15: 20. 17. 

Oeut. 30. 9 ^l^nini, LXX. καί €υλογησ€ΐ (so Codd. B., etc., but 
Codd, Α., etc., πολνωρησα) σβ, Aquil. Theod. ηβμισσ€νσ€ΐ, Symm, 

I Sam. 6. 9 '"T?•!?^, LXX. σύμπτωμα, (^Aquil.) συνάντημα, which is 

the LXX. rendering in Ecclesiastes 2. 14. 15: 3. 19: 9. 2, 3, 
Symm. συγκυρία (cf. S. Luke 10. 31). 

I Safn. 9. 22 nnStpp^ LXX. ets τό κατάλυ/χα, Aquil. γαζοφνλάκων^ 
which is the ordinary LXX. rendering in Nehemiah, Symm. e^edpav, 
which is the ordinary LXX. rendering in Ezekiel. 

I Sam. 19. 14 "^2^, LXX. (νοχΚζίσθαι, Aqutl. άρρωστάν, which is 
a common LXX. rendering of the word. 

I Sam. 21.4 (5) /Π Dn^^ LXX. άρτοι βέβηλοι, Aquil. Symm. Theod. 


1 Sam. 22. 15 V i^^rC} LXX. μη^αμως, Aquil. βζβηλόι/, Symm. 
Theod. ΐλ€ως, which is the LXX. rendering of the same word in 
2 Sam. 20. 20. 

2 Sam. 2. 26 ^V.5<, LXX. els vIkos, Alius βως ίσχάτου. The phrase 
is important in its bearing upon Matt. 12. 20: the same Hebrew 
phrase is rendered els vIkos in Aquil. and Quinius, Ps. 48 (49). 9 = 
LXX. ds TeXos, Symm. ds αΙώνα ; in Aquil. Theod. Is. 33. 20= LXX. 
els τον αΙώνα χρόνον, Symm. els TeXos ', and in Aquil. Is. 57• 16 = 
LXX. διατταιη-όί, Symm. els Tekos. So also in Is. 34. 10 Q^n^J ns:p= 
LXX. els χρόνον ιτο\ύν, Aquil. els vIkos νικ€ων, Theod. els ίσχατα 

Job 6. 8 ""ί?}!?^, LXX. τψ e\mda μου, Aquil. ύπομονην (sO also 4. 

16; 17. 15), which is the LXX. rendering of the same word in 

14. 19. 

Fs. 10 (11). 4, 5 ^^ΓJ^^ LXX. e^eTaCei, Aquil. doK^aCet, which 
elsewhere in the Psalms, viz. 16 (17). 3 : 25 (26). 2 : 65 (66). 10 : 
80 (81). 8 : 94 (95). 9 is the constant LXX. rendering of the same 

It follows from this relation of the other translators to 
the Septuagint that they afford a test of the inferences 
which are derived from the Septuagint itself. Since the 
Septuagint is presumably, it may almost be said demon- 
strably, the work of different persons and different periods, 


it is natural to expect that a new group of translators, 
working under analogous conditions, although at a dif- 
ferent period of time, should stand in the same relative 
position to the several groups of translation of the Sep- 
tuagint in which those groups stand to one another. If, 
for example, it is found that certain words are used inter- 
changeably to translate the same Hebrew word by different 
groups of translators of the Septuagint, it must be pre- 
sumed that a new group of translators will also use those 
words interchangeably. Their not doing so would raise a 
presumption that the variations in the Septuagint were due 
to personal or local peculiarities, and that no general infer- 
ence could be drawn from them. Their doing so affords an 
evidence which almost amounts to proof, that the words 
were in common use as synonyms. This evidence is the 
more important because of the fact that the translators of 
the Hexapla lived after New Testament times. It conse- 
quently shows that, in the case of the words to which it 
applies, the meaning which is gathered from the Sep- 
tuagint lasted through New Testament times. 

This evidence is sometimes of a negative and sometimes 
of a positive kind : it is aegative, when the absence of any 
record of corrections of the LXX. by the other translators 
makes it probable that the latter accepted the translations 
of the former; it is positive, when such corrections are 

The following is an example of the application of this 
test to a group of words of which the LXX. uses have been 
given fully above. It has been shown that the Hebrew 
words h^-\, V^\ t^hj2, SlJ^ ms, tiD^B are translated to 
a great extent interchangeably by the Greek words k^aipeiv^ 
λντρουν, ρν€σθαι, σώζζίν. The negative evidence which the 
other translators afford that the Greek words were regarded 
as practically identical in meaning is that they rarely dis- 
turb the LXX. rendering: the positive evidence which 


they afford to the same effect is that wherever they do 
amend that rendering they do so, with the exception 
mentioned below, by using another member of the same 

(i) In Is. 35. 9 DvlN'S is translated by the LXX. 'Κε\υτρωμ€νοι, 
by Theodotion €ρρυσμβνοι: (2) in Ps. 114 (116). 4 ΠΰρΌ is trans- 
lated by the LXX. βνσαι, by Aquila πβρίσωσον, by Symmachus 
c^eXoC: in Jer. 46 (39). 18 t3^P« t^^n is translated by the LXX. 
σώζων σώσω σβ, by Aquila ρνόμΐνος ρνσομαί σε: (3) in I Sam. 30. 2 2 
IJpifn is translated by the LXX. ίζαλόμεθα, by Aquila (ρρυσάμ^θα : in 
Job 5. 19 ^'^Ψ. is translated by the LXX. e^eXelrat, by Aquila 
ρνσζται : in Ps. 30 (31). 3 ^^'^ϊ] is translated by the LXX. τον i^e- 
λίσθαι, by Symmachus e^eXov : in Ps. 32 (33). 16 b'^^] is translated 
by the LXX. σωθησβται, by Aquila βυσθησβταί, by Symmachus διαφβν- 
ξ€ται : in Ps. 33 (34). 5 ^'''Sfn is translated by the LXX. ^ρρύσατο, by 
Symmachus ίξ^ίλ^το : in Ps. 38 (39) b^^>] is translated by the LXX. 
ρυσαι, by Symmachus βξ^λον : in Ps. 71 (72). 12 ^^ψ_ is translated 
by the LXX. ^ρρνσατο, by Symmachus i^eXelrai : in Prov. 24. 11 
PSfn is translated by the LXX. βΰσαι, by Symmachus σώσον : in Is. 
38. 6 7^-f^ is translated by the LXX. and Aquila βύσομαι, by Sym- 
machus €ξ€\ονμαι, by Theodotion σώσω : (4) in 2 Sam. 4. 9 «TlSl 
is translated by the LXX. βλντρώσατο, by Symmachus βυσάμ^νος : in 

Ps. 43 (44.) 27 ^^"12^ is translated by the LXX. κα\ λντρωσαι ημάς, 

by another translator (Άλλος, ap. Chrysost. ad loc.) κα\ βνσαι ημάς : 
(5) in Ps. 17 (18). 44 ώεΐη is translated by the LXX. and Symma- 
chus βΰσαί (βνστ)), by Aquila διασώσεις: in Ps. 3 1 (32). 7 i2?a is 
translated by the LXX. λντρωσαι, by Aquila 8ιασώζων, 

The exception mentioned above is that the translators of the 
Hexapla introduce into the group of Greek words another word 
which is not found in the N. T., and which is found in the LXX, 
in other senses, viz. άγχιστεύειμ. The use of this word helps to 
confirm the general inference as to the practical identity of mean- 
ing of the other members of the group, and the word itself affords 
an interesting illustration of the light which the fragments of the 
Hexapla throw upon later Greek philology. 

άγχιστβύειΐ' occurs in the LXX. in the active, in Leviticus, 
Numbers, Deuteronomy, Joshua, and Ruth : in all cases as the 
translation of b^\ kal, or ?ϊ<3 ; and in the passive, in 2 Esdr. 2. 62, 


Neh. 7. 64 as the translation of another word ^^^ pu. The mean- 
ing * to be next of kin ' had evidently passed into the meaning 
' to act as next of kin/ with especial reference to the buying back 
of a kinsman's possession (Lev. 25. 25), and exacting the penalty 
of a kinsman's blood (Num. 35. 19, etc.), and 'purchasing/ i.e. 
marrying a kinsman's widow, ' to raise up the name of the dead 
upon his inheritance' (Ruth 3. 12: 4. 5). These derived mean- 
ings had become so thoroughly identified with the word in 
Hellenistic Greek that in time they lost their specific reference, 
and passed into the general meaning ' to redeem ' or ^ set free.' 
Hence it is used commonly by Aquik, and occasionally by 
Symmachus and Theodotion, where the LXX. uses i^aipelv, 

Χυτρονν, ρνίσθαι'. Gen. 48. 1 6 LXX. ρνόμ€Ρ09, Aquila άγχίστεύων: 
Ps. 118 (119). 153 LXX. λύτρωσαΐ pe, Aquila άγχίστ€νσόν pel Prov. 

23. II LXX. 6 XvTpovpevos, Aquila, Symmachus and Theodotion 
ayxiarevs: Is. 35. 9 LXX. λελντρωμίροι, Aquila and Symmachus 
άγχιστ€νμίνοί, Theodotion ippvapepoi : Is. 47. 4 and 54. 5 LXX. 
ρυσάμενος, Aquila άγχιστευων : Is. 60. 1 6 LXX. e^aipovpevos, Aquila 
άγχιστ€ν$ : Is. 63. 1 6 LXX. ρυσαι, Aquila dyxiaTevaat. 

The application of this test seems to show clearly that 
the inference which was derived from the interchange of 
the words in the LXX. is valid : its validity is rather 
strengthened than weakened by the admission of a new 
member into the group of virtual synonyms. 

IV. Inferences which are drawn from the LXX. in 
regard to the meaning, and especially in regard to the 
equivalence in meaning, of certain words may sometimes be 
further checked and tested by an examination of the various 
readings of the MSS. of the LXX. For in those MSS. 
it is not unfrequently found that a word is replaced by 
another of similar meaning : e. g. in Prov. 8. 30, Codd. 
A Β have τριβών, Cod. S^ has όδώζ;, in Prov. 11. 9, Codd. 
A Β have άσ€βών, Cod. S^ has αμαρτωλών. These pheno- 
mena may be explained on more than one hypothesis : 
they may be survivals of other translations : or they may 
be signs of successive revisions : or they may be indications 
that the copyists dealt more freely with a translation than 


they would have dealt with an original work, and that they 
took upon themselves to displace a word for another which 
they thought more appropriate. But whatever be the 
origin of the phenomena, they afford additional data for 
determining the meanings of words, if not in the time of 
the original translators, at least in that of early revisers 
and copyists. They consequently may be used in the 
same way as the fragments of the Hexapla to test 
inferences as to the equivalence of words. 

The following is an example of a partial application of 
the test to the same group of words which has been already 
discussed in its use both in the LXX. and the Hexapla. 
It will be noted that only the historical books have been 

In Judges 6. 9, Codd. IV, 54, 58, 108 al. read ('ρρυσ-άμην, Codd. 
X, XI, 15, 18, 19 αι. read (ξβϊΚάμην (e^etXo^r/i/) as the translation of 
P2fJ: in Judges 9. 17 the same two groups of MSS. vary between 
ippvaaro and ίξζίλατο, and in Judges 18. 28 between 6 βυόμ^νος and 
6 €ξαιρονμ€νος : in 2 Sam. 12. 7 Codd. X, XI, 15, 18, 85 have eppv- 
σάμην, Codd. 82, 93 εξαλάμην I in 2 Sam. 14, 16 Codd. X. 92, 108, 
242 have ρνσάσθω, Codd. XI, 29, 44, 52, 56 al. i^eXelrai: in 2 Sam. 
19. 9 Codd. X, XI, 29, 44, 55 a/, have ^ρρύσατο, Codd. 19, 82, 93, 

108 e^eiXero : in 2 Sam. 22. 18 Codd. X, XI, 29, 44, 55 have (ρρνσατο, 

Codd. 19, 82, 93, 108 e^eiXero : in 2 Sam. 22. 44 Codd. X, XI, 29, 
44, 55 have ρνστ}^ Codd. 19, 82, 93, 108 have i^elXov. 

These instances are sufficient to show that the general inference 
as to the identity in meaning of i^aipeiv and pveaOat is supported by 
their interchange in the MSS., as it was also supported by their 
interchange in the Hexapla. 

If we now put together the several groups of facts to 
which attention has been directed, it will be possible to 
draw some general inferences, and to frame some general 
rules, for the investigation of the meanings of words in the 
New Testament. 

There are two great classes of such words, one of which 

may be subdivided : 



I. (a) There are some words which are common to 
Biblical Greek and contemporary secular Greek, and which, 
since they are designations of concrete ideas, are not 
appreciably afifected by the fact that Biblical Greek is the 
Greek of a Semitic race. The evidence as to the meaning 
of such words may be sought in any contemporary records, 
but especially in records which reflect the ordinary ver- 
nacular rather than the artificial literary Greek of the 

Instances of such words will be found below in ayyapeveiv, γλωσ- 

σόκομον, σνκοφαντύν. 

(δ) There are some words which are common to Biblical 
Greek and to contemporary secular Greek, in regard to 
which, though they express not concrete but abstract 
ideas, there is a presumption that their Biblical use does 
not vary to any appreciable extent from their secular use, 
from the fact that they are found only in those parts of the 
New Testament whose style is least affected by Semitic 
conceptions and forms of speech. The evidence as to the 
meaning of such words may be gathered from any contem- 
porary records, whether Biblical or secular. 

An instance of such words will be found below in Βζίσώαψονία. 

II. The great majority of New Testament words are 
words which, though for the most part common to Biblical 
and to contemporary secular Greek, express in their 
Biblical use the conceptions of a Semitic race, and which 
must consequently be examined by the light of the cognate 
documents which form the LXX. 

These words are so numerous, and a student is so 
frequently misled by his familiarity with their classical 
use, that it is a safe rule to let no word, even the 
simplest, in the N. T. pass unchallenged. The process of 
enquiry is (j) to ascertain the Classical use of a word, 
(2) to ascertain whether there are any facts in relation to 
its Biblical use which raise a presumption that its Classical 


use had been altered. Such facts are afforded partly by 
the context in which the word is found, but mainly by its 
relation to the Hebrew words which it is used to translate. 
It is obvious that the determination of this relation is a 
task of considerable difficulty. The extent and variety of 
the LXX., the freedom which its authors allowed them- 
selves, the existence of several revisions of it, necessitate 
the employment of careful and cautious methods in the 
study of it. As yet, no canons have been formulated for 
the study of it ; and the final formulating of canons must 
from the nature of the case rather follow than precede the 
investigations which these essays are designed to stimulate. 
But two such canons will be almost self-evident : — 

(i) A word which is used uniformly, or with few and 
intelligible exceptions, as the translation of the same 
Hebrew word, must be held to have in Biblical Greek 
the same meaning as that Hebrew word. 

(2) Words which are used interchangeably as transla- 
tions of the same Hebrew word, or group of cognate 
words, must be held to have in Biblical Greek an allied 
or virtually identical meaning. 

D 2 


Of the application of the principles and methods which 
have been described in the preceding essay the following 
short studies are examples. 

Some of the words have been selected on account of the 
interest or importance which attaches to their use in the 
New Testament, some on account of their being clear 
instances of contrast between Classical and Biblical Greek, 
and some also to illustrate the variety of the evidence 
which is available. They fall into two groups, correspond- 
ing to the two great classes into which all words in Biblical 
Greek may be divided, some of them having meanings 
which are common to Biblical Greek and to contemporary 
secular Greek, and some of them having meanings which 
are peculiar to the former, and which, even if suspected, 
could not be proved without the evidence which is afforded 
by the versions of the Old Testament. There has been an 
endeavour in regard to both groups of words to exclude 
evidence which is not strictly germane to the chief object of 
enquiry ; but it will be noted that in some instances 
evidence of the special use of words in Biblical Greek has 
been gathered from sources which have not been described 
in the preceding essay, and which require a more elaborate 
discussion than can be attempted in the present work, viz. 
from writers of the sub- Apostolic age who had presumably 
not lost the traditions of Biblical Greek, and who confirm 

ayyapevetv. 37 

certain inferences as to the meanings of New Testament 
words by showing that those meanings lasted on until the 
second century A. D. 


1. Classical use. 

In Classical Greek this word and its paronyms were 
used with strict reference to the Persian system of mounted 
couriers which is described in Herod. 8. 98, Xen. Oyr. 8. 6, 

2. Post-Classical use. 

Under the successors of the Persians in the East, and 
under the Roman Empire, the earlier system had developed 
into a system not of postal service, but of the forced trans- 
port of military baggage by the inhabitants of a country 
through which troops, whether on a campaign or otherwise, 
were passing. 

The earliest indication of this system is a letter of Demetrius 
Soter to the high priest Jonathan and the Jewish nation (Jos. Ani. 
13. 2. 3), in which among other privileges which he concedes to 
them he exempts their baggage animals from forced service, «eXevw 
be μη8€ αγγαρβύεσθαι τα ^Ιουδαίων νττοζνγια. 

In the important inscription of a.d. 49, Corp. Inscr. Gr. No. 4956, 
A 21, found in the gateway of the temple in the Great Oasis, there 
is a decree of Capito, prefect of Egypt, which, after reciting that 
many exactions had been made, goes on to order that soldiers of 
any degree when passing through the several districts are not to 
make any requisitions or to employ forced transport unless they 
have the prefect's written authorization [μφ^ν Χαμβάνειν μφ& άγγα- 
peueiK ei μη Tives €μα διπΧώματα €χωσι\ 

Epictetus, Diss. 4. ι. 79? arguing that a man is not master of his 
body, but holds it subject to any one who is stronger than it, takes 
the case of a man s pack-ass being seized by a soldier for forced 
service : ' don't resist,' he says, ' nay, don't even grumble. If you 
do^ you'll not only be beaten, but lose your ass as well, all the 


same ' (αϊ' δ' άγγαρβία § και στρατιώτης βπιλάβηται, αφΐς μη avriTfive μηΒ€ 
γόγγυζ^' d Be μη π'Κηγάς Χαβων ovbev ήττον άπόλβϊς και το ονάριον). 

The extent to which this system prevailed is seen in the 
elaborate provisions of the later Roman law : angariae 
came to be one of those modes of taxing property which 
under the vicious system of the Empire ruined both indi- 
viduals and communities. A title of the Theodosian Code, 
lib. 8, tit. 5, is devoted to various provisions respecting it, 
limiting the number of horses to be employed and the 
weights which were to be carried in the carts. 

3. Use in the N. T. 

Hence ayyap^veiv is used in S. Matt. 27. ^2, S. Mark 15. 31 
in reference to Simon the Cyrenian, who was pressed by the 
Roman soldiers who were escorting our Lord not merely to 
accompany them but also to carry a load. 

Hence also in S. Matt. 5• 4i the meaning is probably not 
merely ' whosoever shall compel thee to go one mile,' but 
' whosoever shall compel thee to carry his baggage one 
mile': and there may be a reference, as in S. Luke 3. 14, to 
the oppressive conduct of the Roman soldiers. 

1. Post-Classical use. 

That the word was sometimes used in post- Classical 
Greek of reading aloud with comments is shown by its 
use in Epictetus. 

In Epictet. Diss. 3. 23. ijo, there is a scene from the 
student-life of Nicopolis. A student is supposed to be 
' reading ' the Memorabilia of Xenophon : it is clear that 
he not merely reads but comments. 

Πολλά /CIS (θαύμασα τ'ισι ποτ€ Xoyois . . . Ί have often wondered on 
what grounds . . . ' (these are the words of Xenophon, Afem. i. i, 
upon which the * Reader ' comments). 

αναγινωσ•Κ€ΐν, αττοστοματιζειν. 39 

οϋ' αλλά τίνι ποτέ λόγω, ' Νο : rather, On what ground : this is a 
more finished expression than the other ' (this is the comment of the 

μη yap αΧλως αυτά άμεγμώκατε η ως ωδάρια ; ' Why, you do not lec- 
ture upon it any differently than you would upon a poem, do you ? ' 
(these are the words of Epictetus, finding fault with this way of 
lecturing upon the words of 2, philosopher). 

The students appear to have ' read ' or lectured in the 
presence of the professor, who made remarks upon their 
reading : for which the technical word was Ιτταναγίνώσκζίν, 
Epict. Diss. I. 10. 8. 

2. Use in the N. T. 

It is probable that this practice of reading with com- 
ments explains the parenthesis in S. Matt. 24. 15, S. Mark 
13. 14 άναγίνώσκων νο€ίτω, ' let him who reads, and com- 
ments upon, these words in the assembly take especial care 
to understand them.' It may also account for the co-ordi- 
nation of ' reading ' with exhortation and teaching in S. 
Paul's charge to Timothy, i Tim. 4. 13. 


1. Classical use. 

In its Classical use the word is used of a master dictating 
to a pupil a passage to be learnt by heart and afterwards 
recited : Plat. Euthyd, 1^6 c όταν ovv ns άποστοματίζβι otlovv, 
ov γράμματα αττοστοματίζβι ; ' when, then, any one dictates 
a passage to be learnt, is it not letters that he dictates?' 

2. Post-Classical use. 

But in its later use the meaning of the word widened 
from the recitation of a lesson which had been dictated to 
the answering of any question which a teacher put in regard 
to what he had taught : Pollux 2. lo:^ defines it as νπο του 
διδασκάλου €ρωτασθαι τα μαθήματα» 


a Use in the N. T. 

Hence its use in S. Luke ii' S3 ^ρζο-ντο ol γραμματείς καΐ 
ot Φαρισαϊου . . . άτΓοστοματιζ€ΐ>' αντον ττερΧ ττλείόνων, ' they began 
to put questions to him as if they were questioning a pupil 
on points of theology/ 


1. Use in the LXX. 

The word occurs in the following passages of the 
canonical books : 

(i) In the two following passages it is the translation of Ί\η 
' glory/ 

Had. 3. 3 eKoKv^ev ovpavovs ή άρετη αντον, ' his glory covered the 
heavens': another translator in the Hexapla renders ΊίΠ by τψ 

€νπρ€7Γ€ΐαν της δόξης αντον. 

Zach, 6. 13 fai αντος Χτγ^^ται άρ€τήΐ' (of the Branch), ' and he shall 
bear the glory ' : other translators in the Hexapla render Ί'ΐΠ by 
ίπιδοξότητα, €νπρεπ€ΐαν, δόξαν. 

(2) In the four following passages it is the transladon of 
Π?ΠΓ1 ' praise/ 

Is. 42. 8 την δόξαν μον cTepco ov δώσω ovbe τάς άρ€τά§ μον rots 
γλνπτοϊς, ' my glory will I not give to another, neither my praise to 
graven images ' : tus άρ€τάς is corrected by Aquila to τψ νμνησιν, 
by Symmachus to τ6ν enaivov. 

Is. 42. 12 δωσονσι τω β^ω 8όξαν, τάς aperas αντον iu ταΐς νησοις 
avayy€\ovai, ' they shall give glory to God, His praises shall they 
declare in the islands/ 

Is. 43. 21 \αόν μον ov ττ^ρι^ττοιησόμην τάς άρέτάς μον διηγ^ΐσθαι, 'my 
people which I acquired for myself to show forth my praises ' : 
Symmachus corrects Tas άρ^τάς to τ6ν νμνον. 

Is. 63.7 τον eXeov κνρίον βμνησθην, τά,ς aperas κνρίον, ' I will mention 
the lovingkindness of the Lord, the praises of the Lord ' : another 
translator in the Hexapla corrects τάς άρετάς to αϊνεσιν. 

Outside the canonical books the word occurs once in an 
apocryphal addition to the book of Esther, and three times 
in the Wisdom of Solomon. 

αρετή. 4 ^ 

Fs/h. 4. 17» lin^ 33j ed. Tisch. (Esther prays God for help 
against the efforts which the heathen were making) : άνοΐξαι στόμα 
€θνών eis άρ€τά$ ματαίων, ' to Open the mouth of the Gentiles for the 
praises of vain idols.' The translation of aperas by 'praises' is 
supported by the Vulgate ' laudent/ 

Wz'sd. 4. i; 5. 13; 8. 7: there can be no doubt that in these 
passages άρ^τη has its ordinary Classical meaning, and not the 
meaning which it has in the LXX. : in 8. 7 the aperai are enume- 
rated, viz. σωφροσύνη, φρόνησις, δικαιοσύνη, avdpeia. 

2. Use in the N. T. 

In the N. T. the vi^ord occurs in the Epistle to the 
Philippians, and in the two Epistles of St. Peter. 

Ρ/ΐΐ'ί. 4. 8 TO λοιπόν, ά^ζλφοί, οσα €στ\ν αληθή, οσα σψνά, οσα δίκαια, 
οσα αγνά, οσα προσφιλή, οσα εύφημα, ε'ί τις άρ€ττ) κα\ ei tis έπαινος, ταύτα 
λογίζ€σθ€ : since άρ^τή is here coordinated with ίπαινος and follows 
immediately after εύφημα, its most appropriate meaning will be that 
which it has in the canonical books of the O. T. as a translation of 
^Sn or ^\\}^, viz. ' glory ' or ' praise.' 

1 Pel. 2 . 9 όπως τάξ άρετάς (ξα-γ-γίίλητε τον €Κ σκότους υμάς καλίσαντος. 

It seems most appropriate, especially when the general philo- 
logical character of the Epistle is taken into consideration, to give 
the word the LXX. meaning of ' praises.' 

2 Pe/. I. 3 δια τής ^πιγνωσεως του καλ^σαντυς ημάς ιδ/α ^όζτ] κα\ 

Here also the coordination with Βόξα, as in Is. 42. 8, 12, seems 
to make the meaning ' praise ' more appropriate than any other : 
the use of the singular has its parallels in Hab. 3. 3, Zach. 6. 13. 

2 Pel. I. 5 €πιχορηγησατί iv ttj πίστΐΐ υμών ΤΎ\ν άρετήκ, iv be τή 
apcTrj την γνώσιν. 

This is the most obscure use of the word in the N. T. : nor, in 
the absence of philological indications, can its meaning be deter- 
mined without a discussion of the general scope both of the passage 
and of the whole Episde, which belongs rather to exegesis than 
to philology. 



1. Classical use. 

The word, in the form γΧωσσοκομα,ον, is very rare in Attic 
Greek, being chiefly known to us from a quotation by 
Pollux lo. 154 of a fragment of the Bacchae of Lysippus, 
a poet of the Old Comedy, which however is sufficient to 
show its derivation from γλώσσα in the sense of the tongue 
or reed of a musical pipe or clarionet : αντοί^ avXols ορμά [so 
IBentley, Ad Hemster/i. p. 69, for 6p/xat] καί γλωττοκομείω '(the 
piper) rushes in with his pipes and tongue-case.' 

2. Use in later Greek. 

But of this first and literal use there is no trace in later 
Greek. In the LXX. it is used (i) in 2 Sam. 6. 11, Codd. 
A. 247, and Aquila, of the Ark of the Lord, = Cod. B.and 
most cursives η κιβωτοί, (ζ) in 2 Chron. 24. 8, 10, 11 of the 
chest which was placed by order of Joash at the gate of the 
temple to receive contributions for its repair, = in the 
corresponding passages of 2 Kings 12 η κιβωτός. It is also 
used for the Ark of the Covenant by Aquila in Exod. 25. 
^° • 3^ {^1)' ^ ' ^^^ Josephus, Ant. 6. 1, 2, uses it for the 
* coffer ' into which were put * the jewels of gold ' ' for a 
trespass-offering' when the Ark was sent back (i Sam. 6. 
8 = LXX. Θ4μα), 

In a long inscription from one of the Sporades, probably 
Thera, known as the Testamentum Epictetae, and now at 
Verona, which contains the regulations of an association 
founded by one Epicteta, γλωσσόκομον is the ' strong-box ' 
or muniment-chest of the association, and is in the special 
custody of the γραμματοφνλαξ or ' registrar.' 

This wider meaning is recognized by the later Atticists : 
for Phrynichus, § 79 (ed. Rutherford, p. 18) defines it as 
βιβλίων η Ιματίων η αργύρου η δτωυν άλλον. 

'γΧωσσ-όκομον^ ^€ΐ<τί3αίμονία. 43 

3. Use in the Ν. Τ. 

It is found in the N. T. only in S. John 13. 6 : 13. 39, 
where it is appropriately used of the common chest of our 
Lord and His disciples, out of which were not only their 
own wants provided but also the poor relieved. 

In still later Greek this wide use of it was again narrowed : 
it was used, at last exclusively, of a wooden coffin, σορόξ 
having apparently come to be used only of a stone-coffin or 
sarcophagus. The earliest instance of this use is probably 
in Aquila's version of Gen. 50. 26. In modern Greek it 
means a purse or bag. 

δβίσιδαίμων, δβισιδαιμονία, 

1. Classical use. 

It is clear that the dominant if not the only sense of 
these words in Classical Greek is a good one, ' religious/ 
'religion' : e.g. 

Xenophon, Cyrop. 3. 3. 58, tells the story of Cyrus, before attack- 
ing the Assyrians, beginning the accustomed battle-hymn and of 
the soldiers piously {θίοσββως) taking up the strain with a loud 
voice : ' for it is under circumstances such as these that those who 
fear the gods (ol δ€ΐσιδαιμομ€$) are less afraid of men/ 

Aristotle, Pol. 5. 11, p. 1315 a, says that rulers should be con- 
spicuously observant of their duties to the gods : ' for men are less 
afraid of being unjustly treated by them if they see a ruler religious 
(δεισιδαίμομα) and observant of the gods, and they plot against him 
less because they consider that he has the gods also as his allies.' 

In this last instance the reference is probably to the outward 
observance of religion : and that this was implied in the words is 
shown by a senatus consultum of b. c. 38, which is preserved in 
an inscription at Aphrodisias in Caria {Corp. Inscr. Gr., No. 2737 b). 
The senatus consultum decrees that the precinct (r^evos) of 
Aphrodite shall be held as consecrated, ' with the same rights and 
the same religious observances, ταντω 8ίκαίω ravrfj re δεισιδαιμομια 
(eodem jure eademque religione), as the precinct of the Ephesian 
goddess at Ephesus.' 


2. Post-Classical use. 

In later Greek the words have a meaning which is 
probably first found in Theophrast. Char act. i6, a/xeXet ή 
δεισιδαιμονία bo^€L€v αν elvaL δείλια irpbs rovs Oeovs '. ' no doubt 
^€ίσώαίμονία will be thought to be a feeling of cowardice in 
relation to the gods : ' they are used not of the due 
reverence of the gods, which is religion, but of the excessive 
fear of them, which constitutes superstition. Of this there 
are several proofs : — 

(i) Philo repeatedly distinguishes ^ασώαψονία from ^νσφύαΐ 
e.g. De Sacrif. Abel et Cain, c. 4 (i. 166), where he speaks of 
the way in which nurses foster fear and cowardice and other mis- 
chiefs in the minds of young children 'by means of habits and 
usages which drive away piety, and produce superstition — a thing 

akin to impiety,' bi idatu kcu νομίμων βνσζβείαν μέν ΙΚαννόντων δεισι- 
δαιμομίαΐ' δε πράγμα αδελφοί' άσφξ'ια κατασκΐυαζόντων. Again, in Quod 
Deus immut. c. 35 (i. 297), he defines it more precisely in Aris- 
totelian language as the ' excess ' of which impiety is the corre- 
sponding ' defect ' and piety {^υσφάα) the ' mean ' : cf De Gigan- 
tibus, c. 4 (i. 264): De Ρ latitat. Nee, c. 25 (i. 345): De Jusiitia, 
c. 2 (ii. 360). 

(2) Josephus, Ant. 15. 8, 2, relates that, among the other means 
which Herod adopted for adorning the amphitheatre which he had 
built at Jerusalem, he erected trophies in the Roman fashion with 
the spoils of the tribes whom he had conquered. The Jews thought 
that they were men clad in armour, and that they came within the 
prohibition of the divine law against images. A popular tumult 
was threatened. Herod, wishing to avoid the use of force, talked 
to some of the people, trying to draw them away from their super- 
stition (τη? δεισιδαιμοΐ'ία? αφαιρούμενος), but without SUCCesS, until he 
took some of them into the theatre and showed them that the 
armour was fixed on bare pieces of wood. 

(3) Plutarch has a treatise Περί δεισιδαιμοΐ'ΐας {Moral, vol. ii. pp. 
165 sqq.), which begins by saying that the stream of ignorance 
about divine things divides at its source into two channels, becoming 
in the harder natures atheism (άθεότης), in the softer, superstition 


δεισιδαιμονία ^ διάβολος. 45 

(4) Μ. Aurelius, 6. 30, in painting the almost ideal character of 
his adopted father, speaks of him as ' god-fearing without being 
superstitious ' {θ^οσφης χωρίς 8€ΐσιδαιμοΐ'ία9). 

It seems clear from these facts that in the first century 
and a half of the Christian era the words had come to have 
in ordinary Greek a bad or at least a depreciatory sense. 
That it had this sense in Christian circles as well as outside 
them is clear from its use in Justin M. Aj^ol. i. 2,, where it 
is part of his complimentary introduction to those to whom 
his Apology is addressed that they are ' not men who are 
under the dominion of prejudice or a desire to gratify 

superstitious persons ' (μη ιτροληψεΐ μηb^ ανθρω-παρ^σκ^ία ττ) 

δ€ΐσιδαιμόι/ωμ κατξχομ&ονξ), but that they can form a candid 
judgment on the arguments which are addressed to them. 

3. TTse in the N. T. 

This having been the current meaning, it is improbable 
that the words can be taken in any other sense in the two 
passages in which they occur in the Acts of the Apostles : 
in 17. 22 S. Paul tells the Athenians that they are 
δ^σιδαιμοί/εστερου?, * rather inclined to superstition ' : and in 
25. 19 Festus tells Agrippa that the charges which Paul's 
accusers bring against him are questions irepl rijs ibias 
δεισιδαιμομία?^ ' Concerning their own superstition.' 

διάβολος^ διαβάλλω, 
1. Classical use. 

These words were ordinarily used in reference to slan- 
derous, or at least malicious, accusation : bLaβάλλω is 
sometimes found in the probably earlier sense of setting 
at variance, e.g. Plat. Rep. 6. p. 498 d μη διάβαλλε e//e καΙ 
Θρασύμαχον άρτι φίλους yeyovoras, and, in the passive, of 
being at variance, e.g. Thucyd. 8. 83 καΙ irporepov τω Τισσα- 
(fS€pv€L άτηστοΰντξς ττολλω δή μάλλον ίτι διεβεβληΐ'το : but 


διά/3ολο9, whether as substantive or as adjective, seems 
invariably to have connoted malice. Hence the Atticists, 
e.g. Pollux 5. 18, coordinate Xoibopos, βλάσφημοί, διάβολος, 
and Lucian s treatise, Tie pi του μη βqbίωs ina-TeveLv Βιαβολτΐ', 
gives no trace of any other meaning. 

2. Use in the LXX. 

In Job and Zechariah, and also in Wisd. 2>. 24, δ διάβολος 
is clearly used of a single person, Jtptl?, the ' enemy' of man- 
kind. In the other passages in which it occurs it is used to 
translate either the same word or its equivalent in meaning, 
Ί^, but without the same reference to that single person. 
The passages are the following : — 

I Chron, 21. i άνβστη διάβολος h τω Ίσραηλ, of the * enemy' who 
stirred up David to number Israel (the E. V., following Codd. 1 9, 
93, 108, transliterates the Hebrew, * Satan'). 

J^sik. 7• 4 ού yap άξιος 6 διάβολος της αυλής του βασιλέως. 
Esth. 8. Ι οσα νπηρχ€Ρ Ά/χάι/ τω διαβάλω (Cod. S' OmitS τφ δ. but 
Codd. S'^ 249 add τών Ίονδαίωρ). 

In both these passages the Hebrew has "l? or ^"^^, which have no 
other connotation than that of hostility, and of which the former is 
ordinarily translated by έχθρας. 

Ps. 108 (109). 5 κα\ διάβολος στητω eK δεξιών αυτού. 

In JVumd. 22. 22 where the LXX. translates by άνεστη 6 αγγίλος 
του θίοΰ βνδιαβάλλπν (so Codd. A Β and most cursives, Ed. Sixt. 
διαβολών) αυτάν, Aquila transliterates the Hebrew (eis) σατάν, Theo- 
dotion translates by άντικύσθαι : so in Job i . 6, where the LXX. 
have 6 διάβολος, Aquila has σατάρ, Theodotion άντικ^ίμβρος. Con- 
versely in I Kings 11. 14, where the LXX. transliterates σατάρ, 
Aquila agrees with Theodotion in translating by άρτικΐίμζρος. 

In Numb. 22. 32 where the LXX. has κα\ Ιδού βγώ 4ξηλθορ «s 
διαβολψ σον, Symmachus translates by €ραρτιονσθαι, Theodotion by 


The Hebrew word in both passages is ]ψψ. 

It seems to be clear that the LXX. used διά/3ολο5 and its 

^ίάβολοξ, 8ιαθηκη, Λ 7 

paronyms with the general connotation of enmity, and 
without implying accusation whether true or false. 

3. Use in the N. T. 

In the New Testament διάβολος is invariably used as a 
proper name, except in the Pastoral Epistles, where it is 
also used as an adjective, and when so used has its 
ordinary meaning of 'slanderous' (i Tim. 3. 11 ; ζ Tim. 
3. 3 ; Tit. 2. 3). But when used as a proper name there is 
no reason for supposing that it is used in any other sense 
than that which it has in the LXX., viz. as the equivalent 
of ]tpt? and as meaning ' enemy.' 

διαβάλλω occurs only once, viz. S. Luke i5. i of the ' unjust 
steward ' : the accusation was presumably true, and hence the 
meaning of slander would be inappropriate; so Euseb. 
H.E. 3. 39. 16, referring to Papias and possibly using his 
words, speaks of the woman who was taken in adultery ' in 
the very act ' as yvvaLKos . . . διαβληθείσης iirl του κυρίου. 


1. Classical use. 

The word has at least two meanings, (i) a * disposition ' 
of property by will, which is its most ordinary use, (2) a 
' covenant,' which is a rare meaning, but clearly established 
e.g. by Aristoph. Av. 4^g. 

2. Use in the LXX. 

It occurs nearly 280 times in the LXX. proper, i.e. in 
the parts which have a Hebrew original, and in all but 
four passages it is the translation of ΓΙ'^ΊΙ « covenant ' : in 
those passages it is the translation respectively of Ή^Π^^ 
'brotherhood,' Zech. 11. 14, Ί1"] 'word,' Deut. 9. 5, and 
Π*»Ί2Π ^ΊΙ"! 'words of the covenant,' Jer. 41 (34). 18 ; in 


Ex. 31. 7 την κίβωτον της διαθήκης takes the place of the 
more usual την κιβωτον τον μαρτυρίου. 

In the Apocryphal books, which do not admit of being 
tested by the Hebrew, it occurs frequently and always in 
the same sense of ' covenant.' 

3. Use in the Hexapla. 

The Hexapla Revisers sometimes change it to that which 
is the more usual Greek word for ' covenant/ viz. συνθήκη : 
e.g. Aquil. Symm. Gen. 6. 18 : Aquil. Theod. i Sam. 6. 19 : 
Aquil Symm. Ps. 24 (25). 10. This fact accentuates and 
proves the peculiarity of its use in the LXX. 

4. Use in Philo. 

In Philo it has the same sense as in the LXX. : e. g. De 
Somniis 2. 2)?>'> ^'^' i• P• ^^^3 where he speaks of God's 
covenant as Law and Reason, z^o/xo? ^e Ιση και λόγο^ : cf. 
Justin Μ. Tryph. c. 43, where he speaks of Christ as being 
the cCmvio^ νόμο^ καΐ καινή Βιαθήκη. 

5. Use in the Ν. Τ. 

There can be little doubt that the word must be invariably 
taken in this sense of ' covenant ' in the N. T., and especially 
in a book which is so impregnated with the language of the 
LXX. as the Epistle to the Hebrews. The attempt to 
give it in certain passages its Classical meaning of ' testa- 
ment' is not only at variance with its use in Hellenistic 
Greek, but probably also the survival of a mistake : in 
ignorance of the philology of later and vulgar Latin, it was 
formerly supposed that ' testamentum,' by which the word 
is rendered in the early Latin versions as well as in the 
Vulgate, meant ' testament ' or ' will,' whereas in fact it 
meant also, if not exclusively, ' covenant.' 

SiKaio?, δικαιοσύνη, 49 

δίκαιοι, δικαιοσύνη, 

1. Use in the LXX. and Hexapla. 

Into the Classical meaning of these words it is hardly 
necessary to enter.; that meaning is found also in both the 
LXX. and the N. T. : but intertwined with it is another 
meaning which is peculiar to Hellenistic Greek. The 
existence of this meaning is established partly by the 
meaning of the Hebrew words which δίκαιο?, ^ίκαιοσννη 
are used to translate, and partly by the meaning of the 
Greek words with which they are interchanged. 

(i) ΊΟΠ 'kindness ' is usually (i. e. more than 100 times) trans- 
lated by eXeos, sometimes by Ιλΐημοσννη, ίΚεήμων. but nine times 
(Gen., Ex., Prov., Is.) it is translated by δικαιοσύνη, and once by 

Conversely, Πζ"ΐν 'justice,' which is usually translated by δικαιο- 
3-ύνη, is nine times translated by (λ^ημοσύνη, and three times by 


(2) Sometimes the LXX. δικαιοσύνη is changed by the Hexapla 
Revisers into ΙΚΐημοσννη, and sometimes the reverse : apparently 
with the view of rendering 'IDH uniformly by ίλ^ημοσύρη, and '"^ζΊν 
by δικαιοσύνη : for example — 

Exod. 15. 13 LXX. δικαιοσύνη, Aquil. ΐΚΐημοσύνη. 
Oeut, 24. 13 LXX. (λζημοσύνη, Aquil. δικαιοσύνη, 

I Sam. 12. 7 LXX. δικαιοσύνη, Symm. ίλ^ημοσύνη. So also Ps. 
30 (31). 2 : 35 (36). II : 105 (106). 3. 

Ps. 32 (33). 5 LXX. €λ€ημοσύνην, Aquil., Int. Quint, δικαιοσύνην. 
Is. I. 27 LXX. βλίημοσύνης, Aquil., Symm., Theod. δικαιοσύνης. 
So also 28. 17. 

Is. 56. I LXX. TKcos, Aquil., Symm., Theod. δικαιοσύνη. 

Is. 59• 16 LXX. ζΚζημοσύντ}, Theod. δικαιοσύντ]. 
Dan. 9. 16 LXX. δικαιοσύνην^ Theod. ^Χ^ημοσύντι. 

This revision seems to show that the sense in which 
δικαιοσύη] is used in the LXX. was not universally accepted, 
but was a local peculiarity of the country in which that 



translation was made. The same tendency to the revision 
of the word is seen in some MSS. : e. g. in Ps. 34 (^^). 24, 
where all MSS. (except one cursive, which has eAeos) read 
Βικαιοαύητ]!/^ Cod. S reads ζλζημοσννην, and in Ps. 37 (38). '21, 
where Codd. A Β and many cursives read δικαιοσυη^ί^, Cod. 
S^ and many other cursives read άγαθωσηύνην (-οσννην). 

The context of many of these passages shows that the 
meanings of the two words Ιικαιοσννη and ^λζημοσννη had 
interpenetrated each other : 

(a) Sometimes, where ίλβημοσννη is used to translate •^ij'jy, no 
other meaning than ' righteousness' is possible : e. g. 

Oeut. 6. 25 €Κ€ημοσύνη (σται ήμίρ iav ψυΧασσώμίθα noieiv πάσας τας 

fVToXas ταύτας . . . 'It shall be our righteousness if we observe to 
do all these commandments ... * 

Oeut, 24. 13 (15) . . . και %σται σοι ίΚίημοσννη evavriov κυρίου του 
θ€θϋ σου, 

(' In any case thou shalt deliver him his pledge again when the 
sun goeth down) . . . and it shall be righteousness unto thee 
before the Lord thy God.' 

(δ) Conversely, sometimes, where δικαιοσύνη is used to render 
*lDn, no other meaning than 'kindness' or 'mercy' is possible: 

Gen. 19. 19 (Lot said after having been brought out of Sodom) 
«TTftd^ eiipev 6 παΙς σου eXeos ivavTiov σου και ΐμ€-γάλυνας την Βικαιοσύητ)!' 
σου . . . 

* Since thy servant hath found grace in thy sight, and thou hast 
magnified thy mercy which thou showest unto me in saving my 
life . . .' 

Gen. 24. 27 (when Eliezer is told that the damsel is the daughter 
of Bethuel, he blesses God) os ουκ εγκατίλιπε την δικαιοσύνης αυτοϋ 
κα\ την άληθβιαν άπο του κυρίου μου. 

'Who hath not left destitute my master of his mercy and his 

2. Use in the N. T. 

There is one passage of the N. T. in which this meaning 
of bLκaLoσvuη is so clear that scribes who were unaware of 
its existence altered the text ; in S. Matt. 6. i the estab- 

δικαιοσύνη, ίτοιμάζαν, ζΐ 

lished reading is undoubtedly δικαιοσύμηκ, for which the later 
uncials and most cursives have ^Κ^ημοσννην, and for which 
also an early reviser of Cod. i^, as in some similar cases in 
the LXX., substituted boaiv. 

There is no other passage of the N. T. in which it is clear 
that this meaning attaches to either bUaios or δικαιοσύνη : but 
at the same time it gives a better sense than any other to 
the difficult statement about Joseph in S. Matt. 1. 19 'Ιωσήφ 
δβ ό ανηρ avTtjs δίκαιο9 ωμ καΐ μη Θίλων αυτήν '^ζΐ'γματίσαι^ 
'Joseph her husband, being a kindly man^ and since he was 
not willing to make her a public example . . .' 

έτοίμάζβίρ, ίτοιμασία^ έτοιμοι' 

1. Use in the LXX. 

In the great majority of instances €Τοιμάζ€ΐν, ετοιμασία, 
€τοιμο9 are used in the LXX. to translate p3 or one of its 
derivatives. That word, which properly means 'to stand 
upright,' was used in the meanings ^ to set upright,' ' to 
make firm ' (e.g. 2 Sam. 7. 13 ' I will stablish the throne of 
his kingdom for ever'), and hence in the more general 
meanings ' to make ready,' ' to prepare ' (e. g. Job 29. 7 
' when I prepared my seat in the street,' Deut. 19. 3 thou 
shalt prepare thee the way '). This latter use being the 
more common use of the word, it was ordinarily translated 
by k.τoιμaζζ.ιv^ which in Classical Greek has no other mean- 
ing. But the use of this Greek word in the Septuagint 
affords an interesting illustration of the manner in which 
the meaning of the Hebrew acted upon the Greek ; for it 
is clear that it came to have some of the special meanings 
of the Hebrew ' to set upright,' ' to establish,' ' to make 

(i) The existence of that meaning when the Septuagint 
versions were made is shown by the use of words which 
undoubtedly express it : that is to say, "1^3 is translated by 

Ε 2 


(a) άμορθουΐ' 2 Sam. 7. 13, 16, 26, Prov. 24. 3, Jer. 10. 12 : 40 

(33)• 2. 

(<5) irtia-n\pil€iv Cod. A, Judges 16. 26, 30 ( = Cod. Β ίστάι/αι). 
(γ) θεμελιοΟμ Ps. 8. 4 : 47 (48). 9' ^6 (87). 5 ^ nS (119)• 9°• 
(ί/) κατορθου»' ι Chron. 16. 30, Ps. 95 (9 6)• ^θ• 
(^) στβρεοΟΐ' Ps. 92 (93)• 2. 

(:ζ) In similar passages, and sometimes in the same 
books, the same Hebrew word is translated by €τοίμάζ€ίν, 

e.g. (a) 2 Sam. 7. 13 άι/ορθώσω τον θρόνον αύτον, but ιδ. V. 12 
ετοιμάσω την βασι\€ίαν αυτοϋ : ιδ. V. 24 ήτοίμασας σίαυτω τον λαόν σου 
Ίσραηλ els Χαον «oy του αΙώνος Ι ΐ'δ. V. 20 (Cod. Α) ό oIkos του δούλου 
σου Δανιθ βσται ά,νωρθωμ,ένοζ ενώττιόν σου. 

(δ) Ps. 64 (65). 7 €Τθΐμάζωι/ opTy ev τ^ ίσχυί σον: Ps. 47 (48)• 9 ^ 
Beos έθεμβλίωσεί' ηντην els τον αΙωνα : Ps. 8. 4 σ€\ηνην και aarepas α συ 
έθεμελίωσας : Prov. 3• 1 9 ήτοίμασε δε ουρανούς ev φρονησ€ΐ. 

(c) Ps. 23 (24)• 2 67Γΐ ποταμών ήτοιμασεκ αυτί;!/ (ΧΓ. την οΙκουμ€νην): 
Ps. 95 (9^)• ΪΟ κατώρθωσε TJ71' οίκουμίνην ήτις ου σa\eυθησeτaι : Ps. 92 
(93)• 2 ίστερέωσε την οίκουμίνην ήτις ου σaKeυθησeτaι. 

In other words, €τοιμάζ€ίν is used interchangeably with 
άνορθονν, θ€μξλωνν, κατορΘοϋν, στ€ρ€οϋν as the translation of 

In the same way ^οιμασία is used to translate both the 
verb and its derivatives ]'^2^> "^J^^^j 'base,' or 'foundation/ 
or 'fixed seat'; and ίτοιμο^ is used to translate both ]iD^5 
and ]"^^5 (J)art. niph.) : e.g. 

I Kings 2. 45 θρόνος Δαυίδ «rrat Ιτοιμος evaymov κυρίου eh τον 

1 Kings 8. 39, 43, 49, 2 Chron. 6. 30, 33, 39, Ps. 32 (33). 14 
wll^^'jiDiSp f^ ετοίμου κατοικητηρίον σου. 

2 Esdr. 2. 68 του στηναι αυτόν enl την ετοιμασία^ αντοΰ. 

Ρ^' 5^ (57)• 8 : 107 (ΐθ8). ι : ΙΙΙ (112). 7 «τοίμη η καρδία μου. 
Ps. 88 (89). 15 δικαιοσύνη κα\ κρίμα ετοιμασία του θρόνου σον. 
Ps. 92 (93)• 3 έτοιμος ό θρόνος σον άπο τότε. 
Zach. 5• 1 1 θησουσιν αυτό CKel em την ίτοιμασίαν αυτού. 

It seems clear from these passages that, like ^οιμάζ^ιν, 

ίτοιμάζβιν, €Τθΐμο9» 53 

ΐΓοιμασία and crot/xos had come to have the meaning of the 
Hebrew words which they were used to translate. 

2. Use in the Hexapla. 

This inference that the three Greek words are used in the 
LXX. in the proper sense of ]15 and its derivatives, is 
strongly confirmed by their use in the Hexapla. 

(i) Sometimes they are replaced by words of whose use 
in the proper sense of p3 there is no doubt : 

Έχ. ΐ5• i7 LXX• «s Ιτοιμοκ κατοίκητηριόν σον, Aquil.^ Symm, 

€δρασμα eis καβίδμαν σον. 

Ibid. LXX. ήτοίμασαν, Aquil, ηΒρασαν. 

I Sam. 20. 31 LXX. ετοιμασθήσβται, Symm. ύΒρασβησεται, Alius 

1 Sam. 23. 33 LXX. eh έτοιμοι', Symm. «Vl βφαίω. 

2 Sam. 5• 12 LXX. ήτοΐμασεν, Symm. ηδρασ^ν. 
2 Sam. 7. 12 LXX. ετοιμάσω, Symm. εδράσω. 

2 Sam. 7• 24 LXX. ήτοίμασας, Symm. ηδρασας. 

Ps. 9. 8 LXX. ήτοιμασεμ iv κρίσει τον θρόνον, Symm. ηδρασεν, 

Ps. 9. 39 (10. 18) LXX. την ετοιμασίαΐ' τψ καρδίας ^ Symm. πρό- 

Ps. ΙΟ (ι ι). 2 LXX. ήτοίμασαΐ', Aquil., Symm. ήδρασαν, 

Ps. 20 (21). 13 LXX. ετοιμάσει?, Aquil., Symm. εδράσεις, 

Ps. 23 (24). 2 LXX. ήτοίμασεί', Aquil., Symm. ηδρασεν. 

Ps. 32 (33). 14 LXX. εξ έτοιμου κατοικητηρίον σον, Aquil. άττο 
εδράσματος καθέδρας αντον, Symm. άττό εδραίας (§. έδρας) κατοικίας αντοϋ. 

Ps. 56 (57)• ^ LXX. έτοιμη ή καρδία μον, Symm. εδραία ή κ. μον. 

Ps. 64 (65)• 7 LXX. ετοιμάζων ορη, Symm. ηδρασας ορη. 

lb. ν. 10 LXX. ΟΤΙ όντως ή ετοιμασία, Symm. δτι όντως ήδρασαδ 

Ps. 88 (89). 3 LXX. έτοιμασθήσεται, Symm. εδρασθησεται (but 
ib. V. 4 Symmachus retains ετοιμάσω). 

lb. V. 15 LXX. ετοιμασία τον θρόνον σον, Aquil. το εδρασμα, Symm. 

PrOV. 8. 27 LXX. ψοΐμαζε, Symm. ήδραζε. 

PrOV. 16. 12 LXX. ετοιμάζεται, Symm. Theod. εδρασθησεται. 

(ζ) Sometimes, on the contrary, they are substituted for 


other words which had been used in the Septuagint as 
translations of p5 : 

Gen. 41. 32 LXX. άΧηθΙς earai το βήμα, Aquil. ετοιμομ, Symm. 

Ps. 8. 4 LXX. ίθβμίλίωσας, Aquil. Theod. ήτοίμασαξ, Int. Sextus 


Ps. 86 (87). 5 LXX. και avTos ζθ^μ^Κίωσ^ν αυτήν 6 ύψιστος, Aquil. 
edpaaei, Symm. rjdpaaev, Theod. ήτοίμασει^. 

Prov. 4. 18 LXX. 6ω5 κατορθώστ} ή ημ^ρα, Aquil. (βως) έτοιμης 
ημέρας, Symm. {εως) idpaias ημέρας, Theod. εως ετοιμασίας ημέρας. Int. 
Quintus ετοιμασίας. 

Prov. 12.3 LXX. κατορθώσει, Aquil., Symm. €Τθΐμασθήσ€ται. 

Prov, 12. 20 LXX. κατορθόί, Aquil., Symm., Theod. €τοιμασθήσ€ται. 

Prov. 25. 5 LXX. κατορθώσει, Aquil., Symm. εδρασθησεται, Theod. 

This latter group of facts makes the inference certain that 
in the latter part of the second century €τοιμάζ€ίν was some- 
times used in Hellenistic Greek in the sense of ' to set 
upright/ 'to estabHsh/ 'to make firm/ eVot/xos in that of 
' established/ ' made firm/ and ετοιμασία in that of ' establish- 
ment/ ' firm foundation.' 

3. Use in the N. T. 

In the majority of passages in which the words ίτοιμάζζΐν, 
6Γοι/Λθί occur in the N. T., their ordinary meanings are 
sufficient to cover the obvious sense which is required by 
the context. There are some passages in which the 
secondary meaning which they bear in the LXX. and 
Hexapla is appropriate, if not necessary : for example, 

S. Matt. 20. 23, S. Mark 10. 40 οΧς ήτοίμασται : S. Matt. 25. 34 
την ήτοιμασμ^ΐ'ηΐ' νμλν βασϊΚείαν άπο καταβολής κόσμου : id. V. 41 '"ο ιτνρ 
το αΐώνιον, το ήτοιμασμ^μοί' [Cod. D et al. 6 ητοίμασεν ό πατήρ μου] τω 
διαβάλω κα\ τοις άγγελοις αντου : 1 Cor. 2. ^ ά ητοίμασεν ό θεός τοις 
άγαπωσιν αυτόν: Heb. II. 16 ητοΊμασε yap αυτοΧς ττόλιν. The nearest 
English equivalent in each of these passages would probably be 
'destined/ as in 2 Sam. 5. 12 (= i Chron. 14. 2) «γνω Δαυίδ on 

ετοιμασία, θρησκεία, 55 

ήτοίμασ€>' αντον Κύριος ds βασιλέα eVi Ίσρατ^λ, Tobit 6. ΐ8 /λ^ φοβον 
ΟΤΙ σο\ αυτή ήτοιμασμ^μη ην άττο του αιώνος. 

Ephes. 6. 15 υποδησάμ^νοι τους πόδα? eV Ιτοιμασία του eiayycXiov της 
(Ιρήνης. In this, which is the only instance of the use of ετοιμασία 
in the N. T., it seems most appropriate to take it in the sense 
which it has been shown to have elsewhere in Biblical Greek of 
' firm foundation/ or * firm footing.' This view is confirmed by the 
use of the instrumental eV Avhich, though not without Classical 
parallels (e. g. Hom. //. 5. 368 δήσαν κρατ^ρω iv\ δβσ/Μω), gives to the 
passage a strong Hellenistic colouring. 


1. Classical use. 

The word is used by Herodotus 2. 37 of the ceremonial 
observances of the Egyptian priests: it does not appear to 
occur in Attic Greek. 

2. Use in the LXX. 

In the LXX. it is found in Wisdom 14. 18, 27 of the 
worship of idols, η των ανωνύμων εώάλων θρησκεία : and in 
4 Mace. 5. 6 of the religion of the Jews, in relation to its 
prohibition of the eating of swine's flesh, as r^ 'Ιουδαίων 
θρησκεία. Symmachus uses it in Dan. 2. 46 of the worship 
paid to Daniel by Nebuchadnezzar's orders (LXX. εττέταξε 
θυσίας καΐ aiTovbas ττοίησαί αυτω), and in Jer. 3. 19, Ezek. 20. 
6, 15 as a translation of "^1^. 

3. Use in Philo and Josephus. 

Its use is equally clear in Philo and Josephus, both of 
whom distinguish it from ενσφεία, which = religion in its 
deeper sense, or piety. 

Philo Quod det. potiori insid, c. 7 (i. 195), in substance: *Nor 


if anyone uses lustrations or purifications and makes his body 
clean, but soils the purity of his mind — nor again, if out of his 
abundance he builds a temple or offers ceaseless hecatombs of 
sacrifices, is he to be reckoned among pious men (βυσεβάι/) : nay 
rather he has altogether wandered from the path that leads to piety, 
with heart set on external observances instead of on holiness 
(θρησκεία»' άντι όσιότητος ηγούμενος), offering gifts tO Him who cannot 
be bribed, and flattering Him who cannot be flattered/ 

Josephusyi;//. 9. 13. 3 (Solomon restored the decaying practice 
of giving tithes and firstfruits to the priests and levites) tva άά τη 

θρησκ€ΐα παραμενωσι καΧ ttjs θ^ραπύας ωσιν αχώριστοι του θίον, ' that 

they may always remain in attendance on public worship, and 
might not be separated from the service of God.' 

Ιδ. 12. 5• 4 ηνάγκασε δ' αί/τούς άφΐΐμίνους της nepl τον αυτών Qeov 
θρησκεία? τούί νπ αυτοΰ νομιζομ4νους σ^βεσθαι, * (Antiochus Epiphanes) 
compelled them to abandon their worship of their own God, and 
to pay honour to the gods in whom he believed/ 

Ιδ. 5• ΙΟ• I γυναϊκας τας em θρησκεία παραγινομΐνας, of the WOmen 
who went to worship and offer sacrifices at the Tabernacle. 

Ιδ. 4. 4. 4 (of those who sacrifice at home) ευωχίας evcKa τής αυτών 

«λλά μη θρησκεία?, ' for the Sake of their own private enjoyment 
rather than of public worship.' 

Ιδ. 12. 6. 2 (When a Jew offered sacrifice on an idol altar, 
Mattathias rushed upon him and slew him, and having overthrown 

the altar cried out) ei τις ζηΧωτης eVrt τών πατρίων ίθών και της του 

θΐου θρησκείας ίπίσθω ψοί^ ' whoever is zealous for his fathers' 
customs and for the worship of God, let him follow me.' 

4. Use in sub-Apostolic writers: — 

Clem. R. i. 45. 7 """^^ βρησκ^υόντων την μεγαΧοπρεπη κα\ tvho^ov 
βρησκ€ίαν του υψίστου, * those who practised the magnificent and 
glorious worship of the Most High.' 

Ιδ. 02. I π€ρ\ pev τών ανηκόντων τη θρησκεία ημών, τών ώφβλιμωτάτων 
€ΐς evapcTov βίον τοϊς θίΧουσιν ευσεβώς κα\ δικαίως Βΐ€υθύν€ΐν, ' of the 

things which pertain to our religion, things that are most useful to 
those who wish to guide their life piously and righteously into the 
way of virtue (we have given you sufficient injunctions, brethren).' 

6. TJse in the N. T. 

This contemporary use of θρησκάα for religion in its 

θρησκεία, μναττηριορ. 57 

external aspect as worship, or as one mode of worship 
contrasted with another, must be held to be its meaning 
in the N. T. It occurs in the following passages : 

Ads 26. 5 (in St. Paul's address to Agrippa) κατά τψ άκρφ^στάτην 
mpeaiv rfjs ή)Χ€τ^ρας θρησκεία? (ζησα Φαρισαιος, ' after the StraiteSt 
sect of our religion I lived a Pharisee/ 

Col. 2. 18 €v τα7Γ€ΐνοφροσύνϊ] και θρησκεία των άγγίλων, * by humility 
and worshipping of the angels/ 

James I. 26, 27 .... θρησκεία καθαρά κα\ αμίαντος, 'worship pure 
and undefiled in the sight of our God and Father is to visit orphans 
and widows in their affliction, to keep oneself unspotted from the 

μύστη piov. 
1. TJse in the LXX. and Hexapla. 

The only canonical book of the O. T. in which μνστηριον 
is used by the LXX. is Daniel, where it occurs several times 
in c. 2 as the translation of t"J * a secret,^ which is used of the 
king's dream, i. e. of the king's ' secret ' which had gone 
from him and which was revealed to Daniel. 

The other Greek translators of the O. T. use it in the 
following passages ; — 

Job 15. 8 Theodotion μυστήριοκ, = LXX. σύνταγμα, Aquila απόρ- 
ρητα, Symm. ομιλία, Heb. ΊΊ03Π. 

Ps. 24 (25). 14 Theodotion and the Inferpres Quintus μυστήριο»», 
= LXX. and the Interpres Sextus κραταίωμα, Aquila απόρρητον, 
Symm. ομιλία, Heb. ΠΊο. 

Frov. 20. 19 Theodotion uses it to translate ^So in a passage 
which the LXX. omit. 

Is. 24. 16 Theodotion and Symmachus use it as a translation 
of "Ή in a passage which the LXX. omit (but which has found its 
way into some cursive MSS. from Theodotion). 

It is frequently used in the Apocryphal books. In 
Sirach 22>, 22; 27. 16, 17, 21 of the secrets of private life, 
especially between friends : in Wisd. 14. 15, 23, in con- 


nexion with reXerat, of heathen sacrifices and ceremonies : 
but in a majority of passages of secrets of state, or the 
plans which a king kept in his own mind. This was a 
strictly Oriental conception. A king's ' counsel ' was his 
' secret/ which was known only to himself and his trusted 
friends. It was natural to extend the conception to the 
secret plans of God. 

Tob. 12. 7, II μυστήριομ βασιλέως, 'It IS good to keep close 
the secret of a king, but it is honourable to reveal the works of 

Judith 2. 2 Nabuchodonosor called all his officers unto him and 
communicated to them to μυστήριο»' τψ βονλης, ' his secret plan.' 

2 Mace. 13. 21 of one who disclosed τά μυστήρια, 'the secret 
plans ' of the Jews to their enemies. 

Wad. 2. 22 of the wicked who knew not μυστήρια Geoi, 'the 
secret counsels of God/ and especially that He created man to be 

Ιδ. 6. 24 of the 'secrets' of wisdom. 

2. Use in the N. T. 

This meaning of μυστήριου in the Apocryphal books 
throws considerable light upon its meaning in the N. T. 

Matt. 13. II ( = Mark 4. 11, Luke 8. 10) υμίν δ^δοται yvSovai τά 
μυστήρια της βασιλείας των ουρανών : the word implies not merely 
' secrets/ but rather the secret purposes or counsels which God 
intended to carry into effect in His kingdom. The contrast with 
iv παραβολαΐς which immediately follows is interesting when viewed 
in the light of the further meaning of μυστηριον, which will be 
mentioned below. 

J^om. II. 25 TO μυστήριο»» τούτο . . . . οτι ττώρωσις άπο μέρους τω 
*ΐσραη\ yeyovev, the Secret purpose or counsel of God, by which 
* a hardening in part hath befallen Israel until the fulness of the 
Gentiles be come in.' 

Rom. 16. 25 κατά άπυκάλυψιν μυστηρίου χρόνοις αϊωνίοις σίσιγη- 
μίνου φαν(ρωθ€ντος de νυν, of the secret purpose or counsel 'which 
hath been kept in silence through times eternal but now is mani- 
fested' — that the Gentiles were to be fellow-heirs with the seed of 

μύστη pLOV. 59 

Abraham : and in the same sense i Cor. 2. i (unless μαρτύρων be 
there read with Codd. Β D etc.). 

I Cor. 15. 51 Ιδον μυστήριοι νμίν λβγω, Ί tell you a secret 
counsel of God' for the time that is coming. 

Ephes. I. 9 TO μυστηριον τον θ€ληματο5, ' the Secret counsel of His 
will' : 3. 3, 4 iv τω )χυστηρίω τον XpLorov : 3. 9 τις ή οΙκονομία του 
μυστηρίου : 6. 19 το μυστήριομ τον dayye^iov; all in reference to the 
' secret counsel ' of God in regard to the admission of the Gentiles. 
So also Col. I. 26, 27 : 2. 2 : 4. 3. 

1 Tim. 3. 9 TO μνστηριον της πίστ€ως, probably the secret counsel 
of God which is expressed in the Christian creed : hence ζδ. 3- 
1 6 TO TTjs ζνσφΐίας μνστηριον is expressed in detail in the earliest 
and shortest form of creed which has come down to us. 

Rev. 10. 7 (In the days of the voice of the seventh angel, when 
he is about to sound) και €Τ€\€σθη τ6 μυστήριοι τον Oeov ως ίνηγγΐ- 
λισ€ T0VS iavTov dovXovs τονς προφήτης, ' then is finished the Secret 
counsel which God purposed to fulfil according to the good tidings 
which He declared to His servants the prophets.' 

2 Thess. 2. 7 TO γάρ μυστήριοι ^'δ?; ivepyelTat της ανομίας. In this 
passage the meaning which has hitherto seemed appropriate is less 
obvious in its application : but nevertheless it seems to me to be 
more probable than any other. The passage and its context seem 
to be best paraphrased thus : ' The secret purpose or counsel of 
lawlessness is already working: lawlessness is already in process of 
effecting that which it proposed to effect. But it is not yet fully 
revealed : there is he who restraineth, but he who now restraineth 
will be put out of the way; and then shall that lawless one be fully 
revealed whom the Lord shall consume with the breath of His 
mouth . . . . ' 

3. Use in the Apologists. 

But there are two passages in the Apocalypse, and 
probably one in the Epistle to the Ephesians, for which 
this meaning of μνστηριον does not seem to afford a sufficient 
or appropriate explanation, and for which we have to 
depend on the light which is thrown backwards on the 
N. T. by Christian writers of the second century. 

The word is used several times by Justin Martyr, and in 
almost every case it is in connexion with σνμβολον, ηίποί, 


or Ίταραβολη : and it is used in a similar connexion in a 
fragment of Melito. 

Justin M. Apol. i. 27: in all the false religions the serpent is 
pictured as σύμβοΚον μβγα κα\ μυστήριοι/. 

Id. Jryph. c. 40, with reference to the paschal lamb, το μυστήριοκ 

ovv τον προβάτου .... τύπος ην τον Χρίστου. 

Id. Tryph. c. 44 (some of the commandments of the Law were 
given with a view to righteous conduct and godliness : others 
were given) η els μυστήριοι^ του Xpn /τοΰ η 8ia το σκΚηροκάρ8ιορ τον 
\αοΰ υμών. 

Id. Tryph. c. 68 (with reference to Ps. 132. 11 'of the fruit of 
thy body will I set upon thy throne/ and Is. 7. 14 'Behold a 
virgin shall conceive ...')... to άρημίνον προς Δαυίδ υπο GeoD Ιν 
μυστηριω δια Ήσαιου ως e/^ieXXe γίνβσθαί ^ξηγηθη' el μητι τούτο €πιστασθ€, 
2) φίΧοι^ ?φην^ οτι πόλΧους Χόγους, τους €πικ€καλυμμ€νως και iv παραβοΧάίς 
η μυστηριοις η iv συμβολοις βργων XeXey/xe'i/ov? οί .... προφηται €ζηγη- 
σαντο, ' that which God said to David symbolically was interpreted 
by Isaiah as to how it would actually come to pass : unless you do 
not know this, my friends, I said, that many things which had 
been said obscurely and in similitudes or figures or symbolical 
actions were interpreted by the prophets.' 

Id. Tryph. c. 78 (commenting on Is. 8. 4 'he shall take away the 
riches of Damascus and the spoil of Samaria '), Justin interprets it in 
reference to the Magi, who by worshipping Christ revolted from the 
power of the evil demon which had taken them captive) ψΐν μυστηριω 

€σημαιν€ν ό λόγος οίκύν iv Ααμασκω' άμαρτωλον δε και αδικον ουσαν iv παρα- 
βοΧι^την δύναμιν iκeίvηv καλώς Σαμάρειαν καλεί, ' which pOWer, aS the pas- 
sage indicated symbolically, lived at Samaria : and since that power 
was sinful and unrighteous he properly calls it by a figurative ex- 
l!)ression Samaria.' (The equivalence of iv μυστηριω and iv παραβολή 
is evident.) 

MtMio frag. ix. (ap. Otto Corpus Apolog. vol. ix. p. 417) (Isaac 
is said to be 6 τΰπος του Χρίστου, ' a type of the Messiah,' and one 
which caused astonishment to men), ην γαρ θεάσασθαι μυστήριοι 
καινον . . . ' for one might see a strange symbolical representation, 
a son led by a father to a mountain to be sacrificed.' 

It is evident that μνστηριον was closely related in meaning 
to the vi^ords v^^hich are interchanged with it, tvttos, σνμβολον, 

μνστηριον, 6 1 

τταραβολη : and if with this fact in our minds we turn again 
to the N. T. there will be some instances in which the 
appropriateness of this meaning will be clear. 

I^ev. I. 20 TO μυστήριοι' των ίπτα αστέρων, 'the Symbol of the 

seven stars,' which is immediately explained to refer to the ' angels' 
of the seven churches. 

lb. 17. 7 TO μνστηριον της γυναικός, 'the Symbolical representation 
of the woman,' is in a similar way explained to refer to ' the great 
city which reigneth over the kings of the earth/ 

It is probable that the same meaning is to be given in Ephes. 5. 

32 TO μυστήριοι τοντο μίγα Ιστίν' eyw δε λίγω eis Χριστον και ίΐς την 

€κκ\ησίαν, ' this symbol (sc. of the joining of husband and wife into 
one flesh) is a great one : I interpret it as referring to Christ and 
to the Church/ 

The connexion of this meaning with the previous one is 
not far to seek. A secret purpose or counsel was intimated 
enigmatically by a symbolical representation in words, or 
in pictures, or in action. Such symbolical representations 
played a much more important part in the world in early 
times than they play now : the expression of ideas by 
means of pictures only passed by gradual and slow transi- 
tions into the use of written signs, in which the original 
picture was lost: and every written word was once a 
μνστήρίον. It was by a natural process that the sign and 
the thing signified came to be identified, and that the word 
which was used for the one came also to be used for the 

The meaning of μνστηριον was expressed in early eccle- 
siastical Latin by sacramentum. It has hence resulted that 
the meaning which came to be attached to sacramentum^ 
and which has passed with the word into most Euro- 
pean tongues, is the meaning which is proper not to the 
word itself but to its Greek original, μνστηριον. (The 
instances of the early use of sacramentum in this sense are 
given in detail by Ronsch, Itala und Vulgata^ p. '>)'^'>,., and 


Das Neue Testament Tertullians, p. ^^^) And although 
it is true that Tertullian, as was natural to one who had 
been educated in the rhetorical schools and had there 
dabbled in etymologies, does connect the theological use 
of sacramentum with its Classical use to designate a 
military oath [Ad Mart. c. 19, 24), yet that reference to 
Classical use is probably as misleading as it is insufficient to 
cover the facts which have to be explained : and just as the 
theological use of persona must be explained simply with 
reference to υπόσταση, so the theological use of sacramentum 
must be explained simply with reference to μυστηρων. 


The word was used in later Greek in two special senses, 
each of which appears in the N. T. 

I. It was used of the dispensator or slave who Λvas 
employed to give the other slaves of a household their 
proper rations : it is found in this sense in Corp, Inscr, 
Gr. 1247, 1498. 

Hence in 6*. Luke 12. 42 6 ττιστό? oiKoj'tJfjios 6 φρόνιμος, hv 
καταστησζί 6 Kvpios Ιτά της Oepaireias αντου, του bLbovai kv καφω 
το σιτομίτριον, ' the faithful and wise steward whom his lord 
shall set over his household to give them their portion 
of food in due season.' 

1. It was used of the villicus or land-steward : it is found 
in this sense in an inscription at Mylasa (Le Bas et Wad- 
dington, vol. iii, No. 404), in which οικονόμοι and ταμίαι are 
mentioned together, the former being in all probability the 
administrators of the domain, the latter the treasurers. 

Hence, in 6". Luke 16. i, the οίκοι/όμος is in direct relations 
with the tenants of the lord's farms : and hence the point 
of his remark, σκάιττ^ιν ονκ Ισχύω, ' I have no strength to 
dig/ since a degraded bailiff might be reduced to the status 
of a farm-labourer. 

οικονόμος, ομοθυμαδόν, ST) 

Hence also in Rom. i6. 23 6 οικονόμος τψ πολβω? is probably the 
administrator of the city lands. 


1. Classical use. 

The uses of the word in Classical Greek seem to imply- 
that the connotation which is suggested by its etymology 
was never wholly absent ; it can always be translated 'with 
one accord.' 

2. Use in the LXX. 

In the LXX. {a) it is used to translate Hebrew words 
which mean simply ' together,' [b) it is interchanged with 
other Greek words or phrases which mean simply 'together,' 
{c) it occurs in contexts in which the strict etymological 
meaning is impossible. 

{a) Its Hebrew originals are either 1Π^, e.g. in Job 3. 18, or ΠΠ;;, 
e.g. in Job 2. 11. 

[b) The same Hebrew words are more commonly rendered 
by αμα e.g. in Gen. 13. 6 : 22. 6, eVt το αυτό e.g. in Deut. 
22. 10, Jos. 9. 2, κατά. TO αυτό e. g. in Ex. 26. 24, I Sam. 30. 
24 (by όμου only in a passage which is inserted from Theo- 
dotion, Job 34. 29): the other translators and revisers some- 
times substitute one of these phrases for it, and m'ce versa, e.g. 
Job 2. II : 3. 18 LXX. όμοθυμαδό»', Symm. ό/ζοί, Ps. 2. 2 LXX. 
«ri TO αυτό, Symm. όμοθυμαδόΐ', Ps. 33 (34). 4 LXX. eVi TO αυτό^ 
Aquil. ομοθυμαδόι/. 

[c) Num. 24. 24 αυτοί όμοθυμαδο»' άπολονιαα*, I Chron, ΙΟ. 6 και 
δλοί ο οίκος αυτοΰ όμοθυμαδοί' απίθανα. 

Job 38. 33 ^^ίΟ'τασαι δε τροπάς ουρανού ή τα νπ ουρανον όμοθυμαδο»' 

In these and similar passages any such meaning as ' with one 
accord ' is excluded by the nature of the case. 

3. Use in tlie N. T. 

In the N.T. the word occurs in Acts i. 14 [some Codd., 
not ^ A Β C, of 2. 1], 2. 46, 4. 24, 5. 12, 7. 57 i 8. 6, 12. 20, 
15. 25, 18. 12, 19. 29, Rom. 15. 6. In none of these 


passages is there any reason for assuming that the word 
has any other meaning than that which it has in the Greek 
versions of the O. T., viz. ' together.' 

παραβολή, τταροιμία, 
1. Classical use. 
{a) παραβολή : 

Aristotle, R/iet. 2. 20, p. 1393 δ, defines it as one of the 
subdivisions of ^τapάb€Lγμa, ' example,' and coordinates it 
with λόγου : as an instance of it he gives τα Σωκρατικά : as 
when Socrates showed that it is not right for rulers to 
be chosen by lot by using the illustration or analogous case 
that no one would choose by lot those who should run 
in a race or steer a ship. Quintilian, 5. 11. i, follows 
Aristotle in making τταραβολη a kind of 7τapάb€Lγμa, and says 
that its Latin name is similitudo : elsewhere, 5• n• ^^j ^^ 
says that Cicero called it conlatio : he gives an instance 
of it, the passage from the Pro Murena, about those who 
return into port from a dangerous voyage, telling those who 
are setting out of the dangers and how to avoid them. 

(^) τταροιμία : 

Aristotle, RL•/. 3. 11, p. 1413 a, defines τταροίμίαί as 
μ€ταφοραΙ άττ' eXhovs Ιέ ei6os ; and, ib. I. II, p. 1371 ^3 ^^ 
gives as instances the sayings ^Xif ηλικα repwet, aet koXolos 
τταρα κολοων: in a fragment preserved in Synes. Calvit. 
Encom. c. %%, p. 234 (Bekker's Aristotle, p. 1474 b)^ he says 
of them T:a\aia<s βισι φιλοσοφία? . . . €γκαταλ€[μματα7Τ€ρίσωθ€ντα 
δια σνντομίαν /cat ^^ζιότητα. Quintilian, 5• H• ^^5 says of 
τταροιμία that it is ' Velut fabella brevior, et per allegoriam 
accipitur : non nostrum, inquit, onus : bos clitellas.' 

2. Use in the LXX. and Hexapla. 

τταραβολή occurs about thirty times in the Canonical books 
as the translation of ^'ζ?^» ^^^ ^^ ^^ other word (in Eccles. 

τταραβοΧη, τταροιμία. 6 ζ 

Ι. 1 7, where all the MSS. have it as a translation of ΤΫΌ^Π 
* madness,' it is an obvious mistake of an early transcriber 
for τταραφοράς, which is found in Theodotion). 

The passages in which ^^^ is not rendered by τταραβολη 
are the following : — 

I Kings 9. 7, and Ezek. 14. 8; the Targum ίσται (θησ-ομαι) 
ds άφανισμόν, ' shall be for a desolation/ is substituted for the 
literal translation €σται {θησομαι) eis παραβολην, ' shall be for a 

J^od 13. 12 άποβησ€ται δε νμων τη γαυρίαμα ισα σττοδο), is SO far froin, 

the Hebrew as to aiford no evidence. 

Id. 27. I and 29. i: it is rendered by προοίμιον, which may 
be only a transcriber's error for παροιμία: in 27. i Aquila has 

Prov. I. i: the LXX. have παροιμίαι, Aquila παραβολαί. 

Is. 14. 4 LXX. Χηψει τον θρήκοί' τούτον €π\ τον βασίΚεα Βαβ. 
Aquil., Symm., Ύheoά. παραβόλψ: cf. Ezek. 19. ΐ4> where the LXX. 
combine the two words in the expression els παραβολην θρήνου, and 
Mic. 2. 4 where they are coordinated. 

It will be seen then in a majority of the cases in which 
τταραβολή was not used to translate ^ψ^, τταροψία was used 
instead of it : this is also the case with the following 
passages, in which the LXX. used τταραβολη but the 
Hexapla revisers substituted τταροιμία : — 

I Sam. 10. 12 LXX. παραβολην, "Αλλος' παροιμίαν. 
lb. 24. 14 LXX. παραβολή, Symm. παροιμία. 

Ps. 77 (78). 2 LXX. and Aquil. Iv παραβολαΐς^ Symm. 8ia παροι- 

Eccles. 12. 9 LXX. παραβόλων, Aquil. παροιμίας. 

Ezek. 12. 2 2 LXX. Aquil., Theod. παραβολή, Symm. παροιμία, 

lb. 18. 3 LXX. παραβολή, Aquil. παροιμία. 

Prov. 25. i: Codd. AS'^ of the LXX. have παροιμίαι, Codd. 
BS^ and most cursives παώεΐαι : Aquila, Symmachus, and Theo- 
dotion παραβολαί. 

lb. 26. 7, 9 : in the first of these verses most MSS. of the LXX. 


have παρανομίαν [παρανομίας)) 2L transcriber's βΓΓΟΓ for παροιμίαν 
{παρανομίας), which, is found in Codd. 68, 248, 253; Symmachus 
has παραβολή. In V. 9 the LXX. have, without variant, the impos- 
sible translation dovXeia (possibly the original translation was παώ€ία, 
as in I. i, and this being misunderstood, the gloss dovXeta was 
substituted for it) : there is a trace of the earlier reading in S. Am- 
brose's quotation of the passage in his Comment, in Ps. 35, p. 
^68 d, 'ita et injusti sermone nascuntur quae compungant loquen- 
tem': but in Epist. 37, p. 939, he seems to follow the current 

These facts that τταραβολή and τταροιμία are used by the 
LXX. to translate the same Hebrew word, and that the 
other translators and revisers frequently substitute the one 
for the other, show that between the two words there 
existed a close relationship, and that the sharp distinction 
which has been sometimes drawn between them does not 
hold in the Greek versions of the O. T. If we look at some 
of the sayings to which the word τΐαραβολη is applied, we 
shall better see the kind of meaning which was attached 
to it:— 

I Sam. 10. 12 of the 'proverb' 'Is Saul also among the pro- 

lb. 24. 14 of the * proverb of the ancients,* 'Wickedness pro- 
ceedeth from the wicked.' 

Ezek. 12. 22 of the ' proverb that ye have in the land of Israel, 
saying, The days are prolonged, and every vision faileth.' 

£zek. 16. 44 of the 'proverb' 'As is the mother, so is her 

lb. 18. 2 of the ' proverb' ' The fathers have eaten sour grapes, 
and the children's teeth are set on edge.' 

Deut. 38. 37, 2 Chron. 7. 20, Ps. 43 (44). 15 : 68 (69). 12, 
Jer, 24. 9, Wnd. 5. 3, of men or a nation being made a byword 
and a reproach. 

Intertwined with and growing out of this dominant sense 
of τταραβολή and τταροιμια as a ' common saying' or 'proverb,' 
is their use of sayings which were expressed more or less 

'παραβολή J τταροιμία, 6 J 

symbolically and which required explanation. The clearest 
instance of this in the canonical books is probably Ezek. 
20. 47-49) where after the prophet has been told to speak 
of the kindling of a fire in the ' forest of the south field,' he 
replies μηbaμώs, κνρί€ Kvpie' αυτοί λίγουσι irpos μέ Ονχί τταρα- 
βολη €στί λεγομένη αϋτη ; hence τταραβολη and παροιμία are 
sometimes associated with αϊνυγμα : e.g. Sir. 39. 2, 3 (quoted 
below) iv αΐνίγμασι τταραβολων, and in Num. 21. 27 the 
LXX. have 01 αινιγματισταί, where a reviser ("AXkos) in the 
Hexapla has oi τΐαροιμιαζόμ^νοι as a translation of Ο^'ρφΏΓΤ. 
It appears even more distinctly in Sirach. 

Sir. 13. 26 cvpeais παράβολων 8ιάΚογισμο\ μ€τα κόπου, Ε. V. 'the 

finding out of parables is a wearisome labour of the mind.' 

Sir. 39. 2, 3 (of the man * that giveth his mind to the law of the 

Most High ') ev στροφαϊς παραβοΧων avueiaeXevaerat' απόκρυφα παροιμιών 
€κζητησ€ΐ, κα\ iv αΐνίγμασι παραβόλων αραστραφησεται, Ε. V, ' where 

subtil parables are he will be there also, he will sell out the secrets 
of grave sentences, and be conversant in dark parables.' 

Sir. 47. 17 (of Solomon) eV ωδαΓ? κα\ παροιμίαις καΐ παραβολαίς κα\ 

ev €ρμψ€ίαις άπίβανμασάν σ€ χώραι, Ε. V. ' the countries marvelled 
at thee for thy songs and proverbs and parables and interpreta- 

The reference in this last passage to i Kings 4. 29 (33) may be 
supplemented by the similar reference to it in Josephus An/. 8. 2, 
5 : and it is interesting to note that the words of the LXX. 

(λάλησαν υπβρ των ξύλων άπο της κέδρου . . . are paraphrased by 
Josephus καθ^ €καστον yap eldos δ^ν^ρου τταραβολη I' eXirev άπο υσσώπου 
€ως κέδρου. 

Α review of the whole evidence which the LXX. oft"ers 
as to the meaning of τταραβολη and παροιμία seems to show 

(i) that they were convertible terms, or at least that 
their meanings were so closely allied that one could be 
substituted for the other ; 

{2) that they both referred (a) to ' common sayings ' or 
* proverbs,' and (δ) to sayings which had a meaning below 
the surface, and which required explanation. 

F 2 


3. Use in sub-apostolic writers. 

These inferences are supported by the use of the word in 
sub-apostolic writers and in Justin Martyr : — 

Barnabas 6. lo (quotes the words * into a good land, a land 
flowing with milk and honey,' and then proceeds) (νλογητος 6 κύριος 
ημών, άδ€λφοί, 6 σοφίαν και νουν βίμίνος iv ήμίν των κρύφιων αυτού' Xe'-yet 
γαρ 6 προφήτης παραβολή μ κυρίου' τις νόησα el μη σοφός κα\ επιστήμων 

καΐ αγαπών τον κύριον αυτού, ' Blesscd be our Lord, brethren, who 
hath put into us wisdom and understanding of His secrets: for 
what the prophet says is a parable of the Lord,' i. e. evidently, a 
saying which has a hidden meaning and requires explanation : * who 
will understand it but he who is wise and knowing, and who loves 
his Lord/ 

Id. 17. 2 (' If I tell you about things present or things to come, 
ye will not understand) Sm τ6 iv παραβολαϊς κύσθαι, ' because they 
lie hid in symbols/ 

The Shepherd of Hermas consists to a great extent of παραβο'λαί, 
Vet. Lat. * similitudines ' ; they are symbols or figures of earthly 
things, which are conceived as having an inner or mystical mean- 
ing : e. g. in the second * similitude ' the writer pictures himself as 
walking in the country, and seeing an elm-tree round which a vine 

is twined. The Shepherd tells him αΰτη ή παραβολή eh τους δούλους 

του Oeod κείται, ' this figure is applied to the servants of God ' : and 
he proceeds to explain that the elm-tree is like a man who is rich 
but unfruitful, the vine like one who is fruitful but poor, and that 
each helps the other. 

Justin M. Tryph. c. 36 says that he will show, in opposition to 
the contention of the Jews, that Christ is called by the Holy Spirit 
both God and Lord of Hosts, Iv παραβολί}, i. e. in a figurative 
expression: he then quotes Psalm 24, the Messianic application of 
which was admitted. 

Id. Tryph, c. 52 (It was predicted through Jacob that there 
would be two Advents of Christ, and that believers in Christ would 
wait for Him) : kv τταραβολι^ δε και παρακ€κα\υμμ€νως το πνεύμα το 

αγιον δια τοΰτο αυτά iXeXaXrjKei, ' But the Holy Spirit had said this in 
a figure and concealedly, for the reason which I mentioned,' viz. 
because, if it had been said openly, the Jews would have erased 
the passage from their sacred books. 

τταραβοΧη, τταροιμία, 69 

Id. Tryph, c. 63 : the words of the same last speech of Jacob, 
* he shall wash his clothes in the blood of grapes,' were said Iv 
παραβολτ], ' figuratively,' signifying that Christ's blood was not of 
human generation. 

Id. Tryph. c. 113, 114, Christ is spoken of eV παραβόλαΐς by the 
prophets as a stone or a rock. 

So Tryph. c. 68, 90, 97, 115, 123. 

4. Use in the N. T. 

In the N. T. παραβολή is used only in the Synoptic 
Gospels and in Heb. 9. 9, 11. 19 : παροιμία is used only in 
the Fourth Gospel and in % Pet. %. ι,'Χ. If we apply to 
these passages the general conclusions which are derived 
from the LXX. and confirmed by the usage of sub-apostolic 
writers, their appropriateness will be evident : nor is it 
necessary in any instance to go outside the current con- 
temporary use to either the etymological sense or the usage 
of the rhetorical schools. The majority of passages in 
which τταραβολη is used belong to the common foundation 
of the Synoptic Gospels, and refer to the great symbolical 
illustrations by which Christ declared the nature of the 
kingdom of heaven. They are Matt. 13. 3=:Mk. 4. 2>, Luke 
8. 4; Matt. 13. io = Mk. 4. 10, Luke 8. 9 ; Matt. 13. 13 = 
Mk. 4. II, Luke 8. 10 ; Matt. 13. 18 = Mk. 4. 13, Luke 8. 
II ; Matt. 13. 2,4, Matt. 13. 31 = Mk. 4. 30 ; Matt. 13. ^^, 
Matt. 13. 34, 35 = Mk. 4. 33^ 34; Matt. 13. 36, 53, Matt. 
iZi. 33 = Mk. iri. I, Luke 30. 9 ; Matt. 21. 45 = Mk. in, 12, 
Luke 20. 19; Matt. 22. i, Matt. 24. 32^ — Mk. 13. 28, 
Luke (Zi. 2^, Luke 19. 11. It is also used of the similar 
illustrations which are peculiar to S. Luke, and which do 
not all illustrate the nature of the kingdom of heaven in its 
larger sense, Luke i:^. 16, 41; 13. 6; 14. 7; 15. 3; 18. i, 9. 
In all these instances the requirements of the context are 
fully satisfied by taking it to mean a story with a hidden 
meaning, without pressing in every detail the idea of a 
^ comparison.' 


In S. Luke 4. 23 it is used in a sense of which the LXX. 
affords many instances : ττάντως epelre μοι την τταραβολην 
ταύτην' Ιατρ€, Θζράτϊζνσον σ^αντόν, ' doubtless ye will say to 
me UasJ)roverb' [so e.g. i Sam. 10. 12; 24. 14], 'Physician, 
heal thyself.' 

In S. Luke 6. 39 it is used of the illustration of the blind 
leading the blind : and in S. Mark 3. 23 of that of Satan 
casting out Satan^ neither of which had so far passed into 
popular language as to be what is commonly called a 
' proverb,' but which partook of the nature of proverbs, 
inasmuch as they were symbolical expressions which were 
capable of application to many instances. 

The other passages in which τταραβολή occurs in the N. T. 
are — (l) Heb. 9. 9 tJtls τταραβολή et? τον καιρόν τον €ν€(Γτηκότα, 
* which' [i. e. the first tabernacle] 'is a symbol for the present 
time'; (2) Heb. 11. 19 oOev [sc. €κ ν€κρων'\ αυτόν καΐ h τταρα- 
βολτ} €κομίσατο, 'from whence he did also in a figure receive 
him back.' In both passages the meaning of τταραβολή, 
' a symbol,' is one of which many instances, some of which 
have been given above, are found in Justin Martyr. 

2 I^ef. 2. 22 TO της άληθονς παροιμίας' κύων €Τΐΐστρ€^ας eVt το 'ίδιον 
ίξβραμα . . . . ' the (words) of the true proverb, The dog turning to 
his own vomit/ .... Here παροιμίας is an application of the 
title of the book Παροιμίαι, from which (26. 11) the quotation is 

S. John 10. 6 ταίιτην την παροιμίαν ΐΐπ^ν αυτοίς 6 ^Ιησονς' cKelvoi δε 

ονκ έγνωσαν τίνα ην ά ίΚαλει αντοίς, ' this parable Said Jesus to them ; 
but they did not understand what it was that He spake to them ' : 
the reference is to the illustration of the sheep and the shepherd, 
for which the other Evangelists would doubtless have used the 
word παραβολή I with the substitution of παροιμία for it in S. John 
may be compared the similar substitution of it as a translation of 
7ψΌ by the Hexapla revisers of the LXX., which has been men- 
tioned above. 

S, John 1 6. 25, 29 ονκίτι ev παροιμίαις λαλήσω, παροιμίαν ούδεμίαν 
Xeyetf are contrasted with παρρησία [Codd. Β D eV παρρησία^ άπαγ- 

7Γ6ίράζ€ΐν, ττειρασμός. yi 

γίλω, iv παρρησία \aK(is '. the contrast makes the meaning clear : iv 

τταροιμίαις λαλΰν is equivalent to the ev τταραβόλί) κα\ παρακΐκαλνμμίνως 

of Justin Martyr (quoted above), the substitution of παροιμίαις for 
τταραβολαΊί having its exact parallel in Ps. 77 (78). 2, where Sym- 
machus substitutes Βιά παροιμίας for the ev παραβολαΐί of the LXX. 

(and of S. Matt. 13. 35). 

ΤΓβφάζβίΐ/, π€φασμ09. 

1. Use in the LXX. 

The words are used sometimes of the trying or proving 
of God by men, e. g. Ex. 17. 2>, 7, Num. 14. Z2, : but more 
commonly of the trying or proving of men by God. The 
purpose of this trying or proving is sometimes expressly 
stated : e.g. Ex. 16. 4 ττβιράσω avrovs el τιορ^νσονται τω νόμ(ύ 
μου η ου; Judges Ζ. %2 του τΐ^ιράσαι τον Ίσραηλ el φυλάσσονται 
την obov Κυρίου. The mode in which God tried or proved 
men was almost always that of sending them some affliction 
or disaster : and consequently ' trial ' (as not unfrequently 
in English) came to connote affliction or disaster : hence 
τΓ€φασμόί is used, e.g. with reference to the plagues of 
Egypt, Deut. 7. 19 τουί ΤΓ€ίρασμου9 του9 μ€γάλου9 ots ϊhoσav 
οί οφθαλμοί σου, τα σημύα κα\ τα τίρατα τα μεγάλα €Κ€Ϊνα^ την 
χ€Ϊρα την κραταιαν καΐ τον βραχίονα τον υψηλόν, 'the great 
trials which thine eyes saw, the signs and those great 
wonders, the mighty hand and the uplifted arm ' : so also 
29. 3. In the Apocryphal books this new connotation 
supersedes the original connotation, and is linked with the 
cognate idea of ' chastisement."* 

Wi'sd. 3. 5 κα\ ολίγα παώ^υθ^ντΐς μ^γαΚα €ν€ργ€τηθησονται' οτί 6 0(6: 
€πΐίρασ€ν αντονς κα\ evpev avTovs άξιους eavrov, *And having been a 

little chastised, they shall be greatly benefited: for God proved 
them and found them worthy of Himself.' 

Ιδ. II. 10 (the Israelites are contrasted with the Egyptians) ore 
γαρ €π€φάσθησαν καίπΐρ iv iKefi παώ^υόμίνοι. βγνωσαν πως iv οργή κρινό- 
μενοι άσφζίς iβaσavίζovτo, Ε. V. * For when they were tried, albeit 


but in mercy chastised, they knew how the ungodly were judged in 
wrath and tormented . . .' 

Sir. 2. I TCKvov €1 Trpoaepxij dovXevetv κνρίω de^ ετοίμασαν την ^Ι^νχ^ην 
σον els π^φασμόν, ' My SOU, if thou come near to serve the Lord 
God, prepare thy soul for trial.' 

Judith 8. 24—27 €υχαριστησωμ€ν κνρίω τω θίω ημών os π€φάζ€ΐ ημάς 

καθα κα\ τους πατέρας ημών, ' let US give thanks to the Lord our God, 
who trieth us as He did also our fathers ' (sc. by sending an army 

to afflict us) oTi ου καθώς εκείνους επύρωσεν εΙς ετασμον της 

καρ8ίας αυτών και ημάς ουκ εξεδίκησεν αλλ* εΙς νουθετησιν μαστιγοΐ κύριος 

τους εγγίζοντας αυτω, ' for He hath not tried US in the fire as He did 
them for the examination of their hearts, neither hath He taken 
vengeance on us : but the Lord doth scourge them that come near 
unto Him to admonish them/ 

2. Use in the N. T. 

There are some passages of the N. T. in which the 
meaning which the words have in the later books of the 
LXX. seems to be established : — 

6". Zuke 8. 13 εν καιρώ πειρασμού has for its equivalent in S. Matt. 

13. 21, S. Mark 4. 17 γενομένης θλίψεως η διωγμού, SO that 'in time 

of trial' may properly be taken to mean * in time of tribulation' or 
' persecution.' 

Acts 20. 19 πειρασμών τών συμβάντων /xoi εν ταΐς επιβουΧαΐς τών 
^Ιουδαίων. S. Paul is evidently speaking of the * perils by mine 
own countrymen' of 2 Cor. 11. 26, the hardships that befel him 
through the plots of the Jews against him. 

Hed. 2. 18 εν ω γαρ πεπονθεν αυτός π€ΐρασθεί$; δύναται τοΙς πειρα- 
ζομ^ΐΌΐς βοηθησαι, * for in that He Himself suffered, having been 
tried, He is able to succour them that are being tried.' 

I Pet. I. 6 ολίγον άρτι ει8εον Χυπηθεντες εν ποικίΧοις πειρασμόΐς, 

' though now for a little while, if need be, ye have been put to grief 
by manifold trials,' with evident reference to the persecutions to 
which those to whom the epistle was addressed were subjected 
(so 4. 12). 

JleV. 3• 10 κάγώ σε τηρήσω εκ της ώρας του πειρασμού της μελλούσης 
ερχεσθαι επ\ της οικουμένης όλης, πειράσαι τους κατοικοϋντας επΙ της γης, 

' Ι also will keep thee from the hour of trial, the hour that is about 

ΊΓενης, ττραϋς, ιττωγος^ ταττβινος, 73 

to come upon the whole world to try them that dwell upon the 
earth/ with evident reference to the tribulations which are pro- 
phesied later on in the book. 

This meaning, the existence of which is thus established 
by evident instances, will be found to be more appropriate 
than any other in instances where the meaning does not lie 
upon the surface : — 

S. Matt. 6. 13 = S. Luke 11. 4 /m?) claeveyKjjs ημάς els ΤΓβιρασμόΐ', 
' bring us not into trial/ i. e. into tribulation or persecution ; but, on 
the contrary, ' deliver us from him who — or that 'which — does us 
mischief (see below, p. 79): cf. 2 Pet. 2. 9 olbev KvpLos (νσβββίς 

eK πειρασμού ρνβσθαί ά8ίκονς de els ημίραν κρίσεως κο\αζομ4νονς τηρύν, 

' the Lord knoweth how to deliver the godly out of trial, but to 
keep the unrighteous under punishment unto the day of judg- 

S. Matt. 4. I =S. Mark i. 13, S. Luke 4. 2 Tr€ipaa0T]mi v-rro τον 
διαβόλου, ' to be tried/ i. e. afflicted ' by the devil/ with reference to 
the physical as well as the spiritual distresses of our Lord in the 
desert: cf. Hed. 4 15 'Π•€π€ΐρασμενομ δε κατά πάντα καθ' ομοιότητα 
χωρίς αμαρτίας, ' tried/ i. e. afflicted ' in all points like as we are, 
yet without sin ' : this interpretation is strongly confirmed by 
Irenaeus 3. 19. 3, who says of our Lord ώσπερ ην avθpωπos ha 
πειρασθη οντω5 και Aoyos ίνα δοξασθη, ' as He was man that He might 
be afflicted, so also was He Logos that He might be glorified.' 

π€ΐ/ης, TTpavs, πτωχός, ταπεινοί, 

1. Classical use. 

In Classical Greek these words are clearly distinguished 
from each other, ττ^νης is 'poor' as opposed to rich, τττωχό^ 
is ' destitute ' and in want : cf. Aristoph. Plut. ^^2, : 

πτωχού μ^ν γαρ βίos, ov συ Xeyeis, ζην iaTiv μηΒεν €χοντα' 
του δε πίνητος ζην φίώόμΐνον και τοις epyois πρυσίχοντα, 
π^ριγίγνίσθαι δ' αυτω μηΒ^ν, μη μίντοι μηδ^ έπιΚζ'ιπίΐν. 

TTpavs {iTpaos) is 'easy-tempered' as distinguished from 


6ργίλο9, 'passionate' (Arist. Etk. N. 2. 7, p. 1108 a, 4. 11, 
p. ii35<2), and τηκρός, 'sour-tempered' {Rhet. ad Alex. 38): 
TttTretros is not only 'lowly' but almost always also 'dejected' 
(e.g. Arist. Pol. 4. 11, p. 1395^, of ot κα& υττζρβολην h 
kvMa τούτων^ sc. ισχυο? καΧ τιΚοντον καΧ φίλων, who conse- 
quently submit to be governed like slaves, αρχ^σθαι bovXiKrjv 
αρχήν) and ' mean-spirited ' (e. g. Arist. R/iel. %.η, p. 1384 λ, 
who says that to submit to receive services from another, and 
to do so frequently, and to disparage whatever he himself 
has done well, are μίκροψνχίαί καΐ ταπςινότητοζ σημ^ϊα). 

2. Use in the LXX. 

In the LXX., on the contrary, the words are so constantly 
interchanged as to exclude the possibility of any sharp dis- 
tinction between them : nor can any of them connote, as in 
Classical Greek, moral inferiority. 

(i) They are all four (but irpavs less than the other 
three) used interchangeably to translate the same Hebrew 
words : — 

''jy, 'afflicted,' is rendered by -rrivi]^ in Deut. 15. 11 : 24. 14 (16), 
15 (17). Ps. 9. 13, 19 : 71 (72). 12 : 73 (74). 19 : 108 (109). i6. 
Prov. 24. 77 (31. 9) : 29. 38 (31. 20). Eccles. 6. 8. Is. 10. 2 : by 
•πτωχός in Lev. 19. 10 : 23. 22. 2 Sam. 22. 28. Job 29. 12 : 34. 
28 : 36. 6. Ps. 9. 23 (10. 2) : 9. 30 (10. 9) : 11 (12). 6 : 13 (14). 
6 : 21 (22). 25 : 24 (25). 16 : 33 (34). 6 : 34 (35). 10 : 36 (37). 
15 : 39 (40). 18 : 67 (68). 11 : 68 (69). 30 : 69 (70). 6 : 71 (72). 
2, 4 : 73 (74). 21 : 85 (86). i : 87 (88). 16 : loi /if. : 108 (109). 
22 : 139 (140). 13. Amos 8. 4. Hab. 3. 14. Is. 3. 14, 15 : 41. 
17 : 58. 7. Ezek. 16. 49 : 18. 12 : 22. 29 : by Taireij'os in Ps. 17 
(18). 28 : 81 (82). 3. Amos 2. 7. Is. 14. 32 : 32. 7 : 49. 13 : 
54. II : 66. 2. Jer. 22. 16 : by πραΰς in Job 24. 4. Zach. 9. 9. 
Is. 26. 6. 

IJy, 'meek,' is rendered by π^ι^ης in Ps. 9. 38 (10. 17) : 21. 27 : 
by -πτωχός in Ps. 68 (69). 33. Prov. 14. 21. Is. 29. 19 : 61. i : 
by ταπ€ΐμ05 in Prov. 3. 34. Zeph. 2. 3. Is. 11. 4 : by πραυ§ in 
Num. 12. 3. Ps. 24 (25). 9 : 33• 3 : 36 (37)• u ' 75 (76). 10 : 
146 (147). 6 : 149. 4. 

7Γ€νης, τΓρανς, τττωχο?, ταττεινος, 75 

[^"•^ξ?, 'needy,' is rendered by π^μης in Ex. 23. 6. Ps. 11 (12). 
6 : 34 (35). 10 : 3^ (37)• i5 : 39 (40). 18 : 48 (49)• 2 : 68 (69). 
34 : 71 (72). 4, 13 : 73 (74). 21 : 85 (86). i : 106 (107). 41 : 108 
(109). 22, 31 : III (112). 9 : 112 (113). 7 : 139 (140). 13. Prov. 
24• 37 (30• 14). Amos 2. 6 : 4. I : 5. 12 : 8. 4, 6. Jer. 20. 13 : 
22. 16. Ezek. 16. 49 : 18. 12 : 22. 29 : by ιττωχός in Ex. 23. 11. 
I Sam. 2. 8. Esth. 9. 22. Ps. 9. 19 : 71 (72). 12 : 81 (82). 4 : 
108 (109). 16 : 131 (132). 15. Prov. 14. 31 : 29. 38 (31. 20). 
Is. 14. 30 : by ταπεινός in Is. 32. 7. 

^% *weak,' is rendered by ir^nrjs in Ex. 23. 3. i Sam. 2. 8. 
Ps. 81 (82). 4. Prov. 14. 33 : 22. 16, 22 : 28. 11 : by ιττωχός in 
Lev. 19. 15. Ruth 3. 10. 2 Kings 24. 14. Job 34. 28. Ps. 71 
(72). 13 : 112 (113). 6. Prov. 19. 4, 17 : 22. 9, 22 : 28. 3, 8 : 
29. 14. Amos 2. 7 : 4. I : 5. II : 8. 6. Is. 10. 2 : 14. 30. Jer. 
5. 4 : by Ta,iT€iv6s in Zeph. 3. 12. Is. 11. 4 : 25. 4 : 26. 6. 

^^"^, 'poor/ is rendered by iriv^q in 2 Sam. 12. i, 3, 4. Ps. 81 
(82). 3. Eccles. 4. 14 : 5. 7 : by πτωχός in Prov. 13. 8 : 14. 20 : 

17. 5 : 19. I, 7, 22 : 22. 2, 7 : 28. 6, 27 : by ταττειι/ός in i Sam. 

18. 23. 

(2) They are used interchangeably by different translators 
to translate the same Hebrew word : e. g. 

Ps. II (12). 5 Q*'*^^. is translated by the LXX. and Symmachus 
πτωχών; by Aquila -πενήτων: conversely, i3''J'i''nx is translated by 
Aquila -πενήτων, and by the LXX. and Symmachus πτωχώΐ'. 

Ps. 17 (18). 28 ^^V is translated by the LXX. τα-πεινόν, by Aquila 
πένητα, and by Symmachus πραον. 

Is. II. 4 ^)^V is translated by the LXX. and Theodotion ταπβι- 
KOiisj by Aquila πρα^σι, by Symmachus πτωχούς. 

Is. 66. 2 ^^V is translated by the LXX. ταπεινόν, by Aquila 
πραυν, by Symmachus πτωχόν, by Theodotion συντετριμμένοι. 

(3) In a large proportion of cases the context shows that, 
though the words vary in both Hebrew and Greek, the 
same class of persons is referred to : the reference 
ordinarily being either (a) to those who are oppressed, 
in contrast to the rich and powerful who oppress them ; 
or (δ) to those who are quiet, in contrast to lawless wrong- 
doers : e. g. 


{a) Ps, 9. 31 (10. 9): 

* He lieth in wait secretly as a lion in his den : 
He lieth in wait to catch the poor (πτωχόζ/); 
He doth catch the poor, dragging him with his net. 
And being crushed, he sinketh down and falleth ; 
Yea, through his mighty ones the helpless fall' 

(LXX. iv τω αυτόν κατακνρί€νσαι των πενήτων, 
Symm. ΐπιπίσόντος αυτόν μ€τα των Ισχυρών αύτον tols άσθβν^σινλ 
Ps. 34 (35)• 10 : 

' All my bones shall say, Lord, who is like unto thee, 
Which deliverest the poor (πτωχόν) from him that is too 

strong for him, 
Yea, the poor and the needy {πτωχον κα\ πένητα) from him 
that spoileth him/ 
So also, and with especial reference to God as the deliverer of the 
oppressed, Ps. 11 (12). 6 : 33 (34). 6 : 36 (37). 14 : 39 (40). 18 : 
71 (72). 4, 13 : 75(76). 10. 
(δ) Ps. 36 (37). 10, 11: 

'Yet a little while and the wicked shall not be, 
Yea, thou shalt diligently consider his place, and it shall 

not be: 
But the meek {oi npaeh) shall inherit the earth ; 
And shall delight themselves in the abundance of peace.' 
Ps. 146 (147). 6: 

' The Lord lifteth up the meek (πραβΐς) : 
He casteth the wicked down to the ground.' 

The inference to which these comparisons lead is that 
the Ίττωχοί^ 7τ4νητ€5^ Trpaet?, τατϊζίνοί are all names for one 
and the same class, the poor of an oppressed country, the 
peasantry or fellahin who, then as now, for the most part 
lived quiet and religious lives, but who were the victims of 
constant ill-treatment and plunder at the hands not only 
of tyrannical rulers, but also of powerful and lawless 

3. Use in the N. T. 

It is probable that this special meaning underlies the use 
of the words in the Sermon on the Mount. This is in- 

■* - - / 

ΤΓονηρος, ΤΓΟνηρια, 7? 

dicated partly by the coordination of subjects, which in the 
LXX. are used interchangeably, ol τττωχοί, ol 'npaeis, and 
which are in harmony with the following subjects — ot 
TTevdovvTeSi ol 7T€lvo^vt€S Koi διψώζ^Γβ?, ot ^^^ιωγμίνοί ; and 
partly by the fact that at least one of the predicates comes 
from a psalm in which the contrast between ol ττονηρβνόμξνοί, 
ol αμαρτωλοί, and ot biKaLOL, ol Trpaets is strongly marked, viz. 
Ps. 36 (37)• II ot δ€ Ttpaeis κληρονομησονσι γην. The addition 
in S. Matthew of the modifying phrases ot πτωχοί τω πι/€ύματι, 
ot 7reLV(uVT€S καΐ δίψωντ€ί την δικαιοσύνη κ, ot b€bLωγμ€VOL eveKev 
δικαιοσυμη$, shows that the reference was not simply to the 
Syrian peasantry, as such ; but the fact that those modifying 
phrases are omitted by S. Luke helps to confirm the view 
that the words themselves have the connotation which they 
have in the LXX. 

ΤΓονηρός^ πονηρία. 

1. Classical use. 

The connotation of ττονηρόί in Classical Greek is pro- 
bably best shown by Arist. Etk. N. 7. 11, p. 1152 ^, where 
Aristotle, speaking of the άκρατης, says that what he does 
is wrong, and that he acts as a free agent, but that he is 
not wicked in himself, ^κων μ4ν .... ττονηρος δ' ού' η γαρ 
7Γpoαtpeσts iincLKrjs' ωσθ^ ημίττόνηροί' καΐ ουκ. abiKOS' ον γαρ 
€ΐΓίβουλος, ' He (i. e. the weak man), though he is a free 
agent .... yet is not wicked : for his will is good : he 
may consequently be called "half-wicked." And he is 
not unrighteous : for what he does is not done afore- 

2. Use in the LXX. 

Πονηρός, πονηρία are used frequently, and in various 
relations, to translate 1Π, TlV"^) 


Of wild or ravenous beasts, 

Gen. 37. 20 Koi €ρονμ€ν, θηρίον TTonijpoK κατβφαγίρ αυτόν. So ζδ. 

V. 33 ; Lev. 26. 6. 
Ezek. 14. 15 faf fctt βηρία πονηρά €πάγω eVi την γην καϊ τιμωρή- 
σομαι αντην. So ζδ. V. 21 : 5*^7• 34• 25• 

Of the plagues of Egypt, 

Deut, 7• 15 ττάσα? i/oVovs Αϊγύτττον τάξ wonfjpas as θώρακας. So 
28. 60. 
Of Divine plagues in general, and their ministers, 

J^OS. 23. 15 enamel κύριος 6 Oebs εφ' νμας πάντα τα ρήματα τά. ΤΓομηρά, 

€ως αν €ξοΚοθρ€υστ] υμάς άπο της γης .... 
JPs. 77 (7^)• 49 ^ξαπ€στ€ΐ\€ν fls αυτούς opyrjv θυμού αυτοΰ .... 
άποστοΧην bi άγγίλων ΊΤονΐί\ρων {JSytnm. κακουντων). 
Of unwholesome water or food, 

2 Kings 2. 19 ra υ^ατα ΐΓ<$μηρα (the water which Elisha healed). 

Jer. 24. 2 σύκων ΊΤοντ]ρων σφό8ρα α ου βρωθησβται άπο πονηρίας 

In connexion with blood-shedding, 

/s. 59• 7 °* ^^ ττόδβί αυτών €π\ ττοκηρίαΐ' τρεχυυσι, ταχινοί (κχεαι 

Of the malice or mischievousness of an enemy, 

Szf. 12. 10 μη πιστΐύσης τω ^χθρω σου ίΐς τον αΙώνα' ώς γαρ 6 

χαΧκος Ιουται οϋτως ή πονηρία αυτοΰ. 
Esth. 7. 6 άνθρωπος εχθρός [Cod. t5 επίβουλος κα\ €χθρος~^ *Αμάν 

ό πονηρό? ούτος. 

They are used in similar relations and with equivalent 
meanings to translate other Hebrew words, 

Is. 35• 9 ουκ εσται λέων olde τών πονηρών θηρίων ου μη άναβ^ eh 

αυτήν. Heb. ΓΙ? 'violent.' 
Is. 10. I γράφοντες γαρ πονηριαν γράφουσι : Heb. /'^V 'mischief.' 

In all these cases it seems clear that the words connote 
not so much passive badness as active harmfulness or 

3. Use in the N. T. 

There are several passages in the Synoptic Gospels 
in which this meaning of ' mischievous ' seems to be 
appropriate : 

ΤΓονηρος, ΤΓονηρια, 79 

S. Matt. 5. 39 (Ύ^ h^ve heard that it was said, An eye for an 

eye, and a tooth for a tooth ') €γώ δε λέγω νμιν μη άντιστηναι τω 
πομηρω" αλλ' oaris σε ραπίζα els την de^iav σιαγόνα, στρ€ψον αυτω κα\ 
την αλλην. Whether τω πονηρω be masculine or neuter, the appro- 
priate meaning seems to be, * Resist not him who — or, that which — 
does thee mischief,' and an instance of the kind of mischief referred 
to is at once given, viz. that of a blow on the cheek. 

Ιδ. 6. 13 ρνσαι ημάς άπο του TTonfjpou. Here also, whether τοΰ 
πονηρού be masculine or neuter, the appropriate meaning seems to 
be, ' Deliver us from him who — or, that which — does us mischief.' 
This meaning will be confirmed by the antithetical clause μη 
(laeveyKTjs ημάς els neipaapov, if it be assumed that the meaning which 
is assigned above to els πeφaσμόv is correct (see p. 71): the two 
clauses are probably two modes of stating that which is in eifect 
the same prayer, ' Bring us not into affliction, but on the contrary, 
deliver us from him who — or, that which — is mischievous to us : ' 
hence in the shorter form of the prayer which is given by S. Luke, 
the second of the two clauses is omitted (in Codd. ίί Β L, etc. : 
cf. Origen De Orat. c. 30, vol. i. p. 265, ed. Delarue, δοκεί δε' μοι 6 

AovKcis dia τον μη elvevejKjjs ημα^ els πειρασ/ζόι/ ^υvάμeι dedidaxevai κα\ το 
ρνσοί ημάs άπο του πονηρού) ^. 

S. Mark 12. 45 (^^^ ^• Luke 11. 26) πνεύματα πονηρότερα eavTOv. 
S.Luke 7. 21 : 8. 2 πνεύματα πονηρά. Probably rather ^mischievous' 
or ' baneful spirits', i. e. spirits who do harm to men, than spirits 
who are bad in themselves : so in Tob. 3. 8 of Asmodaeus το πονηρον 
haipovLov, who killed the seven husbands of Sara. 

S. Matt. 5. II μακάριοι ε'στε οταν ονεώΐσωσιν υμάς κα\ 8ιώξωσιν καϊ 
εΐπωσιν πάν πονηρον καθ" υμών ψευδόμενοι ένεκεν εμοΰ. Probably, though 

less clearly than in the previous instances, the meaning is * mis- 
chievous ' or ' malicious accusation! 

S. Matt. 22. 18 yvovs δε 6 *lησoυs την πονηρίαν αυτών ^ * their malice ' 

or ^ evil intent' (=S. Mark 12. 15 τψ υπόκρισιν, S. Luke 20. 23 

την πανονργίαν). 


Another meaning of the words, though of less frequent 

^ The important questions of the gender of τον πονηρού and, if it be mas- 
culine, of the identification of 6 -πονηρό^ with 6 διάβολοι, involving as it does 
theological as well as philological considerations, cannot conveniently be dis- 
cussed here. 


occurrence, is clearly established, and helps to explain some 
otherwise obscure passages of the Synoptic Gospels : 

Sir. 14. 4, 5 has the following pair of antithetical verses, — 

6 συνάγων από rrjs ψνχη^ αυτυν avvayei aWuis 
Koi iv Tois άγαθοϊς αυτοΰ τρνφησονσιν aXXoi' 
6 τΓΟί'ηρός ίαντω τίνι αγαθός earai ; 
και ου μη (νφρανθησ^ται iv rots χρημασιν αντον. 

' He that gathereth by defrauding his own soul gathereth for 

And in his goods shall others run riot: 
He that is niggardly to himself to whom shall he be liberal ? 
And he shall not take pleasure in his goods/ 
Then follow five verses, each containing two antithetical clauses, 
and each dealing with some form of niggardliness : the first clauses 
of vv. 8, 9, 10 are strictly parallel to each other, 
πονηρός 6 βασκαίνων οφθαΚμω .... 

TrXeoveKTOv οφθαΧμος ουκ (μπίπΧατο μ^ρ'ώι 

οφθαλμός ττομηρός φθονερός in αρτω 

* the grudging eye,' ' the eye of the miser,' ' the niggardly eye/ 
being e\ddently different names for the same thing. 
•Sir. 34 (31). 23, 

Χαμττρον eV* αρτοις €νΧογησ€ΐ χ^ίΧη, 

και μαρτυρία της καΧΚονής αυτοΰ πίστη' 

ΊΓομηρω eV αρτω διαγογγύσ^ πόλις, 

καΐ η μαρτυρία τηξ ΤΓΟί/ηρίαξ αυτοΰ ακριβής. 

Ε. V. ' Whoso is liberal of his meat men shall speak well of 

And the report of his good housekeeping will be be- 

But against him that is a niggard of his meat the 
whole city shall murmur. 

And the testimonies of his niggardness shall not be 
doubted of.' 

The Hebrevv^ word 5?*^, which is usually translated by 
ττοι/ηρός, is also sometimes translated by βάσκανοξ, with a 
distinct reference, as in Sirach, to the ' evil * or ' grudging 
eye': e.g. 

ΤΓονηρός, ΤΓονηρΙα. 8 1 

Prov. 23. 6, 

μη (Tvvdeinpet avSpl βασκάνφ 

μηΒέ €πιθνμ€ΐ των βρωμάτων αντον. 

(For βασκάνω Schol. ap. Nobil. and Cod. 161 in marg. have 
* Feast not with him that hath an evil eye, 
Neither desire thou his dainty meats, 
(For he is as though he had a divided soul, [so Ewald] 
Eat and drink, saith he to thee. 
But his heart is not with thee).' 
So Deut. 28. 56 VJ^ LXX. βασκανύ, AquiL iTonf]pcU€Tai. 

This use of ττονηρός in the sense of ' niggardly ' or 
' grudging,' especially in connexion with the idea of the 
' evil eye,' throv^^s a clear light upon a well-known passage 
of the Sermon on the Mount, which, if taken in its context, 
will be seen to refer not to goodness or badness in general, 
but specially to the use of money : 

S. Matt, 6. 19 Lay not up for yourselves treasures upon the 
earth . . . 

20 But lay up for yourselves treasures in heaven .... 

21 For where thy treasure is, 
There will thy heart be also. 

22 The lamp of the body is the eye, 
If therefore thine eye be liberal, 

Thy whole body shall be full of light: 

23 But if thine eye be grudging {πονηρός), 
Thy whole body shall be full of darkness. 

24 Ye cannot serve God and mammon. 

If this meaning does not wholly remove the difficulties 
of the passage, it at least contains elements which any 
exegesis of it must recognize. The same meaning appears 
to be appropriate in two other passages of S. Matthew : 

6". AfaU. 7. II (=S. Luke 11. 13) el oZv υμύς ΊΓΟη^ροΙ ovres oXbare 
δόματα αγαθά 8ι8όναι to7s tckvois υμών . . . (which may be paraphrased 
thus) : 'If ye then, whose own nature is rather to keep what you 



have than to bestow it on others, are still able to give good gifts to 
your children, how much more shall your Father in heaven, who is 
always bestowing and never keeping back, give good things to 
them that ask Him ' ? 

S. Matt, 20. 15^0 οφθαλμός σον ττομηρός co-tlv οτι βγω αγαθός €Ϊμι, 

* Art thou envious at my being hberal' ? 


This word is found in the N. T. only in the Gospel and 
first Epistle of S. John. The facts upon which any in- 
duction as to its meaning there must be sought in the first 
instance in contemporary writings cognate in character to 
those of S. John. They are found in Philo in sufficient 
numbers and in a sufficiently clear connexion to render 
the induction from them free from doubt : they show that 
Philo used the word {a) in a sense closely akin to its Attic 
sense of one who helps or pleads for another in a court 
of law, and hence (ύ) in the wider sense of helper in 

(a) Philo De JosepL• c. 40, vol. ii. p. 75 (Joseph after discovering 

himself to his brethren says to them) άμνηστίαν απάντων παρέχω των 
fls e/ie π€πραγμ€νων' μηδ€ν6ς cTepov δβΐσ^ε παρακΧήτου, Τ grant yOU free 

forgiveness for all that you have done to me: you need no one else 
to intercede for you/ 

Vi't. 3ios. iii. 14, vol. ii. p. 155 (Philo gives the reason why the 
High Priest in going into the Holy of Holies Λvore the symbol of 

the Logos) avayKoiov yap ην τον ί€ρωμ4νον τω τον κόσμον πατρΧ παρακλητω 
χρησθαι reXeiorttro) την άρζτην νΐω προς re άμνηστείαν αμαρτημάτων κα\ 

χορηγίαν άφθονωτάτων αγαθών, 'it was nccessary that he who was 
consecrated to the Father of the world should employ as his inter- 
cessor the Son who is most perfect in virtue, for both the forgive- 
ness of sins and the supply of boundless goods.' 

So JDe Exsecrat. c. 9, vol. ii. p. 436 : in Flacc. c. 3, vol. ii. p. 
519, ih. c. 4, p. 520. 

(3) De Mund. Opif. C. 6, vol. i. p. 5 ovh^v\ hk παρακλήτω, τις γαρ ην 
€Tepo?, μόνω de ίαντω χρησάμξνος 6 θ^ος €γνω delv ^ν^ργίτ^Ιν . . . την 

•τταράκλητος, ττίστις, 83 

φύσιν, * employing not any helper — for who else was there ? — but 
only Himself, did God resolve that He ought to bless the world 
with His benefits.* 

The meaning which is thus established in Philo must 
be held to be that which underlies its use by S. John. 
The meaning * consoler ' or ' comforter ' is foreign to Philo, 
and is not required by any passage in S. John : it may, 
indeed, be supposed that ' comforter ' in its modern sense 
represents the form only and not the meaning of confor- 

In philosophical and later Greek ττίστις may be said to 
have three meanings, — a psychological, a rhetorical, and 
a moral meaning. In Biblical Greek it adds to these a 
theological meaning. 

(i) Its psychological meaning appears in Aristotle: it 
is ' conviction,' and as such is distinguished from νπόληψις 
or 'impression,' for a man may have an 'impression' and 
not be sure of it, Top. 4. 5, p. 125 /5 κατά ταϋτα δ' ovb' η 
ττίστίί νττόληψίζ' evhe-x^eraL γαρ την αυτήν ιητόληψιν καΐ μη 
τηστξνοντα ίχ^ιν : it is used both of the conviction which 
comes through the senses and of that which comes through 
reasoning, Phys. Auscult. 8. 8, p. 160, a η ttiWi? ov μόνον 
€7rt τη^ αΙσΘη(Γ€ω9 αλλά καΐ ΙττΙ του λόγου^ ' the conviction (of 
a particular fact which is mentioned) lies not only in the 
sensible perception of it but also in the reason ' : hence 
it may come either mediately or immediately. Top. 1. i, 
p. 100 b τα μη hi ^Τ€ρων αλλά hi αυτών έχοντα την ττίστιν, 
(of primary truths) 'which force their conviction not 
mediately through other truths but immediately of them- 

(2) Its rhetorical meaning also appears in Aristotle. 
It is not conviction but that which causes conviction in 

G 2 


the mind of a hearer. It is the ' proof ' of a case as dis- 
tinguished from 'statement' of it (which is ττρο^βσι? or 
bLr|γη(ΓLs, the latter word being limited by Aristotle to 
judicial speeches), the relation being similar to that of 
άτΓοδει^ι? to a ττρόβλημα: Rhet. 3. 13, p. 14.14a τούτων be 
[i. e. of the two parts of a speech] to μ€ν TrpoOeais €στί το 
he irCaTLS ώσπβρ h.v et rts dteAot otl to μ€ν ττρόβλημα τό be 

(3) Its moral meaning is also found in Aristotle: it is 
good faith or mutual trust: e.g. Pol. 5. 11, p. 1313^ 
η γαρ yv&ais ττίστιν Tioiel μάλλον irpbs άλληλονξ, ' mutual 
knowledge tends rather to produce mutual trust.' It is 
found more frequently in the later Greek philosophy : 
e.g. pseudo-Aristot. De Virtut. et Vit. c. 5, p. τι^ο b 
άκολονθ€Ϊ be Trj biKaiO(TVvr\ . . . . η πίστίί καΐ η μισοττονηρία^ 
'justice is accompanied by ... . good faith and the hatred 
of wrong-doing,' and Ethic. Eudem, 5. 2, p. 1237 b ουκ ίστι 
δ' avev ττιστεω? φίλια /3e/3ato9, 'there is no firm friendship 
without mutual trust' 

(4) In Biblical Greek it has another or theological mean- 
ing which we shall best understand by first examining 
its use in Philo, who furnishes a connecting link between 
its philosophical and its biblical use, and who, while using 
it in the main in its biblical sense, adds explanations which 
make its meaning clear. 

He sometimes uses it in its rhetorical sense of 'proof* 
or ' evidence ' : e. g. De Mundi Opif. c. 28, vol. 1. p. 20 
τιίστι^ Trjs αρχηί €ναργ€στάτη τα φαίν6μ€να, ' the actual facts 
(of man's relation to animals) are the clearest proof that 
God gave him dominion over them.' But he more com- 
monly uses it in a sense in which the intellectual state 
of mind which is called ' conviction ' is blended with the 
moral state of mind which is called 'trust.' It is trans- 
ferred alike from the conviction which results from sensible 
perception and from that which results from reasoning to 

ττίστις, 85 

that which is based on a conception of the nature of God. 
The mass of men trust their senses or their reason : in a 
similar way the good man trusts God. Just as the former 
believe that their senses and their reason do not deceive 
them, so the latter believes that God does not deceive 
him : and the conviction of the latter has a firmer ground 
than that of the former, inasmuch as both the senses and 
the reason do deceive men, whereas God never deceives. 

This use of the word will be made clear by the following^ 

De Mundi Opif, c. 14, vol. i. p. 10 (God anticipated, before ever 
men were created, that they would be guessers of probabilities and 

plausibilities) κώ. on πιστίύσουσι τοις φαινομίνοις μάλλον η θεά, ' and 

that they would trust things apparent rather than God.' 

Zegl's Alleg. iii. 81, vol. i. p. 132 άριστον ουν τω θ€ω ΊΤζπιστΐυκίναι 
και μη τοις άσαφίσι λογισμοίς και ταϊς άβφαίοις ciKaaiais, *it is best, then, 

to trust God and not uncertain reasonings and unstable conjectures.' 
Quis rer. div. heres c. 18, vol. i. pp. 485-6 (the trust in God with 
which Abraham is credited is not so easy as you may think, because 
of our close kindness with this mortal part of us which persuades us 
to trust many other things rather than God) ro be €κνίψασθαι τούτων 

€καστον κα\ άττιστησαι yeveaei Trj πάντα (ξ έαυτψ άττίστω, μόνω be τηστΐυ- 
σαι θ€<ρ τω κα\ προς άληθίίαν μόνω πιστω, μίγάλης και ολυμπίου biavoias 
epyov eaTi, προς ovbevos ονκβτι b€λeaζoμevηs των παρ* ή^^^, ' tO wash OUr- 

selves thoroughly from each one of these things, and to distrust the 
visible creation which is of itself in every way to be distrusted, and 
to trust God w^ho is indeed in reality the only object of trust, re- 
quires a great and Olympian mind — a mind that is no longer 
caught in the toils of any of the things that surround us.* 

De Migrat. Abraham, c. 9, vol. i. p. 442 (commenting on 

Genesis 12. i ' into a land that I will shew thee,' he says 

that the future tense is used rather than the present in testimony of 
the faith which the soul had in God : for the soul) avevboiaaTa νομί- 

σασα ήδη παρίΐναι τα μη παρόντα bta την του νπυσχομίνον βφαιοτάτην 

πίστιν, αγαθόν τίλ€ΐον αθλον (ϋρηται, ' believing without a wavering of 
doubt that the things which were not present were actually present 
because of its sure trust in him who had promised, has obtained a 
perfect good for its reward ' : (this ' perfect good ' is probably faith 


itself: cp. De praemiis et poems c. 4, vol. ii. p. 412 αβΚον alpeiTai τψ 
ηρ6ς τον Qeov πίστιν). 

De praemiis et poenis c. 5, vol. ii. pp. 412-13 (A man who has 
sincere trust in God has conceived a distrust of all things that are 
begotten and corruptible, beginning with the two things that give 
themselves the greatest airs, sense and reason. For sense results 
in opinion, which is the sport of plausibilities : and reason, though 
it fancies that its judgments depend on unchanging truths, is found 
to be disquieted at many things : for when it tries to deal with the 
ten thousand particular facts which encounter it, it feels its want of 
power and gives up, like an athlete thrown by a stronger wrestler) 
οτω de e^eyevero πάντα pev σώματα πάντα de ασώματα vnepibelv κώ. vnep- 
κΰψαι μόνω δε eπepeισaσ6aι καχ στηρίσασθαι θβω μετ Ισχυρο-γνώμονος 
λογισμού κα\ άκλινονς κα\ βeβaιoτάτης πίστεων, €υδαίμων κα\ τρισμακάριος 
οντος αληθώς, ' but he to whom it is granted to look beyond and 
transcend all things corporeal and incorporeal (objects of sense and 
objects of reason alike), and to rest and fix himself firmly upon 
God alone with obstinate reasoning and unwavering and settled 
faith, that man is happy and truly thrice blessed/ 

It will be seen from these passages that faith is regarded 
as something which transcends reason in certainty, and 
that when spoken of without further definition its object 
is God. It is consequently natural to find that it is not 
only ranked as a virtue, but regarded as the chief of virtues, 
την τβλζίοτάτην άρβτών Quis rer. div. heres c. 18, vol. i. 
p. 485, the queen of virtues, την βασιΚίΙα των άρκτων De 
Abraham, c. 46, vol. ii. p. 39 : in having it a man offers 
to God the fairest of sacrifices and one that has no blemish, 
αμωμον καΐ κάλλίστον Upelov otcrei Θίώ, ττίστιν De Cherubim 
c. 25, vol. i. p. 154. And in one passage he sings its 
praises in the following remarkable enconium : 

De Abraham. C. 46, vol. ii. p. 39 μόνον ovv d^evbes κα\ βίβαιον 
αγαθόν η προς τον Geov πίστις, παρηγόρημα βίου, πλήρωμα χρηστών (λπί- 
δων, άφορία pev κακών, αγαθών de φορά, κακοδαιμονίας άπόγνωσις, (ίσφίας 
γνώσις, ίυ8αιμονίας κλήρος, ψνχης iv άπασι βίλτίωσις, 4πΐρηρζΐσμ€νης τω 
πάντων αΐτίω, κα\ δνναμενω μίν πάντα βουλομίνω δε τα άριστα, ' Faith 

towards God [i. e. trust which has God for its object] is the only 

undeceiving and certain good, the consolation of life, the fulness of 
good hopes, the banishment of evils, the bringing of blessings, 
the renunciation of misfortune, the knowledge of piety, the pos- 
session of happiness, the bettering in all things of the soul which 
rests for its support upon Him who is the Cause of all things, 
and who though He can do all things wills only to do what is 

It v^ill be clear from this use of the v^ord in Philo that 
its use in the N. T. vi^as not a vi^holly new application of 
it : ' trust,' or ' faith,' had already become in the Alexan- 
drian schools an ideal virtue. It v^ill also be clear that, 
assuming it to be used by S. Paul in the sense which 
it bore in the philosophical language with which he was 
familiar, it is not used of a vague and mystical sentiment, 
the hazy state of mind which precedes knowledge, like 
a nebula which has not yet taken a definite outline or 
become condensed into a star, but that it is a state of 
firm mental conviction, based upon a certain conception 
of the nature of God ; hence it is used in close connexion 
with the strongest word for full assurance, viz. Ίτληροφο- 
ρύσθαι : Rom. 4. 20, 11 iv€bvvaμώθη rfj ττίστει, hovs bo^av 
τω Θεώ καΙ πληροφορηθείς otl ο ζττηγγξλταί hwaros €στί καΐ 
ττοίησαί, 'he waxed strong through faith, giving glory to 
God, and being fully assured that what He had promised 
He is able also to perform.' 

Hence in the Epistle to the Hebrews it is used, as Philo 
used it, to designate a state of mind which transcends 
ordinary knowledge, the conviction that the words or 
promises of God have a firmer basis of certainty than 
either phenomena of sense or judgments of reason; it 
believes that certain things exist because God has said 
so, and in spite of the absence of other evidence of their 
existence: and since it believes also that what God has 
promised will certainly come to pass, its objects are also 
objects of hope: hence it is described (11. 1) as ^λΈΐζομίνων 


ΰττόστασίί, ττραγμάτων €λ€γχο9 ον βλ€'πομξνων, ' the ground 
of things hoped for, the evidence of things not seen.* 


The word is used by the LXX. only i8 times in the 
canonical books, but it represents 15 different Hebrew 
words : in some cases it is difficult to avoid the conclusion 
that the LXX. misunderstood the Hebrew words, in other 
cases it must be admitted that the Hebrew text is itself 
both obscure and uncertain. 

In some passages it appears to be the translation of 
Ίψ^ * outpost' or 'garrison,' viz. i Sam. 13. 2^^ (= Theod. 
στάσΐί): 14. 4- That it can bear this meaning is shown 
by its use in a fragment of the Phoenix of Sophocles in 
the sense of hibpa (Iren. ap. Socrat. H. E. 3. 7 τιάρα. Σοφο- 
κλζΐ iv τω ΦοίνίΚί kvihpav σημαίν^ιν την ντιόστασιν : and Pollux, 
Hist. Phys. p. 376). 

The consideration of some of the other passages seems 
to belong rather to Hebrew than to Hellenistic philology : 
but there is a small group of passages which furnish a 
well-established meaning and which throw a clear light 
upon some instances of the use of the word in the N. T. 

Ruth I. 12 oTi ίϊπα on tan μοι υπ<5στασι$ τον γίνηθηναί μ€ avbpl και 
re|o/xai vlovs . . . ' for my Saying (i. e. if I said) that there is ground 
of hope of my having a husband and I shall bring forth sons . . . ' : 
ΐ'πί)στασΐΓ = Π1ί^Γΐ ' hope.' 

Ps. 38 (39). 8 ή νπόστασίς μου πάρα σοι βστιν, ' my ground of hope 
is in thee' : {ιπόστασΐ5=ΠρΠίη * expectation,' which Aquila renders 

by καρα8οκία, Symmachus by αναμονή. 

Ezek. 19.5 ατ:(ύΚίτο ή νπόστασίί αύτψ, * her ground of hope was 
lost': νπόστασΐ5=^)ϊ>Ρι^ which Symmachus renders by προσδοκία, 
Theodotion by ίλπίς. 

This meaning * ground of hope ' probably follows from 
the Classical use of υττόστασις for the ' ground ' or ' founda- 

υττόστασ-ις, συκοψαντεΐν, 89 

tion ' of anything : and it passes by a natural transition 
into the meaning of 'hope' itself. Hence its use in several 
passages of the N. T. 

2 Cor. 9. 4 μήπως .... καταισχυνθώμεν ημίϊς . . . eV τη νποστάσΐί 
TavTjj, * lest by any means ... we should be put to shame ... in 
this ground' (so. of our glorying on your behalf: Codd. N^. Dc. and 
others add της κανχησ€ως, from the following passage). 

2 Cor. II. 17 ο λαλώ ου κατά, κνριον λαλώ αλλ' ως iv άφροσνντ), ev 
ταυττ) τη νποστάσΐΐ της κανχησ€ως, 'that which I Speak I Speak not 

after the Lord but as in foolishness, in this ground of my glorying.' 

ffed. 3. 14 iavirep την άρχην της υποστάσ€ως μ^χρι τέλους βφαίαν 

κατάσχωμ€ν, ' we have become partakers of Christ, if, that is to say, 
we continue to hold the beginning of our hope firm until the end' : 

cf. V. 6 eav την παρρησίαν κα\ το καύχημα της ελπιδο? μίχρι τέλους β€βαίαν 

Heb. II. Ι %στιν δε πίστις ίλπιζομ^νων νπόστασις, ' Faith is the 
ground of things hoped for,' i. e. trust in God, or the conviction 
that God is good and that He will perform His promises, is the 
ground for confident hope that the things hoped for will come to 

(In the same passage ίλεγχος appears to be used in its Hellenisdc 
sense of a fact which serves as the clear proof of another fact: e.g. 
Jos. An/. 16. 8. I Herod's slaves stated that he had dyed his hair, 
thereby κΚίπτοντα τον eXeyxov της ηλικίας, * concealing the clear proof 
of his age': Epict. Diss. 4. 146 speaks of the fears of the Emperor's 
favour or disfavour which were ελ^γχον?, ' clear proofs,' that though 
the professors of philosophy said that they were free, they were in 
reality slaves : so trust in God furnishes to the mind which has it a 
clear proof that things to which God has testified exist, though they 
are not visible to the senses). 

1. Classical use. 

In Classical Greek the word and its paronyms are used 
exclusively of calumnious accusations, especially of such 
as were intended to extort money: e.g. Xen. Mem. 2. 9. i, 
where it is used of those who brought suits against Crito, 


who was known to be rich, because, as he says, νομίζονσ-ιν 
rjhLov av μ€ αργύρων τ^λ^σαι η ττράγματα €χ€ίν, ' they think 
that I would a good deal rather pay money than have 

2. Use in the LXX. 

Its wider range of meaning in the LXX. is made clear 
by several kinds of proof: {a) it is used to translate 
Hebrew words which mean simply either ' to oppress ' or 
' to deceive ' : (3) it is interchanged with other Greek words 
or phrases which mean simply ' to oppress ' : {c) it occurs 
in contexts in which its Classical meaning is impossible. 

(a) In Job 35. 9. Ps. 71 (72). 4: 118 (119). 122, 134. Prov. 
14. 31: 22. 16: 28. 3, 16. Eccles. 4. I : 5. 7 : 7. 8, they are 
translations of Ρψν ' to oppress,' or of one of its derivatives : in 
Lev. 19. II of"^i?^ 'to lie.' 

(f) In Gen. 26. 20 LXX. άΒικία' η^ίκησαν yap αυτόν =zAqm\. συκο- 
φαντία' (συκοφάντησαν yap αυτόν. Lev. 6. 2 LXX. ^δίκί;σ6 = Aquil. 
Symm. Theod. (συκοφάντησε, Deut. 24. 14 LXX. ουκ άπαδικησ€ΐς=: 
Aquil. Symm. Theod. ου συκοφαντήσεις. Job 10. 3 LXX. iav άδι- 
κησω=^''Α\Κθζ' όταν συκοφάντησες. Ezek. 2 2. 29 LXX. εκπιεζουντες 
άδικΙα=ζ Aquil. Symm. ^συκοφάντησαν συκοφαντίαν. Ezek. 2 2. 12 LXX. 
καταδυναστεία, Symm. συκοφαντία, and SO also Aquil. mjer. 6. 6. 

{c) It is used especially in reference to the poor, whereas the 
Classical use related especially to the rich : Ps. 71 (72). 4 'he shall 
save the children of the needy and shall break in pieces the oppressor 
{συκοφάντην) : Prov. 14. 31: 22. 16 'he that oppresseth (συκοφαντών) 
the poor': id. 28. 3 'a poor man (so E. V. but LXX. άνδρύος kv 
άσεβίσι) that oppresseth (συκοφαντών) the poor ' : Eccles. 4. i ' so I 
returned and considered all the oppressions (συκοφαντίας) that are 
done under the sun : and behold the tears of such as were oppressed 
(των συ κοφαντου μένων), and they had no comforter ; and on the side 
of their oppressors (συκοφαντούντων) there was power ; but they had 
no comforter.' 

3. Other Hellenistic uses. 

The meaning of the word which appears in the LXX. 
appears also in some Egyptian documents, which are the 

συκοψαντ€Ϊν, ύττοκριτής. gi 

more valuable for comparison because the social state of 
Egypt under the Ptolemies and afterwards under Roman 
rule was in many respects closely similar to the state of 
Palestine in the corresponding period of its history. 

In Brunei de Presle Notices et textes du Musee du Louvre in the 
Notices et extraits des manuscrits de la Bibliotheque Impe'riale, Tom. 
xviii. 2<ie partie, Paris 1865, papyrus No. 61, p. 351, consists of a 
letter of B.C. 145 from Dioscorides, a chief officer of finance, to 
Dorion, a local subordinate. After reciting the strong desire of 
the king and queen (Ptolemy Physcon and Cleopatra) that there 
even justice should be dealt (δικαιοδοτεΐσ^αι) to all classes of their 
subjects, the document proceeds ττ^ρΐ hi διασβισμώ»/ και παραλίων 
evLCiv de καΙ συκοψαμτεΐσθαι προφ€ρομ€νων βονΧόμ€θ^ υμάς μη hiaKav6av€iv 
ore [ravToj ττάντα iariv άΧΚότρια τψ τβ ημών ά-γω-γης ούχ ησσον δε καί της 
νμΐτύρας σωτηρίας irrav τις e^fXey;^^,^ Χ^λυπηκώς τίνα των κατά μίρος, ' m 

the matter of fictitious legal proceedings and plunderings, some 
persons being moreover alleged to be even made the victims of 
false accusations, we wish you to be aware that all these things are 
at variance not only with our administration but also and still more 
with your safety when any one is convicted of having injured any- 
one in his district.' 

The offences διασεισ/χο?, nrapak^ia, συκοφαντία, are evi- 
dently all offences committed by taxgatherers. 

In the Corpus Inscr. Graec, N°. 4957 consists of a decree of 
Julius Alexander, prefect of Egypt in A. D. 68, and is almost 
entirely concerned with the wrongs done by local au- 
thorities, especially in the matter of the revenue. 

ντΓΟκρίσίς, υποκριτής. 

In the Old Testament υττοκρίτψ is found in two passages 
of Theodotion's translation of Job which have been incor- 
porated into the LXX. text, and in each case it is the 
translation of Γ]Ρ_Π ' impious ' : Job 34. 30 βασιλεύων av- 
θρωττον VTTOKpLTTjv άτΓο bviTKokias λαον, ' making an impious 
man king on account of the discontent of the people ' : 


Job ^6. 13 και νττοκρίταϊ καρδία τάξουσι θνμόν, 'and the 
impious in heart shall ordain (for themselves) wrath.' The 
word ^yn is also translated by νττοκρίτηί by Aquila and 
Theodotion in Job 15. 34, where the LXX. have άσξβουί ; 
by Aquila in Job 20. 5, where the LXX. have τταρανόμων ; 
by Aquila, Symmachus, and Theodotion in Prov. 11. 9, 
where the LXX have ασφ&ν: and by the same three 
translators in Is. ^^. 14, where the LXX. have ασ^βά^. 
Similarly fj^^, which only occurs in Is. 3:?. 6, is there 
translated by the LXX. άνομα, and by Aquila, Symmachus, 
and Theodotion ντ:όκρισιν. 

These facts seem to shew that early in the second 
century, and among Greek-speaking Jews, υποκριτής had 
come to mean more than merely 'the actor of a false 
part in life.' It connoted positive badness. The inference 
is corroborated by its use in the 'Two Ways,' especially 
in the form in which that treatise is appended to the 
Epistle of Barnabas, c. 19. 2 ov κολληθήστ} μβτα ττορ^νομ^νων 
kv όδω θανάτου, μίσή(Τ€υί ττάν ο ουκ ^afiv άρεστον τω Θβω, 
μίσησ€ί9 ττασαν υτΐόκρισιν ου μη ζγκαταλίτττ^^ ivroXas κυρίου, 
' thou shalt not join thyself with those who go in the way 
of death, thou shalt hate whatever is not pleasing to God, 
thou shalt hate all υτιόκρισιν, thou shalt not abandon the 
commandments of the Lord.' The collocation and em- 
phasis can hardly be accounted for unless υττόκρισιν has 
a stronger meaning than that of ' false pretence.' 

The meaning which is evident in the Hexapla seems 
more appropriate than any other in the Synoptic Gofepels : 

iS". MaU. 24. 51 (of the master returning suddenly and finding 
the slave whom he had set over his household beating his fellow 

slaves) διχοτόμησα αυτόν και το μ€ρος αντοΰ μ€τά των υποκριτών θησα^ 

' he will surely scourge him, and will appoint his portion with the 
impious ' : it would be mere bathos to render υποκριτών by ' false 

S. Matt. 23. 28 €(τωθ€ν be cVre μαστοί νποκρίσ€ως καϊ ανομίας^ 

υτΓΟκριτης. 93 

' within they are full of impiety and wickedness ' : and in the 
denunciations of the Scribes and Pharisees which both precede 
and follow this verse the point seems to be not merely that they 
were false pretenders but that they were positively irreligious. 

iS*. Mark 12. 15 ίΖδώ? αντων την νπόκρισιν=3. Matt. 2 2. 18 γνονς 
de 6 Ίί/σοΰί την πονηρίαν αντων, S. Luke 20. 23 κατανοησας δε αντων την 
πανονργίαν : the three words νπόκρισιν, πονηρίαν, πανονργίαν are of 

equivalent meaning: and in S. Mark as in the two other Evan- 
gelists that which our Lord is said to have known was not their 
' false pretence ' but their ' wickedness ' or ' malice/ 



In examining any philosophical terms which are found 
in Hellenistic Greek it is necessary to observe to an in- 
creased degree the caution with which all Hellenistic words 
must be treated. At every step the student is haunted 
by their Classical meanings, and at every step the ghosts 
of their Classical meanings must be exorcised. For Greece 
and the Greek world had come not only under a different 
political rule, and into new social circumstances, but also 
into a new atmosphere of thought and to a new attitude 
of mind towards the questions with which philosophy deals. 
Those questions were,, almost of necessity, stated in their 
ancient form ; the technical terms remained the same : 
but by the operation of those silent changes by which 
all thinking races are constantly elaborating new meanings, 
and finding new points of view, the connotation of those 
terms and the answers to those questions had undergone 
more than one complete transformation. The philosophical 
words of Hellenistic Greek must be viewed in relation not 
to past but to contemporary philosophy. Nor can that 
contemporary philosophy be taken as an undivided whole. 
It is as various in its character as the philosophy of our 
own time, with which it is the more interesting to compare 
it because, as in our modern philosophy, a large part of 
it was syncretistic. 

For the investigation of such philosophical terms as 
are found in the New Testament we possess a mass of 
material of unique value in the writings which are com- 


monly gathered together under the name of Philo. Except 
in relation to the doctrine of the Aoyo?, which is itself 
often misunderstood because it is isolated from the rest 
of the philosophy, those writings are an almost wholly 
unworked mine. Many of the MSS. which contain them 
remain uncollated : no attempt has been made to differen- 
tiate the characteristics of the main group of writings so 
as to afford a criterion for distinguishing between the 
writings of Philo himself and those of his school : the 
philosophy itself, which is more like a mosaic than an 
organic unity, has for the most part not been resolved 
into its elements. But although whatever is now said 
about Philo must be regarded as subject to correction 
in the future when the writings which bear his name have 
been more critically investigated, the study of those writ- 
ings is indispensable for the determination of the meanings 
of Hellenistic words which even touch the circumference 
of the philosophical sphere. It would be unwarrantable 
to assert that the meaning of such words in Philo deter- 
mines their meaning in the New Testament : but at the 
same time no inference as to their meaning in the New 
Testament can be regarded as even approximately certain 
if it leaves out of sight the evidence which Philo affords. 

But the number of words in the New Testament which 
can be regarded simply as philosophical terms with an 
added theological connotation is very small. An instance 
has been given in the preceding chapter in ττίστίί. The 
majority of terms which appear to be philosophical require 
a different kind of caution in their treatment. For Biblical 
Greek is with comparatively rare exceptions not a philo- 
sophical but a popular language. It is not, that is to 
say, the language of men who were writing with scientific 
precision to an inner circle of students, but that which 
was addressed to, and therefore reflected from, the mass 
of the people, to whom, then as now, the minute distinc- 


tions of philosophy are unfamiliar, and to a great extent 
incomprehensible. The tendency of many commentators 
and lexicographers has been to assume the existence in 
Biblical Greek of the distinctions which are found in 
philosophical writers, and to attach to words in their 
popular use meanings which belong to them only in their 
philosophical use. The presumption is that in the majority 
of cases those distinctions and meanings are inapplicable : 
and the presumption is sometimes raised to proof by the 
evidence which the LXX. affords. 

I propose to deal with a special group of philosophical 
terms, viz. psychological terms, partly because of their 
importance in themselves, and partly because they furnish 
a good illustration of the general principle which has been 
stated. In dealing with them I propose to investigate 
(i) their use in the LXX. and Hexapla, (2) their use in 

I. Psychological terms in the LXX. and Hexapla. 

In the case of all but concrete terms, such as horse, fire, 
wood, used in their primary sense, it must be borne in 
mind that a general equivalence of connotation between 
two words in two difierent languages must not be held to 
imply an exact coincidence of such connotation. The domi- 
nant meaning of a word in one language must no doubt 
be held to form at least an integral part of the meaning 
of the word by which it is translated in another language : 
but it is only by adding together all the predicates of the 
two words in their respective languages that an inference 
becomes possible as to the extent to which the spheres of 
their connotation coincide. 

When the two terms are each of them so far isolated 
in their respective languages that the one is uniformly the 
translation of the other, this addition of predicates is the 
only method by which the extent of the coincidence of 


their connotation can be determined. But in dealing with 
groups of allied terms, for example, psychological terms, 
this method may be supplemented by others. If it be 
found that each member of the group in one language 
is rendered uniformly by one and only one member of 
the corresponding group in the other language, it must 
no doubt be inferred that each term had in its own lan- 
guage a distinct and isolated meaning, and no other method 
than that of the addition of predicates will be applicable. 
But if it be found, as it is found in the case of the terms 
with which we are about to deal, that the members of 
the group in the one language are each rendered by more 
than one of the members of the group in the other lan- 
guage, it must be inferred that while the group as a whole 
in the one language corresponded as a whole to the group 
in the other, the individual members of the two groups 
did not so correspond. 

The question which lies immediately before us is that 
of the precise extent of the correspondence or non-corres- 
pondence between the respective members of the two 
groups, and of the light which that correspondence or 
non-correspondence throws upon the meaning of the Greek 
terms. In other words, given a group of Hebrew terms 
ABCi and a corresponding group of Greek terms abc^ 
since it is found that a is used to translate not only A 
but also sometimes Β and C, and that δ is used to trans- 
late not only Β but also sometimes A and C, and that c 
is used to translate not only C but also sometimes A and 
B, and conversely that A and Β and C are each of them 
translated, though in varying degrees, by a and δ and c, 
what may we infer as to the relations of the Greek terms 
a and d and c to each other ? 

It will thus be found necessary to ascertain 

(i) of what Hebrew words each member of the Greek 
group is the translation : 


(ii) what corrections of and additions to the trans- 
lations of the words in the LXX. are found in 
the Hexapla. 
(iii) by what Greek words each member of the Hebrew 
group is translated : 
When these questions have received provisional answers, 
it will be found necessary to ascertain further how far 
those provisional answers are confirmed by (i) the com- 
binations and interchanges of the several words in the 
same or similar passages, (2) the predicates which are 
attached to the several words. 

I. Translations. 

I. καρδία. 

It is ordinarily the translation of I7 or 117• 

i. The other words which it is used to translate are — 

(i) i?|l 'the belly': Prov. 22. 18, Hab. 3. 15. 

(2) TO 'my bowels': Thren. 2. 11, where the MSS. vary 
between κοικΊα and καρδία. 

(s) ^"3^ 'the inward parts': Ps. 5. 10: 61 (62). 5: 93 (94). 
19, Prov. 14. 33 : 26. 24. 

(4) D^i 'the spirit': Ezek. 13. 3. 

In several passages the Hebrew is paraphrased rather 
than translated: e.g. Ps. 31 (32). 5: 84 (85). 9, Prov. 15. 
22 ; and in one instance, Ps. ^6 (37). 14 tovs ^vOeh τγ[ καρδία 
is a mistake of either the translator or the transcriber for 
the less familiar tov9 evO^ls rfj οδω. 

ii. The translation of Iv by Kapbia is almost always ac- 
cepted by the translators of the Hexapla, and the MSS. 
of the LXX. do not greatly vary: the corrections and 
variations are the following : 

Beut. 6. 5 : 28. 47, Jos. 22. 5 MSS. vary between καρΒίας 

{καρδία) and διανοίας [διάνοια). 


2 Sam. 7. 27 LXX. καρ^ίαν, Symm. diavoiav. 

^^• 36 (37). 15 Codd. A. B. Kapbiav, Cod. S\ ^νχψ, S^ ψυχά?. 

Ps. 72 (73). 13 LXX. Aquil. Kapbiav, Symm. Theod. ψνχηρ. 

Prov. 7. 3 LXX. καρ8ίας, Symm. στήθους. 

Eccles. 7. 3 LXX. ds καρδίαν, Symm. rfi biavoia. 

Eccles. 10. 3 "ipn \^h 'his heart faileth him': LXX. καρδία αντον 

vareprjaeif Symm. ανόητος. 

Jer. 5. 21 37 pti] 'without heart' : LXX. και άκάρ8ιος^ Symm. κα\ 


Jer. 38 {31). 33 LXX. καρδίας, Theod. στήθους. 

iii. The other words by which λ^, DIT' are translated 

(i) νους, Jos. 14. 7, Is. lo. 7, 12 : and in the phrase νουν ^φιστά- 
veiv for yp. iT'ti^ 'to apply the heart to . . .'^Kapblav ΐφιστάν^ιν 
Prov. 22. 17: 27. 23, Kaphlav τιθίναι I Sam. 13. 20, Ps. 47 (48). 
14: so Symin. Job 7. 17 νουν Ίτροσ^χ^ιν. and for 1? ^^ Is. 
41. 22•=. Aquil. Symm. Theod. καρΒίαν (ψιστάναν. 

(2), (3) διάνοια, ψυχή : see below. 

(4) σαρξ, Ps. 27 (28). 7 άνίθαΚΐΡ η σαρξ μου, Aquil. Symm. Theod. 
η Kapdia. 

II. πκβυμα. 

It is ordinarily the translation of Γϊ^Ί. 

i. The other words which it translates are — 

(i) D''*n 'Hfe*: Is. 38. 12=. Aquil. Symm. ζωή, as usually in 

(2) nrn^ 'breath': i Kings 17. 17. 

ii. The translation of Π^"^ by ττνζΰμα is almost always 
accepted by the other translators who are included in the 
Hexapla, and the MSS. of the LXX. do not greatly vary : 
but several of the instances of revision and variation are 

/οδ I. 19 LXX. πνεύμα, Aquil. άνεμος: SO id. 30. 15 Symm. 
Ps. 32 (33). 6 LXX. τω πν€υματι, Symm. TTj nvorj. 
Ps. 142 (143). 4 LXX. πνεύμα, Aquil. ψυχή. 

U 2, 


Ps. 148. 8 LXX. 7Γν€νμα, Alius άνεμος. 

Eccles. I. 14 LXX. προαίρβσις πνεύματος, Aquil. νομή άνψου (sO 
Aquil. Theod. ih. 2. 11), Symm. βόσκησις άνψου (so also ib. 4. 16). 

Eccles. 3. 19 LXX. πνεύμα, Symm. αναπνοή. 

Eccles. 6. 9 LXX. προαίρεσις πνεύματος, Aquil. Theod, νομή άνεμον> 
Symm. κάκωσις πνεύματος. 

Eccles. 7} 8 (9). LXX. νψηλον πνεύματι, Symm. υψηλό κάρ8ίον. 

Is. 7• 2 LXX. πνεύματος, Symm. 6 άνεμος. 

Is. 32. 15 LXX. πνεύμα, Symm. άνάψυξις, Theod. άνεμος. 

iii. The other words by which Π^Ί is translated are the 
following : 

(1) άνεμος, PrOV. 30. 4, SO alsO Symm., but Aquil. πνεύμα. 

(2) θυμός, Job 1 5. 1 3, PrOV. 18. 14 {Aquil. πνεύμα): 2g. ii,Ezek. 
39. 29, Zach. 6. 8. 

(3) καρδία, Ezek. 13. 3. 

(4) νους, Is. 40. 13 τίς yap εγνω νουν κυρίου, Aquil. πνεύμα: the 

passage is important on account of its quotation by S. Paul in 
Rom. II. 34, I Cor. 2. 16 : the use of νους rather than πνεύμα in 
the latter passage is especially noteworthy because πνεύμα would 
have followed more naturally from the preceding verses : and since 
this is the only passage in the LXX. in which D^i is translated by 
νους, the presumption is very strong that S. Paul had the LXX. in 

(5) ^Pyhy Prov. 16. 32, Is. 59. 19, Aquil. Symm. Theod. πνεύμα 
(which is used, without any qualifying word, to denote anger in 
LXX. Judges 8. 3). 

(6) πνοή, Gen. 7• 22 πνοψ ζωής: ProV. I. 23 εμης πνοής ρησιν, 

Aquil. Theod. πνευμά μου: id. II. 13 πιστός δε πνοτ}, Aquil. Symm. 
πνεύματι : Is. 38. 1 6 εξηγειράς μου την πνοψ, Aquil. ζωη πνεύματος μου. 

(ι) ψνχν, Gen. 41. 8, Ex. 35• 21. 

(8) φρόνησις, Jos. 5• I. 

In Job 6. 4, Prov. 17. 23: 25. 28, Is. 32. 2 the LXX. translation 
is not literal, and the Greek and Hebrew cannot be balanced word 
for word. 

There are some noteworthy compound phrases into 
which Π^Ί enters, which in the LXX. are rendered by 
όλίγόψνχο^, όλίγοψν)(^ία : 

Ex. 6. 9 Πϊ|Ί Ίνρ < shortness of spirit ' : LXX. oKiyoy\rvxia^ Aquil. 

κοΧοβότης πνΐύματος. 

Ps. 54 (55). 9 ^V^ n^ltt 'from the stormy wind' is rendered 
in the LXX. by the gloss άπο όλιγοψυχίας, Aquil. Theod. άπ6 rrvev- 
ματος Χαΐλαττώδονς. 

Prov. 14. 29 πη Ί^ϊ? < hasty of spirit': LXX. ο\ι^ο^νχο<ϊ, Alius 


Prov. 18. 14 nX3J nn ' a broken spirit': LXX. okiyo^vxov ap^pa, 

Theod. πν€νμα ττΐπΚη-γμίνον, 

/f. 54- 6 Πη nn^Vj; 'pained in spirit': LXX. ολίγόχ/^υχο?, ^^«//. 

Symm. Theod. κατώδυνος πν^υματι. 

III. ψυχή. 

It is ordinarily the translation of tl''D5 • 
i. The other words of which it is the translation are the 
following : 

(i) ^^i^ 'man': Lev. 17. 9, where the MSS. vary between '^νχη 

and άνθρωπος. 

(2) π;π, ύ'^η qife': job 38. 39, Ps. 63 (64). I {Symm. ζω^ν): 
73 (74). 20. 

(3) Dp, 117 'heart': 2 Kings 6. 11, i Chron. 12. 38: 15. 29: 
17. 2 : 22. 9, 2 Chron 7. 11 : 9. i : 15. 15 : 31. 21, Ps. 68 (69). 
21 (Aquil. Symm. Kapdlav), Prov. 6. 21: 16. i (15. 32), Is. 7. 2, 4: 
10. 7 : 13. 7 : 24. 7 : 33. 18 : 42. 25 : 44. 19. In Ps. 20 (21). 2: 
36 (37). 15, Prov. 26. 25 the MSS. vary between yj /νχη and καρδία. 

(4) np 'a dead body': Ezek. 44. 25, Symm. ν€κρω: in Num. 

23. 10 απο^άνοι ^ 'ψ'^Λ'? h^^ ^^ ψυχαίς δικαίων, ψνχαϊς must be Con- 
sidered to be part of a paraphrase rather than a literal translation 
of Τ\)Ό ' death ' : but in Num. 9. 6 eVl ψυχί} {^^^!?) no doubt means 
* by the dead body.' 

(5) ""ja 'look': Prov. 27. 23 (perhaps like the English 'person'). 

(6) n^'' 'spirit' : Gen. 41. 8, Ex. 35. 21 {Aquil. πνεύμα). 

In Ps. 38 (39). 12 τψ ψνχψ is a free gloss for that which is 
more literally rendered by Symmachus τό ^τηθυμψόν, 

ii. The variations in the translation of ITDJ by ψνχη 
in the Hexapla and in MSS. of the LXX. are the fol- 
lowing : 


Ex. 23. 9 LXX. τψ ψνχην, Aquil. {την) θλίψιν. 
Num. 9. 6 LXX. eVl ψνχί}, Alius eVl ν^κρω. 
I Sam. 24. 10 LXX. τψ ψυχψ, Aquil. Symm. Theod. τψ κακίαν. 
Job 6. II ΟΤΙ ανέχεται μου ή ψνχη, Aquil. otl μακροθνμησω. 
Ps. 87 (88). 15 Codd. AS. tmrt απω^βΐ? την ψνχψ μου, SO Aquil. 

Symm. : Cod. Β., ed. Rom., τψ προσ€υχην μου. 

PrOV. 24. 12 ο πλάσα? πνοην πάσιν, Aquil. Symm. διατηρων ψνχην 

Prov. 28. 26 Ιί^33"3ΠΊ literally as in Aquila ιτλατυς •<\rvxri=.Symm. 
πλατύψνχος : the LXX. dropS ^^l and has Cod. A. άπληστος, Cod. B. 

In Prov. 13. 25 δίκαιο? εσθων €μπιπλα την ψνχψ αντον, ψνχαι δε 

άσφών ivhceis, it is possible that there is some confusion in the text : 
ψνχην, as usual, translates It^^S^, but is wrongly amended by a 
reviser (^AXXos) to κοιλίαν, but ψνχαί translates i^f ' belly,' and is 
rightly amended to κοιλίαι {Aquil. Symm. Theod. Quint, in Syriac, 


iii. The other words by which tTDi is translated are the 
following : 

(i) ανηρ, Gen. 14. 21, Prov. 16. id, ■= Aquil. Symm. ψνχη. 

(2) Jos. 10. 28, 30, 35, 39 t^5|n~?3 is translated by πάν έμ,τΐνέον. 

(3:) /s. 43. 4 άρχοντας νπβρ της Κ€φαλή8 σου. 

(4) Gen. 36. 6 πάντα τα σώματα, i.e. slaves, as probably πάσαν 
ψνχψ in Gen. 12. 5• 

In Is. ^9. 8 μάταιον το βννπνιον is a free gloss for that which Aquila, 
Symmachus, and Theodotion render literally by κ€νη ή ψνχη αυτοΟ. 

In Jer. 28 (51). 14 ω/χοσβ κύριος κατά του βραχιόι /os αντον is a 
characteristic periphrasis for της ψνχης, which is not amended in 
the existing fragments of the Hexapla. 

IV. διάίΌΐα. 

It is ordinarily the translation of I7 . 

i. The other words which it translates are — 

(i) ΠΠξ^ΠΟ 'thoughts': Is. 55. 9. 

(2) y]]l 'inward parts' : Jer. 38 (31). 33. 

ii. The variations of the LXX. translation of ^? by δΐίίΐΌΐα in the 
Hexapla are — 

Gen. 34. 3 LXX. κατά τψ Βιάνοιαν, Aquil. eVi Kapbiav, Symm. κατα- 

IN THE LXX. 103 

Ex. 35. 22 LXX. Symm. rfj diavola, Aqiiil. καρδία. 
Lev. 19. 17 LXX. Tji diavola, Ah'us iv rf] καρδία. 
Job I. 5 LXX. eV TTi diavola, Aquil. €π\ καρδίας. 

■^^• 35• 4 LXX. oi ολιγόψυχοι τη διάνοια, Aquil. vols ταπανοΐς rij 
καρδία, Symm. Tois avofjTOis, Theod. ταχνκαρδίοις. 

iii. The other words by which 2/ is translated have been given 
above, under καρδία. 

2. Combinations and interchanges in the same 
or similar passages, 

(1) καρδία and ττΐ'ευμα : Ex. 9. 13 etc. (σκληρννβ 5e κνριος τήκ 
καρδία»' Φαραώ, but Deut. 2. 30 Ισκ\ηρυνε κύριος 6 θξος το T{veO[f.a 
αντου : Jos. 2. II ζξέ'στημβρ τγ]" καρδία ημώρ κα\ ουκ €στη en irveOp,a iv 
ονδίνι ημών. Ps. ^Ο {^ΐ). 19 θυσία τω θεω ττνεΟμα συντετριμμίνον, καρ- 
δίαΐ' συντ€τριμμ€νην και τίταπεινωμίνην 6 ucos ουκ €ζουδ€νωσ€ΐ: Ps. ^6 
(77)• 7 ''ΐ^ίτό? μ€τα τη? καρδία? μου ηδο\€σχουν κα\ €σκαΧΚον το ττν€υμ.ά 
μου: Ps. 77 (7^)• ^ y^vea ήτις ου κατβύθυνζρ iv Tifj" καρδία αντης κα\ ουκ 
ΐττιστώθη. μετά του θίοΰ το πνεύμα αύτης '. Ps. 1 42 (143)• 4 ηκηδίασεν iir 
4μ€ τδ πΐ'€υμά μου, iv iμo\ βταράχθη ή καρδία μου : Ezek. II. 1 9 δώσω 
αυτοϊς καρδία»' irepav κα\ ττνζνμ,α καινον δώσω iv αυτοϊς, SO ϊδ. ^6. 2 0. 
In one instance the words are interchanged between the LXX. 
and the Hexapla, Eccles. 7. 8 LXX. υψηλον πνεύματι, Symm. 


(2) καρδία and ψυχή : (a) Sometimes they are combined : Deut. 

6. 5 ^σται τα ρήματα ταύτα . . , . iv τη καρδία σου κα\ iv ττ} ψυχή σου : 

SO ιδ. II. 1 8, Jos. 23. 14, Ι Sam. 2. 35» ι Chron. 22. 19. {δ) Some- 
times they have the same or analogous predicates : Judges 19.5 
στηρισορ τήκ καρδίαν σου ψωμω άρτου: Ps. 103 (104)• 1 5 "ρτοζ καρδίαΐ' 
ανθρώπου στηρίζει: Ps. 34 (35)• ^3 iτo■π€ίvoυv iv νηστεία Tr\v ψυχή μ μου, 
so Ps. 68 {^g). 11: Ps. 77 (78). 18 βρώματα rals ψυχαϊς αυτών: Jer. 

4. ΙΟ ηψατο η μάχαιρα εως ttJS ψυχή? αυτών, ιδ. V. 1 8 ηψατο εως τηξ 

καρδίας σου. {c) Sometimes they are interchanged in the MSS. of 
the LXX., or in the Hexapla: e.g. Ps. 20 {21). 2, Codd. A. B. 

ψυχής, Cod. S^. καρδίας: Ps. 36 (37). 15, Codd. A. B. καρδίαν, Cod. 

5. ψνχην (^υχάς): Ps. 72 (72). 1 3 LXX. Aquil. καρδίαν, Symm. 

Theod. ψνχην: so 2 Kings 6. 11, Ps. 68 (69). 21, Prov. 6. 21: 16. 
I (15. 32). The most important instance of the combination of 
the two words is in the phrase i$ ΰλης της καρδίας σου και εξ όλης της 


ψνχης σον : Deut. 4. 29 : ΙΟ. 12 : ii. 13 : 13. 3 : 26. 16 : 30. 2, 
6, 10, Jos. 22. 5 [Cod. B.], 2 Chron. 15. 12. The variations of 
this phrase are significant: (λ) Deut. 6. 5, Jos. 22. 5 [Cod. Α.] 
substitute διανοίας for Kapbias : (δ) I Sam. 12. 24, I Kings 2. 4 omit 
the mention of ψνχη and substitute iv άληθεία, the force of the 
phrase being shown in Jer. 3. 10 by a contrast with its opposite, 

ovK επεστράφη προς με .... e^ όλης της καρδίας αντης αλλ' em ψεύδει : 
SO Jer. 39 (S^)• 4^ ^^ ττίστει κα\ iv πάσυ καρδία μου και iv πάστ) ψυχή. 

(3) 'ΐΓΐ'€υμα and ψυχή: (α) of the principle of life, Gen. i. 30 
ψυχήμ ζωής, ih. 6. 1 7 TTfeu^a ζ<ύΎ\ς (Π*•*!!! Π^"ΐ)^ and Ezek. i. 20, 21 : 
10. 17 (<^^DD Ιϋ^"^) : (3) of fainting, i.e. the apparent suspension 

of life, Ps. 106 (107). 5 ή ψυχη όχτων iv αντοΐς i^eXmev, ib. 142 (143). 
7 i^eXiTre το τΐνεΰμά μον: (c) of dying. Gen. 35• 18 iv τω άφύναι 
αντην τη μ ψυχή κ, Ι Kings 1 7. 21 επιστραφητω δη ή ψυχή του παιδαρίον 
τούτον eh αυτόν, Is. 53* ^^ παρεδόθη ets θάνατον ή ψυχή αυτόν, Thren. 
2. 12 eV τω iKxua0aL ψυχά,9 αυτών, Ps. ΙΟ3 (104)• 29 άνται^€λ€Τ5 το 
-π-μευμα αυτών κα\ 6κλ6ίλ//•ουσι, ζ'^. 145 (^46)• 4 «^^λίύσεται το πμευμα 
αντοΟ, Eccles. 12. 7 '"Ο 'Π'ΐ'ευμα επιστρεψη προς τον θεον ος εδωκεν 

In only one instance are the words interchanged between the 
LXX. and the Hexapla, Ps. 142 (143). 4 LXX. πνεύμα, Aquil. 

The elements of the two words are sometimes combined in a 
single phrase: Judges 15. 19 (Co^. A.) iπeστρεψε to ττμεΟμα avTov 
καί ανεψυξεκ, Ps. 76 (77)• 4 ώλιγοψύχησ€ το ττμευμα αυτού, Jer. 2. 24 
iv iπLθυμίaις ψυχή§ αύτοΟ επί'βυματοφορ^Γτο, Ezek. 21. 7 εκψύξει πάσα 
σάρ^ κα\ παν πι^ευμα. 

Cf. Ι Sam. 16. 23 ^Τ}, LXX. άνεψνχε, Aquil. άνεπνεε. 

(4) κφδία and διάμοια: {α) they are sometimes interchanged, 
Ex. 25. 2 ois av δό^χι Ti(] καρδία αυτον•=.ίδ. 35. 2 2 ω εδο^ε τή" διαμοία : 
ib. 28. 3• 35• 9• 3^• ^ τ^ασι rols σοφοΐς τγ^ διαΐΌία = ζ*3. 3Ι• 6 παντί 
συνετω καρδία: SO in Deut. 6. 5 - 28. 47) J^S. 22. 5, Prov. 27. 19 
the MSS. vary between καρδία and διάνοια: {b) they are sometimes 
combined, Gen. 6. 5 πας τις διαιρείται iv τρ καρδία αυτοΰ, Ι Chron, 
29• 18 φυλαζον ταΰτα εν διαι^οία καρδίας. 

3- Predicates of the several words. 

(i) Strong emotion is expressed by ταράσσπν with each 
of the three v^ords : 

IN THE LXX. 105 

(i) J^^ 36. 34 (37. l) ^ταράχθη ή καρδία μον : SO Ps. 37 (38). ΙΟ : 

54 (55)• 3 : i42 (143)• 4> Thren. 2. 11. 

(2) I Things 20 (21). 5 Ti TO -πνίυμά σου τΐταρα-γμίνον ] SO Is. 

19• 3• 

(3) Gen, 41.8 εταράχθη η ψνχη αντοΰ (where, as noted above, the 
Hebrew word is not ^ξ^. but πη) : so also Ps. 6. 4 : 41 (42). 7. 

(ii) Pride is expressed by νψονν, νψηλ05, with each of 
the three words : 

(1) Deui. 17. 20 iva μη νψωθΐι ή mpbia αυτόν : SO 2 Chron. 32. 25, 

Ps. 130 (131). I, Jer. 31 (48). 29, Ezek. 28. 2, 5, 17: so also Is. 

9. 9 €φ' ΰβρξΐ κα\ υψηΧτ} καρδία. 

(2) Eccles. 7. 8 ifnep υψηλορ πν^υματι. 

(3) Λ. 130 (l3^)• 2 ^^ /^^ (ταπΐΐνοφρόνουν αλλά ΰ^|rωσa την ψυχην 

(iii) Humility^ with ταπ€ΐΓ09 and cognate words : 

(i) καρδία : 

Ps. 108 (109). 16 ανθρωπον π€νητα κα\ πτωχον κα\ κατανεννγμζνον ττ} 

(2) •πνευμ.α: 

Ps. 33 (34)• 19 '''^^^ ταπ€ΐνους τω πνίύματι. 

(3) ψυχή : 

/s. 58• 3 ^νατΓβΐί/ώσα/χβι/ τα? ψυχας ημών. 

(iv) Dejection is expressed by άκτ^διαι; with each of the 
three words : 

(1) Ps. 60 (61). 3 eV τω α/ίτ^δίάσαι ττ)»/ καρδίαν μου. 

(2) /'ί. 142 (143)• 4 ηκ^δίασ€Ρ evr' e/xe το πνΐΰμά μου, Is, 61. 3 
πνεύμα άκηδίας. 

(3) Λ. Ιΐ8 (119)• 28 ίνυσταξεν ή ψυχή μου υπο άκηδίας. 

(ν) Contrition and distress are expressed by σνντρίβ6σθαί 
and cognate words with each of the three words : 

(1) I Sam. I. 8 ίνατί τύπτει ae η καρδία σου; Ps. 50 (51)• ^^ καρδίαν 
συντΐτριμμίνην κα\ τβταπ€ΐνωμ€νην, ib. 146 (147)• 3> ^'^' 57• ^3j J^^• 

23. 9• 

(2) Λ. 5θ (51)• 1 9 'Ti/eC/ta συντβτριμμζνον, Is. 65. 1 4 "^^ συντριβής 
πνεύματος υμών. 


(3) Gen. 43. 21 τψ βλίψίν της ψνχης αντον. 

(vi) Sorrow and anguish are expressed by each of the 
three words : 

(1) Deut, 15. 10 oh λνπηβηση rfj καρδία σον, Is. 65. 14 δια τον πόνορ 
της καρδίας νμων. 

(2) Ps. 76 (77)• 4 ^λιγοψνχησξ το πν^νμά μου: ih, 105 (106). 33 
παρεπίκραναν το πνεΐιμα αυτοΰ. 

(3) Ι Sam. Ι. ΙΟ κατώδυρος ψνχη : SO ζ<5. 2 2. 2 : 30• 6, 2 Kings 4• 

27: Is. 38. 15 '''ψ οδννην της ψνχης : 2 Sam. 1 7. 8 κατάπικροι τβ 
Ψ^Χη- Job 7• II : ΙΟ• Ι• 21. 25 πίκ/)ία λ//•υ;(^ί: Job 14. 22 17 δε 
ψνχη αντον ^πίνθησίν. 

(νϋ) The predicates which are found with καρδία and 
ψνχτι, but not with ττνζνμα, are those oi fear and cowardice. 

{a) With τηκζσθαι : 

(1) 2 Sam. 17. 10 ^ καρδία καθώς ή καρδία τον λέοντος τηκομενη τακη- 
σ(ται: Ps. 21 (22). 1 5 €ν€νηθη ή καρδία μον ωσβΐ κηρος τηκόμ^νος. 

(2) DeuL 28. 65 δώσω σοι . . . τηκομίνην ψνχην : SO Ps. Ι06 (107)• 

(<^) With φό/ίίοί, φοβζίσθαι. 

(ΐ) Ζ^^Ζ^/. 20. 8 ό φοβούμενος και δειλός τ^ καρδία: id. 28. 67, JoS. 

7. 15) 2 Chron. 13. 7j Ps. 26 {27). 3, i Sam. 28. 5 εψοβηθη κα\ i^e- 

στη ή καρδία αντον σφόδρα. 

(2) Is. 21. 4 V ψ^Χ^ ^ο^ ίφεστηκεν eU φόβον. 

(viii) 0/ affection with ayanav and cognate phrases : 

(1) Judges 16. 15 ηγάπηκά σε κα\ η καρδία σον ονκ εστί μετ εμον: 
2 Sam. 14. Ι ^ καρδία τον βασιλέως επ\ ^Αβεσσαλώμ: id. Ι ζ. 1 3 «V^- 
νηθη η καρδία ανδρών Ίσραηλ οπίσω ^Αβεσσαλώμ. 

(2) Ι Safn. Ιο. Ι, 3 ηγάπησεν αντον Ίωνάθαν κατά την ψνχην αντον. 
Cant. 3• 1} 2, 3) 4 ^^ ηγάπησεν η ψνχη μον. 

(ix) Of gladness with ayaOvv^iv^ αγαλλιασθαί^ and cognate 
words : 

(l) fudges 16. 25 OTi η-γαβύνθη η καρδία αντων: id. 1 8. 20, I KingS 

8. 66, I Chron. 16. 10, Is. 66. 14, Zach. 10. 7, Ps. 12 (13). 6 άγαλ- 

λιάσεται ή καρδία μον : id. Ι18 (119). Ill άγαλλίαμα της καρδίας μον: 
id. 85 (86). II ενφρανθητω ή καρδία μον. 

IN THE LXX. 107 

(2) Ps. 34 (35)• 9 >? δε '^νχη μου άγαλλιάσεται eVi τώ κυρίω'. SO Is. 
61. ΙΟ, Prov. 23. 24 fVi δε νιω σοφω ευφραίνεται η ψνχη αυτοΰ. 

(χ) 0/ hope, with ΙΚ-ηίζ'ξ.ιν : 

(ι) Ps. 27 (28). 7 f^' α^τω ήλπισεν ή καρδία μου. 

(2) jPj. 129 (13°)• ^ ήληισερ η ψυχή μου eVi τον κυριον. 

(χί) Those which apply to the moral nature as a whole : 

(1) Deuf. 9. 5 δια τήι/ όσιότψα της καρδίας σου, Ι Kings 9• 4 ^'^ 
όσιότητι καρδίας, Ρίον. 2 2. II αγάπα κύριος όσιας καρδίας, Neh. 2. 2 
πονηρία καρδίας. 

(2) Prov. 26. 25 €ττά γά/[) €ΐσί πονηρίαι iv τη ψυχτ} αυτού, Is. Ι. 1 6 
αφελετβ ras πονηρίας άπο των ψυχών υμών. 

(χϋ) Wt/l and intention are expressed by (i) Kapbia, 
(2) ττν€νμα, especially by KapbCa : 

(1) In the phrase πάντα τά iv τη καρδία (τίνος) ποιάν, Ι Sam. 9• 1 9» 
2 Sam. 7• 3> 2 Kings ίο. 20 : the more complete phrase πάντα τα 

iv Tjj καρδία μου καΐ τα iv τη ψυχή μου ποιήσει is probably equivalent 

to ' all that I intend and that I desire.' So in the phrases βεβάρηται 

ή καρδία Φαραώ του μη . . . Εχ. 7• 14? iσκkηpυvθη η καρδία αυτού Εχ. 8. 
19, and frequently in Exodus, άπεστησαν την καρδίαν . . . όπως μη 

είσελθωσιν Num. 32. g, Deut. I. 28: and in the phrases iyeveTo iπϊ 

της καρδίας . . . οίκοδομησαι I Kings 8. 1 7? iyiveTO eVi καρδίαν οίκοδο- 
μησαι I Chron. 28. 2, 2 Chron. 6. 7? 8 : so also τά αρεστά της καρδίας 

Jer. 9• 13 • ι6. ιι : ι8. ΐ2. 

(2) Oeut. 2. 30 iσκ\ηpυvεv . . . το πνεύμα αυτοΰ: 2 Chron. 36. 2 2, 
2 Esdr. I. I iζηyειpε Κύριος το πνεύμα Κύρου βασιλέως ΤΙερσών κα\ 
παρήγγειλε κηρνζαι. 

(χίϋ) Desire is expressed, perhaps exclusively, by ψ"^χ?7 : 

{a) Of food, Deut. 12. 21 φαγη iv ταΐς πόλεσί σου κατά την iπιθυμίav 
της ψυχής σου: SO ιδ. 14. 20, Ι Sam. 2. 16 : 20. 4> 2 Sam. 3• 2Ι, 
Ι Kings II. 37? Job 33• 20, Ps. 68 (69). n : 106 (107). 18, Prov. 
6. 30: 10. 3: 13. 25: 19. 15: 25. 25, Is. 32. 6: 58. II, Jer. 38 

(31). 25 : so εταπείνουν iv νηστεία την ψυχην μου Ps. 34 (35)• ^3> "^^^ 
αίτήσαι βρώματα ταΐς ψυχαΊς αυτών Ps. 77 (7^)• ^^> V ^^ ψ^Χ^ υμών 
προσώχθισεν iv τω αρτω Num. 2 1 . 5• 


(ί>) Of spiritual desire, Ps. 41 (42). 2 (πιποθβΐ η ψυχή μον προς σ/, 
6 eeos: ib. 62 (63). 2 : 83 (84). 3= 1 18 (up)• 20. 

(xiv) Mental powers and operations are predicated of 
all three words : 

(1) Of καρδία: (βπιστημη), Ex. 36. 2 ω debs εδωκ?»/ ΐπιστημψ iv rrf 
καρ8ία: (^ddevai) Deut. 29. 4 θ(6ς εΒωκεν νμ'ιν καρ8ίαν elbevai και οφθαλ- 
μονδ βλtπeιv και ωτα άκον€ΐν: Ι Kings 2. 44 ''"'7^ κακίαν σον ου oldev ή 
καρδία σου: (voeii/, 8ιανο€'ίσθαι) Ι Sam. 4-20 ουκ (νοησεν η καρδία αυτής'. 
Is. 32. 6 η καρδία αυτού μάταια νοήσει, Jer. 7.31' 19-5^•••°^ ^^^" 
νοηθην iv τγι καρδία μον : cf. Hos. 7. II ως περιστερά ανους ουκ έχουσα 
καρδίαν [φρόνιμος, φρόνησις : σοφός, σοφία): 1 Kings 3• 12 δεδωκά σοί 
καρδίαν φρονίμψ κα\ σοφην : ib. ΙΟ. 24 ΤΎ]ς φρονησεως αυτού ης έδωκε 
κύριος ττ] καρδία αυτού : 2 Chron. 9-23 t^s σοφίας αυτοΰ ης εδωκεν 6 
θεός εν καρδία αυτοϋ '. Job 1 7• 4 καρδίαν αυτών εκρυψας άπο φρονησεως : 
{σννιεναι, συνετός) Job 34• ΙΟ? 34 ο^ν^τοί καρδίας [Cod. Α, καρδία^: Is. 
6. ΙΟ /Α^ ποτέ . . . TTJ καρδία σννώσι'. [βονλεύεσβαι) Neh. 5• 7 ^βονλεν- 
σατο καρδία μον επ* ε με. 

(2) Of TTi'eGjxa : Ex. 28. 3 "η^νευμα σοφίας κα\ αίσθησεως'. DeUt. 
34• 9> Jot) 15• 2 πνεύμα συνέσεως: Ι Chron. 28. 12 το παράδειγμα ο 
ειχεν εν πνενματι αυτού : Ps. 76 (77)• 7 εσκαλλον το πνεϋμά μου. 

(3) Of ψυχή : Jos. 23. 14 γνωσεσθε τη καρδία υμών κα\ τη ψυχή 
υμών: Ps. 12 (13)• 2 εως τίνος βησομαι βούλας εν ψυχή μον: Ps. 1 38 
(139)• 14 V ψ^χί fov γινώσκει σφόδρα: Prov. 24. 14 αίσθηση σοφίαν 
τη ση ψνχη: Cant. 6. II ονκ εγνω η ψυχή μου: Is. 44• 1 9 °^'^ ελσγί- 
σατο τη ψνχη αυτοϋ. 

If we gather together the results, it will be seen that in 
the LXX. 

(i) Kaphia, Έν^νμα, ψνχη are capable of being inter- 
changed as translations of the same Hebrew 
words : 

(2) consequently, the lines of distinction between them, 

whatever they may be, are not sharply drawn : 

(3) a survey of the predicates which are attached to 

each of them shows a similar impossibility of 
limiting them to special groups of mental 
phenomena, with the exceptions that {a) Kapbia 

IN PHILO. 109 

is most commonly used of will and intention, 

(d) ψνχη of appetite and desire. 
But this general inference as to Greek words does not 
of necessity apply also to their Hebrew originals. A 
student of the Hebrew terms must no doubt take into 
account the fact that at a certain time those terms con- 
veyed to Greek minds a certain meaning, and that a 
certain group of them was to some extent treated as 
synonymous. But this fact is only one of many data for 
the determination of the meaning of the Hebrew terms 
themselves : and it must be carefully borne in mind that 
the study of the words by which Greek translators ex- 
pressed Hebrew psychological terms is not identical with 
the study of Hebrew psychology. 

II. Psychological terms in Philo. 

The use of psychological terms, such as ιτν^υμα and ψνχη, 
in Philo can only be understood when viewed in relation 
to his psychology as a whole. But that psychology is 
of great complexity. The complexity arises partly from 
the fact that he uses the same terms to designate different 
groups of phenomena, partly from the fact that he uses 
different terms to designate the same phenomena, and 
partly from the fact that he regards the phenomena from 
different points of view, sometimes using the terms or 
conceptions of one system of philosophy and sometimes 
those of another, and sometimes borrowing both terms 
and conceptions not from philosophy but from the Old 
Testament. There is in some cases the additional element 
of uncertainty which arises from the uncertain authorship 
of some of the writings which are attributed to him. 

It would be beyond my present purpose to discuss that 
psychology in detail, or to endeavour to resolve it into 
the elements from which it was formed. I must be con- 
tent to gather together the more important of the predicates 


which he attaches to the chief psychological terms, and 
to add to them only such brief explanations as may be 
necessary to develop their meaning. 

I. σώμα and ψυχή. 

The conception of the duality of human nature runs 
through all Philo's writings. (i) We are compounded 
of two elements, body and soul, which are {2>) allied during 
life, but (3) separated at death. 

(1) ^eg, Alleg. iii. 55 (i. 119) δύο Ιστϊν i^ hv συνίσταμ^ν "^νχη tc 
και σώμα. 

De Ebriet. 26 (i. 372) (ανθρωπον) τό ψνχης κα\ σώματος ύφασμα ή 
ΤΐΚί-γμα η κράμα η 6 τι ποτ ξ χ^ρή καλ^ΐι/ τοντί το συνθ^τον ζωον. 

Oe Cherubim 32 (i. 159) '^y^^y ovu €Κ ψνχψ κα\ σώματος σνν^στώς. 
De Mundi Opif. 46 (i. 32) Ικ σώματος κα\ ψνχης σννίστώς, 

(2) Quod dei. pot. insid. 6 (i. 194) σνζν^η hk και συνβταιρϊς καΧί'ΐται 
Χββρών, συμβολικούς ημών το σώμα οτι συνίζ^υκται και ωσπ€ρ ΐταιρίαν κα\ 
φϊλίαρ προς ψνχην τίθενται. 

(3) Leg. Alleg, i. 33 (^• ^5) ° ^-^^ °^^ άνθρωπου {sc. θάνατος) χωρισ- 
μός ίστι "^υχης κα\ σώματος. 

II. σώμα, σάρ|. 

If we gather together the predicates of σώμα, we find that 
the word is sometimes used in a narrower, sometimes in a 
wider sense. 

i. The body in its strict sense is (i) a compound of earth 
and other elements : (2) it is the passive receptacle of soul, 
its dwelling-place, its temple, its prison, its tomb : (3) it is 
dead, and we carry about, as it were, a corpse with us. 

(l) Leg. Alleg. iii. 55 (i• 1^9) ''^ F^'' ^^^ σώμα Ικ γ\ς δ^δημιούργηται. 
Lbld. το μ\ν €Κ "γης διαπλασθίν σώμα. 

De Migrat. Abraham, ι (i. 436) τ-ό μίν σώμα κα\ sk γης cXajSe την 
σύστασιν καϊ άνοΚν^ται παΚιν ei? γην. 

De Sacrificant. 2 (ii. 252) ^στιν οΰν ημών ή κατά το σώμα ουσία ή γη 

κα\ νδωρ : (and earth and water are conceived as saying to men) 

Tjpeh eV/xcV η του σώματος υμών ουσία' ημάς ή φύσις κ^ρασαμίνη, η θ(Ία 
τβχνη, δκ'πλασίν €Ϊς άνθρωπόμορφον ideav. 


De Mundi Opt/. 51 (i. 35). (In respect of his body man is akin 

to the whole visible world) συ-γκ^κριται γαρ ck των αυτών, γης καΐ ν8ατος 
κα\ aepos κα\ πυρός, εκάστου των στοιχείων ΐΐσ^ν^γκόντος το (πιβάλΧον 
μίρος προς βκπλήρωσιν αυταρκίστάτης ΰλης, ην ίδίΐ λαβείν τον δημιονργόν 
ινα τ€χνιτ€ύσ~ηται την όρατην ταύτην ίΐκόνα. 

(2) De Somniis i. 5 (i• 624) αλλά και on ψυχής εστίν άγγεϊον (sc. το 
σώμα^ ουκ άγνοοϋμεν. 

/did. i. 20 (i. 639) '''ον συμφυα της ψυχής οίκον, το σώμα. 

De Migrat. Abraham. 5 (i• 439) "^ov σωματικον οίκον: ibid. 2 (i. 
438) €κφυγών δίσμωτήριον, το σώμα. 

Quod Deus immut. 33 (i. 295) 6 τής ψυχής οίκος η τύμβος ή ότιουν 
χρή KoXelv. 

De Mundi Oplf. 47 (i. 33) οίκος γάρ τις ή νεώς ΐ€ρος €Τ€κταίν€το 
ψνχής λογικής ην epeWev άγάΚματοφορήσειν αγαλμάτων το Οεοίώίστατον. 

Quis rer. divin. heres 14 (i. 482) 6 μ^νων iv τί} σώματος ύρκτή 

De agricuU. 5 (i. 304) τόΐ' σύνθετον χουν, τον πεπλασμίνον ανδριάντα, 
τον ψνχής ΐγγιστα οίκον, ον άπο γίνεσΐως άχρι τΐλΐυτής, άχθος τοσούτον, 
ουκ άποτίθβται ν€κροψορουσα. 

Leg. Alleg. iii. 22 (i. 100, ιοί) μή γαρ άλλο τι ποιήσεις εκαστον 
ημών ποιύν η νίκροφορύν, το vcKpov i^ εαυτού σώμα Ιγειρούσης και άμοχθϊ 
φερουσης τής ψνχής Ι ibid, του νεκρού οντος σώματος άλογεϊ. 

De Gigant. 3 (i. 264) τον συμφυα νεκρον ημών, το σώμα, 

ϋ. The term body is sometimes used in an extended 
sense: (i) it includes the senses and desires: (2) the pas- 
sions grow out of it : (3) hence it is regarded as evil, the 
seat of the vices, and the enemy of the higher life. 

(1) Leg. Alleg. i. 32 (i. 64) αίσθήσεσι σώματος. 

Quod del, pot. insid. 29 (i. 212) το γεώδες σώμα κα\ τας συγγενείς 

Leg. Alleg. i. 32 (i. 64) το σώμα και τας επιθυμίας αυτού. 

(2) Quis rerum divin. heres 54 (i. 511) νόθα γαρ κα\ ξένα διανοίας 
τα σώματος ως αληθώς πάθη, σαρκός εκπεφυκότα, η προσερρίζωνται. 

De Somniis ii. 39 (ί• 692) το ήμετερον σώμα και τα εν αυτω κα\ δι 
αυτο εγγινόμενα πάθη. 

(3) Leg. Alleg. iii. 22 (i. 100) τον γαρ δερμάτινον ογκον ημών το 
σώμα .... πονηρόν τε κα\ επίβουλον τής ψνχής, ουκ άγνοεΐ, καΐ νεκρον κα\ 
τεθνηκος αεί. 


Leg. Alleg. i. 32 (i. 64) τό hi σώμα ουκ. οίον ου avvepyei προς τούτο 
(sc. the attainment of virtue) αλλά καΐ KoXvaupyei. 

De SomnilS ii. 39 (i. 693) τά? σώματος και δια σώματος κακίας. 

In this extended sense the terms ' flesh ' (σαρξ) and 
' sense ' {αΧσθησι^) are sometimes substituted for body, 
and in addition to the constant antithesis between body 
and soul [σώμα and ψυχτ?) as different physical elements, 
an antithesis is sometimes made not only (i) between the 
same terms, but also between (2) flesh and soul [σαρξ and 
ψνχη), (3) flesh and the divine spirit {σαρξ and rd Θύον 
Ίτνζνμα), as representing different elements of consciousness 
and different aims of human action. 

( 1 ) Quod Deus immut. 11 (i. 281) τών yap ανθρώπων ol pev ψυχής 
oi de σώματος Ύ€γόνασι φίλοι. 

De SomnilS ii. 39 (i. 692) 6 σπονδαίος κληρον έλαχε ψνχψ κα\ τας 
ψνχης άρ€τάς, ώσπβρ ό φαύλος βμπάλιν σώμα και τας σώματος κα\ δια 
σώματος κακίας. 

De Abraham. 41 (ϋ• 34) ο* 'ψ'^Χ.7 φαλλοί.• r\ σώματι ζώντ€ς. 

(2) De Gigantihus 10 (i. 268) άντίθ^ς yap, φησίν, ώ yevvaUy το 
σαρκός αγαθόν τω της ψνχης κα\ τω του τταντος άγαθω' ουκοΰν το piv 
σαρκός βστιν άλογος ηδονή, το de ψυχής κα\ του παντός 6 νους τών όλων, θεός. 

(3) De Gigantihus 7 (i. 266) αίτιον he της αν επιστημοσύνης μεγιστον 
ή σαρζ και ή προς σάρκα οίκΐίωσις' κα\ αυτός he όμολογεϊ φάσκων hia το 
eivai αυτούς σάρκας μη δυνασθαι το θείον πνεύμα καταμεΐναι, 

Quis rer. divin. heres 12 (i. 481) ώστε hiTTov είναι γένος ανθρώπων 
το μεν θείω πνευματι και λογισμω βιούντων το hi αίματι και σαρκός rfhovrj 

III. ψυχή. 
i. The term ψνχη is used sometimes, though rarely, (i) in 
a very wide sense, to designate all life whether conscious 
or unconscious, (2) in a special sense, to designate the 
highest form of mind, that is, the intuitive reason as dis- 
tinguished from apprehension by the senses. 

(i) De Mundi Opif. 22 (i. 15) Nature fashions την μεν iypav 
ονσίαν (i.e. the element water, of. infra c. 45, i. 31) εΙς τα του σώ- 
ματος μέλη κα\ μέρη hιaveμoυσa την he πνευματικήν (ί. e. the element air) 
εΙς τάς TtjS ψ^χή? δυνάμεις, την τε θρεπτικην και την αισθητικήν. Eut 

IN PHILO. 113 

elsewhere he distinguishes between e^ts the power of cohesion 
which holds material bodies together, φνσις the power of growth, 
ψνχη animal life, λογική ψυχή rational life : Quod Deus immut. 7 (i. 
277) των yap σωμάτων τα μ^ν evebvaaro e^ei, τα Se φύσει, τα. be ψυχί), τα 
δε κα\ λογική ψυχτ} : De Somniis i. 2 2 (i. 64 1 ) eVoiet yap αυτόν 6 
τ€χνίτης ακίνητων μεν σωμάτων e^iv κινουμένων be άφαντάστως (i. e. with- 
out power of perception) φύσιν, fjbq bi όρμ?} κα\ φαντασία χρησθαι buva- 
μ€νων ψνχην. 

(2) Quis rer. divin. heres 22 (i. 487) αϊσθησις, which is usually 
included in ψνχη, is made coordinate with it, thus limiting ψνχη to 
reason as distinguished from sensation: so De gigant. 3 (i. 264) 

ψνχην η νουν' το κράτιστον των iv ημίν. 

But in its ordinary use ψυχ?}, though limited to conscious 
life, is made to cover all the phenomena of conscious life, 
sensations, emotions, and thoughts. These phenomena 
are commonly grouped into the two divisions which, in 
the language of the Peripatetics, he calls the irrational 
and rational parts of the soul, or, in language which is 
probably that of the Stoics, sense and mind. Hence ψυχή 
is said to have two meanings, or to be divided into two 

Quis rer. dlVin. heres 1 1 (i. 480) ψνχ)] bix&g λέγεται, η re δλ»; και 
το ηγ^μονικον αντης pepos ο, κνρίως elneh, ψνχη ψνχψ εστί. 

De Migrat. Abraham, ι (i. 43^) "ΐσ^ί^σί? δε σνγγ^νίς και άbiλφόv 
εστί biavoias, αλογον λογικής, ε'ττειδ^ μιας αμφ'ω μίρη ψ^χψ ταντα. 

De Agricult. 7 (i• 3^4) "^Ά^ ψ^ΧΨ ώσπερ άπο μιας ρίζης ερνη biTTa 
άναβλαστοΰσης hv το μεν ατμητον όλον bi όλων iaOev επεφημίσθη vovs, το 
δ' ίζαχη σχισθεν (Is επτά φύσεις πέντε των αισθήσεων κα\ bvolv άλλων 
οργάνων φωνητηρίον τε κα\ γονίμον. 

In some passages Philo substitutes the threefold division 
of Plato for this Aristotelian dichotomy : 

JLeg. Alleg. i. 22 (i. 57) νοητεον ovv OTi εστίν ημών η ψνχη τρίμερης 
και έχει μέρος το μεν λογικον το δε θυμικον το δε επιθυμητικόν. 

Ibid. iii. 38 (i. no) τρίμερη σνμβεβηκε την ψνχψ ημών είναι καΐ εχειν 
μέρος μεν εν λογιστικον bεvτεpov δε θνμικον τρίτον δε επιθυμητικόν. 

De con/us. ling. 7 (i. 408) τρίμερους ημών της ψυχής νπαρχουσης το 
μεν νους και λόγος το be θυμός το δε επιθνμία κεκληρωσθαι λέγεται, 



Quis rer. divin. heres 45 (i. 504) ^υχΐ7 γάρ τριμ€ρψ ian ΒΙχα de 
^καστον των μζρών ως ΐδ^ίχθη (sc. anfe, C. 26, i. 49 1) τβμνεταΐ' μοιρών 
Βη γίνομ^νων ίζ (βδομος (Ικότως τομ^ύς ην απάντων, 6 Upos και Oelos 

In other passages he adopts in whole or in part the 
Stoical division into sense (or the five senses enumerated 
separately), speech, the reproductive faculty, and the 
governing faculty : in some of these passages he combines 
the Stoical and the Aristotelian divisions : in others, 
though he preserves the coordination of speech with sense 
and reason, he omits the reproductive faculty. 

De mundi Opific. 40 (i. 28) r^s ημ^τίρα^ ψνχης το Βίχα τον ηγ€μονικον 
μίρος €πταχη σχίζεται^ προς π€ντ€ αισθήσεις και το φωνητηριον opyavov κα\ 
eVi πασι το γόνιμον. 

Leg. Alleg. i. 13 (i. 51) τοντω (sc. τω νω) μόνω ^μπνΰ 6 Bebs τοΊς de 
αλΧοις μίρίσι ονκ άξιου ταΐς Τ€ αίσθησ^σι κα\ τω λόγω και τω γονίμω '. 

(but immediately afterwards all these are grouped together as τ6 

αλογον μίρος της ψνχης^ 

Quis rer. div. heres 48 (i. 505) το μίν yap oKoyov ψνχης μ^ρος ^ξαχη 
8ί€λων 6 δημιουργός €$ μοίρας elpyaaaTO, δρασιν, yevaiv, άκοήν, οσφρησιν, 
άφην, γόνιμον, φωνην' το de Χογικόν, ο drj νους ωνομάσθη ασχιστον etaae 
κατά την του παντός ομοιότητα ουρανού. 

Ibid. 22 (i. 4^7) T^o-poi^^TeOeTo he σοι αυτω ψνχήν, \oyov, αισθησιν 6 

De congr. erud. grat. 18 (i. 533) ev ήμΐν yap αντοΐς τρία μέτρα eivai 
doKcif αϊσθησις, Χόγος, νους. 

Oe Somniis i. 5 (i• 624) ουκουν τ€τταρα τα άνωτάτω των nepi ημάς 
€στι, σώμα, α'ίσθησις, λόγος, νους. 

But neither the Platonic nor the Stoical psychology 
penetrates his system, or forms to any appreciable extent 
the basis of other parts of his teaching : he adheres in 
the main, with w^hatever inconsistencies, to the division 
of the phenomena of consciousness into rational and ir- 
rational, or mind and sense. 

ii. To each of these parts of ψνχη he assigns (i) a 
different essence, the one blood, the other spirit : (:z) a 
different origin, which is expressed in theological language 



in the assertions that the one is of the earth, and the 
other breathed into man by God, or that the one was made 
by God's ministers and the other by God himself: (3) a 
different destiny, the one being mortal, the other immortal. 

(i) Quis rer. divin. heres 11 (i. 481) 'ibo^e τω νομοθίττ) 8ιπλην elvai 
κα\ την ονσίαν της ψνχης, αίμα μ^ν το της όλης του δε ήγ€μορικωτάτον 
πνίΰμα θύον. 

Quod Deus immut. 10 (i. 279) τοντο της ψνχης το €Ϊ8ος [SC. ό νους] 
ουκ (Κ των αυτών στοιχήων βξ S)v τα αλΧα άπετελείτο 8ί€πλάσθη, καθαρω- 
τ€ρας δε κα\ άμύνονος ΤΚαχ€ της ουσίας. 

De Concupiscent. 10 (ii. 35^) "^^ Ζ^^" ^ν^ο. .... ουσία ψνχης (στ\ν 
ουχί τηί νο€ρας καΐ λογίκης αλλά της αισθητικής .... εκείνης [sc. της 
νθ€ράς~] yap ουσία πν^ΰμα uelov. 

(2) I^eg. A //eg. i. 13 (i. 51) τών γάρ γινομένων τα μβν κα\ υπο θ€θϋ 
yiyovev καΐ δι αυτοΰ, τα δε υπό θΐου μ^ν ου, di αυτού δε" τα μ€ν άριστα 
καΐ υπο BeoO yeyove κα\ bi αυτού .... τούτων κα\ 6 νους ε'στί* το δε 
άλογοι/ υπο θεοϋ pev yeyovev ου δίά θ^ου δε, αλλά bia του XoyiKod του 
άρχοντος τε και βασιλίύοντος iv "ψ^υχη. 

De profugis 13 (i. 5 5 6) ^loKeyeTai μεν ουν [referring to the words 

ποιησωμξν ανθρωπον in Gen. i. 26] 6 των όλων πατήρ ταΐς δαύτου δυνά" 
μ€σιν αις το θνητον ημών της ψνχης μίρος έδωκε διαπλάττίΐν, μιμουμίναις 
την αυτού τίχνην, ηνίκα το λογικον iv ήμιν εμόρφου, δίκαιων υπο piv ηγε- 
μόνας το ήγεμονικον iv ψνχη, το δε υπηκοον προς υπηκόων δημιουργεϊσθαι. 

De Con/us. /ing. 35 0• 43 2) "^Ψ τούτου (sc. of the irrational part 
of the soul) ^εόί περιηψΐ κα\ τοις υπαρχοΊς αυτού λίγων * ποιησωμεν 
ανθρωπον. Ινα αι μεν του νου κατορθώσεις επ* αυτόν άναφίρωνται μόνον ε'ττ' 
άλλους δε αί άμαρτίαι. (He goes οη, as in the preceding passage 
and elsewhere, to account thus for the presence of evil and sin 
among men : God Himself is the direct author only of good). 

(3) Leg. A//eg. ii. 24 (i. 83) δύο γένη φορεί η ψυχή το μεν θείον το 
δε φθαρτόν. 

Quod Deus i?nmut. 10 (i. 279) μόνον των iv ημϊν αφθαρτον εδοξεν 
είναι την διάνοιαν. 

IV. T/ie lower manifestations of ψυχή. 

The lower or irrational part of ψυχτ], of which the essence 
is blood, consists of those phenomena of consciousness which 
are common to man with the brutes, and which may con- 

I ^ 


sequently be regarded as phenomena simply of physical 
life. It is admitted, in language which will be quoted 
below, that those phenomena as they actually occur in 
man are interpenetrated with mind, and could not be 
what they are without mind. At the same time a real 
as well as a logical distinction is drawn between the 
functions and phenomena of sense and those of mind. 

i. The senses have, as mere functions of the animal life, 
(i) a certain dull power of feeling, i.e. of acquiring know- 
ledge of external things: (2) their precise function is to 
present to the mind images of present objects. (3) To 
such objects they are limited : for they neither remember 
the past nor anticipate the future. (4) They are cognizant 
of the presence of objects, but cannot form judgments upon 
them : in Philo's phraseology they know σώματα but not 
ττράγματα. (5) They are so far independent of mind that 
if the mind were to tell them not to act, they would refuse 
to obey. 

(i) InDe congr. at. 25 (i. 539, 540) he uses the difference 
between the senses in themselves, and the senses acting con- 
currently with mind, as an illustration of the difference between 
arts and sciences : of which he says that the former άμυ^ρώς όρώσιν, 

the latter τηλαυγως και σφό8ρα εναργως κατάλαμβάνονσιν. 

ωσπβρ yap οφθαλμοί μεν δρωσιν, δ δε vovs di οφθαλμών τηλανγίστ^ρον 
κα\ ακούει μεν ωτα, 6 be νους δι ωτων αμεινον και οσφραίνονται μεν οι μνκ- 
τηρες, η 8ε ψνχη δια ρινών εναργεστερον κα\ αί αλλαι αΙσθησεις των καβ* 
αυτάς αντιλαμβάνονται καθαρώτερον δε καΐ εΐΚικρινεστερον η διάνοια, κυρίως 
yap εΙπεΐν ηδ* εστ\ν δφθαΚμος μεν οφθαΧμών άκοη δ' άκοης κα\ εκάστης των 
αισθήσεων αισθησις ειλικρινέστερα, χρωμενη μεν εκείναις ως εν δικαστηρίω 
νπηρετισι δικάζουσα δε αύτη τας φύσεις των υποκείμενων ως τοϊς μεν συναι- 
νεΐν τα δε απόστρεφε σθαι, ούτως αι μεν \εyόμεvaι μεσαι τεχναι ταΐς κατά το 
σώμα δυνάμεσιν εοικυΐαι rots θεωρημασιν εvτυyχάvoυσι κατά τινας άπλας 
επιβοΧάς άκριβίστερον δε επιστημαι κα\ συν εξετάσει περιτττ}. 

De mundi Opif. 59 (i• 4^) . « • • τόΐ' νουν ω τά φανεντα έκτος ε'ίσω κομί' 
ζουσαι διayyελλoυσL κα\ επιδείκνυνται τους τεπους εκάστων, εvσφpayιζόμεvaι 
τ6 ομοιον πάθος, 

(2) De SomnilS i. 5 (i• 624) (at αισθήσεις) ayycXot διανοίας εϊσ\ν 

IN PHILO. 117 

^ιαγγίΧλονσαι χρώματα, σχήματα, φώνας, ατμών και χυλών Ιδιότητας, 
συνόλω? σώματα και οσαι ποιότητα kv τούτοις. 

Leg. Alleg. iii. 19 (ΐ. 99) ^^"^ Ί^Ρ V αίσθησις €πιβά\λονσα τω αίσθητω 
πΚηρωθίΙ της αύτοΰ φαντασίας €νθύς κα\ 6 νους σνμβ€βληκ€ και άντ€Χάβ€το 
και τρόπον τινά τροφής της απ' έκύνου π€π\ηρωται. 

(3) /δζ(ί. ϋ. 12 (i. 74) V οϊσθησις φύσ€ΐ νυν ζστΊ, κατά τον 
€ν€στώτα χρόνον υφισταμίνη μόνον, δ μ^ν yap νους τών τριών βφάπτβται 
χρόνων κα\ yap τα παρόντα voei κα\ τών παρ^Χηλυθότων μίμνψαι και τα 
μίΧλοντα προσδοκά' ή §€ α'ίσθησις οϋτ€ μελλόντων αντιλαμβάνεται ουδ' 
avaXoyov τι πάσχει προσδοκία η εΧπίδι οϋτ€ παρεΧηΧυθότων μίμνηται αλλ' 
νπό του ηδη κινοΰντος καΙ παρόντος μόνον πάσχειν πεφυκεν, οϊον οφθαΧμος 
Χευκαίνεται νυν υπο του παρόντος Χευκου υπυ δε του μη παρόντος ουδέν 

Ibid. iii. 16 (i. 97) ο^^ y^P ν ορασις οϋ& η ακοή ούτε τις τών αΧΧων 
αισθήσεων διδακτή, ώστε ου δύναται κατάΧηψιν πpayμάτωv ποιήσασθαι' 
μόνων yap σωμάτων διακριτικήν εlpyάσaτo αυτήν ό ipyaσάμεvoς Ι cf. infra 
C. 18. 

(4) /did. iii. 35 (i. 109) τυφΧον yap φύσει ή αισθησις άτε 
aXoyoς ούσα επει το XoyiKov εζομματοΰται' παρ ο και μόνω τούτω τα 
πpάyμaτa καταΧαμβάνομεν αίσθήσει δε ούκετι' μόνα yap τά σώματα φαντα- 
σιούμεθα δι αίσθήσεως. 

(5) Ibid. iii. 18 (i. 98) «αρ yoiv βουΧηθη 6 νους προστάζαι τη δράσει 
μή ιδεϊν, ονδεν ήττον αυτή το ύποκείμενον οψεται. 

' ϋ. On the other hand there is in sensation a mental 
element : the senses, even as powers of the physical 
organism, are set in motion by mind, and cannot act 
without it. 

Leg. Alleg. ii. 12 (i. 74) "^όντα yap οσα πάσχει ή α'ίσβησις ουκ άνευ 
νου υπομένει. 

Ibid. iii. 65 (i. 124) ^.πδ yap τούτου (sc. του νου) καθάπερ τινδς πηyης 
αί αίσθητικαί τείνονται δυνάμεις, μάΧιστα κατά τον ιερώτατον Μωϋσήν ος εκ 
του Άδά /i πεπΧάσθαι φησΧ την yυvaiκa, τήν αισθησιν εκ του νου. 

Ibid. C. 67 ^ρχή δε ήν αίσθήσεως δ νους. 

De poster it. Cain. 36 (i. 249) ή ουκ αν είποι τις τών αισθήσεων εκά- 
στην ώσπερ άπδ ττηγης του νου ποτίζεσθαι . . . . ; ούδε\ς y ούν εύφρονών 
είποι αν οφθαλμούς δράν αλλά νουν δι οφθαλμών οι»δ' ώτα άκούειν αλλά δ*' 
ωτων εκείνον ούδε μυκτηρας οσφραίνεσθαι αλλά δίά μυκτήρων το ήyεμovικόv. 

Leg. Alleg. i. 11 (i. 49) God 'rains' the objects of sense upon 


the senses, i. e. He causes images from those objects to fall upon 
the senses ; but there would be no use in His doing this, i. e. the 

senses would not act iav μη πηγψ τρόπον δ vovs reivas ίαυτον άχρι της 
α.1σθησ€(ύς κίνηση τε αυτήν ηρεμούσαν κα\ άναγάγη προς άντιΚηψ^ιν τον υπο- 

De profugis 32 (i. 573) ''^ ηγεμονικον ημών, εοικοί πηγτ]^ δυνάμεις 
πολλάί οία δια γης φλεβών άχρι τών αισθήσεων οργάνων άνομβρονν, τάς 
δυνάμεις ταύτας οφθάΚμών, ωτων, ρινών, τών αλΧων άποστεΧΚει. 

This relation of subordination between the physical 
and the mental elements is expressed by several meta- 
phors : the senses are described as marionettes moved by 
mind, as its messengers, its handmaidens, its helpmates, 
its satellites, the purveyors of its food : in one passage 
roCs is spoken of as being a God to the senses, as Moses 
was to Pharaoh. 

De mundi opif. 40 (i. 28) α δ^ πάντα (sc. the senses and speech) 

καθάπ(ρ εν τυΐς θανμασιν (i. e. in puppet-shows) νπό του ηγεμονικού 

νενροσπαστούμενα (i.e. worked by Strings, like puppets or marionettes) 

τότε μ€ν ηρεμεί τότε δε κινείται. 

Hid. 59 (i. 40) The senses offer their gifts to their master, reason, 

θεραπαινίδων τρόπον. 

Leg. Alleg. ii. 3 (i, 68) πώς ημών ό νους καταλαμβάνει οτι τουτί Χενκον 
η μίλαν εστίν εΐ μη βοηθώ χρησάμενος δράσει ) 

De plantat. Noe 32 (i. 349) τό τρεφον τον νουν ημών εστίν αϊσθησις. 

Quod det. pot. insid. 23 (i. 207) τάς δε νου δορυφόρους αισθήσεις. 

De Somniis i. 5 (ί• 624) «α* οτι άγγελοι διανοίας εΙσ\ν διαγγελλονσαι 
χρώματα .... και δτι δορυφόροι ψνχης είσιν οσα &ν ΐδωσι κα\ άκούσωσι 
δηλοϋσαι .... 

Leg. Alleg. i. 13 (i. 5^) <^σαν€Ϊ γαρ θεός εστί του αλόγου δ νους, παρ* 
t και Μωϋσην ουκ ωκνησεν ειπείν θ(ον του Φαραώ. 

But there is a metaphor sometimes used which seems 
to express more exactly than the preceding the relation 
in which the physical and mental elements stand to each 
other. It is that of a marriage : and it is interwoven with 
an allegorical interpretation of the history of Adam and 
Eve. Mind is represented as leaving its father, the God 

IN PHILO. 119 

of the Universe, and its mother, the virtue and wisdom 
of God, and, joining itself to the body, becomes one flesh 
with it. 

Leg. Alleg. ii. 14 (i. 75) «Ί'βκα r^? αΙσθησ€ω5 6 vovs όταν avrfj δουλωθη 
κατάΚ€ΐπ€ΐ και τον πατέρα, τον ο\ων θβόν, και την μητέρα των συνπάντων 
την άρίτην κα\ σοφίαν του θΐοΰ και προσκολλάται κα\ ivovTai τη αίσθηση 
και αναλύεται ety αισθησιν ινα γίνωνται μία σαρζ καΐ ev πάθος οι δυο. 

iii. In itself sensation, whether acting alone or with mind, 
is neither good nor bad. 

Leg. Alleg. iii. 21 (i. 100) Xexreoj/ ουν οτι η αίσθησις οΰτ€ των φαύλων 
ovTe των σπουδαίων eVrtj/ αλλά μίσον τι αυτή και κοινον σοφοΰ re κα\ 
αφρονοί και γ^νομβνη μ€ν iv αφρονι γίνεται φαύλη iv άστζίω be σπουδαία. 

But sensation gives not only knowledge but also pleasure 
and pain. Out of it the passions grow : the statement that 
the passions are rooted in the body and spring out of it 
(above p. iii) is modified into the statement that they 
are the products of irrational consciousness. 

Leg. Alleg. ii. 3 (i. 67) τ6 be oXoyov (SC. μ€ρος της ylrvxrjs) α'ίσθησίς 
€στι και τα ταύτης cKyova πάθη. 

Ibid. ρ. 68 μιας γάρ 4στι ψνχής μ€ρη και -γεννήματα η re αισθησις κα\ 
τα πάθη. 

Quod Deus immut. 11 (i. 28) τα ψυχής άλογα πάθη, 

Quis rer. divin. heres 13 (i. 482) ... . ^τίρου ψυχής τμήματος οπ€ρ 
aXoyov υπάρχον αίματι πΐφύραται, θυμούς ζίοντας κα\ π^πυρωμ^νας επιθυ- 
μίας ανάφλεγαν. 

Hence the sense, 'the more corporeal element of the soul' 
(to σωματο€ώ€στ€ρον ψνχτ]9 /xepoy, De congr. erud. grat. 5, i. 
^'Xl) may become the same as ' flesh,' σαρξ, [Leg. Alleg. ii. 14, 
i. 75), and is in one passage described by the phrase ' the 
soul of the flesh ' (σαρκόί ψνχή Quod det. pot. insid. 23, 
i. 207). 

Leg. Alleg. ii. 14. (i. 75) ^"^^^ y^P ''o κρείττον, 6 νους, ίνωθιι τω 
χΐίρονι, TTj αίσθησει, αναλύεται εΙς το χείρον το σαρκός γένος, την παθών 
αΐτίαν αισθησιν' όταν δε το χείρον, ή α'ίσθησις, άκολουθηστ] τω κρείττονι, τω 
νώ, ουκετι εσται σαρζ αλλά αμφότερα νους. 


The sense is not merely logically and physically distinct 
from mind but at constant variance with it. Sometimes the 
mind wins the battle, and then sense is merged in mind : 
more frequently the flesh proves the stronger, and mind 
is lost in sense. This latter contingency is sometimes 
described by the expressive phrase 'the death of the 
soul ' : for there are two kinds of death, he says, the death 
of a man, which is the separation of soul and body, and 
the death of the soul, which is the loss of virtue and the 
acquisition of vice. 

Leg. Alleg. ii. 14 (i. 75) ^''^^^ y^P ''o κράττον, 6 vovs, ίνωθη τω χύ- 
ρονι, rfj αίσθησ^ι, avaXverai els το γ/ιρον, το σαρκός yivos, την παβων 
αΐτίαν αισθησιν' όταν δε το χεΊρον, η αϊσθησις, άκοΚονθηστ] τω κρ(ίττονι, τω 
νω, ονκίτι βσται σαρξ αλλά άμφότβρα νους. 

Leg. Alleg. i. 33 0• ^4? 65) διττός Ιστι θάνατος ό pev άνθρωπου ό δε 
ψυχής Χ8ιος' δ pkv ουν άνθρωπου χωρισμός €στι ψυχής άπο σώματος, 6 δε 
ψνχης θάνατος άρβτης pev φθορά €στι, κακίας δε άνάληψις' παρ' ο και φησιν 
ουκ άποθανΐΐν αυτό μόνον αλλά ' θανάτω αποθανόν ' ^Gcn. 2. 1*7), ^ηλών ου 
τον κοινόν, αλλά τον ΪΒιον και κατ (ζοχην θάνατον δς €στι ψυχής €ντυμβ€υο- 
μίνης πάθ^σι Kcxi κακίαις άπάσαις. 

De poster. Caini 21 (i. 239) ψνχης θάνατος ος κατά πάθους αλόγου 
ίστίν αύτης μ€ταβολη. 

Quod del. poi. insid. 20 (i. 205) τίθνηκΐ δε ... . τον ψυχικον θάνατον, 
άρ€της κα& ην άξιος μόνην ε'στι ζην άποσχοινισθ^ις. 

Frogm. ap. Joh. Damasc. sacr. parall. p. 748 a (ii. 653) επειδ?) δε 
η8ονην €ζητησ€ bC ης ψυχικός θάνατος ετΓίγίι/εται Trj yrj προσ^νίμηθη (with 

reference to Gen. 3. 19). 

QutS rer. divin. heres 11 (i. 480) .... αϊσθησιν ψ κα\ ό γήινος νους, 
δνομα *Αδά/χ, Ιδών διαπλ ασκείσαι/ τον ίαντοΰ θάνατον ζωην 4κάνης ώνόμασ^ν 
* ε'κάλεσε * γάρ, φησιν, ' Άδά/Α όνομα γυναικός αυτού Ζωην, δτι αυτή μητηρ 
πάντων των ζώντων ' των προς αληθίίαν τον ψνχης δηπου τίθνηκότων βίον. 

V. T/ie higher manifestations of ψυχή. 

But although the higher elements of consciousness are 
usually so blended with the lower as to be sometimes over- 
powered by them, they are in their essence independent 

IN PHILO. 121 

of them. It is a cardinal point of Philo's psychology that 
pure intelligence, ψνχη or vovs in its highest form, is not a 
phase or development of animal life, but an element infused 
into animal life from above and separable from it. 

The nature of this higher element is expressed some- 
times in the terms of physical philosophy and sometimes 
in the terms of theology. It is described sometimes as 
a part of the ' quinta essentia,' the purest of all modes 
of existence : and sometimes as a part of the divine 
nature. The terms which are used to describe its relation 
to God are derived from several sources : some of them 
come from Greek philosophy, for the belief that the mind 
is a part of God was not peculiar to Judaism ; but the 
majority of them embody and combine the statements 
of the book of Genesis, that man was made ' in the image 
of God/ and that God breathed into man ' the breath of 
life.' Sometimes Philo himself expressly distinguishes 
between the philosophical and the theological modes of 
stating the same facts {e.g. De plantat. Noe 5, i. '^'^1^ see 
below) : and sometimes also in adopting a philosophical 
term he attaches to it a theological sense, e.g. in adopting 
the Stoical term άττόσττασμα he guards himself against the 
inference which might be drawn from it that the essence 
of man is separate from that of God, τίμν^ται yap ovhlv 
τον Odov κατ ατϊάρτησιν (i.e. SO as to be detached) άλλα 
μόνον €κτ€ίν€ταί Quod det. pot, insid. 24 (i. 209). 

(i) In the following passages he speaks of it in the terms 
of philosophy : 

Qui's rer. divin. heres 57 (i. 514) to hk voepov κα\ ούράνιον της ψυχής 
"γίνος προς aWepa τον καθαρώτατον ως προς πατέρα άφίξζται' πβμπτη γά/>, 
ως ό των αρχαίων Χόγος, Ζστω τις ουσία κνκλοφορητικη των τεσσάρων κατά 
το κρζίττον διαφέρουσα, (ξ ης οΐ re άστίρ^ς κα\ 6 συμπάς ουρανός edo^e 
Ύ(γ€νησθαι ης κατά το άκόλουθον θ^τίον και την άνθρωπίνην ψυχην από- 

Quod Deus immuL ίο (i. 279) τοντο της ψυχής το etSos ονκ ίκ των 


αυτών στοιχείων i^ ούν τα αλΧα aTreTeXeiTO δίίπλάσθη, καθαρωτ€ρας Se και 
άμ^ίνονος €λαχ€ της ουσίας e^ ης αι θα,αι φύσβις (^ημιουργοΰντο. 

De profuglS 24 (i• 5^5) *^°^ ° voGs, %v6ip\iov κώ. τχ^ττυρωμίνον πν€ϋμα. 

De decern orac. 25 (ϋ. 202) άνθρωπος δί ζώον άριστον κατά το κρύττον 
των iv αντώ, την ψυχην, συγγ^νίστατος τω καθαρωτάτω της ουσίας ουρανω, 
ως be ό πλίίστων λόγος, και τω του κόσμου πατρίς των em γης απάντων 
οΙκ€ΐότατον άπακόνισμα κα\ μ'ιμημα της άώίου και ΐυ^αίμονος 18βας τον νουν 

(2) In the following passages he speaks of it in the 
terms of theology, or in the terms of philosophy and 
theology combined. 

De mundi Opif. 46 (i. 32) το γαρ ^€ν€φύσησ€ν' ονδει/ ην eTepov ή 
πνεύμα θείον άπο της μακάριας κα\ ευδαίμονος εκείνης φύσεως άπϋικίαν την 
ενθάδε στειΚάμενον eV ωφελεία του γένους ημών. 

Ibid. 51 (i. 35) """^ άνθρωπος κατά μεν την biavoiav ωκείωται θείω λόγω 
της μακάριας φύσεως εκμαγεΐον η απόσπασμα η απαύγασμα γεγονώς. 

Quod det. pot, insid. 23 (i. 207) η μ^ν ουν κοινή προς τα άλογα ^ύναμις 
ουσίαν ελαχεν αϊμα ή δε εκ λογικής άπορρυεΐσα πηγής το πνεύμα, ουκ άερα 
κινούμενον άλλα τύπον τίνα κα\ χαρακτήρα θείας δυνάμεως ην ονόματι κνρίω 
Μωΰσης ' εικόνα * καλεί, δηλών δτι άρχετυπον μεν φύσεως λογικής ό .θεός 
εστί μίμημα δε κα\ άπεικόνισμα άνθρωπος. 

Ibid. 24 (ΐ• 2θ8) ψυχην οΰδεμίαν τω σώματι δ ποιών ειργάζετο ικανην 
i^ εαυτής τον ποιητην ΙδεΙν' λογισάμενος δε μεγάλα όνησειν το δημιούργημα 
εΐ λάβοι του δημιουργησαντος εννοιαν, ευδαιμονίας γαρ κα\ μακαριότητος ορός 
ούτος, άνωθεν επεπνει της Ιδίου θειότητος. 

De plantat. Noe 5 (i. 332) οί μεν αΚΚοι της αίθερίου φύσεως τον ήμε- 
τερον νουν μοίραν είπόντες είναι, συγγενειαν άνθρώπω προς αιθέρα άνηψαν' 
ό δε μέγας Μωϋσης ουδενι τών γεγονότων της λογικής ψυχής το είδος ομοίως 
ώνόμασεν, αλλ' είπεν αυτήν του θείου κα\ αοράτου εΙκόνα. 

Quis rer. divin. heres 12 (i. 481) θείας εικόνος εμφερες εκμαγεΐον. 

Ibid. 13 (ί• 4^2) 6 καταπνευσθεΊς άνωθεν ουρανίου τε κα\ θείας μοίρας 
επιλαχών, ό καθαρώτατος νους. 

Ibid. 38 (ΐ. 49^) ["οί^ϊ] ο,π^ ουρανού καταπνευσθεϊς άνωθεν. 

De mutat. nomin. 39 (i• 612) λογισμός .... της του παντός ψυχής 
απόσπασμα η όπερ όσιώτερον εΙπεΙν τοις κατά Μωϋσήν φιλοσοφοϋσιν, 
εΙκόνος θείας εκμαγεΐον εμφερες. 

Vita Mosis iii. 36 (ii. 176) ό γάρ νους ουκ &ν ούτως ευσκόπως ευθυβό- 
λησεν ει μη καϊ θείον ην πνεύμα το ποδηγετούν προς αύτην την αλήθειαν. 

IN PHILO. 123 

De Concupiscent. 11 (ii. 356) ro Se €μφνσώμ€νον Βηλον ως aWepiov ην 
Ίτν^νμα και ei bx} τι αίθ^ρίον τΐν^υματος Kpelrrop are της μακάριας και τρισμα- 
Kapias φύσβωί άττανγασμα. 

This divine and immortal part of us is not only separable 
in its nature from the fleshly and mortal part, but it some- 
times even in life disentangles itself from the body, sense, 
and speech, and contemplates the realities to which it is 
akin. The mist is dispersed and it sees clearly {De 
migrat. Abraham, ofi^ i. 467). The mind is constantly 
emancipating us from our captivity {Quod Deus immut 
10, i. 379 TO €ξαιρονμ€νον ets iKevOepLav, vovs). Its life in 
the body is but a temporary sojourn. The true home 
and fatherland of the soul is not the body but heaven : 
and to that home and fatherland the philosopher is always 
trying to return. 

De Somniis i. 8 (i, 627) κινίιται yap ημών η ψνχη πολλάκις μ^ν ίφ' 
€αντης^ όλον τον σωματικον ογκον €κ8νσα κα\ τον των αίσθησΐων οχλόν 

De migrat. Abraham. 35 (i• 466). The power of our mind to 
rid itself of the senses, whether in sleep or when awake, is an argu- 
ment for the separate personality of the Creator : d μη νομίζ€Τ€ τ6ν 

pev ημίτ^ρον νουν άττοδνσάμίνον σώμα, αισθησιν, λόγοι/, ^Ίχα τούτων γνμνον 
δννασθαι τα οντά οραν, τον de των όλων νουν τον Qebv ουκ (ζω της υλικής 
φνσβως πάσης έστάναι, π^ρύχοντα ου π^ρι^χόμ^νον. 

De Gigantibus 4 (i• 264) αύται μ^ν ουν (Ισι ψυχαΐ των άνωθεν πως 
φιλοσοφησάντων, e^ ^ΡΧΨ άχρι τέλους μίλ€τωσαι τον μετά σωμάτων άπο- 
θνησκ€ΐν βίον Ινα της ασωμάτου και άφθαρτου παρά τω ayevvr)T(u κα\ άφθάρτω 
ζωής μ^τάλάχωσιν. 

De agricult. 14 (i• 3^°) '"'? 7°Ρ '^^"^^ πάσα μίν ψυχή σοφού πατρίδα 
μ€ν ουρανον ζίνην δε γην Ζλαχ<ε. 

De confus. ling. 17 (i. 416) (παδάν ουν βνδιατρίψασαι σώμασι τα 
αΙσθητά και θνητά di αυτών πάντα κατίδωσιν,-βπανβρχονται βκεΐσε πάλιν 
οθΐν ωρμηθησαν το πρώτον, πατρίδα pev τον ονράνιον χώρον iv ω πολιτεύον- 
ται ^evov δε τον περίγ^ιον iv ω παρωκησαν νομίζουσαι. 

Quis rer. divin, heres 57 (i. 514). The bodily parts of us are 
resolved into the four elements, το δε voepov και ούράνων της ψυχής 
yίvoς προς aWepa τον καθαρωτατον ως προς πατέρα άφίζΐται. 


VI. ψυ)(ΐκ(>5. 

It is so reasonable to expect that the adjective ψυχικό? 
should follow in Philo the varieties of meaning of its sub- 
stantive, that the word would not need a separate notice 
if it were not for the special senses in which it is found in 
both the New Testament and later Greek. It is clear 
that although those special senses of ψυχικοί are not in- 
consistent with its use in Philo, the word had not yet 
become narrowed to them : it is used, as ψυχή is used, in 
reference (i) sometimes to animal life, (2) sometimes to 
the common human life of feeling and passion, (3) some- 
times to spiritual life or the highest activity of thought. 

(1) Leg. Alleg. ii. γ (i. γι) b γυμνός και avevderos σώματι νους .... 
τΓολλά? €χ€ΐ δννάμ€ΐ9, €κτικην [?'. e. the power of cohesion], φντικην, 
ψυχική I', XoyiKTjv, διανοητικην, oKXas μνρίας κατά re Λδη και γίνη. 

Ihid, 13 (i. 74) ^ Ί^Ρ ^°^^ καθάπ^ρ βδηλωσυ, 6τ€ eyevvaTo, συν πολλαΤί 
8υνάμ€σι και e^caiv eyevvaTo, λογική, ψυχικ-ρ, φντικτ), ώστε και αΙσθητικη. 

(2) Leg. Alleg. ii. 21 (i. 81, 82). Solitude does not necessarily 
give a man freedom from the stings of sense and passion, and, on 

the other hand, eVn δε οτε και iv τνληθίΐ μυριάνδρω 4ρημω την διάνοιαν, 

τον ψυχικον οχΧον [the crowd of sensations and passions] σκ€δάσαντος 

θίον και διδάξαντος μ€ οτι ον τόπων διαφοραϊ τό τε ευ κα\ χείρον αργάζονται 
αλλ' 6 κινων θ(ος και άγων fj αν προαιρηται το της ψνχης όχημα. 

Ibid. iii. ι 7 (i• 9^) °^ φοβούμενοι κα\ τρψοντβς νη άνανδρίας κα\ δειλίας 

De Cherubim 24 (i. 154) ^^ effeminate men whose strength is 

broken before its proper time, μετ εκλυσεω? ψυχικών δυνάμεων. 

Ibid. 30 (i• 158) as frescoes and pictures and mosaics adorn 
a house, and minister delight to its inmates, ούτως ή των εγκυκλίων 

επιστήμη tov ψυχικον οίκον άπαντα διακοσμεί, each kind of knowledge 

having some peculiar charm. 

(3) Leg. Alleg. ii. 15 (i. 75) of the soul which, putting off the 
sights and sounds of sense, είσελενσεται σπεϊσαι το ψυχικό»' αίμα κάί 
θυμιάσαι όλον τον νουν τω σωτηρι και ευεργεττ] θεώ. 

Decongr. erud. grat. 19 (i. 534) "τουτ εστί, κυρίως είπεϊν, το ψυχικοί' 
ΤΙάσχα, η παντός πάθους κα\ παντός αισθητού διάβασις πρ6ς το δεκατον δ δη 
νοητόν εστί και θείον. 

IN PHILO. 125 

VII. vods. 

For the term ψνχη, in all its senses, Philo sometimes 
substitutes the term vovs. The distinctions which exist 
between the terms in both earlier and later philosophy 
sometimes wholly disappear : and although vovs is used 
for the highest manifestations of thought, it is also used, 
as both ψνχη and τιν^νμα are used, for purely physical 

(i) It is simply convertible with ψνχη : e.g. — 

Oe Gigant. 3 (i. 264) ψυχψ η νουν τό κράτιστον των ev ημίν. 

Quis rer. divin. heres 22 (i. 487): Philo enumerates ψνχψ^ αίσθησιν, 
λόγον, and immediately afterwards substitutes rod vod where rrjs ψυχής 
would be expected. 

De congr. erud. grat. 25 (i. 540) in a co-ordinate enumeration 

we find ο hk νοΐις . . . . 6 de νους . ... η be ψνχη. 

(2) It is used, like ψνχη, of the highest powers of thought, 
those by which we have cognizance of τά νοητά and of 

Qm's rer. divin. heres 22 (i. 488) νω γαρ 6 Oebs καταλάμβαναν τον /xeV 
νοητον κόσμον δι' έαντοΰ τον 8e όρατον δι αΙσθήσ€ως €φηκ€ν : but imme- 
diately below he substitutes ψυχή for vovs, dta μεν αίσθησίων els τα 
αισθητά διακίιψας eveKa του το ά\ηθ€ς evpeiv δια be τψ ψυχής τά νοητά κα\ 
οντά ούτως φίλoσoφήσas. 

(3) It is used, like ψνχη, of the cognizance of the sensible 

Quod det. poi. insid. 26 (i. 210), φαντασία, i.e. perception, is a 
function of vovs: but in Quod Deus immut. 9 (i. 278, 279) it is a 
function of ψνχή. 

Leg. Alleg. ii. 10 (i. 73) sensation is one of the powers of vovs : 
ibid. iii. 90 (i. 137), and elsewhere, the senses are collectively a 
part of ψνχΎ\. 

(4) It is used, like ψνχη, not only for all the forces or 
powers of both animal and vegetable life, but also for the 
force of cohesion. 


The two passages in Leg. Alleg. ii. 7, 13, which show this most 
clearly, are quoted above under § VI (i), p. 124. 

VIII. ΊΓΐ^ευμα. 

It will have appeared from several passages which have 
been already quoted that ΐϊν^νμα is used with no less 
a width of meaning than ψυχτ) or vovs. There is the broad 
general distinction between the terms that 'πν€νμα is re- 
garded as the underlying cause which gives to the several 
forms of ψνχη not their capacity but their energy. The 
conception of ττνζΰμα may be regarded as being closely 
analogous to the modern conception of 'force,' and espe- 
cially to that form of the conception which makes no 
distinction of essence between ^mind-force' and other kinds 
of force, such as light or electricity. It is analogous but 
not identical : for force is conceived to be immaterial, 
whereas ττν€ϋμα, however subtle, is still material. 

(i) It is used, like ψνχη and povs, of the force which holds solid 
bodies together : cohesion is a ' force which returns upon itself/ 

Quod Deus immut. 7 (i. 277, 278) Χίθων μ€Ρ ovv κα\ ξύλων .... 
δ€σμ6ν κραταιότατον e^iu (Ιργάσατο' ή δβ 4στι ττν^νμα άνα(Γτρ€ψον εφ' 

(2) It is used of the physical basis (ουσία) of growth and 

De mundi Opif. 22 (i. 15) ^ δε {sc. φύσις) .... ζωοπλαστ€ΐ τψ μ€Ρ 
vypav ονσίαν els τα του σώματος μίλη κα\ μέρη ^ιανίμουσα, την ττνευματικην 
els τας της ψνχη5 ^υvaμ€ιs την re Θρΐπτικην κα\ την αίσβητικην. 

(3) It is used of both (a) reason and (δ) sensation. 

{a) Quod det.pot. insid. 23 (i. 207) άνθρωπου δε ψνχήν 6voμάζeι πν€ϋμα, 
ανθρωτΐον ου το σνγκριμα καλών ως Ζφην αλλά το ueoeides eKclvo δημιούργημα 
ω Χογίζομίθα. 

{δ) De profugis 32 (i. 573)• Each of the senses owes its activity 
to the πν€ΰμα which the mind infuses into it, τ6 μ^ν όρατικον πνεύμα 

TeivovTos els όμματα, το δε ακουστικοί/ ει$ ους, eh δε μυκτηρας το 6σψpησeως, 
το δε αυ yeύσeως ει? στόμα κα\ το άφης els άπασαν την eπιφάveιav. 

Leg. Alleg. i. 13 (i. 51) God Himself breathes only into the 
highest part of man, and not into the second rank of human 

IN PHILO. ' 127 

faculties : νπο τίνος ovv Koi ταντα ενβπΐ'βύσθη ; υπό του νου δηλονότι' ου 
γαρ μίΤ€σχ€ν ό vovs πάρα τον Seov τούτου μ^ταδίδωσι τω άΚόγω pepei της 
'^Xrjs, ωστ€ τον pev νουν έψυ)(ώσθαι ΰπο θβοΰ, το δε αΚογον νπο του νου. 

(4) So far, the senses in which Philo uses Ένδυμα are 
senses in which it was also found in current Greek philo- 
sophy. To these senses he added another which comes 
not from philosophy but from theology, and is expressly 
based, on the statement of Moses that God breathed into 
man the ' breath ' of life. So that while, in some passages, 
by using the current philosophical language which spoke 
of ττν€υμα as the essence of mind, he implies that mind 
could not exist without it, he elsewhere implies that mind 
existed anterior to it and may now exist without it. He 
speaks of ττν^νμα being infused into mind by a special 
act of God, or, by another metaphor, of mind being drawn 
up to God so as to be in direct contact with Him and 
moulded by Him. 

Leg. Alleg. i. 13 (i. 5^) "^9^^ J^P ^^^'- ^"j ''o ^ρπν^ον, τό Bexopevov, 
TO €ρπν€Ορ€νον' TO pev έρπνίον 4στ\ν 6 Beov, τό δε δεχόρίνον 6 νους, το δε 
€ρπν€Ορ(νον το πν€νρα. τι ονν €Κ τούτων συνάγεται' ενωσίί γίνεται των 
τριών, τείνοντος του θεοϋ την αφ' εαυτού δυναριν δια του ρεσου πνευρατος 
άχρι τον νποκειρενον, τίνος ένεκα η όπως εννοιαν αντοϋ λαβώρεν ', επεϊ πώς 
αν ενόησεν η ψνχη θεον εΐ ρη ενέπνευσε κα\ ηψατο αντης κατά δύναριν ; ου 
γαρ αν επετόλρησε τοσούτον άναδραρεΊν 6 ανθρώπινος νους ώς άντϊΚαβεσθαι 
θεού φύσεως εΐ ρη αύτος 6 θεός άνεσπασεν αύτον προς εαυτόν, ώς ενην 
άνθρώπινον νονν άνασπασθηναι και ετύπωσε κατά τας εφικτας νοηθηναι 

(5) The conception of this special form of τΐν^νμα seems 
to be required on the one hand by philosophy in order to 
account for the fact that some men have a knowledge or 
intellectual power which others have not, and on the other 
hand by theology, since the Pentateuch speaks of men being 
filled, in some special sense, by a divine spirit. The word 
is therefore used for ' the pure science of which every wise 
man is a partaker,' and especially for the knowledge of 


God : and it is sometimes regarded, especially in treatises 
which probably belong to a generation subsequent to 
Philo, as an external force acting upon men and leading 
them to the knowledge of God. 

(a) De Gig ant. 5 (i. 265) Xiy^rai δε θ^ον πν€νμα .... καθ^ erepov Be 
τρόπον η ακήρατος επιστήμη ης πάς 6 σοψος (Ικότως μ€Τ€χ€ί (the instance 

given is that of Bezalel, who was filled πνεύματος Beiov, σοφίας, 

σνν€σ€ως, επιστήμης, Exod. 3 1. 3). 

Vl/a Afosi's 3. 36 (ii. 176) ό γάρ νους ουκ αν ούτως €υσκόπως €νθνβό- 
Χησίν el μη κα\ θεΐον ην πν€νμα το ποΒηγΐΤοΰν προς αντην την aXiiOeiav. 

De Somniis 2. 38 (i. 692) νπηχύ he μοι πάλιν το elωθ6ς άφανως 
€νομιλ€'ιν πν€υμα άόρατον και φησιν' ω ούτος, €θίκας άνΐπιστήμων eivat καΐ 
/χβγάλου κα\ πepιμaχητoυ πράγματος .... 'ίσθί δη, yevvaU, δτι θεός μόνος ή 
ά\ΐΑ€υδ€στάτη και προς aKrjeeiav earTiv ίΐρήνη η 8e γ€ννητη και φθαρτή ουσία 
πάσα συνεχής πόλ€μος. 

It follows that ιτν^υμα in its theological as well as in its 
philosophical sense, is not a part of human nature but 
a force that acts upon it and within it. The dichotomy 
of human nature remains. There is a single body with 
many members ; there is a single mind with many func- 
tions. But the mind may be drawn in either of two ways, 
yielding to the allurements of pleasure or to the special 
force of the divine spirit. There are thus two kinds of 
men. (a) On the one hand, though all men have mind 
and, so far, have an element within them which is not 
merely spirit but divine spirit, yet in another sense there 
are men in whom the divine spirit does not abide, (δ) On 
the other hand there are the prophets, men in whom the 
manifestation of the special force of the divine spirit is 
so strong that the human mind for a time migrates from 
them, ' the sun of the reason sets,' and in the darkness of 
the reason the divine spirit carries them whither he wills. 
In other words, just as, though the material world is held 
together, and animals live, by virtue of a ττν^νμα, and yet 
men are differentiated from animals by the presence of 

IN PHILO. 129 

a higher degree or special form of ττνζΰμα : so men are 
differentiated from one another by the presence of a still 
higher degree or more special form of it. The conception 
becomes more intelligible if it be remembered that all the 
forms of τΐν^νμα are regarded as being material, being in 
fact different degrees of the purity or rarefaction of the 
air. The lowest form is moist air near the surface of the 
earth, the highest is the clear ether beyond the starry fir- 
mament, (r) It must also be noted that Philo does not 
confine the expression ττνζνμα Oeov to the highest form, but, 
following Genesis i. 2, applies it to the lowest. 

(a) De Gtgant. 5 (i. 265) iv drj τοΊς toiovtois {i.e. m men of 
pleasure) άμηχανον το του θευΰ καταμβΐναι και διαιωνίσαι πν^νμα ώί δηΚοΙ 
και avTos 6 νομοθέτης' fwre, γαρ, φησί^ κύριος ό ueos' ου καταμ^νύ το 
πι/€νμά μου iv τοΙς άνθρώποις els τον αΙώρα bia το eivai αυτούς σάρκας, 
μίνα μίν yap Ζστιν οτ€ καταμένίΐ Be ονδ' (Ις άπαν τιάρα τοΙς ττ6)\ΧοΙς 

[δ) Quis rer. divin. heres 53 (i. 511) τω δε προφητικω yevei φιλ€Ϊ 
τοϋτο (Γυμβαίν€ΐν ^ξοικίζίται μέν yap iv ήμΐν ό νους κατά την του βίίου 
πνεύματος αφιξιν, κατά de μ^τανάστασιν αυτού τΐοΚιν €ΐσοικίζ(ται' θέμις yap 
ουκ εστί βνητον άθανάτω συνοικησαι. δια τούτο ή δύσις του \oyLσμoυ κα\ 
το nepX αυτόν σκότος Ζκστασιν κα\ θ€οφόρητον μανίαν €γεννησ€. 

(c) De Glgant. 5 (i. 265) Xeyerai δε θίου πνεύμα καθ' eva μβν τρόπον 
ό ρέων άηρ eVi y^y, τρίτον στοιχείον Ιποιχουμενον υδατι, παρ' 6 φησιν ev Tjj 
κοσμοποιί^. πνεύμα θεού επεφερετο επάνω του ύδατος. 

General Results, 

The chief importance of this discussion of the psycho- 
logical terms of the Septuagint and Philo is in relation 
to the New Testament. It will be clear that the fine 
distinctions which are sometimes drawn between them in 
New Testament exegesis are not supported by their use 
in contemporary Greek. Into the large subject of the 
psychological ideas of the several writers of the New 
Testament as indicated by the use of psychological terms 



I do not propose now to enter : but I believe that two 
points may be clearly gathered from the facts which have 
been mentioned, — 

(i) That the use of such terms in the Synoptic Gospels 

is closely allied to their use in the Septuagint. 
(a) That the use of such terms in S. Paul differs in 
essential respects from the use of them in 
Philo, and that consequently the endeavour to 
interpret Pauline by Philonean psychology falls 
to the ground. 


The textual criticism of the LXX. is a subject which 
has hitherto received but slight attention from scholars. 
It has naturally been postponed to that of the New Tes- 
tament : and on even the textual criticism of the New 
Testament it is probable that by no means the last word 
has been said. The materials have been collected, and 
are being collected, with singular care : but, so far from 
the final inductions having been made, the principles on 
which they should be made have not yet been finally 

In the case of the LXX. we are at least one step further 
back. The materials have yet to be collected. They are 
of three kinds (i) Greek MSS., (ii) Versions, (iii) Quotations. 

i. The MSS. of the whole or parts of the LXX. enu- 
merated by Holmes and Parsons, and wholly or partially 
collated for their great Thesaurus^, amount to 313, of. 
which 13 are uncials. Since the publication of that work 
many additional MSS. have come to light, and among 
them several uncials of great importance : of the 29 MSS., 
including fragments, in Lagarde's list of MSS. written 
before A. D. 1000 ^, 13 were unknown to Holmes and 
Parsons. The addition of this new material to the appa- 
ratus criticus would be a work of moderate compass, if 

^ Vettts Testamentum Graecum cum variis lectionibus : Editionem a Roberto 
Holmes inchoatam continuavit Jacobus Parsons : Oxonii, mdccxcviii- 

^ Lagarde, Genesis Graece (Lipsiae, 1868), pp. 10-16. 

Κ 2 


the existing basis were trustworthy : but it is unfortunately 
the case that Holmes and Parsons entrusted no small part 
of the task of collation to careless or incompetent hands : 
consequently before any final inductions can be made the 
whole of the MSS. must be collated afresh. 

The extent and nature of the deficiencies in Holmes and Parsons 
will be seen from the following comparison of a few verses, chosen 
at random, of the collations made for Holmes and Parsons with 
the collations made by Lagarde. 

The passage chosen is Gen. xxvii. 1-20 : in it Holmes and 
Parsons mention various readings from, and must therefore be 
presumed to have collated, 36 cursives : of these Lagarde has 
collated three, viz. a Munich MS., H. and P. No. 25 ; a Venice 
MS., H. and P., No. 122 ; and a Vienna MS., H. and P., No. 130. 
This more accurate collation requires the following additions to be 
made to the apparatus criticus of the Oxford edition. 

v. I : Cod. 130 reads *ΐσαάκ for Έσαυ, and omits vlk μου κα\ elnep 
Ιδον €γω Koi einev. 

V. 4 : Cod. 25 fiXoyrjaci. 

V. 5 : Cod. 122 ηκου€ for rJKOvae : 1 30 *1σαακ \(ΐΚονντος. 

V. 6 : Cod. 122 omits του before Ίακώ/3 : 130 reads ίδου for iSe. 

V. 9: Cod. 130 adds re after απαλούς. 
V. ϊο: Cod. 25 eiXoyrjaei. 

V. 14 : Cod. 130 adds αυτοΰ after ττ} μψρΧ and reads καθώς for 


V. 15: Codd. 122, 130 omit 011x171' after eVeSvafv. 
V. 16 : Codd. 25, 130 read 'ίθηκΐν inl τά γυμνά, omitting eVi τους 
βραχίονας αυτοϋ και. 

V. 1 8 : Cod. 122 has evcyKC for darjveyKe. 

V. 1 9 : Cod. 2 5 KoX π(ποίηκα I 12 2 omitS από. 

This comparison gives eighteen corrections in the space of 
twenty verses in one-twelfth of the MSS. collated. 

To these corrections of MSS. which were actually collated may 
be added, as an example of the additions which may be expected 
from a further examination of the MSS., Lagarde's collation of the 
same passage in the Zittau MS. which Holmes and Parsons men- 
tion in their list as No. 44, and which was partly collated for their 
edition, but of which no various readings appear in Genesis. 


The following is the collation of the Zittau MS. : — 

V. I : TOP vlov αυτού Ήσαω : om. μον after vie. 

V. 2 : om. 6i7re Se αυτω Ισαάκ : ιδού εγώ γ^γηρηκα. 

V. 4 • €ν\ογησ€ΐ : ττρίρ ή. 

V. 6 : *'Ρ(β4κκα δε ήκουσβ \aXovvTos ταντα και : om. roi) before ^Ιακωβ : 
Vfayrepov for ελάσσω : eyo) ήκονσα : λαλουιαοί του ττατρός σου : 0Π1. toj/ 
ά^ίΚφόν σον. 

V. 7 : και for ιι^α : μ€ άποθανΐΐν. 

V. 8 : cm. μου after υ«ν. 

νν. 9"~ΙΟ • om. ως φιλεί και €ΐσοίσ(ΐς τω ττατρί σον. 

V. ΙΟ : om. ενλογτ^σβι: om. αντοί). 

ν. 1 1 : om. irpos 'νφ4κκαν την μητέρα αυτού and Ήσαυ. 

V. 12 : om. in. 

V. 13 I άκουσον for Ιπάκονσον. 

V. 14 : τ^ Μ'ίτρ' αύτου : καθώς for κα^ά. 

V. 1 5 : om. αυτι^ΐ' after ^νί^νσ^ν. 

ν. ΐ6 : Trepi τους βραχίονας. 
V. ΐ8 : και dne for eiTre δβ. 
V. 19 : τω πατρϊ αυτόν '. ^ποίησα : om. από τ^ί ^Jjpay μον. 

η. The Latin and Eastern versions of the Old Testament 
were made not from the Hebrew original but from the 
LXX. version. They have now to be used reversely, i.e. 
as indicating the LXX. text at the time at which they 
were written : and from the critical study of them more 
light is likely to be thrown upon the early recensions of 
the LXX. than from any other source. With the Eastern 
versions, i.e. the Egyptian (Sahidic, Memphitic, and Bas- 
muric), Ethiopian, Armenian, Arabic, and Syriac, I am 
not competent to deal:, the Latin versions are collected 
with singular care in the great work of Sabatier, nor, 
except in the cases of Cyprian and Lucifer of Cagliari, 
has modern criticism as yet improved to any considerable 
degree the texts which Sabatier used. 

iii. The quotations from the LXX. in the Greek Fathers 
are an almost unworked field. With the Greek even more 
than with the Latin Fathers the texts require to be criti- 
cally edited before the comparison of the quotations with 


the MSS. of the LXX. can be satisfactorily made : but 
the corroboration of the discovery of Lucian's recension, 
which will be mentioned below, by the agreement of the 
MSS. which are believed to contain it with the quotations 
in Chrysostom and Theodoret, shows how much help may 
be expected from this source. 

The next step after collecting the materials is to group 
the MSS. into classes or families. For this our chief 
guide is the statement of Jerome that there were three 
recensions of the LXX. in his time, — that of Hesychius 
which was accepted in Egypt, that of Lucian which was 
accepted from Constantinople to Antioch, that of Origen 
which was accepted in Palestine^. The first step is to 
recover, if possible, the texts of these several recensions. 
And in the case of one of them, that of Lucian ^, we have 
a remarkable clue. In a Paris MS. there is appended to 
some marginal readings of several passages of the Fourth 
Book of Kings a sign which is most probably interpreted 
to be the Sy riac letter Lomad: but this letter is said by 
a tradition which comes through two channels, Greek and 
Syriac, and contains no internal improbability, to have 
been appended to the readings of Lucian's recension : it 
is consequently inferred that these readings furnish a test 
for the determination of the MSS. which contain Lucian's 
recension. It is found that they coincide with the readings, 
in the several passages, of Codd, 19 (Chisianus R vi. '>^^^ 
Lagarde's h), 8:j (Parisinus Coislin 3, Lagarde's f ), 93 (Arun- 
delianus I D 2, Lagarde's m, in his later notation), 108 
(Vaticanus 330, Lagarde's d, the basis, with 248, of the 
Complutensian edition). These four MSS. are found to 
hang together, and to have a peculiar text, throughout 
the LXX. : their readings are also found to agree with 

^ S. Hieron. Apol. adv. Ruffin. Tom. ii. p. 522. 

* It is unnecessary to repeat here the details respecting Lucian's edition 
which are clearly and exhaustively given by Dr. Field, Prolegomena in Hexapla 
Origenis, pp. Ixxxvi sqq. 


the quotations from historical books in Chrysostom and 
Theodoret, who may reasonably be supposed, assuming 
Jerome's statement to be accurate, to have used the text 
of Lucian. To the above-mentioned MSS. several others 
are found to be cognate, viz. 44 (the Zittau MS. mentioned 
above), 118 (Parisinus Graecus 6, Lagarde's p), ^6 (Paris- 
inus Graecus 5, Lagarde's k) : and a MS. in the British 
Museum (Add. 30002, Lagarde^s E). A comparison of 
these MSS. gives a single text which may reasonably be 
taken to represent Lucian's recension: and Lagarde has 
published it as such \ 

The next task of LXX. criticism will be to discover in 
a similar way the texts of the two other recensions. There 
are many indications of the path which research in that 
direction must follow : and the research would be full of 
interest. I do not propose to engage in it now because 
an even greater interest attaches to the question with 
which I propose specially to deal in this chapter, namely, 

What can we learn about the text, or texts, of the LXX. 
before the three recensions of which Jerome speaks were 

The answer to this question does not depend on the 
restoration of the text of those recensions. It is true that 
if we had the three recensions complete we should be able 
to infer that the readings in which they agreed probably 
formed part of a text which was prior to' them : but we 
should still be unable to tell whether any given variant, 
i.e. any reading in which one of the three differed from 
the two others, or two of the three from the third, was 
part of an earlier text or a revision of it. We should 
also find that some of the existing MSS. and versions 

^ A specimen appeared in his Ankundigung emer neuen ausgabe der griechi- 
schen iibersezung des alien testaments^ Goettingen, 1882 : and the first volume 
(Genesis-Esther) of a complete edition in 1883. 


had readings which did not belong to any of the three 
recensions : and we should be in doubt whether these 
belonged to an earlier text or to a revision of it. It is 
consequently not necessary to possess the current texts 
of the third century in order to discover the text or texts 
of the preceding centuries. The discovery is not only in- 
teresting but important : and it is important in relation 
not only to textual criticism but also to exegesis. It is 
important in relation to textual criticism, because it may 
enable us to recognize in some existing MSS. the survivals 
of an earlier text than that of the three recensions : it is 
important in relation to exegesis : for as each recension 
reflects the state of knowledge of Hebrew, and the current 
opinion as to the interpretation of the Hebrew text, in 
the country in which it was made in the third century of 
the Christian era : so the texts which precede those re- 
censions reflect the state of philology and of exegesis, in 
both Egypt and Palestine, during the first two centuries 
of the Christian era, and the two, or three, centuries which 
preceded it. 

I have spoken of earlier texts, in the plural, rather than 
of the original text of the LXX., because there are many 
indications that the first and second centuries were no 
more free from variations of text than was the third. It 
was natural that it should be so. In the case of an original 
work like the Aeneidj or like the New Testament, there 
is a presumption that the scribe would endeavour to copy 
as accurately as he could the text before him, emending 
a passage only in the belief that it had been wrongly 
written by a previous scribe and in the hope of represent- 
ing more accurately by his emendation what the author 
wrote. But in the case of a translation there is a constant 
tendency to make the text of the translation a more 
accurate representation of the text of the original. It 
may be assumed that a certain proportion, though perhaps 


only a small proportion, of the scribes of the LXX. were 
acquainted Avith Hebrew : it would be almost a religious 
obligation on such scribes, when they saw what they 
believed to be a mistranslation, to correct it. This was 
probably the case in an especial degree when certain texts 
came to have a dogmatic or controversial importance. 
Hence there is an a priori probability of the existence of 
varieties of text : and the probability will be found to be 
strongly confirmed by the detailed examination of some 
passages of the LXX. in the following pages. 

What data have we for determining the question that 
has been proposed? How can we go behind the recen- 
sions of which Jerome speaks, and to one or other of 
which it may be presumed that the great majority of the 
existing MSS. belong? 

The data consist partly in the quotations from the LXX. 
in early Greek writers, especially in Philo, in the New 
Testament, and in the Apostolic and sub- Apostolic Fathers, 
and partly in the quotations from the Latin versions which 
are found in early Latin writers. This statement assumes 
in regard to the Greek writers that they made use of the 
LXX. and not of another translation : but the assumption 
will be proved to be true when the quotations are ex- 
amined. The points of similarity between them and the 
text of the LXX., the structure of the sentences, and the 
use of peculiar words and idioms, are altogether too 
numerous to admit of the hypothesis of the existence of 
another translation : the points of difference are, with 
hardly an exception, such as may be accounted for by 
the hypothesis of varieties of text and mistakes in trans- 
mission. The statement assumes also that the early Latin 
versions were made from the LXX. : this assumption also 
will be proved when the quotations are examined. The 
use of each of these classes of data, though more in the 
case of Greek than of Latin writers, is attended with the 


preliminary difficulty that the texts of the quotations have, 
in many instances, been altered by scribes in order to bring 
them into harmony with the Biblical texts of a later time. 
The difficulty is sometimes removed by the fact that the 
writer comments on a particular phrase and therefore 
establishes the fact of his having read it: and the prob- 
ability of its existence in such a writer as Philo, in short 
passages which have no dogmatic importance, is very 
small : but at the same time there is no doubt that the 
data must be used with some degree of caution, and that 
the final results of the examination of them cannot be 
obtained until the texts of the several writers have them- 
selves been critically studied. 

These data may be dealt with in two ways, (i) The 
MSS. readings of a given passage may be compared with 
the quotations of it : the special use of this method is 
twofold : {a) it enables us to classify MSS., and to estimate 
their value, according as they do or do not agree with 
such early quotation? ; {b) it enables us also in certain 
cases to detect, and to account for, the recensions of the 
passage, and so obtain a clue to the history of its exegesis. 
(2) The quotations in a given writer may be gathered 
together : the special use of this method is also twofold : 
{a) it enables us to ascertain approximately the text 
which was in use in his time; {b) it enables us, upon 
a general estimate of the mode in which he quotes Scrip- 
ture, to appreciate the value of the contributions which 
his quotations make to textual criticism. 

The following pages contain examples of each of these 

(i) In the first portion a text of Genesis or Exodus is 
quoted from the Sixtine text : it is followed by {a) a short 
apparatus criticus^ taken from Holmes and Parsons, and 
from Lagarde ; (b) an account of passages in which it is 


quoted in Philo, the New Testament, the Apostolic Fathers, 
and Justin Martyr; (c) an account, where useful, of the 
early Latin versions : to this is appended a short account 
of the conclusions to which the data point in regard to 
the criticism of the passages. 

(2) In the second portion, the quotations of two books, 
the Psalms and Isaiah, in Philo, Clement of Rome, Bar- 
nabas, and Justin Martyr, are gathered together : and the 
bearing of each quotation upon the criticism or exegesis 
of the LXX. is estimated. 

The following pages contain only examples of these 
methods, and not an exhaustive application of them : their 
object is to show in detail the help which the methods 
afford in the criticism of particular passages, and to 
stimulate students to pursue them further. 

It may be convenient for those who are not familiar with the 
notation of MSS. of the LXX. to mention that in the following 
examples the MSS. are quoted according to their number in the Hst 
of Holmes and Parsons : Roman numerals (or capital letters) 
denote uncials, Arabic numerals denote cursives. The MSS. 
which have been more recently collated by Lagarde are quoted 
according to his notation: h=i9, m = 25 (in Lagarde's later 
notation, not in his Genesis Graece, m = 93), χ = 29, ζ = 44, 
y=i22, 1=130, r=i35. The Codex Alexandrinus is usually 
here denoted by A instead of by the numeral III; and the Bodleian 
Codex of Genesis (Auct. T. infr. ii. i) is denoted, as in Lagarde's 
Genesis Graece, by Ε (in his later notation E=the British Museum 
MS. Add. 20002). The Roman or Sixtine text is designated 
by R. 

The quotations from the early Latin versions are for the most 
part due to the great collection of Sabatier, Bihliorum Sacrorum 
Laiinae Versiones aniiguae^ Remis, 1743. 


I. Quotations from Genesis and Exodus, 
Genesis i. i, 2. 

'En λρχΗ επο'ίΗςεΝ ό Geoc ton ογρΑΝΟΝ και την γην* η λέ γη ην ΛορΑτοε 
ΚΑΙ AKATACKfeYAQTOC ΚΑΙ ζκότοο βΠΑΝω THC ΑΒγςςογ' ΚΑΙ ΠΝεγΜΑ θεογ έπεφε- 
ρετο έπΛΝω τογ γλΑτοε, 

Cod. 75 σκότος + ^1', Codd. 68, 120, 121 σκότος ■\-ζ'πίκ(ΐτο. 

Philo Quis rer. divin, heres 24 (i. 490) eV αρχτι ^ποίησβν : id. c/e 
Mundi Opif. 7 (i. 5) eV αρχτι . . . .την yrju=R. i id, de Incorrupt. 
Mundi 5 (ii. 491) eV άρχΐι .... ακαΓασκ€ύαστοί=Κ. : id. de 
Mundi Opif. 9 (i. 7) σκότος ην €πάνω της αβύσσου : id. Zeg, 
A /kg. i. 13 (i. 50), de Gigant. 6 (i. 265) και πνβνμα .... νΒατος 

= R. 

Justin Μ. Apol. i. 59 = R. except των υδάτων: id. ApoL i. 64 has 
the variant ίπιφ^ρομ^νου (probably a scribe's error for (πιφ€- 
ρόμ^νον) as well as των υδάτων. 

The insertion of ψ after σκότος is supported by the early 
Latin versions, all of which have 'tenebrae erant:' its omis- 
sion may be due to a Hebraizing revision of which there are 
further traces {a) in Justin's substitution of ίπιφ^ρόμ^νον (ΠεΠ"]ρ 
pres. par/.) for €π€φ€ρ€το, (δ) in his use of the plural τών υδάτων 
(0]βΠ) which is supported by Excerp/. Theod. 47, Clem. Alex. ed. 
Pott p. 980, and by the Latin 'super aquas' of Tertull. de Bap/ismo 
3, 4 pp. 256, 257, adv. Hermog. 32 p. 282, adv. Marc. 4. 26 p. 
546 : on the other hand, August, de Gen. c. Manich. i. 5 (i. 648), 
de Gen. ad /it/, i. 11, 13, 14 (iii. 120, 121), Serm. 226 (82) (v. 972), 
and Philastr. 109 p. no have 'super aquam! 

Genesis i. 4, 5. 

Και εΤλεΝ ό θεόε το φώε δτι kaAon* και λιεχώριςεΝ ό θεόο ανλ ΜεςοΝ τογ 
φωτόε ΚΑΙ ΑΝΑ ΜεςΟΝ τογ ςκότογε' και έκΑλεςεΝ ό θεόε το φώε ΗΜερΑΝ και 
ςκότοε έκΑλεςε ΝγκτΑ' και έπεΝετο έςπέρΑ και έρεΝετο πρωί πΜερΑ μια. 

The variations of the MSS. are merely orthographical. 

Philo de Somniis i. 13 (i. 632) δκχωρισ^ν .... σκότους=^. : id. 
Quis rer. divin. heres 33 (i. 496) κα\ δί^χωρισ^ν .... νυκτα=.Κ. 
except that 6 θώς is omitted after €κάλ€σ€ν, and εκάλβσί after 
σκότος : id. de Mundi Opif. 9 (i. 7) Ισ-ηίρα re και πρωία (dis) : 
ibid, του χρόνου μίτρον άτητίΚύτο €υθύς ο κα\ ημ^ραν 6 ποιών (κάΚ^σΐ 


Koi ημ€ραν ονχ\ ττρώτην αλλά μίαν ή XeXcKrai οΰτως Βιά την του νοητού 
κόσμου μόνωσιν μονα^ικην €χοντοί φύσιν (cf. Joseph. Antt. I. I και 
αντη μίν αν ΐΐη η πρώτη ημ€ρα Μωϋσης δε αυτήν μίαν elire). 

Genesis i. 9• 
Καϊ emeN ό Geoc ςγΝΑχθΗτω το γλωρ το γποκΛτω τογ ογρΑΝΟγ eic ςγΝΑ- 

ρωΓΗΝ ΜΙΑΝ Κα! ΟφθΗΤω Η 2ΗρΛ. 

Philo de Mundi Opt/. 1 1 (i. 8) προστάττα ό θίος . . . . τό μ€ν ϋδωρ 
.... ίττισυναχβηναι .... την de ζηραν άναφανηναι. 

Philo's quotation is indirect : but άναφανηναι is supported by the 
Latin '■ appareat' in S. August, de Gen. c. Manic h. i. 12 (i. 652), 
while the MSS. reading οφθητω is supported by Tertull. c. Hermog. 
29 p. 243, ^ videaiur arida.' 

Genesis i. 10. 

Και τα qyqthmata τωΝ γλΑτωΝ εκΑλεςε θΑλΑςςΑΟ. 

Philo de Mundi Opif. 1 1 (i. 8) τψ μΙν ξηραν κάλων γην το 8e άπο- 
KpiOev ύδωρ θάλασσαν. 

Philo's use of the singular θάλασσαν is supported by S. August. 
de Gen. c. Manich. i. 12 (i. 652) : but, as elsewhere, it is an open 
question whether the plural is due to a Hebraizing revision of an 
original θάλασσαν, or the singular to a Hellenizing version of an 
original θάλασσας {^'^ψ!). 

Genesis i. 24. 
"ΕΐΑΓΑΓέτω Η ΓΗ ΨΥχΗΝ ζώςΑΝ ΚΑΤΑ reNOC τετρΛΠΟλΑ ΚΑΙ έρπετΑ και Θηρ'ια 


So Codd. A, χ, i6, 68, 72, 73, 77, 120, 121, 128, 129. Cod, 

76 ^ώσαιζ + και τα κτήνη κα\ πάντα τα €ρπ(τά της γης Ι Cod. 75 
om. κατά γίνος .... της γης : Cod. 55 ^^• '^°^^^ γίνος prior. '. 

Cod. 59 '^"'^ τετράποδα: Cod. 135 (γ) om. καί ante θηρία: Cod. 
Ε om. κα\ θηρία: Cod. 108 om. της γης: Codd. 15, ΐ7> ΐ9> 
2θ, 25, 37> 555 56, 6ι, 63, ιο6, 107, ιο8, 134, ΐ35» ζ> τ^? 

γης-\-κα\ τα κτήνη κα\ πάντα τα €ρπ(τά της γης : Cod. 74 ^^^ yV^ 
-{•κα\ πάντα τα ίρπίτά: poSt κατά γένος poster. Codd. 1 4, 3 1, 
32, 78, 79? I3I> tj add. και τά κτήνη κατά γίνος κα\ πάντα τα 
€ρπ€τά της γης κατά γένος : Cod. 25 add. κα\ πάντα τά ερπετά της 
γης κατά γένος : Cod. 83 add. και τά κτήνη κατά γένος : Cod. Ζ 
add. καΐ τά κτήνη κα\ πάντα τά ερπετά της γης κατά γένος. 


Philo de Mundi Opif. 21 (i. 14) l^ayayira η γη κτήνη και θηρία κηΐ 
€ρπ€τα. καθ* (καστον yevos : id. Z'^g. Alleg. 2. 4 (i. 69) i^ayayiro) 
.... ^i;pta=R. 

Tertull. c. Hermog. 22, p. 241, 'producat terra animam viventem 
secundum genus quadrupedia et repentia et bestias terrae 
secundum genus ipsorum ': ibid. 29, p.' 244 'vivam' is read 
for 'viventem/ and 'ipsorum' is omitted: S. Ambros. Hexaem. 
6. 2 (i. 114) adds after "bestias terrae" et pecora secundum 
genus et omnia reptilia,' and S. August, de Gen. ad litt. lib. 
imperf. 53 (iii. iii) and de Gen. ad litt. 2. 16 (iii. 151) adds in 
the same place ' et pecora secundum genus.' 

The variations in the text may probably be explained by the 
hypothesis that in very early times τίτράποΒα was substituted for 
the more usual κτήνη as the translation of ΠϋΠ3. That the two 
words were both found in very early times is shown by the fact 
that they both occur in Philo: and it seems less probable to 
suppose that the translators varied their usual translation of the 
Hebrew word than that τετράποΒα came in as an early gloss or 
targum to emphasise the distinction between the ' winged fowls ' 
of V. 21 and the land animals {τα χερσαία Philo i. 1 4) which were 
not created until the following day. This hypothesis that κτήνη 
rather than τβτράποΒα was the original word is confirmed by the 
quotation of the passage in S. Basil in Hexaem. Horn. ix. 2 (i. 81) 
e^ayayeTO) η γη "^νχην ζώσαν κτηνών κα\ θηρίων καί €ρπ€των, and in S. 
Cyril of Jerusalem Catech. 9. 13, p. 132 θηρία και κτήνη και epncTo. 
κατά yivos. This hypothesis also explains the other variants of the 
MSS. : for it clears the way for the further hypothesis that a 
scribe or reviser finding τετράποδα in some copies and κτήνη in 
others, and not noticing, or not knowing, that they were both 
admissible translations of the same Hebrew word, combined the 
phrases, adding after τής yrjs, or after κατα γίνος, either the words και 
τα κτήνη what would give the original of Augustine's quotation ' et 
pecora,' or the words κα\ τά κτήνη κα\ πάντα τα €ρπ€τά, which are 
found in many cursives and are evidently the basis of the Latin 
' et pecora secundum genus et omnia reptilia.' 

Genesis i. 26. 
ΠοίΗςωΜΕΝ ΑΝθρωποΝ κατ είκοΝΑ ΗΜετέρΑΝ και καΘ' όνιοίωςίΝ. 

So all Codd. 
Philo de Mundi Opif. 24 (i. 17) and de con/us. ling. 35 (i. 432) 


ηοίησωμ^ν ανθρωπον : id. de Mundi Opif. 24 (i. 16) ποιησωμ^ν 
ανθρωπον κατ (Ικόνα ημ^τίραν και καθ* όμοίωσιν : ihld. C. 23 • • • • 
Ίτροσίπ^σ-ημηνατο (Ιπων τω κατ €ΐκόνα το καβ' όμοίωσιν eh €μφασιν 
ακριβούς €κμαγ(ίον τρανον β-υπον έχοντος Ι id. de mutat. ηΟΜ. 4 (ί- 
583) 7Γ0ΐησωμ€Ρ ανθρωπον κατ ίϊκόνα ημ^τίραν'. id. de COnfus. ling. 
33 (ί• 4 3 ο) τ^οιησωμΐν ανθρωπον κατ (Ικόνα rjpcTepav κα\ καθ* 
Clem. R. i. 33 ττοιησωμβν ανθρωπον κατ (Ικόνα κα\ καθ* όμοίωσιν ημ€- 
Tepav : Barnab. 5 ποιησωμ€ν κατ (Ικόνα κα\ καθ* όμοίωσιν ημ^τβραν : 
id. 6 ποιησωμίν κατ* (Ικόνα και καθ" όμοίωσιν ημών τον ανθρωπον '. 
Justin Μ. Tryph. 62 = R.: Clem. Alex. Paedag. i. 12, p. 156 

ποιησωμ^ν ανθρωπον κατ €Ϊκόνα και καθ* όμοίωσιν ημών '. id. StrOM. 
55j Ρ• 662 . . . . κατ ΐίκόνα και όμοίωσιν ημ^τίραν. 

The majority of early Latin quotations (Tertullian, Cyprian, 
Hilary, Interpr. Irenaei, frequently Ambrose, Augustine) have 
' Faciamus hominem ad imaginem et similitudinem nostram ' ; 
the chief exceptions are S. Ambros. Hexaem. 6. 7 (i. 127) 
' ad nostram imaginem et ad similitudinem nostram ' : id. de 
Offic. I. 28 (ii. 35) 'ad imaginem nostram et secundum simili- 

The passage is critically interesting on several grounds : 
(i) The change in the position of the pronoun in Clement, 
Barnabas, and the early Latin Fathers can hardly be ascribed to 
accident or inexact quotation. The controversial importance of 
the pronoun is shown by the Gnostic controversies, Epiphan. 
Haeres. 23. i, 5. The critical importance of the passage lies in 
the indication which it furnishes of the existence of well-established 
readings outside the existing MSS. of the LXX., and of the small 
influence which early patristic citations exercised upon MSS. of the 

(2) The Hebrew has the pronoun with both words, and there 
is a trace of a Hebraizing revision of the LXX. in the Paris and 
Vatican MSS. of Origen m Joann. 13. 28 (iv. 238) κατ cluova ημ^τίραν 
κα\ Kaff όμοίωσιν ήμ€Τ€ραν : SO also in the Coptic, Sahidic, and some 
MSS. of the Arabic, and in the quotation in S. Ambros. Hexaem. 
6. 7 given above. But of this revision there is no trace in existing 
MSS. of the LXX. 

Genesis i. 27. 

Κλ! επο'ίΗςεΝ ό θεόο ton ΑΝθρωποΝ κατ eiKONA θεογ έπο'ίΗςεΝ ΑγτοΝ" 
ΛρςεΝ ΚΛί θΗλγ eno'mqeN Αγτογο. 

Cod. 135 (γ) '"^^ ανθρωπον -|- eV άκόνι αντον. 


Philo Leg. Alleg. iii. 31 (i. 106) κώ. ΙττοΙησ^ν ό β^ος τον άνθρωπον 
κατ ΐϊκόνα deov : id. de Somniis i. 13 (i. 632) βποίησεν .... αυτόν = 
R. : id. Qui's rer. divin. heres 33 (i. 496) €ποίησ€ .... avTovs= 
R. : id. told. 49 (i. 506) (ποίησα, yap, φησίν, 6 deos τον άνθρωπον, 

ουκ ΐΐκόνα αλλά κατ εΙκόνα, where it is conceivable that there 
may be an implied criticism of Wisdom 2. 23 κα\ eUova της 

Ιδίας Ιδιότητος (ποίησβν αυτόν. 

It is possible that the quotation in Philo i. 106, which connects 
κατ elKOva θίοΰ with the words that precede rather than with those 
that follow may go back to an earlier text, which followed the 
Hebrew in repeating the phrase κατ €ΐκόνα θεού \αυτου\ : so Aquila 

and Theodotion έκτισαν 6 θεός συν [Theod. om^ τον άνθρωπον iv ΐΐκόνι 

αύτον, iv (Ικόνι θβοϋ cKTiaev αυτούς. Of such a text, or revision, there 
is a trace in Cod. 135, see above, and in Euseb. Praepar. Evang. 
ii. 27. 3, where Codd. C Ε F G I (Gaisf.) have the same version 
as that of Cod. 135. 

Genesis i. 31. 
Ka! e'^tN ό Geoc ta πάντα oqa εποίΗςε ka! ίλογ καΛα ΛΊαν. 

Cod. 19 om. ό θΐός'. Codd. Ε. 15, 19, 2o, 25 (m), 75, 127, 129, 
om. τά. 

Philo de migrat. Abraham, 8 (i. 442) tiSey 6 θώς τα πάντα οσα €ποίη- 
σ€ν : id. z5td. 24 (i. 457) eldev .... λίαI/=R. : id. Qm's rer. divin. 
heres 32 (i. 495) ciSfv ό ^eos τά πάντα οσα εποίησαν και Ιδού άγαθα 

σφόδρα (so Mangey : some MSS. πάντα). 

Philo's reading σφόδρα is also the translation of Aquila and 
Symmachus, and hence may have been that of an earlier revision : 
'and it is confirmed as a current reading by Sirach 39. 16 τά %pya 
κυρίου πάντα δτι καλά σφόδρα : of its variant πάντα there is also a trace 
in Gregory of Nyssa Hexaem. p. 84 (ed. Migne Patrol. Gr. XLIV) 
who has Ιδου τα πάντα καλά λίαν: SO Philastrius 79) Ρ• 74 'ecce 
enim omnia valde erant bona.' 

Genesis ii. i. 

Και ςγΝ€7€λ€ςθΗςΑΝ ό oypANOC και η γη και hac ό koqmoc ΑγτωΝ. 

Codd. 19, 106, 107, ζ, σννετίλίσθη. 

Philo Leg. Alleg. ι. ι (i. 43) Cod. Medic. κα\ ίτΐλίσθησαν oi ουρανοί 
κα\ ή γη κα\ πας ό κόσμος αυτών, Codd. rell. , . ^ . η ψ\ καί πασαι αί 
στρατιαϊ αυτών. 

The plural οΐ ουρανοί is a closer translation of 0^^^ than the 


singular 6 ovpavos : but the latter is the almost invariable form in 
the LXX. : στρατιά {στρατιαί) and κόσμ,ος are both found as transla- 
tions of i< J^ but the former is more usual : hence it is probable 
that an, early form of the text had both ουρανοί and στρατιαί : cf. 
Neh. 9. 6, where the two words are used in combination to translate 
the same Hebrew words as here, κα\ σο\ προσκυνουσιν αί στρατιαί τών 

Genesis ϋ. 2, 3• 

Κα! ςγΝετέλεςεΝ ό θεόε εν τη ΗΜερΛ τη εκτΗ τΑ ερΓΛ Αγτογ λ έπο'ίΗςε• κλ! 
ΚΑτέπΛγςε τη ΗΜερΑ τίϊ εΒλοΜΗ λπο ΠΑΝτωΝ τωΝ ερρωΝ Αγτογ ωΝ έποίΗςε, και 
εγλόρΗςεΝ ό θεόο την πΜερΑΝ την εΒλοΜΗΝ και ΗρίΑςεΝ αυτήν δτε εΝ ΑγτΗ 
ΚΑτέπΑγςεΝ λπο ΠΑΝτωΝ τωΝ ερρωΝ Αγτογ ωΝ πρίΑΤΟ ό θεόο ποιπςΑΐ. 

So Codd. A, Χ. 15, 25, 68, 72, ΐ2θ, 128, 129, ΐ3ο> ΐ3ΐ• 
Codd. 59j 79 ^^^• ^^ before r.7 w^pa: Codd. 37, 108, ζ κατ€- 
τίανσ€ν•\•6 6eos : Codd. 16, 19, 3^j ^^^ κατίπανσ^ν 6 6ibs iv : 

Codd. 14, 20, 31, 32, 55, 57, 73, 76, 77, 78, 79, 83, 106, 

134, 135 κατ€πανσ€ν -\- €V. 

Philo Alleg. i. 2 (i. 43, 44) και σννίτβλζσ^ν 6 θώς iv τη ήμ^ρα 
τγι ζκτχι epyov αυτοΰ ο €ποίησ€ν, but immediately afterwards, δταν 

ονν \ίγη συνετίΚίσεν ^κτη ημίρα τα fpya, νοητίον οτι ου ττΧηθοί 
ήμερων παραΚαμβάν^ι TeXeiov δε αριθμόν τον e^ : idld. i. 6, 7 (i. 46) 
κατ(πανσ€ν ουν Tfj ^βδόμτ] ημ^ρα άπο πάντων των έργων αυτού ων 
€ΐτοίησε .... καΐ ηυλόγησ^ν ό uebs την ήμεραν την ίβ^όμην κα\ ηγίασεν 
αυτήν .... την ίβ^όμην ηυΚογησί re και ηγίασεν οτι iv avTfj κατί- 
παυσεν απο πάντων τών έργων αυτού Ziv ήρζατο 6 θεός ποιησαι \ id. 
de poster it. Cain. 18 (i. 237) και κατεπαυσεν ό θεός iv ttj ήμερα 
εβδόμτ] άπο πάντων .... ποιησαι ϊεβΒόμτ] . . . ποιησαι = R.J. 

Philo's agreement with the LXX. in reading iv ttj ήμερα ττ} Ικτη 
is remarkable because (i) most MSS. of the Masoretic text have 
''j;"'2^n Di*n 'on the seventh day,' (2) Aquila, Symmachus, and 
Theodotion have τη εβδόμη, (3) Barnab. 15 has συνετελεσεν τη [Cod. 

Sin. : Cod. Const. cV] ήμ^ρα τη εβδόμη καΐ κατεπαυσεν iv αύτη. The 

early Latin versions agree, as usual, with the LXX. : and the first 
indication of a variation is in Jerome ad loc. {Hebr. quaest. in libro 
Genes, p. 4, ed. Lagarde) ' pro die sexta in hebraeo diem septimam 
habet ' : the Syriac and Samaritan also agree with the LXX., and 
in two of Kennicott's MSS. T??^0 is absent. 

The balance of external evidence must be held to be in favour 
of ' sixth ' as opposed to ' seventh ' : but since both readings are of 



great antiquity, and also since, from the nature of the case, the 
external evidence for both readings is scanty, the question of the 
priority of the one reading over the other cannot be decided 
without regard to internal probability. It would be difficult to 
suggest a strong reason for changing ' sixth ' to ' seventh ' : but 
the use which Jerome /. c. makes of the reading ' seventh ' as an 
argument against Jewish Sabbatarianism suggests the probability 
of ' seventh ' having in very early times been changed to ' sixth ' to 
avoid the apparent sanction which would be given to working on 
the Sabbath, if God were stated not to have ceased working until 
the seventh day had actually begun. In other words, the Masoretic 
text is probably correct, and the reading ' sixth ' for ' seventh ' is 
probably the earliest instance of a dogmatic gloss. 

Philo's reading κατίπανσ^ν ό Ocos ci' ttj ήμερα is supported not only 
by several excellent MSS. of the LXX., but also by the Latin 
version in Aug. de Gen. ad Hit. 4. i, 20, 37 (iii. 159, 166, 172) 
' requievit Deus in die septimo ' : on the other hand, Irenaeus Vet. 
Interpr. 5. 28. 3 (i. 327) and Ambrose Epist. 44 (ii. 978) omit 
'Deus' : in Aug. c, Adimant, i (viii. 112) it is both inserted and 
omitted in the same chapter. 

Genesis ii. 4, 5. 

ΑγτΗ Η ΒιΒλοε τεΝεςεωο ογρΑΝΟγ κλ! thc δτε ereNcro η ΗΜερΑ επο'ίΗςε 
KYpioc ό θεόο ΤΟΝ ογρΛΝΟΝ ΚΑΙ ΤΗΝ ΓΗΝ κλΙ ΠΑΝ χλωροΝ ΑΓρογ προ τογ ρεΝε- 
ςθΑΐ επί thc thc και πλντα χόρτοΝ Αρρογ προ τογ ΑΝΑτεΐλΑΓ ^γ γ^Ρ εΒρεΣεΝ 
ό θεόο επί την γη ν και ΛΝθρωποε ογκ ην έρΓΛζεςθΑΐ αϋτην. 

So Codd. 68, 120. 

Cod. 75 W^9^ t ^ΤΓοίησβ : Cod. 129 ^ ήμερα § (ποίησε: Codd. A 
32, 38, 56, 57. 59. 72, 74. 107. 120, 128, 135 ετΓοίησε κύριος 
6 θ€Ος=Κ. : Codd. Χ, 14, 15, i6, 19, 20, 25 (m), 31, 37, 61, 
73. 75. 76, 77. 78, 79. 82, 83, 106, 108, 127, 128, 129, 131, 
134, tz, om. κύριος : Codd. X. (marg.), 19, 25 (m), 32, 57, 

61, 73. 78, 79. 83, 108, 127 (marg.), 131, rt, εβρεξεν κύριος 

6 θεός: Codd. III. 14, 15, 16, 20, 37, 38, 55, 56, 59, 68, 72, 

74) 75. 76, 77, 82, 106, 107, 120, 121, 128, 129 0??L κύριος 

= R.: Codd. AE 14, 15, 16, 20, 25 (m), 32, 38, 55, 56, 57, 
59. 72, 73. 74. 78, 79. ^3. 127, 128, 129, 131, 134, rt, 

ΐργάζεσθαι την yrjv. 
All early Latin versions, e.g. S. Ambros. m Luc. 15 (i. 1464), 


S. Aug. de Gen. c. Manich. 2. i (i. 663) read 'fecit Deus/ 
not ' Dominus Deus.' S. Aug. ibid, has ' cum factus esset 
dies quo fecit Deus/ which supports the readings of Codd. 

75, 129 i7^€/?a or 17 ημίρα. 

Philo Leg. Alleg. i. 8 (i. 47) avrr\ η βίβλος γ€ν€σ€ως ουρανού και γης 
οτ€ iyivcTo [Cod. Vat. eyeVoi/ro] : id. de Mundi Opif. 44 (i. 30) 
αντη ή βίβλοί .... άρατ€'Ίλαι=Κ. except that Kvpios is omitted 
after €ποίησ€ : id. Leg. Alleg. i. 9 (i. 47) ^ ημ^ρα €ποίησ€Ρ .... 
€ργάζ€σθαι την γην=Κ. except that κύριος is also Omitted, and 
την γην is read instead of αντην : these readings are repeated in 
the shorter citations which form the text of his commentary 
in the following page. 

Genesis ii. 6. 
ΤΤηγη he Ar^eBAiNGN εκ thc thc και έπότιζε παν to πρόςωπΟΝ thc thc. 
Cod. 16 από της y^S. 

Philo i. 31 =R. except από τψ γης : i. 249, 573 = R. 

από is more commonly used than ex as a translation of Γ?, and 
the uniform translation de terra shows it to have been the reading 
of the text from which the early Latin versions were made. 

Genesis ii. 7. 

Και επλΑςβΝ ό θεόο ton ΛΝθρωποΝ χογΝ λπο thc chC και €NeφγςHc;εN eic 
το πρόςωπΟΝ Αγτογ πνοην ζωίο και έπεΝετο ό ΑΝθρωποο είε ψγχΗΝ ζώςΑΝ. 

Codd. 15, 16, 18, 19, 31. 37. 59. 61, 68, 72, 75, 79. 82, 106, 

107, 108, 120, 121, ζ, χονν•\-\αβίύν. 

Philo de Mundi Opif. 46 (i. 32) tiikaa^v δ β€6ς ανθρωπον χουν λαβών 
από της γης κα\ €ν€ψυσησ€ν (Ις το πρόσωπον αυτού πνοην ζωής (but 

in the following commentary he interprets πνοην by πνεύμα, 
TO yap 4ν€φυσησΐν ουδέν ην eTcpov η πΐ^βΰμα θύον από της 
μακάριας κα\ €υΒαίμονος 4κάνης φύσεως άποικίαν την evOade στ€ΪΚά- 
μ€νον . . . ) : id. Leg. Alleg. i. 12 (i. 50) καΙ (πλασ^ρ .... ζωσαν 
= R. except that λαβών is added after χουν : (in the following 
commentary he lays emphasis on the use of προην instead of 
πνΐυμα, πνοην Se αλλ' ου πνεύμα Λρηκ^ν ως διαφοράς οϋσης' τό μ^ν 
yap πνεύμα Ρ€ΡΟηται κατά την Ισχυν κα\ ίυτονίαν και δύναμιν η δέ 
πνοή ως αν αύρα τις ioTi κα\ άναθυμίασις ηρεμαία και πραύα) '. id. 
Leg. Alleg. iii. 55 (i. 119) (ν^φύσησε yap (Ις το πρόσωπον αυτοΰ 
πνεύμα ζωής 6 θεός κα\ eyeveTO 6 άνθρωπος εις ψνχην ζωής', id. Quod 
del. pot. insid. 22 (i. 207) ενεφύσησεν eZs TO πρόσωπον αυτού πνεύμα 


ζωψ και iyiv€To 6 άνθρωπος ds ψνχην ζώσαν, where there is a 

following commentary on the use of πι/εν/ια) : id. Qm's rer. 
divin. heres 11 (i. 481) €νεφύσησε γάρ, φησίν, 6 ποιητής των όλων 
€ίί το πρόσωπον αντον πνοην ζωής και iyevcTO 6 άνθρωπος et? ψνχην 
ζώσαν (but the preceding remarks imply that either he read 
πνεύμα OF considered πνοψ to be its exact equivalent) : id. de 
planiat. Noe 5 (i. 332), and (ps.-Philo) de ?nundo 3 (ii. 606) 
€ν€πν€υσ€ γάρ, φησίν, ό θίος €ΐς το πρόσωπον αυτού πνοην ζωής. 

The variants which are found in Philo, ^νίπν^υσίν and ίνβφύσησίν, 
πνοην and πνεύμα, have parallels in the Latin versions, which show 
that they existed side by side in very early times. Augustine not 
only mentions the fact of variation between flavit or sufiavtt, and 
spiravit or inspiravit, and between flatum viiae and spiritum vitae, 
de Gen, ad litt, 7. 2 (iii. 211), Epist. 205 (146), ad Consent, c. 9 
(ii. 770), but himself also varies, cf. de Gen. ad litt. 6. i (iii. 197), 
ib. 7. 5 (iii. 213), de Gen, c. Manich. 2. 10, 11 (i. 668, 669), Upt'st. 
205 (146) ut supra, de Civit. Dei 13. 24 (vii. 346). He regards 
flatum as the more usual and correct word, and it is uniformly 
used by Tertullian, who also avoids spiravit and inspiravit, though 
he varies httwetn flavit, de Anima 26, p. 284, afflavit, Hermog. 26, 
31, pp. 242, 244, inflavit, adv. Marc. 2. 4, p. 383, and insufflavit, 
de Resurr. carnis 5, p. 328. Spiritum is found in Ambrose in 
Ps. cxviii. 10. 15 (i. 109 1), de bono mort. c. 9 (i. 405), (but elsewhere 
flatum), and in Hilar, in Ps. cxviii. p. 299. 

Symmachus and Theodotion have ^πνΐυσ^ν, Aquila has eVe^u- 
σησίν : and the hypothesis that the two readings coexisted in the 
earliest forms of the LXX. is supported by their combination in 
Wisdom 15. II, where there is an evident reference to this passage, 
δτί ηγνόησε τον π\άσαντα αντον κα\ τον €|υιπκ€υσαι/τα αντω ψνχην ivcp- 
γονσαν κα\ €)ΐφυσήσα>'τα πν^νμα ζωτικόν. It may be further noted that 
€μπν€ΐν is not elsewhere used to translate nsj, but that βμφνσαν is 
so used in Ezek. 22. 21: 37. 9: and that there is probably a 
reference to this passage in S. John 20. 20 κα\ τοΰτο άπων €ΐ'€φύ- 

σησεί' και \eyei αυτοΐς Xa/Sere πν^νμα αγιον : SO also JuStin Μ. Dial. 40 

uses του εμφυσήματος in reference to Adam's creation. 

The addition of λαβών to χουν, though probably no more than 
the epexegesis of a Hebraism, is probably very ancient, since it is 
found not only in Philo and many of the best MSS., but also in 
some early Latin versions, viz. Iren. Vet. Interp, 4. 20. i j(i. 253) 
' limum terrae accipiens ' : and in a more expanded form Iren. 5. 


15. I, i. 311 'et suinpsit Dominus limum de terra et finxit homi- 
nem': Philastr. 97, p. 93 'et accepit Dominus terrain de limo et 
plasmavit hominem': so Hilar, in Ps. cxviii. p. 299, Ambros. in 
Ps. cxviii. 10. 15 (i. 1091). Another epexegetical variant in early 
Latin was 'de limo terrae ' Tert. Hermog. 26, p. 242 (but else- 
where, e.g. adv. Marc. i. 24 p. 378 'limum de terra'): Augustine, 
though he sometimes uses the words 'de limo terrae,' not only 
speaks of them as an epexegesis of the Hebrew, but also states 
expressly that in the Greek MSS. which he used (as in the Sixtine 
text), λαβώί/ was omitted, de Civit. Dei 24. 13 (vii. 345) 'et formavit 
Deus hominem pulverem de terra .... quod quidam planius inter- 
pretandum putantes dixerunt Et finxit Deus hominem de limo 
terrae ' : after giving the reason for the interpretation he again 
quotes ' et formavit Deus hominem pulverem de terra, sicut Graeci 
codices habeni, unde in Latinam linguam scriptura ista conversa est.' 

Genesis ii. 8. 

Και εφγτεγςεΝ ό Beoc πΛρΛ^^ειςοΝ In ' Ehm κλτλ anatoAjIc. 

Codd. AE 16, 19, 20, 25 (m), 32, 55, 57, 59, 73, 77, 78, 79, 
106, 127, 128, 131, 135 [? not (r) Lag.], t, Kvpios 6 Scos. 

Philo Leg. Alleg. i. 14 (i. 52), de plant. Noe 8 (i. 334), de confus. 
ling. 14 (i. 414) και €φύτ€νσ€ν .... άρατο\άς=Έ^. 

The omission of κύριος is supported by the early Latin versions 
(except S. Aug. de doctr. Christ. 3. 52 (iii. 62) 'Dominus Deus/ 
elsewhere simply ' Deus '). But it would be difficult to frame any 
theory to account for the omission or insertion of κνριος in this 
part of Genesis. For example, Π^Π^ occurs eleven times in this 
chapter, viz. in vv. 4, 5, 7, 8, 9, 15, 16, 18, 21, 22 ; no existing 
MS. of the LXX. translates it in every passage : and all MSS. 
omit it in vv. 9, 19 : one small group of MSS., viz. 25 (m), 73, 
130 (t) agree in omitting it in vv. 4, 9, 19, 21 and inserting it 
elsewhere: Codd. 82 (f) and z, omit it in vv. 4, 5, 7, 8, 9, 19, 21, 
Cod. 106 agrees with them except as to v. 8, Cod. 108 (d) except 
as to vv. 4, 5 and Cod. 19 (h) except as to vv. 5, 8. There is a 
corresponding variety in the early Latin versions : but Π}Π^ is 
uniformly translated by Jerome wherever it occurs, except in v. 16, 
where the subject of l^^l is continued from the preceding verse. 


Genesis ii. 19. 
Και παν δ εαν έκΛλεςεΝ Αγτό'ΑλΑΜ ψγχΗΝ ζώςΛΝ τογτο όνομα Αγτω. 
Codd. ΑΕ, 38, 127, 129 αυτοί), Codd. 15, 18, 37, 61, 72, 75» 

106, 107, ΓΖ, αντοΐς. 

Philo Leg. Alleg. ii. 4 (i. 68)= R. : id. de mutat. mm. 9 (i. 588) 
δ hv CKoKeaev ό *Αδάμ, τούτο όνομα τον κΧηβίντος ην. 

Philo's reading τον κληθίντος is epexegetical : but it confirms the 
reading αύτον, which is further confirmed by the uniform ' ejus ' of 
the early Latin. 

Genesis ii. 24. 

"ENfcKGN τογτογ ΚΑΤΑλείψει ΛΝθρωποο ton ΠΑτέρΑ Αγτογ και την ΜΗτέρΛ και 
τιροςκολλΗθΗςετΑΐ πρόε την γυναίκα Αγτογ και gqontai οι ^γο eic ςΛρκΑ 

Codd. ΑΕ, 14, ΐ5, ι6, 31, 5^, 57, 59. 6ι, 73, 75, 7^, 77, 78, 

82, Ιθ6, 127, 1^8, 129, ^3° W, 131, Ι34, ΓΖ, μητίρααυτον: 

Codd. AD (Grab.) Ε 25 (m), 31, 59, 68, 83, 120, 121, rtz, 
πρόί την yvvaiKa : Cod. A τη yvvaiKi. 

Philo Zeg. Alleg. ii. 14 (i. 75)= R., but omits αντον after πατβρα : 
id. de Gigant. 15 (i. 2 72)=R. except iyivovTo yap for κα\ έσον- 
ται: id. Fragm. ap. Joan7i. Damasc. ii. 653, 654 = R. except 
δυο for oi δύο. 

The omission of αντον after πατέρα is supported by Codd. Κ BDZ 
and other authorities in Matt. 19. 5, and by Cod. D in Mark 10. 7, 
.and by the early Latin versions here, except only that Aug. de 
Gen. ad It'll. 6 (iii. 198) has ^patrem suum' The addition of αυτού 
to μητ€ρα is Supported by Codd. ^< DM and other authorities in 
Mark 10. 7, but has against it all good MSS. in Matt. 19. 5, and 
all the early Latin versions here. The reading τί} ywaiKi for προς 
την yvvaiKa is Supported by all uncial and most cursive MSS. in 
Matt. 19. 5, and by Codd. ACLN in Mark 10. 7 : also by the 
early Latin ' mulieri suae ' or ' uxori suae : ' it may be noted in 
reference to it that although the text of the quotation in the MSS. 
of Philo i. 75 is irpos τψ y., his commentary has the dative . . . 
ΊτροσκόΚΚαται κώ. evovrat ttj αΙσβησ€ΐ (which is his exegesis of τη yvvaiKi) 
.... ovK η yvvT] κολλάται τω avbpi. 


Genesis iii. 15. 

Και εχθρΑΝ θπςω ana ΜεςοΝ ςογ και ana ΜεςοΝ thc γυναικοο και ana 
ΜεςοΝ τογ ςπέρΜΑτόο ςογ και ana ΜεςοΝ τοΰ ςπέρΜΑτοο aythc* aytoc ςογ 
ΤΗρπςει κεφΑλΗΝ και ςγ ΤΗρΗςειο Αγτογ πτέρΝΑΝ. 

So Codd. ΑΕ, 14, ΐ5, ι6, ι8, ιρ, 2θ, 25 (m), 31, 32, 37, 3^, 
55, 56, 57, 59, 61, 64, 68, 72, 73, 74, 1^, 77, 7^, 79, 82, 
83, 107, 108, 120, 121, 128, 129, 130(0, 131, 134, 135 

(γ) : Cod. 75 '^"* €χθραν θησω άνα μίσον σον καί ανά μίσον του 
σπ€ρματο5 αυτψ' αυτός σου τοφησΐΐ την κίψάλην σου δε αυτοΰ την 
TTTepvav : Codd. Ι06, Ζ, τηρηστ] and τηρηστ]^, 

Philo Leg. A/leg. iii. 21 (i. 99)= R. except that he omits ανά 
μίσον before the second του σπέρματος: ibid. cc. 64-67 (i. 123, 
124) he has the same omission, and the following comments : 

(l) Trjpd be. OTL ουκ ecnev ' βχθραν θησω σοι κα\ τη γυναικί' αλλά ανά 

μίσον σου κα\ της 'γυναικός, the Hebraistic repetition of ανά 
μίσον being omitted : so also, a few lines below, το be ' ανά 
μ€σον του σπέρματος σον και του στΓ6ρματος αυτής ' €Ϊρηται πάΧιν 
φυσικώς. {2) Το δε ' αυτός σου τήρησα κεφαλήν κα\ συ τηρήσεις 
αυτοΰ πτερναν ttj μεν φωνή βαρβαρισμός εστί τω δε σημαινομενω 

κατόρθωμα : and, a few lines below, the commentary leaves no 
doubt that he read τηρήσει, since he explains it το be ' τηρήσει ' 
bio bη\oϊ' εν μεν το οίον bιaφυ\άζει κα\ bιaσώσει, έτερον δε το Ίσον τω 
επιτηρήσει προς άναίρεσιν. 
Justin Μ. Tryph. 102 και εχθραν θήσω ανά μέσον αυτού καΧ της 
γυναικός κα\ του σπέρματος αυτού κα\ του σπέρματος αυτής. 

The early Latin versions, e.g. Lucif. Calar. de S. Athanas. i. i, 
p. 67, ed. Hart., Ambros. de fug. saec. 7. 43 (i. 434) translate ^^"^ 
by 'observabit,' with the exceptions of Tert. de cult. fern. i. 6, p. 
152, Iren. Vet. Interp. 4. 40 who have ' calcabit.' In Cypr. Testi'm. 
2. 9, p. 74, the MSS. vary between ' calcavit ' (Codd. AB ; so ed. 
Hartel) and ' observabit ' ' observavit,' (Codd. LM ; so ed. Fell). 
Notwithstanding this variant the text of the LXX. seems to be 
certain : the difficulty is in the interpretation : almost all Hebrew 
scholars maintain that the Hebrew word requires some such 
translation as that of Aquila προστρί\Ιτει or Symmachus θλίψει : 
and in the only two other passages in which ^^^ occurs the 
LXX. render it by εκτρίβειν, Job 9. 17, and καταπατεΐν Ps. 138 
(139). 10. 


Genesis iv. 3. 

Και ereNexo Μεθ' HiwepAC ΗΝερκε Κλιν απο τΛν ΚΛρπωΝ thc thc θγςίΑΝ τω 

Cod. 72 κνρίω τω θίω, Codd. Ε, 1 29 τω θεω. 

Philo de sacrif. Abel, et Cain. 13 (i. 171) και iyivero μζ& ημίραί 
rjveyKe Κάιν από τον καρπού της-γης 8ωρορ τω Κνρίω. 

It is clear from the comments which immediately follow this 
quotation, and also from p. 176, that Philo read, as all MSS. of 
the LXX., από των καρπώι/ : the only other traces of the singular 
are in Tertull. adv. Jud. 5, p. 187, Lucif. Calar. de S. Athan. i. i, 
p. 67, ed. Hart. The substitution of ^ωρον for βυσίαν does not 
involve any change of meaning, the words being commonly inter- 
changed in the LXX. as translations of '^^P?, e.g. in the two 
following verses of this passage : and in p. 180 Philo himself uses 
βνσίαν in an indirect quotation of this passage τον Κάιν μ^β ημβρας 
φέροντος την θνσίαν : the early Latin versions vary here, in sympathy 
with the Greek, between ' munus ' {' munera ') Tert. adv. Jud. 5, 
p. 138, Ambros. de Cain et Abel i. 7 (i. 195), and ' sacrificium ' 
Lucif Calar. /Γ(9 S. Athan. i. i, p. 67. 

The reading of Codd. E, 129, τω θ^ώ, though not that of the 
quotation in Philo, is supported by Heb. 11. 4 πλείονα θνσίαν "λβίκ 
τταρα Κάιν προσην€γκ€ν τω θ^ω : but in I Clem. Rom. 4 there is the 
same difference as in the MSS. of the LXX. for Cod. A. reads 

τω θ(ω, Cod. C. τω κνρίω. 

Genesis viii. 21. 

"EfKeiTAi Η λίΛΝΟΐΛ τογ ΛΝθρώτιογ έπίΜβλώο έπϊ τλ πΟΝΗρΑ έκ ΝεότΗτοο 

Codd. 61, 78 των ανθρώπων, Cod. 83 οηΐ.ΐπιμΑως, Codd. ΑΕ, 1 5, 

20, 37. 55, 6ι, 64, 68, 74, 83, ΐ2θ, ΐ2ΐ, 129, 130, ΐ34, ζ, 

ΟΜ. αντον. 

Philo Quis rer. divin. heres 59 (i. 5i6)=R. but om. αντον: id. 
Fragm. ap. Joann. Monach. (ii. 663) opa yap ah eyKcxapoKTai 
πάντων ή 8ιάνοια eVt/xfXeoy. 

The omission of αντον is confirmed by the early Latin versions. 
The words iyK^xapaKTai η διάνοια in the fragment of Philo are 
remarkable as being an alternative translation of 2? "*>*^. which 


others rendered by τό π'^άα-μα της Kapdias (Euseb. Emis. m Cat. 
J^eg. = 'Procop. in Gen. p. 253, ap. Y\ud!^ Hexapla in loc). eyicet- 
ται βπιμΐλώς are a gloss rather than a translation, and neither word 
is elsewhere used to render '^'^l or its derivatives : and although 
^γχαράσσαν, like (γκζίσθαι, does not occur elsewhere in the LXX., 
yet the metaphor which it contains is in harmony with the other 
translations of i^J, e.g. πλάυ-σαρ (frequently), καταπλάσσ€ΐν (Jer. 

I. 5), κητασκ€νάζ€ΐν (Is. 45. 7, 9), χων^ύην (l Kings 7. 3 (15)). 

Genesis ix. 25. 

' ΕπιΚΛτΑρΑΤοε Χαναλν haTc οίκετΗΟ εςτΛΐ toTc Λλελφοΐο Λγτογ. 

Cod. 59 ^^• ''■"^^J Cod. 72 OPl. οίκβτης. 

Philo de SOhriet. 7 (i• 397) ^πικατάρατος Xavaav nais οΐκίτης boiXos 
δοίιλωρ ζ'σταί tols αδβλφοίν αυτού, but idid. II (i. 400) = R. 

The text of Philo, i. 397 E, incorporates a gloss, δούλος Βουλών, 
Λvhich is Aquila's translation of the Hebrew text here : it helps to 
show that παις οίκζ'της are to be taken together as in the Old Latin, 
Ambros. Έρ. 37 (ii. 931) ' servus domesticus erit fratribus suis.' 

Genesis ix. 27. 

ΠλΑτγΝΑί ό Geoc τωΊάφεθ κλι ΚΛτοιΚΗςΛτω eN to?c oTkoic τογ Σημ" κλι 
ΓεΝΗθΗτω Χαναλν παΪο Λγτογ. 

Codd. plur. τοΊς σκψωμασι τον [Codd. 1 5, 64, Ιθ6 οηΐ.^ '^W '• 

Codd. D, 19, 58, 59j 108 ^'o"^"' 'Χ.αναάν: Codd. AD, 31, 57, 
58, 59. 71. 73. 75. 78, 83. 108, 128, 129, 130, r, αΙτων. 
Codd. 14, 16, 18, 25 (m), 32, 38, 76, 77, 79, 131, 134, t, 


Philo de sohriet. 12 (i. 401)= R. except the last clause γ^^/σ^ω 
Xavaav δούλος αντοΐς. 

The texts from which the Old Latin versions Λvere made 
evidently varied between ο'ίκοις and σκψωμασι, the former being 
represented by ' domibus ' in Ambros. de Noe 32 (i. 276), and the 
latter by ' tabernaculis ' in Philastr. 121, p. 128. That Philo read 
οίκοις is clear from his comment on the word p. 402. 

Philo's reading αυτο\ς, which finds no support elsewhere, may be 
due to the transcriber and not to Philo himself, since in comment- 
ing upon it he substitutes the genitive, δονλον τον άφρονα των της 
άρίτης μ^ταποιου μίνων, ρ. 4^3 • 


Genesis xii. 1-3. 

Και enre κγριοε τω'ΆΒρΑΜ"Ε2&λθε εκ thc thc ςογ και εκ thc qvpreNe'iAC 
ςογ ΚΑΙ εκ τογ οίκογ τογ ΠΑτρόε ςογ και λεγρο eic την γην ην αν ςοι λε'ιΣω* 
ΚΑΙ ποιπςω ςε e\c εθΝοε ΜερΑ και εγλΟΓΗζω ςε κΆί ΜερΑλγΝω το όνομα ςογ 
καϊ εςΗ εγλορΗΜεΝΟΟ" και εγλορΗςω τογο εγλορογΝΤΛΟ ςε και τογε ΚΑΤΑρω- 
ΜεΝογο ςε ΚΑΤΑρΛςοΜΑΐ* και εΝεγλορΗθΗςοΝΤΑΐ εΝ ςοί ΠΛςΑΐ αι φγλΑΐ thc thc. 

Codd. A [D. Grabe], 15, 55, 74, 76, 129, 134 om. καΐ devpo: 
Codd. A [D. Grabe] Ε 14, 15, i6, 18, 25 (m), 57, 72, 73, 

11, 78, 19, 82, 128, 129, 131, 135 (r), t, ίστ} βύλογητός. 
Philo de migrai. Abraham, i (i. 436) και ehe .... τψ y^s = R. 

except (l) aneXOe for ξξίλθβ, (2) om. καϊ devpo, (3) (υλογητός for 
€ν\θΎημ€νος : ihid. 16 (i. 449) μ€γα\ννώ το ονομά σον : ibid. 1 9, 
20, 21 (i. 453> 454) ^^1} Ί^Ρ, Φν^'^^, €υλογητ09 .... (νλογησω, 
φησί, TOVS evKoyovvras ae και τυύς καταρωμίνονς ere καταράσομαι 
.... €ν€υλογηθήσονται ev σοι πάσαι αί φνΧάί της yrjs : id. Quis 
rer. divin. heres 56 (i. 513) βΓττε κνριος .... %θνος μ€γα=Κ. 

except πρόί for δευρο ds. 

Acts 7• 3 '^Qt eiVe irpos αυτόν, "Έ^ίΚθζ ck ttjs yrjs σου και €Κ της συγ- 
γενείας σου και deijpo els την γην ην αν σοι δίΐ|ω [Cod. D αττό της 
γης '. Codd. BD κα\ της συγγενείας σον : Cod. Ε add. post συγγε- 
νείας σον, και εκ τον οΧκον τον πατρός σου], 

Ι Clem. R. 10. 2 απήλθε εκ της γης σου .... της γής:^!^. except 
(ΐ) άπελθε for έξελθε, (2) om. κα\ 8ενρο, (3) ενλογηθησονται for 

The reading άπελθε, which was certainly in Philo's text, inasmuch 
as he comments upon it, p. 437, though not found in any MS. of 
the LXX. is supported by Clement, and by the fact that εξερχεσθαι 
is very rarely, and not once in the Pentateuch, used to translate 
"Φι, while άπερχεσθαι is frequently so used (18 times in Genesis): 
but in the quotation of this passage in Acts 7. 3 all the MSS. have 
έξελθε, which however is followed in Cod. D by άπό. 

The omission of κα\ δεϋρο is also supported both by Clement /. c. 
and by the fact that the words have no equivalent in the Hebrew : 
but they also are found in all MSS. of Acts 7. 3. They are an 
early and graphic gloss. 

The reading ενλογητός is emphasized by Philo i. 353 εστ] γάρ, 
φησίν, ενλογητος ον μόνον ενλογη μένος, distinguishing the former as a 
permanent and real quality, the latter as contingent on human 
voices and opinions. 


Genesis xiv. 14 (xvii. 23). 

ΉριθΜΗςε joyc i^'ioyc oiKoreNeTc Αγτογ τρίΑκοςίογο λεκΑ και οκτώ. 

Cod. 129 om. καί : Codd. D (Gr.), 1 4 δ/κα και οκτώ και τριακό- 
σιους : Codd. 15, 16, 18, 25 (m), 38, 55, 57, 59^ 7^, 77, 79ι 

82, 128, I3Ij 134, t, όκτω κα\ δίκα κα\ τριακόσιους Ι Cod. 78 
οκτώ κα\ Se/ca τριακόσιους. 

Bam. 9 '^c*• πίριίτβμ^ρ Άβρααμ €Κ του οίκου αυτού [Cod. C om. €Κ . . . . 
αυτοίη avdpas δίκα οκτώ [ita Codd. NC, cett. δ/κα καΐ οκτώ] κα\ 
[Cod. ρ. om.^ τριακόσιους. 

The first part of the quotation in Barnabas is a summary of 
Gen. 17. 23, the material point of the reference being not the 
mention of circumcision but the number of persons circumcised, 
upon which the writer founds an argument : τις ουν η δοθάσα αιτώ 

■γνωσις * μάθ€Τ€ οτι τους δεκαοκτώ πρώτους κα\ διάστημα ττοιησας Xeyei 
τριακόσιους, το δεκαοκτώ [Codd. bcn δβκα κα\ οκτού^' Ι δ/κα, Η οκτώ' 
(χ^ις Ίησοΰν [Cod. ^? om. I . . . οκτώ : Cod. C om. ^χίΐς *1η.~\' οτι δε 6 
σταυρός iv τω Τ ήμ^ΧΚβν €χ€ΐν την χάριν, \eyei και τριακόσιους. δηΧοΐ ουν 
τον μζν ^Ιησουν iv τοΊς δυσίν γράμμασιν κα\ iv τω iv\ τον σταυρόν, 'What, 

then, was the knowledge given to him?' Observe that he mentions 
the eighteen first, and then, with a pause, three hundred. In the 
eighteen, i.e. I=ten, Η = eight, you have (the initials of) Jesus 
(1ΗΣΟΥΣ). And because the Cross was to have its grace in (the 
form) T, he mentions also three hundred : he thus indicates Jesus 
in the two letters and the Cross in the third. 

This shows that in the text which Barnabas used (i) the numbers 
were probably expressed by the symbols ιητ ; (2) that, whether so 
expressed or written in full, τ or τριακόσιους came last. There is a 
similar variety in the MSS. in other enumerations of numbers, e.g. 
Gen. 5. 6, 7, 8, etc., and it is difficult to determine whether the LXX. 
originally followed the Hebrew in placing the larger number last 
so that the text of the uncial MSS. and R here is due to Hellenizing 
copyists, or followed the Greek usage in placing the larger number 
first, so that the text of Barnabas, and of the MSS. which agree with 
him, is due to a Hebraizing revision. 

Genesis xv. 5, 6. 

ΈΣΗΓΑρε he ΑγτοΝ εΐω και emeN Αγτφ, ΑΝΛΒλεψΟΝ Κη eic ton ογρΛΝΟΝ και 
ΑρΊθΜΗςοΝ Toyc ΑςτέρΑΟ ει ^ynhch εΣΑριθΜΗςΑΙ Αγτογε* και εΤπεΝ, ογτωο 


εςτΛΐ TO ςπέρΜΑ ςογ* και επίςτεγςεΝ "ΑΒρΑΜ τω θεώ και ελορΊςθΗ Αγτφ είε 


Codd. 15, 19, 31, 38, 61, 72, 77, 108, 129, 135 (Ο, ζ. ^^• ^ι?: 

Codd. 19, 108 eniarevae 8e for και €πίστ€νσ€. 

Philo Ζ<?^. Alleg. iii. 13 (i. 95) i^riyay^v aWov (ξω και einev, am/SXe^oi/ 
6ts τον ovpavov και αρίθμησαν Tovs aarepas '. id. Qu/s rer. dlVltl. 
heres 15—19 (i. 483—486) (15) i^rjyayev αυτόν βξω κα\ etnev 
άνάβλζ•\1/ον els τον ονρανόν .... (ΐ 6) (ξηγαγ€ν αυτόν ^ξω {dl's) .... 
(17) άνάβλ^λίτον els τον ουρανον κα\ αρίθμησαν tovs aarepas iav 
bvvηθrjs ζ^αριθμησαι avTOvs' ουτωε ίσται το σπίρμα σου .... (19) 
iev δε το φάναι) Χογισθήναι την πίστιν els δικαιοσννην αυτω Ι id. de 
migrat. Abraham. 9 (i. 443) l^τίστevσev * Αβραάμ τω θ€ω : id. de 
mutat. nomin. 33 (i. 605) e^ΐίστevσe he 'Αβραάμ τω dea κα\ (λο-γίσθη 
αυτω els δικαιοσύνην. 

Rom. 4. 3 (τί yap η γραφή Xeyei) eπίστeυσev Be ^Αβραάμ τω θεω καΐ 
€λογισθη αυτω els δικαωσύνην (so Codd. ^< ABC αι. *. Codd. DFG 
om. δε). 

Rom. 4. 18 {κατά το elpημevov) oZtws eorai το σπ€ρμα σου. 

Gal. 3. 6 καθω5 \\βρααμ επί'στβυσεϊ/ τω ^βώ κα\ €\ογΙσθη αυτω els 

James 2. 23 {και ΙττΚηρωβη η γραψη ή λ€γονσα^ eπiστeυσev de * Αβραάμ 
τω ^βώ κα\ ίΧογίσθη αυτω els δικαιοσύνην. 

1 Clem. Rom. 10. 6 e^rjyaye de [Cod. A om. be] δ eebs τον ^Αβραάμ 
και einev αυτω' άνάβΧ€•\Ιτον els τον ουρανον και αρίθμησαν Toi)s aarepas 
el δυνηση €ξαριθμήσαι αυτούς' oυτωs €σται το σπέρμα σου' 4πίστενσεν 
be Άβρααμ τω θεω κα\ εΧογίσθη αυτω els bικaιoσύvηv. 

Justin Μ. Dial. 92 Ιπίστενσε be τω θεω ^Αβραάμ και ίΧογίσθη αυτω 
els bικaιoσύvηv : z'dl'd. 119 {αν yap τρόπον eKelvos TJj φωνή του θεού) 
€πιστευσ6 και εΧογίσθη αυτω els δικαιοσύνην. 

Philo's omission of brj after άνάβΧεψον is confirmed by i Clem. 
Rom. 10. 6 : which also agrees with Rom. 4. 3, James 2. 23, 
Justin. M. Dial. 92 in reading επίστενσε be. Though the variation 
is exegetically unimportant, the consensus of five early quotations 
as against all existing MSS. except 19 (Cod. Chisianus) and 108 
(=Cod. Vatican. 330, which forms the basis of the Complutensian 
edition) is a remarkable testimony to the text which those MSS. 

The common origin of all the quotations is indicated by the fact 
that they agree in translating the active, C.iF'?-» '^^ counted/ by the 
passive εΧαγίσθη. 


Genesis χ v. 13, 14. 

ΓίΝωςκωΝ pn^qh oti π^ροικοΝ eqrAi το ςπέρΜΑ ςογ eN ρπ ογκ ίλ'ΐΑ και 
λογλώςογςΐΝ Αγτογα και ΚΑκώςογςΐΝ Αγτογο και ΤΑπείΝώςογςΐΝ Αγτογε τετρΑ- 
κόςίΑ Ι'τΗ• τό λε ΙΘνοο ω gan λογλεγςωςι κρίΝω ερώ- ΜετΑ λέ ΤΑγτΑ είελεγ- 
ςοΝΤΑΐ ώλε ΜετΑ ΑΐτοςκεγΗΟ πολλΗΟ. 

Cod. 72 eV γη αλλότρια: Cod. Α, κακώσουσιν avrovs κα\ δουλώσουσιν 
αυτούς: Codd. Χ, 37> ^Ι» Ι°7' Ϊ0^5 2, omit avTovs after κακώ- 
σονσιν: Codd. 1 9, 725 81, omit κα\ ταπ. αυτούς I Codd. Χ, 1 9, 
37» 75) 77? 106, 108, 129, 130, ζ, €τη τετρακόσια: Codd. 14, 

18, 19, 25 (m), 32, 57» 73. 75» 77> 78, 79. 1313 t> '^"^ ^^ 


Philo ^///j rd-r. <//ζ;ζ>/. keres 54 (i. 511) -γινωσκων .... t5ia, = R. : 
/(52ί/. 55 (i. 512) το be ίθνος .... πολλ^9, = Κ. 

-i4r/$• 7. 6 εσται τό σηίρμα αυτού [Cod. Χ σον\ πάροικορ iv yrj αλλό- 
τρια κα\ δουλώσουσιν αυτό [Cod. D αυτούς~\ κα\ κακώσουσιν [Cod. C 
adds αύτοΊ €τη τετρακόσια' και το (θνος, [Cod. C τό 8e eOvosl ω iav 
δουλίύσουσιρ [Codd. i< BE al. Βουλεύσωσιν^ κρίνω εγώ, 6 θεός είπεν, 
κα\ μετά ταντα εζελείισονται (κα\ λατρενσουσίν μοι εν τω τόπω τοντω^. 

The critical interest of the passage Ues chiefly in the evident 
tendency to harmonize the LXX. text and that of the Acts, which 
is shown (a) in the MSS. of the LXX. (i) in the substitution of 

αλλότρια for ουκ Ιδια, (2) in the omission of κα\ ταπεινώσουσιν 

αυτούς, (3) in the variant κα\ τό for τό Βε : (δ) in the MSS. of the 
Acts (i) in the substitution of σοΟ for αυτού, which is unquestionable, 
inasmuch as αύτω both precedes and follows, (2) in the addition 

of αυτούς and αυτό tO δουλώσονσιν and κακώσουσιν, {3) poSsibly in the 
variants τό δε for kcu to and δουλεύσωσιν for δουλεύσουσιν. 

The quotation of the passage in Clementin. 3. 43, p. 48 = R. 
except in omitting αυτούς after κακώσουσιν : but in the continuation 
of the quotation it reads μετ ειρήνης with AX, 14, 15, 19, 25 (m), 
32, 37, 38, 555 57» 73» 74» 76, 77» 7^, 106, 107, 108, 129, 134, 
rtz, and confirms the view that these words should be substituted 

for the εν ειρήνη of R. 

Genesis xviii. 1-3. 

"ΩφθΗ λέ Αγτω ό θεόο npoc τη λργΐ τη ΜαμΒρη ΚΑθΗΜεΝογ Αγτογ επί thc 
θγρΑΟ THC QKHNHC Αγτογ ΜεςΗΜΒρ'ΐΑΟ• ΛΝΑΒλέψΑΟ λέ ToTc οφθΑλΜοΐε Αγτογ 
elhe ΚΑΙ ίλογ τρεΐο ΛΝλρεο είςτΗκειςΑΝ έπΑΝω Αγτογ• κα! \L•ώt< προςέλρΑΜεΝ 
eic ςγΝΑΝΤΗςίΝ Αγτοΐο λπο thc θγρΑΟ thc qkhnhc Αγτογ και προςεκγΝπςεΝ 


επί ΤΗΝ ΓΗΝ ΚΑΙ etne Κγριε, ei ΑρΑ εγροΝ χλρίΝ εΝΑΝτίοΝ ςογ, μη πΑρέλθπε 
ΤΟΝ πΑΪλΑ ςογ. 

Cod. 25 (πι) προς τγι θύρα : Cod. 82 eVt rfj θύρα : Cod. 106. om. 
αντον after σκηνής. 

Justin Μ. Dml. 86 προς rrj δρνϊ rrj Μαμβρ^ : z'dl'd. 126 ωφθη .... 
μ€σημβρίας=Κ. eXC. (ΐ) καθημενω, (2) <9/;ζ. αντον after οφθαλμοίς, 
(3) (Tvvedpapev for προσίδραμ^ν'. ihld. 56 ωφ^τ; .... eVt r^v γ^ι/ 
/cat €1776= R. except (l) errt T7 ^i^p?, (2) i>/?Z. αυτοί) after σκηρης 
and after οφθαλμοίς, (3) συν^^ραμ^ν for προσ€^ραμ€ν. 

At the end of this quotation in c. 56 the text of Justin goes on κα\ 
τα λοιπά μέχρι τον "Ωρθρισζ δβ, ί. e. the intervening words are omitted 
as far as c. 19. 28. But since, lower down in the same chapter, 
p. 278 b, Justin excuses himself from repeating some of the inter- 
vening words on the ground that they had been written down 

before, ού yap γράφαν πάλιν τα αντα των πάντων π poycy ραμμένων boKfi 

μοί, it is clear that the omission is due to the copyist 

Genesis xviii. lo. 

ΈπΑΝΑςτρεφωΝ Η2ω npoc ςέ κατλ ton ΚΑΐροΝ τογτοΝ είο ώρΑΟ και είει 
γίοΝ ΣλρρΑ Η ΓΥΝΗ ςογ. 

Codd. 14, 16, 18, 25 (m), 38, 57, 73, 77, 7», 79. 128, 131, 

135 (γ) (HP) + t άναστρίφων. 

Philo de migrat. Abraham. 22 (i. 456) = R. : de Abrah. 25, (ii. 20) 
Ιπανιων ηξω προς σ€ κατά τον καιρόν τοντον €ΐς νεωτα και e^ei νίον 
Σάρρα η yvvq σον. 

Rom. 9. 9 {lπayyiklaς yap b λόγο? οντος'^ κατά τον καιρόν τοντον 
€λ€νσομαι κα\ Ζσται Tjj lappa νΙός. 

The use of the classical eU νέωτα, 'next year,' is remarkable as a 
translation of Π*Π ny3 (which occurs infra c. 14, and 2 Kings 4. 16, 
1 7, where it is rendered ως η ωρα ζώσα). There is no trace of either the 
reading or the interpretation in the MSS. of the LXX. or in the 
early Latin versions : and it is a probable inference that the writer of 
the treatise de Abrahamo, whether Philo or another, had access to 
a revised, and otherwise unknown, edition of the LXX. : so in the 
same treatise, c. 32 (ii. 26), iepehv is substituted for πρόβατον in Gen. 
22. 7, 8. 

The quotation in Rom. 9. 9 is partly from v. 9, partly from 
V. 14, but not exactly from either. 


Genesis xviii. 20-23. 
Erne λε κγριοο κρΑγΓΗ ΣολοΜωΝ και foMoppAC πεπλΗθγΝΤΑΐ πρόο iwe και 


ΑγτωΝ την ερχοΜεΝΗΝ npoc Me ςγΝτελογΝΤΑΐ" ei λέ-ΜΗ Γνλ ΓΝω" και Λποςτρε- 
ψΑΝτεο έκεϊθεΝ οί ΛΝλρεο ηΛΘον είο ΣόλΟΜΑ* ΆΒρΑΑΜ he ετι ην έςτπκώε 
εΝΑΝτΙοΝ κγρΊογ ΚΑΙ err'iQAC ΆΒρΑΑΜ εΤπε Μη ςγΝΑΠολέςπο λ'ιΚΑΐΟΝ ΜετΑ 
ΑςεΒογο και εςτΑΐ ό λίκΑΐοο ώε ό ΛςεΒΗΟ. 

Codd. AD, 15, 59? ^^j 7 2, 82, 120, 121 Om. προς με after πεπλη- 

θννται: Codd. 14, 16, 18, 19, 25 (m), 57, 13, 77, 7^, 79. 

108, 128, 131, t ot άνδρες εκύθεν I Codd. AD, 3 1, 37, 75, 76, 
106, 107, 108, Ζ om. €Ti before ην : Cod. 132 εστως ην. 

Philo de Cherub. 6 (i. 142) eVt, γάρ, φησίν, ην εστηκως εναντίον κυρίου'. 
id. de Somniis 2. 33 (i. 688) (^Αβραάμ) εστίν εστως εναντίον κυρίου: 
id. de poster. Cain. 9 (i. 231) εστως ην εναντίον κυρίου κα\ εγγίσας 

Justin Μ. ΌίαΙ. 56. ρ. 278 είπε 8ε κύριος . . . . 6 άσεβης=Κ. except 

(ΐ) om. irpbs με after πεπΧηθυνται, (2) ot άνδρες εκείθεν for εκείθεν 
οί άνδρες, (3) om. ετί before ην. 

Genesis xviii. 27. 

Και Αποκριθείε'ΑΒρΑΛΜ εΤπε, ΝγΝ ηρΙαμην AaAhqai npoc ton κγριΟΝ Μογ, 
έρώ λέ είΜΐ γη και ςπολόε. 

Codd. 19, 59 ^^• τό''• Codd, 76, 129 τ6ν θεόν : Codd. ADE, 
14, 15, 16, 18, 19, 25 (m), 56, 57» 59. 61, 68, 73, 7^, 79» 

82, 108, 120, 121, 128, 13Ι5 135 W> t, om. μου. 

Philo Qm's rer. divin. heres 7 (i. 477) εγγίσας, yap, φησίν, ^Αβραάμ 
είπε Ννι/ ηρξάμην ΧαλεΙν προς κυριον, εγω δε εΙμι yrj και σποδός '. id. 
Quod Deus immut. 34 (i. 296) {ευθύς εψω) ^ψ και τεφραν 

1 Clem. Rom. 1 7 εγω δε εΙμι yrj καϊ σποδός. 

The text of Philo i. 477 is sufficiently supported by the MSS. of 
the LXX., and by its agreement with the Hebrew, to be probably 
correct, with the exception of εγγίσας for αποκριθείς ; but it may be 
almost certainly inferred that εγγίσας existed in the text which Philo 
used, and that it is not a mere accidental transfer of phrase from 
V. 23, from the fact of his laying stress upon it in introducing the 
second of the above two quotations i. 296 κα\ yap Άβρααμ Ιγγιστα 
τω θεω eauro»' ττοιήσα?, ευθύς εγνω κ.τ.λ. The USe οί τέφρα for yr] in 


the second quotation is less probably correct, because the word 
does not occur in the LXX. except in the Apocryphal Books. 

Genesis xxi. lo. 

Ka! eme τφΆΒρΑΛΜ^ΈκΒάλε την πΛίλίςκΗΝ τΑγτΗΝ και τον γίοΝ aythC ογ 
ΓΛρ ΜΗ κλΗροΝΟΜπςει ό γίόο THC πΛίλίςκΗΟ TAYTHC ΜετΑ τογ γίογ ΜΟγ 

Codd. AD 15, 19, 20, 3Ι: 32, 55. 56, 68, 74, 76, 11, 83, 
108, 120, 121, 129 και €Ϊπ€=Κ.: Codd. χ, 14, ι6, ι8, 
25 (m), 38, 57, 59: 7ΐ, 73, 75, 76, 7^, 19, 82, ιο6, ιογ, 
128, 130 (t), 131, 134, 135 (γ), ζ, om. και. 

Codd. AD, Χ, 15, 55, 56, 57, 68, γι, 74, 75, 76, ιο6, icy, 

120, 121, 131, 134, Ι35+Ζ την παώίσκην ταυτην : Codd. 14, 

16, 18, 19, 20, 25 (m), 31, 32, 38, 59> 73, 77, 78, 82, ιο8, 

128, 129, t, ^^^^' Τ"-^τψ• 

Codd. D, Χ, 59, 72, ιο6 + ζ, om. μη post γάρ: Codd. 

cett. = R. 
Codd. ι8, 2θ, 25 (m), 32, 55, 131, ΐ34, ΐ35 W κληρονομηση : 

Codd. cett. = R. 
Codd. Ill, 68, 108, 120, 121, opi. ταύτης: Codd. cett. = R. 

Philo de Cheruhim 3 (i. 140) Xeyet δε αντικρνς €κβα\€Ίρ την Ίταώίσκην 

κα\ τον νίυν. 
Gal. 4. 30 e/c/3aXe την τταώίσκην [Cod. Α add. ταντην~\ κα\ τον viov 

αντψ' ου γαρ μη [Codd. FG, 37, Om. μη\ κληρονόμησα [ita Codd. 

δ< BDE αι.: Codd. ACFGKL αι. κληρονομηστ]] 6 vl6s τής 

παώίσκης μ€τα του νίοΰ της ίΚζνθβρας [Codd. DEFG αΐ., add. μου 

Justin Μ. Dm/. 56. ρ. 276 κα\ ewre .... 'Ισαάκ = R. except om. κα\ 
before elne, and μη after ου. 

It is uncertain here, as elsewhere, whether the omission of καί 
before ehe is due to the Hellenizing tendencies of the copyists, 
or its insertion is due to a Hebraizing revision of the text. 
The latter is the more probable hypothesis, because there are other 
instances in Genesis in which the LXX. translators seem to ignore 
this use of ), i. e. as introducing an apodosis or virtual apodosis : 

e.g. 3. 6 διανοιχθησονταί for κα\ diav., 1 3. 9 εγώ ct? δβ^ιά for καί «γώ 
(Cod. 75 ή iyo>, Codd. E, 14, 16, 18, 31, 57, 73, 128 iyio de). 

The omission of ταύτην in some MSS. of the LXX. and its 
insertion by Cod. A in Gal. 4. 30 are probably harmonistic. The 


same hypothesis will account for its omission in the Latin versions 
quoted by Ambrose and Augustine (ap. Sabatier) : and the harmonistic 
tendency is certainly shown in the addition μου Ισαάκ. 

Genesis xxii. i, 2, 11, 12. 

V. I ΚΑΙ ereNexo mgta ta ρήματα ΤΑγτΑ ό 9eoc έπε'ιρΑςε ton ΆΒρααμ και 
t?n€N ΑγτφΆΒρΛΑΜ ■ ΑΒρΑΑΜ" ΚΑΙ €?π€Ν Ίλογ ερώ. ν. 2 ΚΑΙ eme ΛαΒε τον 

γίοΝ ςογ ΤΟΝ ΑΓΑΠΗΤΟΝ on ΗΓΑΠΗςΑΟ τον ΊςΑΑΚ .... ν. II ΚΑΙ €ΚΑλ€ς€Ν 

ΑγτοΝ ΑΓρελοε κγρΊογ εκ τογ ογρΑΝογ και εΤπεΝ ΆΒρΑΛΜ ΆΒρΑΑΜ* ό he emeu 
ίλογ έπώ. ν. 12 καϊ εΐπε μη έπιΒΑλπε την χεΤρΑ ςογ επί το ΠΑίλΛριΟΝ ΜΗλέ 


ν. 1 Codd. χ, 7ΐ> 14) ^3 ^π€ίρασ€=Κ.: Codd. cett. eWpa^ij/. 
Codd. 19, 20, 25, 31, 32, 56, 68, 71, 74, 75, 83, 107, 120, 121 

€tn€V ανΓω:=Κ. : Codd. CCtt. eiTre npos αυτόν. 
Codd. 19, 31, 38, 61, 68, 71, 74, 76, 79, 83, 106, 107, 120, 

121, 128, ζ καϊ €Ϊπ€ν Ίδου = Κ. : Codd. cett. 6 δβ ehev Ί8ού. 
V. 11 Codd. 14, 16, 18, 25 (m), 38, 57, 77, 79, 128, t λίγων 

post ουρανού I Codd. cett. κα\ €ln€V=K. 

Philo de Somniis 1. 34 (i. 650)=R. except (i) eneipaCe for 
€7Γΐίρασ€, (2) προς αυτόν for αυτω, (3) ό δε €Ϊπ€ν for κα\ ζϊπίν Ίδου in 
V. Ι, (4) Xeycov for και ehev in V. II. 

It may be noted that the text of Philo agrees throughout with 
that of Codd. 14, 16, 18, 57, 77, 130 (t), and differs throughout 
from that of Codd. 71, 74, 83 : that it agrees in three out of four 
cases (l) with Cod. 25 (m) ίπ^ίραζςν, 6 δε ehev, λίγων, (2) with 
Codd. 38, 79, 128 (πείραζαν προς αυτόν, λίγων, (3) With Codd. 
129, 134, 135 επείραζεν, προς αυτόν, 6 δε einev. 

Genesis xxii. 3? 4• 
Και ΗλθεΝ έπϊ τον τόποΝ on εΤπεΝ Αγτω ό θεόε τη ΗΜερΛ τη τρ'ιτΗ 

ΚΑΙ ΑΝΑΒλέψΑΟ'ΑΒρΑΑΜ TO?C ΟφθΑλΜοΤο ΑγΤΟγ ε?λε ΤΟΝ τόποΝ ΜΑΚρόθβΝ. 

Codd. 19, 37> 1^) ^2, ιο6, ΐ34) ζ εΙς τον τόπον: Codd. cett. ίπΐ 
τον τόπον =R. 

Philo de poster. Cain. 6 (i. 229) Άβρααμ ελθων eU τον τόπον ov 
ίΐπεν αυτώ 6 θεός τη τρίτη ήρίρα άναβλεψας δρα τον τόπον μακρόθεν Ι 

(the following words ποΊον τόπον ; ap' εΙς ον ήλθε ; show that' he 
certainly read cis τ6ν τόπον) : de migrat. Abraham. 25 (i. 457) 
{όταν) επ\ τον τόπον ον εΐπεν αυτω 6 βεος τη ήμερα τη τρίτη πάρα- 



yeinjTai '. ihld, 30. i. p. 462 [άμφότίροι άνηΧθον) erri τον τόπον ov 
€Ϊπ(ν 6 Seas : de Somniis i. 11 (i. 630) ηΧθ^ν ds τόν τόπον ov (ΐπ€ν 
αντώ ό θ(ός. και άναβλίψας τοΐς οφθαλμοΐς αυτού eide τον τόπον 

Philo's testimony is evenly balanced between eVl τον τόπον and eh 
τον τόπον: and between the quotations in i. p. 229 and i. p. 457 
there is the further difference that whereas the former connects tjj 
τρίττ) ημίρα with άναβΚί-^ας, as in the Hebrew, the latter connects it 
with the preceding clause. A presumption in favour of the former 
having been the current Alexandrian reading is afforded by 
the repetition of Philo's quotation in Clem. Alex. Strom. 5. 11 

p. 690, ed. Pott. 6 * Αβραάμ (Χθων els τον τόπον ov ηπ^ν αντω 
6 θίος τί} τρίττ] ημ€ρα άναβλίψας Spa τον τόπον μακρόθεν. The early 

Latin verss., on the other hand, clearly connect r^ τρίττ] ήμβρα with 
the preceding clause: Ambros. de Cain, et Ab. 1. 8 (i. 197); de 
Ahrah. i. 8 (i. 305); so Jerome Hebr. Quaest. p. 33, ed. Lagarde. 

Genesis xxii. 16, 17. 

Κλτ έΜΛγτογ ωΜοςΑ, Aefei K-ypioc, ογ είΝεκεΝ έπο'ίΗςΛΟ το ρΗΜΑ τογτο και 
ογκ έφείςω τογ γίογ ςογ τογ αγαττητου h\ εμΈ, η μην εγλΟΓωΝ εγλορΗςω ςε 
ΚΑΙ πλΗθγΝωΝ πλΗθγΝω το ςπερινίΑ ςογ ώε τογε ΑςτέρΑΟ τογ ογρΑΝΟγ και ώε 
ΤΗΝ ΑΜΜΟΝ ΤΗΝ πΑρΑ ΤΟ χεΐλοο THC θΑλΑςςπο. 

Codd. AD χ, 75j ι 35 " f^v^- 

Philo Leg. Alleg. 3. 72 (i. 127) = R. (except the Attic h^Ka, for 

the Ionic άν^κ^ν, but ibid. p. 129 eheKo). 
Heb. 6. 13, 14 &μοσ€ν κα& eavTov λβγων cl μην ίυλογων ευλογήσω 
σ€ κα\ πΧηθννων πΧηθυνω σβ [Codd. KL αΐ. η μην^. 

Genesis χχν. 21-23. 

Έλέετο λέΊςΑΛΚ κγρΊογ περί 'ΡεΒεκκΑΟ thc tynaikoc Αγτογ δτι ςτεΤρΑ ην. 
έπΗΚογςε λέ Αγτογ ό θεόε και ςγΝελΑΒεΝ ρν ρΑςτρί 'ΡεΒεκκΑ η γΥνη Αγτογ* 
έςκίρτωΝ λέ τα ΠΑΐλίΑ εΝ ΑγτΗ' εΤπε λέ, ει ογτω μοι Μέλλει ρΝεςθΑΐ Γνα τί 
Μοι τογτο ; έπορεγθπ λέ πγθέςθΑΐ πΑρΑ κγρΊογ* και ε?πε κγριοε ΑγτΗ, λγο 
εθΝΗ εΝ ΓΑςτρΊ ςογ είςί και λγο λΑοί εκ thc κοιλ'ΐΑε ςογ λίΑςτΑλπςΟΝΤΑΐ• και 
λΑΟΟ λΑΟγ γπερέ2ει και ό Με'ιζωΝ λογλεγςει τωελΑςςοΝί. 

Codd. ΑΕ, 15, 30, 31, 59. 82, ιο6, 107, 129, 130, ΐ34, ζ 
eSetro: Cod. 75 fi'ptV, Codd. 31, 135 κνριον, Codd. 19, 108 
του κυρίου: Cod. 72, Ζ, om. κυρίου: Codd. 106, ζ υπηκουσ€ 


«e: Codd.EX, 16,18, 25 (m), 57, 59, 72, 73, 79,128, 131,1 
αντω 6 ueos : Codd. ADE, 14, 15, 16, 18, 20, 25 (m), 30, 31, 
38> 55, 57, 59, ^8, 72, 73, 75, 77, 78, 79, 82, 83, 106, 107, 
120, 121, 128, 129, 130 (t), 131, 134, 135, ζ ?λαβ€ν: Codd. 
19, 32, 56, 71, 74, 76, 108 συι/6λαβίΐ' = Κ. : Codd. ADE, 15, 
16, 18, 25 (m), 30, 32, 56, 57, 59, 72, 75, 79, 82, 83, 106, 
107, 128, 130 (t), 131, 134, 135, ζ eV Tfi γαστρί: Codd, 15, 
72, 82, 106, 107 eVrt. 

Philo Leg. Alleg. iii. 29 (i. 105) δύο ^θνη iv rfj γαστρί σον €ση και 
δυο \αο\ εκ της κοιλίας σον 8ιαστάλησονται κα\ λαό? Χαον vnepe^a κα\ 
ό μείζων δουλενσει τω ελάσσονι: id. de sacrif. Abel, et Cain. 2 
(i. 164) δύο Wvr] ev τη γαστερι σον εστί .... κα\ δύο Χαοι εκ της 
κοιλίας σον διασταΧησονται. 

Rom. 9. 12 ό μείζων δονλενσει τω ίλάσσονι. 

Barnab. 13 eSetro Se 'Ισαάκ ττερί ^Ρεβέκκας της γνναικος αντον οτι στείρα 

ην' κα\ σννελαβεν [so Codd. ^ζ and all Others, except Cod. C, 

which has ov σννελαβεν^. είτα εξηλθεν 'Ρεβέκκα πνθίσθαι πάρα 
κυρίον' κα\ είπεν κνριος προς αντην, δυο έθνη εν ttj γαστρί σον καΐ δυο 
\αο\ εν τγι κοιλία σον και νπερεξει λαός λαοί) [sO Cod. ί< : Codd. C 
and all others λαός λαον νπ€ρεζει^ καΐ ό μείζων δονλεύσει τω 

The general correspondence of the quotation in Barnabas with 
the text of the LXX. suggests that he was acquainted with it : but 
the omission of several clauses, including those which have the 
distinctive words εσκρίτων and διαστολή σονται, suggests also that 
either (i) he ρηφ08^7 abbreviated the narrative, or (2) quoted 
from a current manual of Scripture History. 

Genesis xxvii. 30. 

KaI €reN€TO ώο AN έΐΗλθεΝ'ίΑκώΒ atto προςώπογ'ΙςΑΑΚ τογ πΑτρόο Αγτογ 
ΚΑΙ ΉςΑγ ό Αλελφόο Αγτογ ίίλθεΝ αττο thc θΗρΑΟ. 

So Codd. Χ, 31, 32, 68, 83, ΐ2θ, ΐ2ΐ, 131, ΐ34 • Codd. 71, 

106, 107 Om. κα\ εγενετο '. Codd. AD, Ι9, 20, 56, 59, 7 1, 72, 

82, 107, 108, 129 om. αν. Codd. Ε, 14, i5, 16, 18, 25 (m) 
[but with ως written above], 37, 55 [but with -σον erased 
and -T6 written above], 57, 58, 73, 75, 77, 78, 79, 130 (t), 

135, yz 00-01/ : Cod. 106 μετά το εξελθεϊν : Cod. 1 28 ore [but 

ως όσον in margin] : Cod. 106 om. Ιακώβ and Ισαάκ του 

πατρός : Cod. Ε om. άπο της θήρας : Cod. A add. αύτον. 
Μ 2, 


■ Philo de ehriet. 2 (i. 358) iyevcro yap, φησίν, όσον e^rjXBev 'Ιακώβ, 
ηκ€ν Ήσαν 6 αδελφό? αύτον. 

The text of Philo supports the reading oaou, of which ώ? αν was 
probably a corruption and ως a subsequent emendation : but its 
chief importance lies in its agreement with the shorter form of the 
Hebrew, which appears to underlie Jerome's translation 'et egresso 
Jacob foras venit Esau.' The hypothesis of the existence of a cor- 
responding shorter Greek text would account for the MSS. 

omissions of κάΙ iyevero, Ισαάκ του πατρός, and από της θήρας. 

Genesis xxviii. 11-19. 
V. II Και ΑΠΗΝΤΗςε τόπφ και εκοιμηΘη έκεΤ* ελγ r^p ο hAioc" και 
eAABeN απο τωΝ λ'ιθωΝ τογ τόπογ και εθΗκε πρόε κεφΑλΗΟ Αγτογ και έκοΐΜΗθΗ 
eN τφτόττω εκε'ΐΝφ. 

Cod. ζ υπηντησζ, Cod. 56 fV τόττω, Codd. 59? 7^) 134 ^^ '"? τόττω, 
Cod. 72 eis τόπον, Codd. 2 Ο, 82, Ιθ8, 130 προς Κ€φάΚην. 

Philo de Somn. ι (i. 62i) = R., except eV τόττω, ηυλίσθη eW for 

€Κθΐμήθη (Kel, otl eiafjXOev ό ηΧιος for eSu yap δ ήλιος, and προς 
κεφαλήν ΐοτ προς κεφαΧής : lb. 1. 11. i. p. 630 νπηντησ^ν iv τω 
τόπω, but p. 631 νπ^ρφυίστατα be Ζχ^ι το μη φάναι iXBelv ch τον 
τόπον αλλά νπαντησαι τόπω Ι ih. 1. 19. i. p. 638 νπήντησ^ τόπω' 
«δυ yap ό ηΧίος. 

Justin Μ. Dial. 58 = R. 

V. 12 ΚΑΙ εΝγπΝίΑςθΗ καΙ ιλογ κΑΤμαΞ έςτΗριρΜεΝΗ εΝ τη γη hc η κεφΑΑπ 
ΑφικΝεΐτο ειε τον ογρΑΝΟΝ και οί ΑΓΓ^λοι τογ θεογ ΑΝεΒΑΐΝΟΝ και ΚΑτεΒΑΐΝΟΝ 
έπ ΑγτΗ. 

Cod. 59 «"■* "^Ψ τιν'. Codd. Ill, 20, 58, 59> 72, 75> 76^ 82, 
129, 134, 135, +Ε eV αΐτης, Codd. 19, 37» 76\ 79^ ιο6, 
ιο7, +ζ eV αντην, Codd. Ι, 14, 15, ι6, ι8, 25 (m), 30, 31, 
32, 55) 56, 57. 68, 71, 73. 77) 78, 79 S 108, 120, 121, 128, 
130 (Oj 131 f^r' «^'•.ν• 

Philo ibid. i. p. 620 = R. except ^νυπνιάσθη Ιακώβ, and eV αύτης : 
z'iJii/. 1. 22. i. p. 641 = R. except eh την yrjv, and eV αύτ^?. 
Justin M. zbid. = R. except eV αντης. 

V. 13 ό λέ κγριοε έπεςτΗρικτο έττ aythc και ε*πεΝ Έρώ είΜΐ ό θεόο 
ΆΒρΑΑΜ τογ ΠΑτρόε ςογ και ό θεόεΊςΑΛΚ, μη φοΒογ* η γη εφ' hc κΑθεγλειε 
επ Αγτίίο ςοι λώςω ΑγτΗΝ και τφ ςπέρΜΑΤι ςογ. 

Codd. 25 (m), 134 αστήρικτο: Codd. I, III, 15, 31, 37, 58, 


72, 82, 83, 106, 108, 129, 130, +Eyz, om. βΐμί, Codd cett. 
= R. : Codd. Ill, 15, 56 (marg.), 58, 76, 82, 129, 130, 134 
Kvpios 6 eeosf Codd. cett. = R. 

Philo idl'd. i. p. 620 και Ιδού κΚΐμαξ ίστηρι-γμίνη iv rrj yrj και 6 κύριος 
€στήλωται eV αυτής και elnev K.r.X. = R. except την γήν ίφ' ης συ 
καθΐύδζΐς σοι δώσω: I'dl'd. 1. 25. i. p. 644 (ίμψυε το οναρ) ίστηριγ- 
μίνον €7Γΐ της κΚ'ιμακος τον άρχάγγίΧοι/ κυριον et paullo infra μηδίίς 
de άκούων on Ιττ^στηρικτο . . . . : ibid. pp. 644, 646, 647 κύριος 6 
βεος * Αβραάμ .... 

Justin Μ. i'did. = R. except (ι) «V αυτηρ, (2) κύριος 6 θ^ός, (3) om. 
ό θίός before 'Ισαάκ. 

ν. 14 ΚΑΙ εςτΛΐ το ςπέρΜΑ ςογ ώο η ammoc thc thc και πλΑτγΝθΗςετΑΐ έπί 
θΑλΑςςΑΝ ΚΑΙ λΊΒα και ΒορρΛΝ ΚΑΙ έπί anatoAac" και εΝεγλΟΓΗθΗςοΝΤΑΐ In ςοί 
TiAQAi ΑΙ φγλΑΐ THC ["HC ΚΑΙ Ιν τφ ςπέρΜΑτί ςογ. 

Codd. Ill, 20 της θαλάσσης for της γης : Codd. 16, 17 πληθυνθη- 
σ€ται for π\ατυνθησ€ται '. λίβα και em βορραν Codd. Ι, III, 14, 

ιΒ, 25 (m), 38, 56, 57» 58, 59, 73, 7^, 128, 129, 131: «Vi 

λίβα και €7Γΐ βορραν Codd. 15, Ι9, 55, 72j 7^, 77, Ι08, 134• 

Philo ibid. i. p. 620 = R. except ό χους for η άμμος, πληθυνθησβται 
for πλατυνθησξται, and συγγίνειαι for ώυλαί : lb. 1. 28. i. p. 647 

(continuing the commentary on v. 13) το Be σοφίας γίνος αμμω 

γης ζξομοιοϋται .... Xeyerat γαρ οτι ττλατυνθήσίται enl θάλασσαν 
καΐ λίβα κα\ βορραν και ανατολάς .... €ν€υλογηθησονται yap iv σοι, 

φησί, πασαι αί φυλαί [both άμμος and φυλαί are repeated in 
subsequent sentences, so as to leave no doubt that Philo had 
them in his mind]. 
Justin M. ibid.—K. except νότον for λι/3α, and om. eVi before 

V. 15 ΚΑΙ ίλογ έρώ eiMi μετά ςογ 2^ΐΑφγλΑςςωΝ qe In τη όλω haqh ογ αν 
πορεγθΗΟ και Αποςτρέψω ςε είο την γην ΤΑγτΗΝ* δτι ογΜΗ ςεεΓΚΑΤΑλίπω εωο 
τογ ποίΗςΑΐ Με πάντα oqa έλΑλΗςΑ ςοι. 

Codd. Ill, 14, 16, 18, 25 (m), 30, 32, 37, 38, 55, 57, 58, 59, 

73, 78, 79, 106, 107, 108, 128, 129, 130 (t), 131, 134 
+ Ez, om. (Ιμί: Codd. I, X, 15, 19, 20, 31, 56, 68, 71, 72, 

74, 75, 76, 77, 82, 83, 120, 121, 135 eya ei>=:R. 

Philo ibid. i. p. 620 om. άμί, § αν for ου αν, Ιιηστρί^ω for αποστρέψω, 
a for δσα : ibid. 1. 30. i. p. 637 ιδού yap, φησίν, lya> μξτα σου : 
ibid. C. 31. i. p. 648 άηοστρίψω σε eh την yrjv ταύτην. 

Justin Μ. ibid. om. ^ίμί, om, τχ^ before όδώ, .^ av for ol av. 


vv. l6, 17 KAi elMfepGH 'ΐΑκωΒ i• ιογ γπΝογ Λγτογ και etneN δτι "Εςτι 
κγριοε €ν τφ τόπφ τογτω ερώ λε ογκ nheiH' κα! ΙΦοΒηΘη και εΐπβΝ Ώε 
φοΒερόο ότόποο ογτοο' ογκ εςτι τογτο Αλλ' η oTkoc θεογ και ΑγτΗ η πγλΗ τογ 

Codd. 1, III, 20, 72, 75> Si + Z από roO υττί/ον. 

Philo 2*3/(ί/. 1. 31. i. p. 648 €ξηγ€ρθη γάρ, φησίν, "Ίακωβ και CLTrev οτι 
€στι κύριος ev τω τόπω τούτω, εγώ 8e ουκ §deLV .... C. 32 δικαίως 
ονν ίφοβηθη και ewre θανμαστικως ως φοβίρος ό τόπος οντος : de 
migrat. Abraham. 1. i. p. 437 ονκ '4στι τοντο αλλ' η οίκος θβοϋ. 

Justin Μ. idid. = R. 

νν. ΐδ, 19 ΚΑΙ ΑΝεςτΗ 'ΐΑκώΒ το πρωί, και ελΑΒε τον λΊΘον on γπέθΗκεΝ 
έκεΤ πρόο κεφΑλπο Αγτογ και εςτΗςεΝ ΑγτοΝ ςτΗλπΝ και έπέχεεΝ ελΑΐΟΝ επί το 
ΑκροΝ ΑγτΗΟ. ΚΑΙ έκΑλεςε το όνομα τογ τόπογ έκε'ΐΝΟγ OTkoc θεογ" και 
ΟγλΑΜλογζ ΗΝ ΟΝΟΜΑ ΤΗ πολει το πρότεροΝ. 

Codd. 18, 32, 55j 75j ι3ι> + t τω πρωί: Codd. 71, 76, 106, 
107, 1345 + ζ το άκρον αντον : Codd. Ι, III, 14, i5j 16, 18, 

25 (m), 30. 55, 57, 58, 59, 72, 73. 75, 77, 7^, 79, 82, 106, 

107, 129, 130 (Ο, ^31, ^34, 135, +Ζ €KaXeaev Ιακώβ: 

Codd. I, 31, 55, 56, 58, 59, 68, 72, 75, 7^, 82, 83, 106, 

107, 108, 120, 121, 130, 134 ούλαμμαονς, Cod. 20 υύλαμμαούζ, 
Cod. Ill ούλαμμαύς, Cod. 74 ονλαμαούς, Codd. 1 4, 1 6, 1 8, 

25 (ni), 38, 57, 73, 77, 78, 79, 128, 131, + t ούλαμ, 

Justin M. z'dld. τω πρωί, TO eXaiov, το άκρον αντον, om. Ικήνον after 
τόπον, ΟνΧαμμαούς. 

In V. II Philo's ηνλίσθη for (κοιμήθη points to a coordinate 
translation or revision of the LXX., for although φ is always 
elsewhere translated by κοιμάσθαι in the Pentateuch, in the other 
historical books it is uniformly translated by ανλίζ€σθαι. ^ΙσηΚθ^ν for 
εδυ also points to a coordinate translation or revision, for whereas 
^^i3 is only rendered three times in the Pentateuch by δύ^ιν, it is 
frequently (about 150 times) rendered by (Ισίρ^^σθαι : the corre- 
sponding phrase for sunrise is 6 ηλως έξηΧθζν Gen. 19. 23. 

In V. 12 eis την yijv reccives no support from the MSS. of the 
LXX., except the partial support of Cod. 59 eVl την γην, which is 
itself favoured by the Old Latin ' super terram,' Aug. de Civit. Dei 
16. 38 (vii. 449); on the other hand kv τ^ γη is confirmed by 
' in terra,' Tertull. adv. Marc. 3. 24. p. 412. The concurrence of 


Philo and Justin in the reading eV αντψ gives to it a strong 

V. 13, Philo's reading ^στηΚωται for Ιττ^στηρικτο also points to a 
coordinate translation or revision, inasmuch as στηλονν is elsewhere 
found as the translation of ^T,, e.g. Codd. A Judges 18. 16, 17; 
I Sam. 17. 16 ; 2 Kings 17. 10, but not €πιστηρίζαν and only once 
στηρίζων. The revision to which βστηλωταί may be presumed to 
have belonged was apparently Hebraistic, for στηλονν is in several 
places used by Aquila where the LXX. have a more colourless 
word, e.g. Ps. 73 (74). I7) LXX. σν εττοίησα? πάντα τα όρια της yr]S, 
Aquila €στήλωσα$. 

In ν. 14 Philo's reading χους for άμμος points in the same 
direction : the former word is the ordinary translation of "^SV, 
whereas the latter is only found as such in Gen. 13. 16, where it is 
probably transferred from 22. 17, in which passage the Hebrew 
word is not "^^V but bSri, 

The reading πληθυνθησΐται also points in the sanie direction: this 
is the only passage in which Ρ 2 is translated by ηλατύν^ιν, but it is 
translated by πληθύν€ΐν in i Chron. 4. 38, Ps. 105 (106). 29. There 
is a trace of a revision of the same word in Ps. 24 (25). 17 (where 
it is used to translate not Π?" but 31")) : the MSS. reading in that 
passage, ^πληθύνθησαν, could hardly have been the reading when the 
extant extracts from the Hexapla were made, inasmuch as a dis- 
tinction is drawn between Theodotion and Interpres Sextus, who 
have that reading, and Aquila and Interpres Quintus, who are said 
to read the same as the LXX. : hence ζττΚατύνθησαν must there 
be considered to be the original reading, and ^πΚηθύνθησαν to be a 
revision of it. 

The reading avyyeveiat for φνλαί is another instance of the same 
kind. Both words are found as translations of '^^^^^, but while 
the latter is more frequently so used in the Pentateuch, the former 
is more frequent in the other historical books. 

In V. 15 the concurrence of Philo and Justin in the omission of 
€ΐμί makes that omission probable : and the probability is supported 
by its omission in Clem. Alex. Faed. i. 7. p. 131. But there is a 
great want of uniformity of practice in the several groups of MSS. 
as to its insertion or omission here and in v. 13. Some MSS. 
agree with Philo and Justin in inserting it in v. 1 3 and omitting it 


here, viz. Codd. 14, 16, 18, 25, 38, 55, 57, 59, 73, 78, 79, 107, 
128 : some MSS. insert it in both places, viz. Codd. 19, 20, 32, 56, 
68, 74, 75, 76, 77, 120, 121, 135: some omit it in both places, 
viz. Codd. Ill, 37, 58, 106, 108, 129, 130, Ez. 

It may be added that the variants of Philo in this passage help 
to support the hypothesis, to which many other facts lead, that the 
treatise De Somniis belongs to a generation subsequent to that of 
Philo himself. 

Genesis xlix. 10. 

Ογκ εκλείψει ΛρχωΝ έΐΊογλΑ και ΗρογΜεΝΟΟ εκ τωΝ ΜΗρωΝ Λγτογ εωε εΑΝ 
ελθΗ ΤΑ ΑΠΟκε'ΐΜεΝΑ Αγτψ• και Αγτόε προςλοκ'ΐΑ έθΝωΝ. 

Codd. 20, 37? 5^? 72 ουδ€ ήγονμίνος. 

Codd. Ι, III, VII, 15, ι8, 19, 2θ, ^Β, 56, 5^, Ιΐ, 74, 16, 76, 

82, Ιθ8, Ι20, 121, 129 '"" αποκ(ίμ€να αντω Ι Codd. 30, S^, 

37. 38, 57, 59, 73, 75, 7», 79, 83, 107, 127, 128, 134 
ω άπόκαται, SO also, but in the margin, Codd. X, 29, 64 : 

Codd. 32, 84, 135 ο απόκειται αντω'. Codd. 14, 16, 25 (m), 
77, 85, 106, 131, + tz 6 απόκειται: Cod. 72 το άποκ^Ιμβρον 
αντω ο απόκειται. 

Justin Μ. Αροί. i. c. 32. p. 73 (Cod. A) (i) = R., except δ 
άπόκ€ΐται, (2j .... βως αν cX^»/ ω απόκειται το βασίΧειον : idld. 
C. 54. p. 89, = R.5 except δ απόκειται : Dial. c. 52. p. 271 εως 
αν ελβτ] τα άποκείμενα αντω '. Cod. Α. marg. ο απόκειται : ibid. 
Ο. 120. ρ. 348, (ι) €ω£ άνελθτ] τα άποκείμενα αΰτώ = Κ., (2) {μ^χρι 
yap της παρονσίας του Χρίστου ή προφητεία προεκηρνσσεν^ εως &ν 
εΧθυ ω απόκειται, (3) δυνατόν δε ην μοι, εφην, δ άνδρες, μάχεσθαι 
προς υμάς περί της λέξεως ην ΰμεΐς εξηγείσθε λέγοντες είρησθαι' 
"Εως &ν ελθτ] τα άποκε'ιμενα αντω' επειδή ονχ ούτως εξηγησαντο οι 
εβδομηκοντα αλλ'* "Εως αν ΤΚΘτ] ω απόκειται. 

It is clear from the third of the three quotations in Dial. c. 120, 
(i) that there was a difference of opinion in Justin's time between 
Jews and Christians as to the interpretation of the passage, (2) that 
notwithstanding the reading τα άποκείμενα in the chief existing MS. 
of his writings, Justin himself not only read ω απόκειται, but held that 
to be the true reading of the LXX. This fact is of much import- 
ance in relation to the question of the trustworthiness of the 
quotations in Justin's MSS. : it shows that no sound argument can 
be based upon them except in cases where Justin's own commentary 
makes it certain that they contain the text which he used. 


The varieties of reading may perhaps be explained on the 
hypothesis that the original version followed a common Hellenistic 
idiom in reading ω το άττοκίίμζνον {τα άποκύμξνα) αυτω, and that 
δ απόκειται was a gloss or alternative translation for το άπυκβίμ^νορ 
which found its way into the text : hence the readings δ άπόκαταί 
αντω and δ άπόκαταί come from an earlier reading ω δ απόκειται 
αυτφ. This hypothesis is supported by the combination of the 
original reading and the gloss in the remarkable Venice Cod. 72 τό 
άττοκίίμίνον αυτω ο απόκειται. There is a different survival of the 
original reading in Epiphanius i. 332 ω τά άττοκύμβρα: and there is 
a noteworthy rendering in the Clementines, 3. 49. p. 50, ed. Lag. 
ecus av βλθτ] ου εστίν. 

The early Latin versions, with the exception of Cyprian Testim. 
I. 21. p. 55, who has ' deposita illi/ are in favour of ω α7ΓΟκ£ΐται : 
viz. Novatian de Trinit. 9 (p. 7 1 1 in Tertull. ed. Rig.) ' cui repro- 
missum est/ Ambros. de bened. Pair. 4 (i. 5 1 8), ' cui repositum est/ 
Iren. Vet. Interp. 4. 10. p. 239, Hilar, in Ps. lix. p. 158, Hieron. 
Hebr. Quaes t. p. 69, ed. Lag., and in several other passages, e. g. 
in £sai. \ih. 4. c. 11 (iv. 162, Vail.); Rufinus de bened. Pair. i. 3. 
p. 9 has ' veniant ea quae reposita sunt,' but adds ' et velut in aliis 
exemplaribus habetur Veniat is cui repositum est.' Augustine de 
Civit. Dei 16. 41 (vii. 452), ibid. 18. 6 (vii. 492) has 'quae reposita 
sunt ei.' 

Exodus ii. 13, 14. 

Και λέπει τω ΑλικογΝτι ΔιατΊ ςγ τγπτειο τον πΛηοΊον ; ό hk εΤπε Tic ςε 
ΚΑτέςτΗςεΝ ΛρχΟΝΤΑ και λιΚΑςτΗΝ έφ* ΗΜωΝ ; μη ΑΝελείΝ Με ςγ θέλειο on 
τρόπΟΝ ΑΝεΐλεο χθέε τον ΑίργπτιΟΝ ; 
Cod. VII ^ 8ικαστην. 

Codd. 14, 16, 25, 30» 32, 37» 52, 63, 64, 56, 72, 73. 14, 75, 
77, 78, 82, 108, 118, 130 €φ' ημάς Ι Codd. II, III, VII, Χ, 
ι8, 19, 29, 53, 57, 5^, 59, 1^, 1^, ^4, ιο6, 107, 128, 129, 

131, 134, 135 ^φ* ημων = Κ. 

Codd. Ill, VII, Χ, ι6, ι8, 25, 29, 32, 52, 54, 55, 5θ, 57, 73, 
76, 78, 85, 129, 13©, ΐ3ΐ, ΐ35 η aueXdu : Codd. Π, 14, ΐ9, 
30, 37, 53, 58, 59, 1h 72, 74, 75, 77, 82, 84, 106, 107, 

108, 118, 128, 134 Μ ave\€7v = K. 
Acts vii. 26-28 (the narrative portion of the text differs from 
that of Exodus, but the dialogue nearly agrees and is probably 
a quotation) : (avdpes αδελφοί eWe•) Ινατί άδίΚ€ΐτ€ αλλήλους ; (6 


δ€ άδικων τον πΧησιον άπωσατο αυτόν βιττώι/)* Τι? ae κατβστησίν 
άρχοντα κα\ [Cod. Laud. ^] δικαστην εφ' ημών [Codd. DE αι. εφ' 
jy/iiisj J μη aveXuv μ€ σι) ^e'Xeis ον τρόπον avelXes e^ues τον 
Αίγύπτίον ', 
1 Clem. Rom. 4 tIs σε κατύστησ^ν κριτην η [ita Cod. Alex., κα\ 
Cod. Constant.] δικαστην ε'φ' ημών ; μη αι/ελείι/ /ζε συ ^ελει? 
δ τρόπον aveiXes €χθίς τον Αίγύπτιην ', 

There is a probable reference to the passage in Luke xii. 14, 
where the MSS. vary as follows : — 

Cod. χ τις μ€ κατίστησίν κριτην η μ^ριστην ε'φ' ΰμων ; 
Codd. BL ^/. ,, „ „ ,, ε'φ' ν/χα?; 

Codd. Κ αΐ. ,, ,, δικαστην ,, ε'φ' νμας ', 

Codd. Υ> αΐ. ,, ,, κριτήν οτη. ε'φ' υμάς ; 

Cod. 157 ,, ,, άρχοντα καϊ δικαστην ε'φ' υμάς; 

If the reading of Cod. 157 be dismissed, as being obviously 
harmonistic, the chief importance of this reference in Luke, when 
taken together with the quotation in Clement, lies (i) in its substi- 
tution of κριτην for άρχοντα, and of μ€ριστήν for δικαστην ; (2) in itS 

use of ή for καί. In regard to (i), there is no instance in the LXX. 
of the use of κριτής to render "^ψ, but the combination κριτην καϊ 
δικαστην is found in I Sam. 24. 1 6, i Esdr. 8. 23 : the word μ^ριστήν, 
which is not found elsewhere in Bibhcal Greek, is omitted here not 
only by Cod. D, but also by the Curetonian Syriac and by TertulHan 
adv. Marc. 4. 28. p. 445, who, in quoting the Gospel, has 'quis me, 
inquit, judicem constituit super vos ? ' but in quoting Exodus in the 
same place has ' quis te constituit magistrum aut judicem super 
nos?' In regard to (2), the agreement of the Gospel and Clement 
in reading η is supported by the quotation in TertulHan /. c. 

That both the Acts and Clement are quoting the LXX. is shown 
by their use of «x^ey, which word is not in the Hebrew. 

Exodus iii. 2. 

'ΌφθΗ hi Αγτω ΑΓρελοε κγρΊογ eN πγρί φλΟΓΟΟ έκ τογ ΒΑΤογ" και όρΑ δτι ό 
Batoc ΚΑίετΑΐ πγρί, ό λε Batoc ογ ΚΑτεκΛίετο. 

Codd. Ill, VII, 14, 16, 25, 29, 30» 32, 52, 54, 57^ 58, 64, 72, 
73, 74, 75, 76, 77, 78, 83, 84, 106, 107, 130, 132, ΐ34 f'" 
ή>Χογι πυρός: Codd. II, Χ, II, 19, 53, 66i 5^, 59, 7^, 82, 
128, 129, 131, 135 ^^ '^ν* φΧογ05, = Κ. 

Codd. 53» ^2 ου κατακαίεται. 


Philo de profugis 29 (i. 170) {φάσκων 6τι) 6 βάτος KaUrai και ου 


Acts 7. 30 ωφθη αντώ iv Tjj (ρημω του ορούς Σινά άγγελος [ita Codd. 

i< ABC : Codd. DEHP al. add. κυρίου] iv φλογΐ πυρός [ita 

Codd. Ν BDHP αι. : Codd. ACE αι. eV πυρί φλογός] βάτου. 
Justin Μ. Dia/. 60. p. 283 = R., except εκ βάτου. 

The reading eV φλογΐ πυρός in Exodus has in its favour (i) the 
fact that it is supported by MSS. of diiferent groups : (2) the fact 
that, although the passage is not quoted directly by Philo, the 
phrases (6 βάτος) περισχεθείς πολλή φλογί, and το φλέγον πυρ, Vit. 
Mos. I. 12, ii. p. 92, point to iv φλογΙ πυρός. On the other hand 
the reading iv πυρΧ φλογός is supported by Justin not only in the 
quotation given above, but also by the more important paraphrase 
Apol. I. 63. p. 96 : (3) the early Latin versions, which have 'in 
(de) flamma ignis,' e.g. Cypr. Testini. 2. 19. p. 86: Ambros. </c 
Spirit. Sand. i. 14 (vii. 629) : August, de Trin. i. 23 (viii. 785). 

Exodus vi. 2-4. 

ΈλΛλΗςε hk 6 6eoc npoc ΜωγςΗΝ κλι eme npoc ΛγτοΝ Έ[ώ κγριοε και 
ώφθΗΝ πρόοΆΒρΛΛΜ ΚΛίΊςΛΛΚ ΚΑΙ ΊάκώΒ, θεόε ωΝ ΛγτωΝ, κλ! τό όνομα 
Μογ KYpioc ογκ έλΗλωςΑ Αγτοΐε. 

Codd. 19, 108, 118 εγω κύριος 6 θεός, Cod. 55 ^7^ ^ ^^^^) Cod. 

ζ^ om. και before ωφθην. 

Cod. 118 το υνομά μου κύριος ων, Codd. 25, 3^3 ^^'^• κύριος. 

Philo de mutat. ηοηΐ. 2 (i. 580) τό ονομά μου ουκ εδηλωσα αυτοΐς. 

Justin Μ. Dial. 126. ρ. 355 ^'λάλτ^σβ Se κύριος προς Μωσ^ι/ και (ΐπ€ 
προς αυτόν Έγώ εΙμι κύριος καΐ ωφθην προς τον Άβρααμ κα\ Ίσαακ 
καΐ ^Ιακωβ θεός αυτών, καΐ το ονομά μου ουκ εδηλωσα αυτοΐς. 

Justin's omission of ων after θεός may belong to an earlier text 
than that of any existing MS. of the LXX., inasmuch as it follows 
the Hebrew in making θεός an essential part of the predicate (i.e. Ί 
appeared to Abraham . ... as their God, yet my name I did not 
disclose to them'), and not an additional clause. 

His omission of κύριος after τό υνομά μου is apparently, but 
not really, supported by Philo, for Philo's commentary, /. c, makes 
it clear that κύριος (or κύριον) was in his text. For he plays upon 
the grammatical sense of κύριον όνομα, i.e. a ' proper name,' and 
quotes this passage to prove that God had never revealed His 


' proper name/ and he immediately goes on to say, τον yap νπ^ρ- 

βατοϋ μ€τατζθ(ντος e^rjs &.v τοιούτος €ίη \oyos' "Ονομά μου το κύριορ ονκ 
(δήλωσα αντοίς αλλά το ev κατάχρηση 8ια τας ίίρημίναί αίτιας : ' Remov- 
ing the transposition, there will result such a sentence as the 
following : My proper name I did not declare to them, but my 
wrongly applied name, for the reasons stated.' The transposition 
can only be that of to ονομά μου κύριοι/ in the original sentence to 
ονομά μου TO κύριομ in the new sentence which Philo forms: and 
this makes it clear that κύριορ was in his text. 

The reading of Cod. ii8 κύρως ων may be a survival of an 
original ων, without κίιρως, transferred from 3. 24 as the translation 
of the Tetragrammaton. 

2. Quotations from the Psalms and Isaiah in Philo^ 
Clement^ Barnabas, and Justin Martyr, 

1. PhHo. 

I. Quotations from the Psalms. 

The quotations from the Psalms in the Philonean litera- 
ture so nearly correspond with the LXX. version in its 
current form, as to make it certain that the writer or writers 
used that version. 

In some passages there are no variants worthy of note: — 

Ps. 36 (37). 4 is quoted without variant in De Plantatione Noe 7 
(i. 335) and De Somniis ii. 37 (i. 690). 

^^' 74 (75)• 9 is similarly quoted in Quod Deus immut. 17 
(i. 284). 

Ps. 79 (So). 5 is similarly quoted in De Migrat. Abraham. 28 
(i. 460). 

In some passages the variants are only of grammatical 
forms : — ■ 

Ps. 22 (23). I is quoted (twice) in De Agricultura 12 (i. 308), 
and in De Mutatione Nominum 20 (i. 596), in each case with 
νστΐρψΐΐ for the current νστ^ρψΐ}. [So Codd. S 165, 277, 278.] 


Ps. 30 (31). 18 is quoted in De Con/us. Ling. 11 (i. 410), and 
Ps. 41 (42). 4 in Z>^ Migrat. Abraham. 28 (i. 460) with the variants 
respectively of ^^νίσθω, iyivero for the later forms γενηθητω [γ€νηθη- 

τωσαι/], ίγ^νηθη of the existing MSS. of the LXX. 

Ps. 100 (loi). I is quoted in Quod Deus immut, 16 (i. 284) with 
the Hellenistic Τκ^ον [as in S^ and 95 cursive MSS.] for the current 

Attic eXeoy. 

Even when the variations are greater they are not im- 
portant : — 

In Ps. 45 (46). 5 all existing MSS. of the LXX., but one, agree 
with the Hebrew in having the plural του ποτάμον τα όρμηματα 
(νφραίνονσι την πολιν του θ^ον. But in De Somm'is ii. 38 (i. 691) 
Philo has the singular το ορμημα τον ττοτάμον (νφραίν^ι : as in Cod. 
184. There is an indication that he here follows an earlier text of 
the LXX. than any that has come down to us in the fact that the 
Cod. Sangermanensis of the Old Latin, and also Hilary and Ambrose 
have ' Fluminis impetus laetifi<ra/ ' : and it is to be noted that the 
Latin of the Verona Psalter has the singular, though the Greek has 
the plural. 

Ps. 93 (94). 9 is quoted in De Planiat. Noe 7 (i. 334) with three 
variants, viz. (i) the present participles <5 φντίίιων, 6 πλάσσων are 
substituted for the aorists 6 φυτ^νσας, 6 πλάσας which are found in 
all MSS. of the LXX. : (2) the plural οφθαλμούς is used instead of 
the singular οφθάλμόν [so Codd. BS^ of the LXX.] : (3) βπφλ^παν 
is used for the LXX. κατανο^ΐν, and in the future instead of the 
present : in this last point Philo follows the Hebrew more closely, 
and agrees with Jerome's Psalter as against the Old Latin. The 
same passage is also quoted in the treatise De Mundo (ii. 608) 
without the two former of the variants just mentioned, but with 
€πιβλ€π€ΐ for κατανοεί. 

In Ps. 26 (27). I, where all MSS. of the LXX. have Kvpios 
φωτισμός μου, De Somniis i. 13 (i. 632) has φω?: and in this he 
agrees with Aquila and Symmachus. 

Ps. 113. 25 (115. 17) is quoted indirectly, but in harmony with 
the current text, in De Profugis 11 (i. 555) νεκροί 8e . . . . ού< 
αΐνΐσονσι κύριον: and Ps. 83 (84). 1 1 is clothed in a philosophical 


paraphrase in Quis rer. divin. heres 58 (i. 515) μιαν yap ημίραν .... 

βονΚζσθαί βιωναι μ€τα aperrjs η μνρία '4τη ev σκιά τον θανάτου. 

It may be noted that Philo in quoting the Psalms never 
uses the word ψαλμόξ or its compounds, but always v^vos or 
one of its compounds: e.g. i. 59'^) quoting Ps. 2:2 (23). i, 
abcTai be καϊ kv νμνοι^ άσμα tolovtov: i. ^^^, quoting Ps. ^6 {^y), 
4, 6 τοϋ Μωνσ€ω9 θιασώτης . . . lu νμνω^ίαις αν^φθί-γζατο : i. 460, 
quoting Ps. 41 (42). 4, Iv νμνοις άρηται : i. 284, quoting Ps. 
100 (loi). I, 6 νμν(ύΙος el-ni που : i. ^^^ (quoting Ps. 113. 25 
(115. 17) as given above), ώ? καΐ kv νμνοις kiyeTau And that 
νμνοις was the older designation is shown by the subscription 
to the Second Book of Psalms, which is found in most MSS., 
efeAtTToi; ol νμνοι Δαυίδ του υΙου Ίεσσαι. 

II. Quotations from Isaiah. 
Philo appears to quote Isaiah only twice : — 

In De Somniis ii. 25 (i. 681) he quotes the figure of the vine 
from Is. 5• 7? «Α'ΤΓίλώι/ κυρίου παντοκράτορας oikos του Ίσραηλ, the only 
variant being that, as is the case in many passages of the LXX., 
especially in the Minor Prophets, Γli^ζ^V is translated instead of 
being transliterated. The passage is quoted as having been said 
by Tis των πάλαι προφητών, and by him eVi^eiatras•, ' under in- 

In De Muiat. Nom. 31 (i. 604) he quotes Is. 57. 21 χαίρ^ιν ουκ ^στι 
τοΊς άσ€β€σιν che θ€Ος Ι that the quotation is from the LXX. is shown 
by the rendering of ^κ>ψ by χαίρων : it is ordinarily translated by 
€ΐρηνη, Aquila and Symmachus so translate it in this passage, nor is 
it rendered by χαίραν in any other passage of the LXX., except the 
parallel passage Is. 48. 22. 

In De Exsecrat. 7 (ii. 435) η γαρ €ρημος, fj φησιν 6 προφήτης, 

€ντ(κνός τ€ κα\ πόλύπαις may be an echo of Is. 54• I• 

But the resemblance of words is slight : and it may be inferred 
from I Sam. 2. 5, Ps. 113. 9, that the phrase was a conventional 
and even proverbial one. 


2. Clement of Rome. 
I. Quotations from the Psalms. 

In the majority of passages in which the Psalms appear 
to be quoted in Clement of Rome there is a precise agree- 
ment with either the current text of the LXX., or the text 
of existing MSS. : i.e. the variations are only such as exist 
between different MSS. of the LXX., and the quotations of 
Clement must be reckoned to be an additional item of great 
value for the determination of the text of the LXX. 

For example : — 

Ps. 50 (51). 3-19 is quoted in c. 18 with only the following 
variants from the Sixtine text : στηρισον is read in v. 12 for στψιξον, 
as in Codd. BS, 27, 55: τα χίίλη and τό στόμα are transposed 
in V. 15. 

Ps. 61 (62). 5 is quoted in c. 15 with the Hellenistic βύλογονσαι/, 
as in Codd. BS^ 27, 55, Verona Psalter, for the current classical 


Ps. 31 (32). I, 2 is quoted in c. 50 with ov ου μη λογίσηται, as in 

Codd. ABS^ and 12 cursives, for ω ou of Cod. S'^, the majority 

of cursives, and the Sixtine text. 

^^' 36 (37). 35-37 is quoted in c. 14 with (i) the variants άσςβη 
[Cod. Alex.], τον άσφη [Cod. Const.] as in the LXX. where Codd. 
BS^ omit and Cod. A inserts the article : (2) (^(ζψησα as in Codd. 
99, 183 for the current ^ζψησα, 

Ps. 49 (50). 16-23 is quoted in c. 35 with a few unimportant, 
and two important, variants: (i) in v. 21 the current text of the 
LXX. (i.e. Cod. Β and all cursives except 188 : the long lacuna in 
Cod. A begins two verses earlier) has the phrase υπίλαβίς άΐ'ομία»', 
the word άνομίαν having no equivalent in the Hebrew and spoiling 
the sense. Clement agrees with Cod. S^ in reading ανομ^ which, 
though without a Hebrew equivalent, is in entire harmony with the 
spirit of the passage and adds to its force. The Latin of the 
Verona Psalter has ' inique,' which is retained in the Vulgate : but 


this word appears to have been taken not as a vocative but as an 
adverb : hence the translation in the Prayer-Book version ' Thou 
thoughtest wickedly that ....': it may be noted that the only 
variant in the MSS. of the LXX., Cqd. 188, also substitutes an 
adverb, αδικώ?: (2) in v. 22 Clement adds after άρπάση the words 
ως λ/ωι/ in which he is supported by both the Greek and the Latin of 
the Verona Psalter : but the words are probably only a reminiscence 
of Ps. 7. 2. 

The general fidelity of Clement to the text of the LXX. 
is sometimes shown by his reproduction of its mistransla- 
tion : for example in Ps. 50 (51). 8 the Hebrew clearly means 
(as it is translated in the English Revised Version) : 

' Behold thou desirest truth in the inward parts ; 
And in the hidden part thou shalt make me to know wisdom.' 

But the LXX., which is followed by Clement, c. 18. 6, 
translates ΠίΠΙώΙ by τα άbηλa, and appears to destroy the 
paralleHsm of the verse by joining it to the second member, 
viz. : 

18ού γαρ αληθααν ηγάττησας' 

τα α^ηλα και τα κρύφια της σοφίας σου ΐδηλωσάς μοι. 

(At the same time it is conceivable that the original LXX. 
version may have been els τά abηka, and that it was misunderstood 
and altered by a scribe.) 

But in at least one case there are variations from the 
LXX. text which suggest the same hypothesis which is 
suggested by some of the quotations in Barnabas, viz. that 
of the existence of ' revised ' or ' adapted ' editions of the 

JPs. 3. 6 ί'γώ ΐκοιμηθην και ύπνωσα^ 

€ξη'γ€ρθην ΟΤΙ κύριος avTiXrjyjrfTai μου 

[Codd. S^ 210 άντ€\άβ€το ftov] 
is quoted in C. 26 in the form €Κ0ΐμήθην καΐ νπνωσα, (ξηγίρθην δτΐ συ 

μ€τ cfjiou 61, where the last phrase is probably incorporated from 
Ps. 22 (23). 4 (οΰ φοβηθήσομαι κακά) οτι συ μ€τ €μοΰ ci. 


II. Quotations from Isaiah, 

Several of Clement's quotations from Isaiah are com- 
posite, and will be considered separately in the next chapter. 
The other quotations are for the most part faithful repro- 
ductions of the LXX. text, and in several cases afford in- 
teresting contributions to the criticism of it. 

Is. I. 16-20 is quoted in c. 8 : (i) Cod. Const, follows the great 
majority of MSS. of the LXX., and the Old Latin, in reading 
λούσασθί, καθαροί yeveaOe : Cod. A agrees With two cursives 93, 144, 
in reading καί before καθαροί : (2) Cod. A reads άψ€λ€σθ€ for άφ€\€Τ€, 
in agreement with Justin M. Tryph. 18, but against all MSS. of the 
LXX. and Justin M. Apol. 44, 61 : (3) Cod. A reads xr\pa for χηραν^ 
in agreement with Codd. B\ 144, 147* of the LXX. but against 
all other MSS. : (4) Cod. Const, follows Cod. Β and the majority 
of cursives of the LXX., and the Old Latin, in reading SeCre 
διβλ€γχθωμ€ν {8ια\€χθωμ€ν), Cod. A of Clement agrees with Codd. AS 
and 1 6 cursives of the LXX. in inserting καί after devre. 

Is. 29. 13 as quoted in c. 15 affords many points of interest. 

In the LXX., Cod. Β and the majority of cursive MSS. (with 
many minor variants in the cursives) read (-γγίζα μοι 6 Xaos οΰτος iv 
τώ στόματι αντον κα\ iv τοΊς χίίΚίσιν αυτών τιμώσί μ€ ή δε καρδία αυτών 
πόρρω άπ€χ€ΐ απ βμον. Codd. AS, 26, 49j ^7) 9^» 97» ^9^^ S^^j 3°9 
read eyytXct μοι 6 λαό? ουτοί τοις χύλ^σιν αυτών τιμώσί μ€ ή Se καρ8ία 
αυτών πόρρω άπίχ^ι αττ' ipov. 

In Clement, Cod. A has ούτος 6 Xaos τοις χ^ίλ^σί μ€ τιμά η be καρδία 
αυτών πόρρω απίστιν απ €μοΰ : Cod. C has ό Χάος ούτος τω στόματι μ€ 
τιμζ η be κάρδια αυτών πόρρω άπίχίΐ απ* €μου. 

In the Ν. Τ., the following is, except where otherwise noted, the 
reading of the chief MSS. of Mark 7. 6 : ο{)τος 6 Χα6ς [Codd. BD δ 

Χάος dittos] τοΊς χ^Χίσίν μ€ τίμα [Cod. D, a, b, C, αγαττδ] ή δε καρ8ία 
αυτών πόρρω άπίχ(ΐ [Cod. D άφίστηκίν, Cod. L αττεστίν] απ ίμου. In 

Matt. 15. 8 some MSS. viz. CEF, and the Peschitta, have the 
longer form which is found in Cod. Β of the LXX. ; and Cod. D, 
which is supported by most early Latin quotations, has εστίν αττ* 
ε'/ΑοΟ for άπίχα απ (μου. 

It is a legitimate inference that, before the time of 



Clement, the quotation had become detached from its con- 
text, and that ovtos 6 Xaos, having lost its proper predicate 
eyytfet, and having assimilated the following predicate 
τιμωσι (which thereby became rtjuta), the antithesis was ac- 
centuated by the loss of one or other of the phrases kv 
τω στόματί or Iv Tols χίΐλεσι. The quotation is one which 
naturally became common in a time of religious revival, and 
it not less naturally tended to become so in its shortest 
form. Hence it was so written by many of the scribes of 
the LXX., and became the current text of one of its re- 
cognized recensions. 

Hence the shorter form is found 

(i) In all MSS. of St. Mark: while some good MSS. of 
St. Matthew give the longer form. 

(2) In Clement, though the shorter form is found in both MSS., 

Cod. A has rois χίΐλίσι, Cod. C τω στόματί. 

(3) Justin Μ. shows by his repeated indirect quotations of it that 
the shorter form was in frequent use in the Judaeo-Christian con- 
troversies, Tryph. 27, 39, 80 : and at the same time he alone of early 
writers goes behind the quotation to its original meaning, and in 
Tryph. 78 quotes the whole passage in accordance with the 
Hebrew, omitting only τω στόματί αυτών (or equivalent words) 
eyyt^et μοι δ Xaos ovtos' toIs x^iKeaiv αυτών τιμώσί μ€, ή de καρδία αυτών 
πόρρω άπίχ^ί απ €μον. 

(4) Almost all the early Latin quotations of the passage give it 
in the shorter form, indicating that the current version was based 
upon the corresponding recension of the LXX. : e. g. Iren. Ve/. 
Interp. 4. 12, Cypr. Ep. 67. 2, p. 736, Ambros. in Psalm. 36, vol. i. 
810 d. But at the same time it is clear from Jerome in Isai. 29, 
torn. iv. 393, that a version of the longer form was also in existence. 

Is. 53 is quoted entire in c. 16. 

The following are the more noteworthy variants: (i) In v. 2, 
Clement agrees with Codd. AS, 22, 26, 36, 48, (62), 86, 90, 93, 

106, 144, 147, 198, 233, 306, 308, in placing Ιναντίον αυτού 

immediately after άνηγγ^ιλαμ^ν : so Tertull. c. Marc. 3, pp. 671, 676, 
Annuntiavimus de illo [coram ipso] velut [sicut] parvulus, Cyprian 
Tcstim. 2. 13. p. 77, Lactant. Instit. 4. 16, and the majority of early 


Latin writers. (2) In v. 3 Clement reads (κλ^Ίπον παρά το elBos των 
ανθρώπων : the LXX. has many variants, chiefly, €κλ(ϊπον, or (κλύπον 
το (Ιδος [so Codd. 22, 48, 51, 62, 90, 93, 106, 144, 233, 308] πάρα 
τους vlovs των ανθρώπων or παρά πάντας ανθρώπους [so Codd. A, 26, 

198, 239, 306]. None of these translations, in either Clement or 
the LXX., correspond to the Hebrew of this verse : but the 
difference between Clement and the LXX. affords a remarkable 
proof that the translation has been transferred to this place from 
c. 52. 14, for each of the translations is a possible translation of 
the latter half of that verse. Consequently they must have been 
made independently, and this fact suggests the hypothesis that the 
Greek of this verse, whichever of the two translations be adopted, 
represents an alternative, but now lost, Hebrew text. (3) In v. 6 
Clement reads νπ^ρ των αμαρτιών ημών : all existing MSS. of the 
LXX. read rat? άμαρτίαις ημών, but the early Latin quotations, 
e.g. Cyprian Tesiim, 2. 13. p. 77, Lactant. Instii. 4. 16 support 
Clement by reading propter peccaia nostra : so Jerome in Isai. 53, 
tom. iv. 615 propter iniquitates nostras. 

Is. 60. 17 is quoted in c. 42 with the variants (a) επισκόπους for 
the άρχοντας of all MSS. of the LXX., and ib) διακόνους for επισκόπους. 
In regard to {a) it may be noted (i) that Clement and the LXX. 
agree in rendering the abstract n^fS by the concrete words άρχοντας, 
επισκόπους, whereas Aquila has επίσκεψιν, Symmachus βπισκοπήν : 
(2) that the same word is translated by επισκόπους in 2 Kings 1 1. 18, 
and by επισκέψεως in I Chron. 26. 30 : (3) that the concrete 'Τ'ί?^ is 
rendered in LXX., Gen. 41. 34 by the local Egyptian word 
τοπάρχας, in Symmachus by επισκόπους, in LXX., Judges 9. 28 by 
επίσκοπος, in LXX., 2 Chron. 24. II by προστάτης, in LXX., Esth. 
2. 3 by κωμάρχας. It follows that Clement may very possibly have 
had before him a revised text of the LXX. in which επισκόπους was 
used in the present passage. In regard to {b) it may be noted that 
the Hebrew '^ϊ\ which Clement here renders by διακόνους, the LXX. 
by επισκόπους, Aquila and Theodotion by πράκτορας, Symmachus by 
επιστάτας, is rendered in Job 3. 18: 39. 7 by φορολό-^ος. 

Ν 2 


3. Barnabas. 

I. Quotations from the Psalms. 

In three cases the quotation agrees with the Sixtine text 
of the LXX., and there is no important variant from that 
text in the MSS. of the LXX. itself: viz. Ps. ΐτ {i^). 19, 
117 (118). 12 and 2% are all quoted in Barn. 6. 

In four unimportant cases the text of Barnabas differs 
from the Sixtine text, but is supported by good MSS. of 
the LXX. 

In Ps. I. I, quoted in c. 10, Cod. S of Barnabas agrees with 
Codd. BS and 42 cursives in reading Ι-Λ καθβδραν for enl καθέδρα. 

In Ps. I. 5, quoted in c. 11, Barnabas agrees with Codd. A, 268 
of the LXX. in omitting the article before άσ^ββΐς. 

In Ps. 17 (18). 45, quoted in c. g, Barnabas agrees with Codd. 
S^, 179, 286 of the LXX. in reading νττήκονσαν for νπηκονσ€ν, and 
with S**, 205, 206 in reading μον for μοι. 

In Ps. 21 (22). 17, quoted in c. 6, Barnabas is supported by 

two cursives, 81, 206, in reading nepUaxe for nepieaxov. 

Some cases suggest the hypothesis that a Greek text of 
the psalms was in existence, which was based upon the 
LXX. but altered by a Greek hand in the same way as, 
for example, in modern times hymns are sometimes altered 
by the compiler of a hymn-book. 

JPs. 21 (22). 23 ^ιηγησομαι το ονομά σου τοΊς άΒελφοΐς μον, ev μίσω 
εκκλησίας υμνήσω σι is quoted in C. 6 in the form (ξομολο-γήσομαί σοι 
iv ΐΚκΚησΙα iv μέσω άδίΚφων μον και yf /αλω σοι άνα μέσον ίΚκΧησίαί άγιων. 

The fact that elsewhere in the same chapter Barnabas quotes 
exactly the LXX. text of the same psalm seems to show that he is 
not using another translation of the Hebrew : but it must be noted 
(i) that (ξομολογύσθαι does not occur in the LXX. as a translation 
of "ISD, (2) th3,t\jra\\fiv does not occur in the LXX. as a translation 

of ^i!'•?. 

Other cases suggest the hypothesis that psalms were in 


existence which breathed the spirit, and adopted the Greek 
phraseology, of the existing psalms, but which were never 
incorporated into the psalter and only exist in these frag- 
ments ; 

Ps. 33 (34). 13 ris iariv άνθρωπος 6 θελων ζωην^ άγαπωρ ημέρας Idelv 
άγαθάς ; is recalled by C. 9 ris ianv 6 Θί\ων ζησαι ds αιώνα ; 

jRf. 41 (42). 3 7τότ€ ηξω κα\ οφθησομαι τω προσώπω τυΰ θ€θΰ ', is 
recalled by C. 6 ev τινι οφθησομαι τω κνρίω θίω και 8οζασθησομαι ; 

I^S. 50 (51)• 19 ^*^^'" ^<Ρ ^^Φ πνεύμα συντετριμμίνον, καρ^ίαν συντε- 
τριμμένη ν κα\ τ€ταπ€ΐνωμ€νην ό θβος ουκ ονδίνώσίΐ is recalled by C. 2 
θυσία τω θ(ω πνεύμα συντετριμμενον, οσμή ευωδιάς τω κνρίω καρδία 
δο^άζουσα τον πεπΧακότα αυτήν. 

Ps. 89 (90)• 4 χίλια ^τη iv όφθαλμοίς σον ως η ήμερα η εχθές ήτις 
διήλθε is recalled by C. 15 *δού σήμερον ήμερα εσται ώς χίλια ετη. 

In at least one case, in c. 5, there is a cento from several 
psalms, which will be discussed separately in the next 

It must be noted that there is no difference in the mode 
of quotation between passages which are undoubtedly from 
the LXX. and other passages which are best explained by 
the hypothesis of the existence of altered versions or centos : 
undoubted quotations are introduced by e.g. Δαυίδ . . . Xeyet 
ομοίων C. 10, Aeyct Kvpios €v τω ΊΓροφητ-τ} c. 9, Aeyet ττάλίν 6 
ττροφητηξ c. 6, other quotations by e.g. Aeyet ττάλιν Kvpios c. 6, 
ττάλίν TO ττνζνμα του Κυρίου Aeyet C. 9, Aeyet ττροφητ^ύων €π 
αντω c. 5, avTos he [sc. 6 Kuptos] μοί μαρτυρά λίγων C. 15. The 
point is of importance as an indication of the current opinion 
in regard to the limits of the Canon of Scripture. It seems 
likely that as any writer or speaker of exceptional spiritual 
force was regarded as a ττροφήτη^, so what he wrote or said 
was regarded as the utterance of the Spirit of God through 


Π. Quotations from Isaiah. 

In most cases the quotations follow the current text of 
the LXX., with only such variations as are found in existing 
MSS. of the LXX. ; but in some cases the original mean.- 
ing is clearly disregarded and the quotation adapted to the 
immediate point in hand. 

Is.i. 2 is quoted in c. 9 with the addition ταντα etp μαρτνρίαν after 

Kvpios €λά\ησ€ν. 

Is. I. 10 is quoted in c. 9 with the substitution of τον λαον re που 

for 2οδόμων. 

Is. I. 11-14 is quoted in c. 2 with [a) the omission, in Cod. Sin., 

of κριωρ after ολοκαυτωμάτων, ((5) the Omission of και rjpepap μ€γά\ην 

after τα σάββατα. V. 1 3 is also quoted in c. 15 with the same 

omission of καΐ ημ. μΐγ. 

Is. 3. 9 is quoted in c. 6 with the variant δτι for διότι. 

Is. 5.21 is quoted in c. 4 : Cod. Sin., as also Cod. 9 1 of the 
LXX., omits. Cod. Const, retains iv in the phrase 01 aweTol ev 


Is. 33. 13 άκονσονται οι ττόρρωθίν α βποίησα, yvoxrovTai οΐ eyyt^oin-ef 
την Ισχνν μου is quoted in c. 9 with a Hebraistic addition to 
άκονσονται and with the omission of the second subject, viz. aKofj 
άκουσονται ol πόρρωθ(ν α (ποίησα γνωσονται, which showS that the 

words are quoted without reference to their original meaning and 

Is. 33. 16, 17 . . .TO νδωρ avTov ττιστόν' βασΐΚβα μ€τα δόξης οψ^σθβ, 
οΐ οφθαλμοί υμών οψονται yrjv πόρρωθ^ν, η ψυχή υμών μ(λ€τησ€ΐ φόβον IS 
quoted in C. 11 in the form τ6 ΰδωρ αυτόν πιστόν' βασιλέα μ€τα δόξης 
οψίσβί κα\ ή ψυχή υμών μ€λ€τησ€ΐ φόβον κυρίου : here also the 

severance of τό ΰδ. αυ. πιστόν from the preceding sentence to which 
they belong, and the addition of κυρίου to the last words, show that 
the words are quoted as words pertinent to the point in hand, 
without reference to their original meaning and application. 

Is. 40. 3 φωνή βοώντος iv τι} €ρημω is quoted in c. 9 with the prefix 
άκουσατΐ τίκνα, and it is clear that, as in Matt. 3. 3, Mk. i. 3, Luke 3.4, 
iv TT] ipήμω is taken with βοώντος rather than with the following 


ίτοιμάσατζ : Cod. Sin. of Barnabas reads φωνή as in the LXX., but 
Cod. Const, reads φωνής, making the word depend on άκούσατ€. 

Is. 42. 6, "J is quoted exactly in c. 14, with the exceptions (a) 6 θεός 
σου for 6 θίός: [β) Cod. Sin. has Ισχνσ-ω for ενισχύσω: SO Justin M. in his 
three quotations of the passage, Tryph. 26, 65, and 122 : (γ) καί is 
read before i^ayayelv: so Cod. XII and most cursives of the LXX. : 
(δ) πεπώημίνονς is read for δε^εμενονς : SO Justin M. in the three 
quotations just mentioned: this change points to a revised text 
since πεηεδημενος is a more frequent translation of "^''Di? : (e) καί is 
omitted, with most MSS. of the LXX., with Justin M. Tryph. 26, 
65, and in agreement with the Hebrew, before καθήμενους. 

Is. 45. I λέγει κύριος 6 θεός τω χριστώ μου Κύρω is quoted in C. 12, 

probably (i. e. in Codd. Sin.^ Const, as against Codd. Barb. Med. 
Sin^.) with the change of Κίιρω into κυρίω, obviously on apologetic 

Is. 45. 2 is quoted in c. 11 with the variants (a) in Codd. Sin. 
Const, πνλας for θύρας, a change in the translation of Πρ*! which is 
sometimes found in the LXX., (d) αόρατους is omitted, as in Cod. A^, 
{c) γνώσιν for γνως, a middle term between the two readings existing 
in the γνώση of Cod. A. 

Is. 49. 6 (Cod. A) Ιδού τεθεικά σε [Codd. BS, αι. add εΙς διαθηκην 
γένους^ εις φως εθνών του είναι σε εΙς σωτηρίαν εως εσχάτον της γης' ούτως 
\εγει κύριος ό ρυσάμενός σε ό θεός *1σραηλ is quoted in C. 1 4 aS in the 

Alexandrine text with (a) the substitution of λυτρωσάμενος for βυσά- 
μενος; (δ) the omission of the article, as in Codd. Β S'', and six cursives, 
before θεός ; (c) all MSS. of Barnabas, except Cod. Sin., also omit 
Ίσραηλ after θεός. It may be also noted that here, as elsewhere, 
the clause οΰτως λέγει . . . is detached from its proper context and 
adapted to the immediate purpose of the writer. 

Is. 50. 6, 7 is quoted in c. 5 with the omission of 6 <5, ^ a : 
i. e. the final clause of the antithesis, being sufficient for the 
purpose, is given instead of the whole : the only variant is τεθεικα 
for έδωκα, as in the preceding quotation. 

Is. 50. 8, 9 (Cod. B) τις 6 κρινόμενος μοι ; άντιστητω μοι άμα' κα\ τίς 
ό κρινόμενος μοι' Ιδού κύριος κύριος βοηθήσει μοι' τίς κακώσει με ', Ιδού 
πάντες ύμεΐς ώς Ιμάτιον παλαιωθησεσθε κα\ σης καταφάγεται ύμας is quoted 

in c. 6 with omissions and with an apologetic adaptation to Christ: 


the variants are {a) αμα is omitted, {c) η ris is used for κα\ τίς, [c) the 
second κρινόμενος is changed to δικαιούμενος in Codd. Sin. Const. : so 
also Cod. 26 of the LXX., δικαζόμβνος Codd. cett., (d) the clauses 
Idov κύριος . . . . , τις κακώσει με are omitted, as not being pertinent 
to the purpose of the quotation, (e) oval υμίν on is substituted for 
Ιδού : but it is possible that these words are meant not to be part 
of the quotation but only to call the attention to what follows: 
Woe ioyou, for {as the prophet says) ' Ye shall all wax old ... .* 

Is. 58. 4-10 is quoted in c. 3 with the following variants : — 
In V. 4 Barnabas inserts the words \(yn κύριος after νηστεύετε : 
the insertion of the words in MSS. of the LXX. is somewhat 
arbitrary, e.g. they are inserted in the next verse by Codd. 239, 

In v. 5 Barnabas agrees with 13 cursives and the Old Latin, as 
against the other MSS., in inserting εγώ before εξελεζάμην: he reads 

ουκ ανθρωπον ταπεινονντα την ψνχην αντον for και ημεραν ταπεινουν 

ανθρωτΓον την ψνχην αύτον, in which he is supported, against all the 
MSS. of the LXX., by Cypr. Testim. 3. i, p. 108 diem humiliare 
hominem animam suam, Hieron. in Zach. 7, torn. vi. 833 neque ut 
humiliet homo animam suam : he reads the plurals κάμψητε, νπο- 
στρωσητε [Cod. Const. omits] for the singulars κάμψης, υποστρωστ], 

and he gives the special predicate ενδύσησθε to σάκκον. 

In V. 6 the words ουχί τοιαύτην νηστείαν εγω [most cursives omit 
εγω] εξελεξάμην are expanded into the more emphatic form Ιδον αυτή 

ή [Cod. Sin. omits ή] νηστεία ην εγω εξελεξάμην, in which he is 

supported, against all existing MSS. of the LXX., by Clem. Alex. 
Paed. 3. 12, p. 305. 

In V. 7 (l) the order of the clauses πτωχούς άστεγους εΙσαγε εΙς τον 

οικόν σου, and γυμνον εαν ϊδης περίβαλε is inverted : SO also in the Old 
Latin in Hieron. in Zach. tom. vi. 833 si videris nudum operi eum et 
pauperem et absque tecto indue in tabernaculum tuum : but all the 
other quotations of the passage in early Latin writers follow the 
current order of the clauses, with the exception of Auct. Quaest. V. 
T. ap. S. Aug. tom. iii. append, p. 145^, which omits the translation 
of the clause πτωχούς .... οΧκόν σου. (2) πτωχούς is omitted, as in 
Tertull. c. Marc. 4, p. 651 r, 730 3 (but elsewhere mendicos is 
inserted) : possibly because of the practical difficulty of a literal 
observance of the injunction, which may also account for the 


substitution oi peregrinum in Iren. Vet. Interp. 4. 17. (3) A new 
clause is added, eai/ "ώ^ς rarreivov, and the predicate of the follow- 
ing clause, viz. ουκ vnepoyfrrj is placed as its apodosis : the use of 
TarreLvov here, and the omission of πτωχούς in the preceding clause, 
may be explained on the supposition that in some editions of the 
LXX. the former word rather than the latter was used, as in five 
other passages of Isaiah, to translate ^^V. 

The text of the passage in Barnabas is evidently ' conflate ' : the 
quotations in the early Latin writers mentioned above indicate that 
in one text, as in Barnabas and perhaps through the influence of 
the cognate passages, Ezek. 18. 7, 16, the clause about clothing 
the naked was placed next to that about feeding the hungry, 
probably without any further change : and that another text 
followed the Hebrew order. When Barnabas, or a reviser whom 
he followed, put these two texts together, in order to avoid the 
repetition of γνμνόν, he used ταπανόν, which some texts contained in 
the preceding clause, as the object of the repeated eav ϊδτ]ς and 
made the predicate ονχ vnepo^rj αυτόν common to the two last 

In v. 8 it is almost certain, although the reading is corrected, 
perhaps by the original scribe, in Cod. Sin., that Barnabas read 
Ιμάτια for Ιάματα : it is obviously a scribe's error, but it is found in 
Codd. S^ and ^ 91^ io6^ 147 of the LXX., and, in the translation 
vesn'menfa, in Tert. de Resurr. Carnis, pp. 576 r, 577^2, Cyprian 
Tesii'm. 3. I, p. 108, de Orat. Domtn. 33, p. 291, de Op. et eleem. 4, 
P• 376. Jerome notes it as the current Latin reading. In Isai. 58, 
tom. iv. 693. 

In v. 9 the MSS. of Barnabas vary between βοησίΐς and βοησί], 
and between ^πακονσ^ται and (Ισακονσεται : in each case the latter of 
the two readings mentioned is the reading of all the MSS. of the 
LXX. except one. 

In V. 10 Barnabas agrees with Codd. A, 26, 49, 106 in adding 
σου to Tou αρτον : SO also aU the early Latin quotations. 

/f. 61. I is quoted in c. 14 almost exactly as in the current text 
of the LXX., from which there are no important variants : but both 
in the LXX. and Barnabas there is an interesting instance of the 
interchange of πτωχοίς and raTjcivols as translations of ""^V (see 
above, p. 73) : in the LXX. Codd. AB and most cursives have 


πτωχοΊς, Cod. S^ has ταπ€ΐνοϊΐ, in Barnabas the fragmentary MSS. 
have ταπίΐνοϊς and add χάριν, Cod. S. has πτωχοΐς. 

Is. 65. 2 Cod. Β ζξ^τΐίτασα τας χ^ΐράς μου οΚην την ημίραν προς λαόν 
άπ€ΐθονντα και άντΐλ^γοντα^ toIs nopevopevois όδω ου κάΧτ} is quoted in 
C. 12 in the form ολην την ημίραν ξ^οΐΐτασα τας χΐ^ράς μου προς Χαον 
άπ€ΐθη [so Cod. Sin., Codd. Const, cett. άπ€ΐθονντα^ κα\ avrCKeyovTa 
όδώ δικα/α μου. The insertion of the words 6δώ bLKaia μου, which are 
obviously suggested by the following clause of the LXX., is probably 
a rhetorical softening of the harshness of the absolute use of 


In at least tv^o passages the resemblance to the text of 
Isaiah is hardly strong enough to warrant the supposition 
that they are directly quoted from it : viz. 

C. 16 ίδού oi KaOeXovTes τον vaov τούτον αυτοί αυτόν οίκο^ομησουσιν 
recalls Is. 49. ι γ κάΙ τάχυ οίκο^ομηθησ-τ] ύφ' oiv κατγιρίθης \ C 6 και 
Ζθηκίν μ€ ως στβρΐάν πίτραν recalls Is. 50• 7 "'"^ ^^ πρόσωπον μου ίθηκα 

ως στ€ρ€αν πίτραν (which is quoted exactly in c. 5 ; see above, 
p. 186). 

It is a hypothesis for which there is no direct evidence, 
and which at the same time is not contrary to analogy, to 
suppose that besides the canonical books themselves, there 
were manuals of prophecy as well as anthologies, which had 
a certain authority and were accordingly quoted as of 
authority, in the same way as e.g. Clement of Alexandria 
{Strom. 3. 20) quotes the ' Two Ways' as y\ γραφή. This 
hypothesis will serve also to explain the quotations in c. 6. 
13 Ibov ττοίώ τα έσχατα ώί τα ττρώτα, C. II. ίο καΐ δ? hv φάγτ} 
€ξ αυτών ζήσβται els τον αΙωνα (which appears to be a sum- 
mary of Ezek. 47. 12). 

4. Justin Martyr. 

It is desirable, before considering any of Justin's quota- 
tions, to point out that the text of his genuine works prac- 
tically rests upon a single MS. of the fourteenth century, 


Cod. Paris 450, dated 1364. The value of that MS. can 
be tested in two ways: (i) the same MS. contains other 
works of which other and earlier MSS. remain : three of 
these works, ps-Justin Epistola ad Zenam and Cohortatio 
ad Gentiles^ and Athenagoras de Resurrectione, it has in 
common with another Paris MS., No. 451, which was written 
in 914, i.e. 450 years earlier. Omitting unimportant ortho- 
graphical variations, it differs from these three treatises in 
169 passages, in only a small proportion of which (according 
to Otto 17, according to Harnack 5 or 6) is it probable that 
the later MS. has the better reading. In other words, in 
that part of the MS. which admits of comparison with these 
three works there are not less than 150 passages which 
require emendation. If the mistakes in the two Apologies 
and Trypho be in the same ratio, as they may fairly be 
presumed to be, the number of such mistakes will be ver)^ 
large. (2) In a few passages we can compare the MS. with 
quotations from Justin in other works which have well- 
attested texts : e. g. Justin, Apol. ii. 2 with Euseb. H. E. 4. 
17 : this comparison gives the same results as the preced- 
ing : the number of mistakes is considerable. In other 
WOrds the Paris Codex 450 contains a careless and inac- 
curate text which a critic need not scruple to altera 

The only other complete MS. of Justin's genuine writings 
is one which was once in the Jesuits' Library at Paris, and 
hence is known as the Codex Claromontanus, but which is 
now in the Middlehill collection at Cheltenham. It was 
written in 1541, and is merely a copy of the Paris Cod. 


There are two late MSS. which contain fragments of 

^ This account of the MSS, of Justin is entirely based upon Professor 
Hkmack's elaborate account of them in the Texte und Untersuchungen ztir 
Geschichte der altchristliche Literaiur, Bd. i. Leipzig, 1882, entitled Die 
Ueberlieferung der griechischen Apologeten des II Jahrhunderts in der alien 
Kirche und im Mittelalter. 

2 See, for details, the Theologische Literaturzeitung iov 1876, No. 13. 


Justin's genuine works : (i) in the Vatican Library, Cod. 
Ottobonianus Gr. ^^74, written in the fifteenth century, con- 
tains chapters 65-67 of the Apology : (2) in the National 
Library at Paris, Cod. Supplem. Gr. 190, is only a worthless 
transcript, made in the seventeenth century, of some extracts 
from one or other of the earlier printed editions. 

It thus appears that our only authority for almost all 
Justin's text is the Paris MS. 450, of 1364: and considering 
the character of that MS. it will not be necessary for a 
student to treat the text of Justin, as it exists in that MS., 
with the same reverential respect, and the same reluctance 
to assume the existence of an error, which he would feel in 
the case e.g. of the Alexandrine MS. of Clement. 

This account of the existing MS. evidence for Ju.stin's 
text forms a necessary preface to an examination of his 
quotations, because some untenable arguments have been 
based upon the correspondence or non-correspondence of 
those quotations with the existing MSS. of both the Old and 
the New Testaments. The most important of such argu- 
ments are those of Credner's Beitrdge zur Einleitimg in die 
biblischen Schriften: the agreements and differences be- 
tween Justin's text and the biblical texts are stated in that 
work with great minuteness : but the arguments which are 
based upon them are practically without value because they 
assume that the text of the Paris MS. represents Justin's 
own quotations from the biblical texts of his time. It may 
be shown, in disproof of that assumption, that the scribe of 
that MS., or of its original, neglected Justin's own quotations 
and copied them for himself from some other MS. : some- 
times, indeed, as in the quotation from Psalm 71 (72) in 
Tryph, 64, he was not at the trouble to copy out more than 
the beginning and ending of the passage, but after tran- 
scribing a few verses wrote ' . . . and so forth until the 
words . . . .' (καΐ το. λοιττα άχρι τον . . .) 


The following three instances will be sufficient to estab- 
lish this point : — 

(i) In Ps. 18 (19). 6 it is clear from two short quotations in 
Tryph. 69, Apol. i. 54 that Justin read iaxupos (ω? ylyas ^ραμάν 
όδοι/), because in each case he comments upon the word : the same 
inference may be drawn from Tryph. 76. But in the MS. of Tryph. 
64, in which the first six verses of the psalm are quoted at length, the 
word Ισχνρό^ is omitted. It is thus evident that in transcribing Tryph. 
46 the scribe did not follow Justin's text. The insertion of the word 
in the text which Justin used is to be noted because there is no 
trace of it in any existing MS. of the LXX. : it was probably used in 
some recension as a gloss of ylyas or as a substitute for it, yt'yas 
being a rare word, which Hesychius s.v. explains by Ισχυρός. It is 
possible that the true text of Justin himself may be not that of the 
MS. as given above, but ώ? Ζσχυρό? hpa^uv όδοΊ/, and that γιγα? may 
be an interpolation : but however this may be, the fact remains 
that Ισχνρό^ was in his text of the Psalms and that it is not in the 
text of the Psalms which is transcribed in the MS. 

(2) In Ps. 95 (96). 10 it is clear from Justin's words in Tryph. 73 
that he read ό κύρως €βασίλ(υσ€ν άπδ του ξύλου, because he comments 
upon the fact that the Jews omitted those words on account of their 
evident reference to the crucified Jesus. But in the quotation of 
the psalm which immediately follows the words are omitted, as they 
are in all existing MSS. of the Psalter, except the Verona Psalter 
and Cod. 156 (a Basle MS. of uncertain date). It is obvious that 
the scribe did not follow Justin's own text, but transcribed the Psalm 
from a MS. which contained the current text. The absence of the 
words from all MSS. of the LXX., except the two mentioned 
above, is a fact of great importance in regard to the textual tradi- 
tion of the LXX., especially in face of the facts (i) of the use which 
was made of them in the Judaeo-Christian controversies, for they 
are used against the Jews not only by Justin but also by Tertullian, 
adv. /ud., pp. 144, 146 : (2) of the words a ligno being found in 
almost all early Latin quotations of the passage (Hilary is probably 
the only exception). The existence of the words in the two Greek 
MSS. which contain them may be accounted for by the fact that 
both those MSS. are accompanied by a Latin version : and the 
form in which they occur in the Basle MS., viz. aizo τω $νλω, 


suggests the hypothesis that they are there only an attempt at 
retranslation by a mediaeval scribe. 

(3) Ps. 71 (72). 17 is quoted twice in Tryph. 121 in the form 

Wf/) τον ηλιον avareXel (sc, το όνομα αυτού). There Can be nO doubt 

that this was Justin's reading, for he supports his quotation of 
the passage by a quotation from Zach. 6. 12 ανατολή όνομα αντον, 
and his commentary is πνρωδίστ^ροί γαρ αντον 6 τής άληθίίας κα\ 
σοφίας Χόγος και φωτεινότερος μάλλον τον ήλίον δυνάμεων εστί. But in 
the quotation of the whole psalm in Tryph. 34, and in the similar 
quotation (which the scribe has shortened) in Tryph. 64, the scribe 
follows the current reading of the LXX., tt/jo τον ήλιου διαμενεΙ το 

όνομα αντου. 

It is clear from these instances that the longer quotations 
in the Paris MS. of Justin cannot be trusted as repre- 
sentatives of Justin's own text, and that arguments based 
upon them alone fall to the ground. But it is also clear 
that the untrustworthiness of the longer quotations does 
not affect the shorter quotations which form an integral 
part of Justin's own text, and which are in many cases 
confirmed by his comments. 

The following is an examination of some of these shorter 
quotations, with one longer quotation which invites special 
treatment, in order to ascertain what light they throw upon 
the text of the. LXX. 

I. Quotations from the Psalms. 

Ps. 3. 6 is quoted in Tryph. 97, and in Apol. i. 38 : in both 
quotations άντελάβετο is read, with Codd. S^, 210, as against the 
common reading άντιλήψεται. There is a similar variation of tenses 
in the early Latin quotations : but the preponderance of testimony 
is in favour of the past as against the future : the former is found 
in Lactant. Instit. 4. 19, and in the Codex Sangermanensis : the 
latter is found first in Hilar, in Psalm. 131, tom. i. 505 ; in Cypr. 
Testim. 2. 24, p. 91 the MSS. vary: both are found in Ambrose 
and Augustine. 

Ps. 21 (22). 3 is quoted not only as part of the long quotation in 
Tryph. 98, but twice separately in Tryph. 99. Li each case the 


reading is that of the current text of the LXX. κα\ ούκ eh avoiav ζμοί : 
but Justin seems to have read not avoiap but ayvoiav^ for his Avords 

are {Tryph. 99) αλλ' Ινα μή τα Xeyrj ΉγνΟ€ί ovu qti /ueλλeι πάσχειι/, 
«TTiiyet ep τω ψάΚμώ βυθνς. Και ουκ els ανοιαν €μοί. ovnep τρόπον ovde τω 
Becu eli ανοιαν ην το €ρωτάν τον Άδά/χ που eVrti/ ovdi τον Καιν που "Αβζλ 
αλλ' els το €καστον eXey^ai όποϊό$ €(ττι κα\ els ημά5 την γνώσιν πάντων δια 
του άναφανηναι eKuelv .... The whole point turns not upon folly 
but upon knowledge or ignorance : and rjyvoei would be unintelligible 
unless ayvoiav followed. 

The passage raises a wider question than that of Justin's 
reading : neither ds ανοιαν nor ets ayvoiav gives any intel- 
ligible meaning, or is an approximate translation of the 
Hebrew. The meaning of the Hebrew rT^O^l't^fpl Π^"•^*) 
^7 is clearly that there was no cessation of his crying 
in the night. The alteration of a single letter would give 
this meaning to the Greek, and I do not hesitate to suggest 
that the LXX. wrote not ets ανοιαν but ets avdav (i.e. re- 
mission or cessation, from αν'ιημι). But the word was a rare 
one: the only recorded instance of it is in a Paris MS. 
(Colbert, No. 4:^49) of ps-Athanas. Praecepta adAntiochum 
{0pp. ed. Bened. ii. 2^^, and, separately, ed. G. Dindorf, 
Lipsiae, 1857), c. 5, in a passage based upon Hermas, Mand. 
5. I, where it is probably a scribe's error for ayvdav. It 
was consequently unknown to the early scribes of the LXX., 
who substituted for it, with a complete disregard of the 
meaning of the passage, one or other of two words, ανοιαν 
and ayvoiav, which they knew better. A single MS., Cod. 
.167 (British Museum, No. SSS?,)^ has the reading €ts aviav, 
which may be a survival of eh aveiav. 

Fs. 23 (24). 7 is quoted in Trypk 85, Apol. i. 51 in the form 
€πάρθητ€ πνΚαι αιώνιοι Χνα. €ΐσ€λθτ] 6 βασιλ€ν5 ttjs Βόξη5. The reading 
of all existing MSS. of the LXX. is και eiaeXeuaerai : and this 
current reading is found both in the quotation of the whole psalm 
in Tryph. 36, and in the shorter quotation in Tryph. 127. But Iva 
elaeXur] is a closer rendering of the Hebrew : and Jerome's Psalter 
has e/ mgredi'a/ur, for which u/ ingrediatur may reasonably be con- 


jectured, as opposed to the et introibit of the Verona Psalter and 
the Codex Sangermanensis. In other words Iva (ϊσ^Χθυ may be 
supposed to be the reading which existed in the recension of the 
LXX., which was followed not only by Justin but also by the Old 
Latin versions. 

Fs. 8i (82). 7 is quoted in Tryph 124 with a comment on the 
difference between the Jewish and the LXX. interpretation. As 
the text stands it is not clear wherein the difference lies : the longer 
quotation has probably undergone the fate of most of the longer 
quotations in Justin, and is no longer in the form in which he 
wrote it. But the reading of the shorter quotation Ihov drj ως 
άνθρωποι άποθνησκ€Τ€, upon which emphasis is laid as being the 
reading of the LXX., though not found in any existing MS., is 
probably supported by the reading of Cod. S^ Be δη ως άνθρωποι, 
which may be conjectured to be an imperfect transcription of ϊδε 
drj ως άνθρωποι .... If this be SO, it must be supposed that the 
LXX. followed the Hebrew in connecting ίμ^ϊς with the preceding 
clause : and this view is supported by Jerome's Psalter ώϊ estis et 
filii excehi omnes vos. 

It will be seen from these instances that the shorter 
quotations present in almost every case some point of 
interest in regard to the critical study of the LXX. : this 
fact makes the untrustu^orthiness of the longer quotations 
more to be regretted, and leads the student to anticipate 
with hope the possible discovery of a MS. of Justin which 
shall preserve his quotations from the LXX. in their 
original form. 

There is at least one instance, that of Psalm 95 (96). i-io, 
in which it seems likely that this original form has been 
preserved: and it invites examination because the psalm 
is not only quoted twice by Justin, viz. in Apol. i. 41 and in 
Tryph. 73, but also exists in two forms in the LXX., in the 
Psalter and in i Chronicles 16. 23-31. In regard to the 
quotation in the Trypho it was pointed out above that it 
cannot be a transcription of the text which Justin used : 
but since the two phrases, €Ϊδωλα hai^oviiuv and άττό τον 


ξνλον, which, were certainly in Justin's text, though they 
are absent from the longer quotation in the Trypho are 
found in the quotation in the Apology, it may be assumed 
(i) that the two texts were originally the same, (2) that the 
Apology represents the text which Justin used. It may 
further be noted that the text in the Trypho corresponds, 
almost exactly, to the Vatican text of the LXX. Psalter, 
and represents the same tradition as that text : whereas 
the text in the Apology corresponds more nearly to that 
of I Chronicles. (In addition to the longer quotations, 
vv. 1-3 are quoted in Tryph. 74, v. 5 in Tryph, ^^^ 73, 79, 
83, V. 10 in Tryph, 73.) 

The following is a detailed examination of the quota- 
tions : 

vv. I, 2. The form of these verses in the Psalter (= Trypho) is 

ασατ€ τω κυρίω άσμα καινόν, ασατ€ τω κνρίω πάσα ή γη' ασατ€ τω κυρίω, 
ζνΚογησατ€ το όνομα αυτού, emyyeXi^ea^e ημβραν i^ ημίρας το σωτηριον 
αυτοί). There is no noteworthy variant. 

The form in i Chronicles and the Apology is shorter : ασατε τω 

κυρίω πάσα η -γη' avayyeiXaTe ίξ ημίρας els ημύραν το σωτηριον [sO Codd. 

AS and most cursives : Cod. Β and some cursives σωτηριον] αυτού. 
V. 3• The form in most MSS. of the Psalter (= Trypho), is 

avayyfi\aTe Γατταγγ^ίλατε] iv toIs Ζθν^σι την bo^av αυτοΰ, iv πάσι τοϊς Xaols 

τα θαυμάσια αυτοϋ : Cod. Α\ the Verona Psalter, and Tryph. 74, omit 
the first half of the verse, making Iv πάσι .... θαυμάσια αυτού coordinate 

with TO σωτηριον aS an object of €ύαγγ€\ίζ€σθ€ in V. 2. 

The whole verse is omitted in the Apology, and in Codd. ABS, 
and several cursives, in i Chronicles : the MSS. which contain it 
read as in the Psalms with the substitution of €ξηγεϊσθ€ for avay- 

V. 4 is the same in all four passages : except that i Chronicles 
and Justin agree with about 80 cursive MSS. of the Psalter in 
reading υπ^ρ πάντας instead of eVi πάντας, 

V. 5. The form in almost all MSS. of the Psalter (= Trypho) is 
oTi πάντ€$ οί θ€θ\ των (θνων δαιμόνια, 6 δί κύριος τους ουρανούς ΐποΊησΐν^ 



The form in I Chronicles is on πάντ€ς oi deoi των 4θνων είδωλα κα\ 6 
θίος ημών ουρανούς [ABS ουρανοί/] (ποίησαν : the Apology (so also 
Tryph. 55, 73, but not 79, 83) substitutes βϊδωλα δαιμονίων for «δωλα, 
and follows with 6 be deos rovs ουρανούς ίττοίησ^ν. The phrase Λδω\α 
δαιμονίων is Supported by Iren. Ve/. Interp. 3. 6 alone among early 
Latin authorities, and by Clem. Alex. Protrept. c. 4 alone among early 
Greek authorities : άδωΚα is used elsewhere, but δαιμόνια is not, as a 
translation of ^^ V νξ?.. The phrase in Justin, if notwithstanding its 
absence in Tryph. 79, 83 it be really his, is perhaps an intentional 
combination of the two readings. 

V. 6. The form in the Psalter (=Trypho) is ^ξυμολόγησις κα\ 

ωραιότης ενώπιον αυτοϋ, άγιωσύνη και /χεγαλοττρβττ^ια ev τω άγιάσματι 

The form in most MSS. of i Chronicles and in the Apology is δόξα 

κα\ έπαινος κατά πρόσωπον αυτοϋ, Ισχύς κα\ καύχημα iv τόπω αυτού [Apol. 
(V τόπω αγιάσματος αυτοΰ, Codd. 1 9, 93? ^^^ ^^ ''<? αγιάσματι αυτοΰ, 
Codd. 106, 120, 134? 144? 236, 243 ^^ τόπω άγιω αυτοί)]. The form 

of the last clause in Justin seems to be a combination of the readings 
of the Psalter and of Chronicles : as in the preceding verse. 

V. 7 is the same in the Psalter and i Chronicles, except that the 
former reads eWy/care and τίμην where the latter has δότε and Ισχύν. 
But in the Apology, which otherwise agrees with i Chronicles, 
Justin lias the remarkable reading δο'τε τω κυρίω τω πατρί τίαν αΧώνων 
for δότΐ τω κυρίω αί πατριαΧ των εθνών. The origin of this reading 
may probably be traced in Codd. Β S of the passage in i Chronicles, 
which read πατρί for ai πατριαί. Justin may have found a similar 
reading in the copy which he used : and πατρϊ των εθνών being an 
unusual expression was changed to τω πατρϊ των αΙώνων, a phrase 
which may be compared with the current philosophical phrase τω 
πατρϊ των όλων. 

In νν. 8, 9? ΙΟ the form in the Psalter (=Trypho) is — 

8 ενεγκατε τω κυρίω δόξαν ονόματι αυτοΰ, 

άρατε θυσίας καϊ είσπορεύεσθε εις τας αυλας αυτοΰ' 

9 προσκυνήσατε τω κυρίω εν aiikfj άγια αυτοΰ, 
σάΚευθητω άπο προσώπου αυτοΰ πάσα η γη. 

Ι Ο είπατε εν τοις εθνεσιν Ό κύριος εβασίλευσε, 

καϊ γαρ κατώρθωσε την οϊκουμενην, ήτις ού σαΚευθησεται, 
κρίνει λαούς εν εύθύτητι. 


The only noteworthy variant is in v. 10, where AS^ and most 
cursives read δτι κύρως : BS^ are supported in reading ό κύριος by 
the short quotation in Try ph. 73, and by the Old Latin. 

The form in most MSS. of i Chronicles is — 

8 Cod. A : [Codd. BS omit] hare τω κνρίω bo^av ονόματι αυτού, 
\άβ€Τ€ 8ωρα και iveyKaTC κατά πρόσωπον αύτου. 

κα\ προσκυνήσατε κυρίω [Cod. Α τω κ.] ev aiXals ayiais 

9 φοβηθητω άπο προσώπου αυτού πάσα η yrj, 
κατορθωτητω [S^ κα\ κατ.Ι ή γη κα\ μη σαΧίυθήτω. 

ΙΟ €υφρανθητω ό ουρανός κα\ άγαΧλιάσθω ή γη 

καΐ είπάτωσαν iv toXs ίθνεσιν Κύριος βασιλεύων [Cod. Α 

The form in the Apology is — 

8 Χάβετε χάριν κα\ εΙσίΧθετε κατά πρόσωπον αυτου^ 
καΐ προσκυνήσατε εν ταΐς αύΧαΙς άγίαις αύτοΰ' 

9 φοβηθητω άπο προσώπου αυτοϋ τιάσα ή γη, 
κα\ κατορθωτητω κα\ μη σαΧενθητω. 

ΙΟ ενφρανθητωσαν εν τόΐς εθνεσιν' 

6 κύριος εβασίΧευσεν άπο του ξύΧο\κ 

The noteworthy points in this text of the Apology are (i) the 
agreement with Codd. BS in the omission of the first clause of v. 8, 
(2) the use of χάρις for 8ώρον or θυσία as a translation of ΠΠ^Ό : this 
would be even more important if it were certain that Justin knew 
Hebrew : {3) the omission of ε'ίπατε in v. 10, which it is certain that 
Justin read, inasmuch as he twice quotes είπατε εν τοΙς εθνεσιν in 
Try ph. 73 : if this be restored, it may be assumed that the subjects 
of ευφρανθητωσαν in his text were ό ουρανός και η γη, as ίπ Ι Chroniclcs : 
(4) the reading από του ξύλου, for which see above, p. 189. 

It will be noted that, in the form of the psalm in the 
Psalter, (i) the two members of vv. 8, 9 respectively give 
an intelligible antithesis, (2^) the words καΐ γαρ . . . (ταλ^υ- 
θη(Τ€ταί in v. 10 not only destroy the poetical structure of 
the passage, but also introduce an idea which is not germane 
to the rest of the verse. It will also be noted that the 
clause of v. 8 which is found in Cod. A in i Chronicles 
similarly destroys the parallelism of that verse, and that its 

Ο 2 


omission, as in Codd. BS and the Apology, gives to vv. 8, 9 
a perfect poetical structure and an intelligible sequence of 
ideas. It seems very probable that the words came into 
this place in the Psalter from the similar passage in Ps. 28 
(29). 2 : that when they had become an ordinary part of 
the text, the second clause of v. 9 was omitted to restore 
the lost parallelism : and that subsequently the second 
clause of v. 9 was reinserted, in a wrong place, between the 
two clauses of v. 10. The antithesis which is found in 
I Chronicles, and probably also in Justin, between the two 
clauses of v. 10 is confirmed by Ps. 96 (97). i. 

II. Quotations from Isaiah. 

The quotations are very numerous, as may be expected 
in a writer who deals so largely with the Messianic con- 
troversy. They are almost always worth study, and in 
some cases will be found to make material contributions to 
the textual criticism of the LXX. Some of the more im- 
portant quotations occur more than once : but it is rarely 
the case that such double or triple quotations agree through- 
out : in some instances the scribe has apparently copied out 
a current text, in others he has preserved Justin's own text. 
It may be noted that the very fact of such variations in the 
case of double quotations confirms the view which has been 
advanced above as to the inexpediency of drawing in- 
ferences from the existing MS. of Justin's text in the case 
of single quotations, except where Justin's commentary 
makes his readings certain. 

The following are examples of the contributions which 
Justin's quotations make to the textual criticism of Isaiah : 

Is. 3. 10. The LXX. reading is δησωμ€ν top dUaiov on δύσχρηστος 

ημΊν €στί: there is no variant. Tryph. 17, 133, both of which are 
long quotations, have δήσωμ€ν, but Tryph. 136, 137, both of which 


are short quotations, have αρωμβρ, and in 137 Justin remarks upon 
the reading, saying that αρωμ^ν is the true reading of the LXX. and 
8ησωμ€ν the Jewish reading : he adds a remark, which is important 
for the consideration of other passages besides this, that earher in 
his treatise, i.e. in c. 17, he had himself quoted the Jewish reading 
by way of concession to those with whom he was arguing. It may 
be noted that Barnabas c. 6 has δησωμςρ ; Hegesipp. ap. Euseb. 
I/. E. 2. 23, 15, and Clem. Al. Strom. 5. 14, p. 714, have αρωμ^ν: 
TertuU. c. Marc. 3. 22 has au/eramus, but Jerome in Isai. 3, torn. iv. 
p. 57, has alligemus. Neither reading is a translation of the Hebrew 
text as we have it : but the fact that the Jews had and insisted upon 
a translation which implies another text, is an indication that the 
Hebrew text of the passage as we have it is not identical with the 
Hebrew text of the second century. 

The fact that there are no variants in the MSS. of the LXX. is 
important in its bearing upon the tradition of the LXX. text : it 
confirms the view that we owe that text to Jewish rather than to 
Christian scribes. 

Is. 7. 10-17 ^s quoted at length in Tryph.- ^'^, 66 : v. 14 also in 
Apol. 33, and v. 14 λ in Try ph. 67, 71, 84. 

In V. 10 there is no variant: in v. 11 Justin's MS. supports the 
reading τον β^ον of Cod. S and 10 cursives as against B^ov : in v. 12 
there is no variant : in v. 13 the addition of Ήσαία? to etTrej/ is sup- 
ported, and ακοΰίτΐ is read for ακούσατε. 

In V. 14 Tryph. 43 reads mXeaerai (perhaps by a not uncommon 
scribe's error for fcaXeVere, which is found in Cod. XII and several 
cursives, and in the Old Latin), and Tryph. 66 reads κάλεσονσι (which 
is found in several cursives and is the common reading in the Greek 
Fathers, no doubt on account of its being the reading of Matt. i. 23) : 
the same two quotations in the Trypho, and also the short quota- 
tions in 67, 71, 84 have iv yaarpl λη-\ρ•€ταί, which is read in Codd. 
AS, XII, 26, 41, 90, 106, 144, 239, 306. But Apol. 33 has the 

singular reading Ιδού ή παρθένος iv γαστρί €^ei Koi re^erai viov και ipovaiv 
erri τω ονόματι αυτοΰ Me^' ήμων ό θ^ός. The reading iv γαστρί e^ei is 

repeated in the same chapter in a way which shows that Justin 
must have read it, for he uses συλλαβ€'Ίν to explain it: and the 
passage is the more remarkable because Justin lays stress on giving 
it αντοΚφί, ' word for word.' The ipovai is perhaps the source of 
the καλ/σουσί in Matthew : but otherwise there is no trace of this 


translation of the second clause of the verse, which is perhaps a 
tinique survival of a lost Targum. 

In V. 15 Tryph. 43 agrees with the current text of the LXX. in 
reading και ΙκΚί^ασθαι, but Tryph. 66 agrees with AS^ and 17 cursives 

in reading eKKe^erai. 

In v. 16 both quotations agree with AS^ and 14 cursives in read- 
ing τυν before (κΚίξασθαι : in the same verse Tryph. 43 reads aireiBei 
πονηρά for the current LXX. reading air^Ldei πονηρία : only two cursives 
have a variant, viz. Codd. 93, 305 which read πονηρίαν, and the early 
Latin quotations read non credit {credet, credidii) malitiae, or (Iren. 
Vet. Interp. 3. 21) non consentiet nequitiae. But the translation in 
August. Uh. 8 de Gen. ad lit., torn. 3. 237 contemnet malitiam^ taken 
in connexion with the use of the accusative case in Justin and two 
MSS. of the LXX. and with the fact that άπωθάν is frequently used 
as the translation of ^Ψ^, ' to despise,' gives a plausibility to Wolfs 
conjecture that άπ€ΐθ€Ϊ is a scribe's mistake for άπωθβΐ. 

But in v. 16 both quotations agree in inserting c. 8. 4, and it is 
evident from Tertull. c.Jud. 9, p. 141, c. Marc. 3. 12, p. 673, that 
the insertion existed in the text which Tertullian used. It may be 
that the insertion is due only to a scribe's reminiscence of the 
inserted passage, which has part of the same protasis, πρ\ν η γνώναι 
το τταώίον .... , as a clause of v. 16 : but this does not altogether 
explain the fact of its being so far recognized as to be used with 
emphasis in the Judaeo-Christian controversy. 

Is. 29. 14 is quoted thrice, Tryph. 32, 78, 123 : in each case with 
a slight variation which may be compared with both the LXX. and 
with the quotation of the passage in i Corinthians i. 19. 

LXX. ατΓολώ τψ σοφίαν των σοφών [several CUrsivcS add αντοΰ, or 
αντών~^ και την σννεσιν των συνετών [the same CUrsiveS add 
αντον or αντών'^ κρύψω [Cod. 30Ι α^^τ^σωΐ. 

I Cor. I. 19 ατΓολώ την σοφίαν των σοφών κα\ την σύνεσιν των σνν^τών 

Tryph. 32 άφίΚώ την σοφίαν τών σοφών κα\ την σννεσιν τών συνετών 
αυτών κρύψω. 

id. 78 άφεΚώ την σοφίαν τών σοφών αυτών την δε σύνεσιν τών συνετών 

id. 123 ατΓολω την σοφίαν τών σοφών καϊ την σύνεσιν τών συνετών 

The reading άφελώ is supported by Tert. c. Marc. 3. 6, p. 670 



auferam sapientiam sapieniium illorum, ibid. 5. ιι,ρ. 793: but the 
same writer also shows the existence of various readings, for ibid. 
4. 25, p. 719 he hdi^perdam sapientiam sapieniium : at the same time 
it must be noted that άπολλνω is the ordinary translation of "Ι?ζ5, and 
that άφαφίω is never elsewhere used as the translation of it. The 
addition of αύτώρ to σοφών, in c. 78, and to συνετών in c. 32, is in 
harmony with the Hebrew, and is supported by good cursives of 
the LXX. : the omission of the words both in i Corinthians and in 
the uncials of the LXX. is probably due to an adaptation to the 
immediate purpose of the writer. 

Is. 42. 1-4 is quoted in Tryph. 123, 135, and the quotations 
which diifer in many respects from each other, so that they cannot 
both be due to the scribe's transcription from a current text, have 
some points of interest in relation to the similar quotation in 
St. Matt. 12. 18-21. 

The following is a detailed comparison of the four texts : 


St. Matt. 12. 

Tryph. 123. 

Tryph. 135. 

Ίακ^β [Codd. 



106, 302, 305 

I80V Ίακωβ] 6 

Ihov παΐί μον 

6 rrals μου αντι- 

ό παΪ5 μου αντι- 

TTois μου άντιΧη- 

ον τ]ρ€Τίσα' 

λήψομαι αυτού, 

ληψομαι αυτού' 

■ψομαι αυτοί)' 

Ισραήλ 6 ckXck- 

6 αγαπητός μου 


κα\ Ίσραηλ ό e«- 

Tos μου προσδεδ^'- 

[(ls\ ον ηυΒόκησ€ν 


XcktOs μου προσ- 

^ατο αυτόν η ψνχη 

η ψνχη μου' 

Βίξζται αυτόν η 


ψυχή μου' 

(δωκα το πν^υμά 

θησω το πν^ϋμά 

θησω το TTi'eC/xci 

δεδωκα το πν€ΰ- 

μου €7Γ αυτόν, 

μου in αυτόν 

μου cV αυτόν 

μά μου eV αυτόν' 

κρίσιν τοις eOve- 

κα\ κρίσιν τοΙς 

καϊ κρίσιν τοϊς 

και κρίσιν τοις 

σιν (ζοίσζΐ. 

ίΘν^σιν απαγγβλβΐ 

€θν€σιν (ζοίσίΐ 

ίθνζσιν €ξοίσ€ΐ. 

It will be noted (ι) that both quotations in Justin agree with the 
LXX. in asserting, what St. Matthew agrees with the Hebrew in 
omitting, the names Jacob and Israel. That the insertion of the 
words in Justin is not accidental is proved by his quoting them 
separately, c. 123, and giving them a Messianic interpretation: 
(2) that Tryph. 123 agrees with St. Matthew in reading βησω, but 
that the passage has not been altered to harmonize with St. Matthew 



is made probable by the retention in both Justin's quotations of the 
LXX. i^oiVfi as against a7ra77eXei. 

It may also be noted that while the translation of ">''Π? by 
άγαττητός is peculiar to St. Matthew, the rest of St. Matthew's 
phrase is identical with Theodotion's translation of ''ΨΡ1 "^^Vl. 

LXX. S/. Matt. 12. Tryph. 123. Tryph. 135. 

ov Κ€κράξ€ται 

ovbe άκονσθη- 
aerai ξζω η φωνή 

ου Κίκράζ^ται ουκ epiaei ουδέ ουκ iplaei οντ€ 

ovbe άνησΐΐ \βοη- κραυγάσει, κράζα, 
ση Cod. 308], 

οΰδί άκονσθη- ovbe άκονσ€ΐ tls ovtc άκονσ^ταί 

σ€ται (ξω η φωνή ev rais πλατ^ίαις ης eV rats πλα- 

αυτον' την φωνην αντον' τβίαις την φωνην 


It will be observed that the LXX. άνηση does not exist in any of 
the other quotations : that it was the original LXX. translation is 
made probable by the fact (i) that ^ψ^ is rendered by άνίημι in three 
other passages of Isaiah (more commonly, both in Isaiah and else- 
where, by αίρω), iz) that it underlies the Old Latin versions dimittet 
and relinquet^ Hieron. Ep. 121 adAlgas. qu. 2, tom. i. 848, in Isai. 42, 
tom. iv. 506, and cessabit August, de Civit. Dei 20. 30. That it 
was felt to be a difficult expression may perhaps be inferred from 
its omission not only in Tryph. 135, above, but also in Tertull. 
c. Marc. 4. 23, p. 717, Cypr. Testim. 2. 13, p. 78. And that the βοηση 
of Cod. 308 was an early variant is shown by Tertull. c.Jud. 9, 
p. 143 neque contendit neque clamavit, where the quotation must be 
from Isaiah and not from St. Matthew, because /oris and not 
in plateis follows. 

καΚαμοντΐ&Κασ- κάΚαμον σνντ€' καΚαμον συντζ- καΚαμον τ^βραν- 
μίνον [Codd. Α τριμμίνον τριμμίνον σμ4νον 

23, 41, 87, 91, 
97, ιο6, 228, 
3θ8, 309, σ^»'- 

Τ€θΚασμ4νον\ ου ου KaTea^ei κα\ ου κατίάζη κα\ ου σύντριψη κα\ 

συντρί'^Ιτη, κα\ \ί- \ivov τνφόμΐνον Χίνον τυφόμ^νον \ivov τυφομ^νον 

νον καπνιζόμ^νον ου [D ου μη\ ου μη σβίση αλλά ου σβ^ση (ως οβ 

ού σβ€ση αλλ* eis σβ^ση €ως &ν e<- ets• άληθίΐαρ €ζοί- νΐκος (ζοίση κρί- 

αΚηθηαν e^oiVft βαΚη ης νΐκος την ση κρίσιν. σιν. 

κρίσιν. κρίσιν. 


The variations between (<2) τ^&Κασμίνον, σνντ€θλασμ€νον, συντ€τριμ- 

μίνον^ and τ€θρανσμ€νον, {d) συντρί^\τ€ΐ and κατίάξζΐ, correspond to 
variations in the early Latin versions between {a) fractam^ con- 
fractam^ contusam, and quassatam, {b) conteret, comminuet, /regit, 
confringet : they must therefore be taken to mark an early diffi- 
culty, and a consequent early variety, in the rendering of the contrast 
between Γ^^ and "i?f . 

The variations in the rendering of the last clause may perhaps be 
best explained by noting that els νΊκος is interchanged with els reXos 
as a translation of nvj or n^Jp, * for ever/ i.e. utterly or completely : 
it is consequently conceivable that it may have come to be used as 
an equivalent for els άληθ€ΐαν or eV άληθ^ία, ' truly' or ' really.' 

άvaXάμψeι και ου άναΧηψίΐ και ου άναληψει και ου 

θρανσθησ€ται [S μη 6pavaer]aeTai θρανσθησεται eωs 

σβeσθησeτaι^ βωί ews αν 6rj eVi τψ αν θτ} en\ ttjs yrjs 

αν θτ] eni Trjs yrjs Ύηί κρίσιν' κρίσιν' 


κα\ iin τω ονό- και τω ονόματι 
μάτι αυτόν ίθνη αυτοΰ έθνη eX- 
ίΚπιοΰσι' πιοΰσι 

The reading of Justin's MS., άνα\ηψ€ΐ, would no doubt be in an 
earlier MS. άνάλημψ€ΐ, which was originally only a scribe's error for 


The omission of the clause άva\άμψeι .... κρίσιν in St. Matthew 
is perhaps best explained by the hypothesis of a homoioteleuton 
κρίσιν .... κρίσιν in an early MS. 

The absence of any trace either in the MSS., or in the quotations, 
or in the early Latin versions, of any variation in the last clause, in 
other words the fact that all early recensions of the LXX. agreed 

in translating νΓΓ^ D"'JN ^ΠΊίΠρ^ by (eVi) τω ονόματι αυτοΰ €θνη ^λπιουσί, 

whereas the later revisers, Aquila, Symmachus, and Theodotion, 
agreed with modern scholars in translating the passage by τω νόμω 
αυτοΰ νήσοι ίλπιοΰσι, seems to point to a lost variant in the Hebrew 

Is. 53 is largely quoted, and some of the quotations are useful 
contributions to the criticism of the LXX. The following are the 
more noteworthy. y 

V. 2 is quoted in Apol. i. 50, Tryp/i, 13, 42, in each case placing 

και em τω ονό- 

και eTTi τω ονό- 

ματι αυτοΰ eXni- 

ματι αυτοΰ eXm- 

οΰσιν €θνη 

οΰσιν ΐθνη 


the words ως παώίον immediately before cos ρίζα. This is the reading 
of Codd. AS, XII, 22, 26, 36, 48, 86, 90, 93, 106, 144, 147, 198, 
233, 306, 308, and of Clem. Rom. i. 16. 2. 

V. 8 3 is quoted in ApoL i. 51, Try ph. 13, with the variant 17x61 for 
r\χθ^, and in Tryph. 43 ηχβψ. ηκα is found also in Codd. 62, 90, 
144, 147, 233, and in Clem. Rom. i. 16. 9 : but the Latin versions 
all have ductus est or adductus est. 

V. 9 is quoted in Apol. i. 51, Tryph. 13, with the reading ovhk 
{ονχ) ΐνρίθη bokos iv τω στόματι αντοΰ, in agreement with Codd. AS^, 
XII, 26, 36, 41, 49, 51, 86, 90, 91, (93), 104, 106, 144, 147, 198, 228, 
233, 239, 306, 308, 309, [Codd. 87, 97 have ovdi doXos, Cod. Β has 
ovde doXov, without €νρ€θη]. It seems probable that the original 
reading was ovbe doXos, which is a literal rendering of the Hebrew, 
and that {a) δόλον arose from assimilation to the preceding άνομίαν^ 
(δ) (νρβθη was supplied by way of exegesis. The antiquity of 
the accusative doXov is shown by its translations insidias in Cypr. 
Testim. 2. 15, p. 80, and do/um in August, de Civit. Dei 18. 29, 
torn. 7. 510, and elsewhere : Faustin. de Trinit. 3. 4, further proves 
its existence by the reading neque dolum in ore locutus est. But 
Tertull. c. Jud. 10, p. 144, has nee dolus in ore ejus inventus est. 

v. 12 is quoted in Apol. i. 51, Tryph. 13, with only a slight 
variation from the current text of the LXX. : but at the beginning 
of Apol. i. 50 it is prefixed to the quotation of c. 52. 13 — 53. 8, and 

instead of the current text αντος αμαρτίας πολλών άνην^γκί καΐ δια 
τάς ανομίας αυτών παρεδόθη is the important variant αυτός αμαρτίας 
πολλών €ίληφ€ και τοις άνόμοις εξίλάσεται. This last clause brings the 

Greek into harmony with the Hebrew ^32! η^νψΐΟ\ ' he made in- 
tercession for the transgressors,' but there is no trace of the reading 
elsewhere : it must be taken to be part of a lost revision of the LXX. 
of which Justin made use but which is otherwise unknown. 


It would be improbable, even if there were no positive 
evidence on the point, that the Greek-speaking Jews, who 
were themselves cultured, and who lived in great centres of 
culture, should not have had a literature of their own. It 
is no less improbable that such a literature should have 
consisted only of the Apocalyptic books, and the scanty 
fragments of other books, which have come down to us. It 
may naturally be supposed that a race which laid stress 
on moral progress, whose religious services had variable 
elements of both prayer and praise, and which was carry- 
ing on an active propaganda, would have, among other 
books, manuals of morals, of devotion, and of controversy. 
It may also be supposed, if we take into consideration the 
contemporary habit of making collections of excerpta^ and 
the special authority which the Jews attached to their 
sacred books, that some of these manuals would consist 
of extracts from the Old Testament. 

The existence of composite quotations in the New Testa- 
ment, and in some of the early Fathers suggests the hypo- 
thesis that we have in them relics of such manuals. The 
passages which are examined in the following chapter are 
more consistent with such a hypothesis than with any 
other. The view that they are mere misquotations in which 
the several writers have, through defect of memory, blended 
several passages into one is rendered improbable by the 


whole character of the quotations which they make from 
the Old Testament : it will be clear from the preceding 
chapter that such quotations were ordinarily made with 
great accuracy, and that the existence of a discrepancy 
between them and the existing MSS. points not to an in- 
accuracy on the part of the writer but to a variation in the 
current text. The view, which might otherwise be tenable, 
that such passages are combinations, such as might be 
made by any writer who was familiar with the text of the 
Old Testament, is set aside by the fact that in some cases 
the same, or nearly the same, combinations occur in dif- 
ferent writers. Two instances of this will be found below, 
viz. (i) the composite quotation, Jer. %. 12, 13, Is. 16. i, 2, 
which is found in both Barnabas 11, and in Justin M. Tryph. 
114: (%) the composite quotation from the Psalms and 
Isaiah, which is found in the New Testament, Romans 3. 
10-18 and in Justin M. Tryph, ιη. 

1. Clement of Borne. 

(l) C. XV. 

In c. 15 there is a passage which is composed of Ps. 77 
{78). 36, ?,1 : 30 (31). 18:11 (12). 4*5-5 : 

Ps. 77 (78) ηγάπησαν αυτόν iv τω στόματι αυτών 

κα\ TTJ γλωσση αντων εψ^ύσαντο αυτω [sO Cod. Alex. 

and Clem. Alex. : Cod. Const, e^c^av αυτόνΐ • 

ή de Kapbta αυτών ουκ eiiOela /xer αυτοΰ 

ονδε Ιπιστωθησαν iv Tjj διαθηκτ] αύτοϋ. 
Ps. 30 (31) (δια τούτο) άλαλα γίνηθητω τα χ(1λη τα ^όλια, 
Ps. II (12) γλώσσα μεγαλορημων [sO Cod. ConSt. I Cod. AlcX. 
γλώσσαν μεγάλο ρήμοναί, 

τους elnovTos την γλώσσαν ημών μεγαλννοϋμΐν 

τα χείλη ημών παρ' ημίν εστίν' τις ημών κύριος εστίν ; 

άπο της ταλαιπωρίας τών πτωχών κα\ άπο του στεναγμού 
τών πενήτων, 

νυν άναστησομαι, λέγει Κύριος, 

θησομαι εν σωτηρίω' παρρησιάσομαι εν αυτφ. 


The text of Clement is not certain : recent editors, Lightfoot, and 
Gebhardt and Harnack, insert the first clause of Ps. 11 (12). 4 α 

ΐξοΚοθρζνσαι κύριος πάντα τα χ^ίλη τα. δόλια after τα χ^ίλη τα δόλια, and 

follow Cod. Alex, in reading the accusative γλώσσαν μίγάλορημονα: 
this gives a good grammatical construction for τους Σπόντας but 
destroys the parallelism. The harshness of the construction without 
a governing verb was evidently seen by the scribe of Cod. Const, 
for he prefaces τους elnovTas by the words κα\ πάλιν, as though it 
were a separate quotation. But this confirms his reading. 

Whether the words be inserted or not, the sense of the cento 
is consecutive. 

The same cento is also found in Clement of Alexandria, 
Strom. 4. 5, p. 577 : that it comes from the same source is 
shown by the use of the words δια τούτο, which are not 
found in the LXX., in introducing the half verse from Ps. 
30 (31) : and it is to be noted that whereas in Clement of 
Rome the quotations from Is. cn^. 13, Ps. 61 (62). 5, which 
precede it, are separated from it and from each other by 
the introduction of the words τιάλιν Xiy^i . . . . καΧ ττάλιν 
Aeyet, in Clement of Alexandria there is no such distinction 
between the quotations, and the whole series of passages 
forms a single cento. 

(2) c. XXII. 

In c. 2,2, after quoting Ps. ^^ (34). 12-18 with great fidelity 
to the existing text of the LXX., instead of the following 
verses of the Psalm, Clement adds Ps. 31 (32). 10, 

πολλαι ai μάστι•γ€ς τον αμαρτωλού, 

Tovs δε (λπίζοντας eVt κύριον eXeoy κυκλώσει, 

which preserves the sequence and antithesis of the passage 
so well that the whole quotation may be taken to be a 
separate current poem, formed of the second part of Ps. 
3'i (34) — the psalm is divided by the bLάψaλμa after v. 11 — 
with an abridged ending, which has been transferred from 
Ps. 31 (32). 


(3) C. XXXIV. 

In c. 34 there is a passage in which Daniel 7. 10 and 
Isaiah 6. 3 are blended together. 
The passage in Daniel is — 

'Thousand thousands ministered unto him, and ten thousand 
times ten thousand stood before him.' 

The passage in Isaiah is (after the description of the 
seraphim with six wings) — 

' And one cried unto another and said Holy, holy, holy is the 
Lord of hosts ; the whole earth is full of his glory.' 

The passage in Clement is — 

μύριαι μυριάδίζ παρ€ΐστηκ€ΐσαν αύτω κα\ χίλιαι xiXiabes eXeirovpyovp 
αντω και €Κ€κραγορ' "Αγιος, ayios, άγιος κύριος σαβαώθ, πΚηρης πάσα ή 
κτ'ισις της δόζης αυτόν. 

(4) C. L. 

In C. 50 there is a passage in which Is. 26. 20 and pro- 
bably either Ezek. ^y. 12, 13 or 4 Esdr. 2. 16 are blended 

The passage in Isaiah is — 

* Enter thou into thy chambers and shut thy doors about thee : 
hide thyself for a little moment, until the indignation be overpast.' 

The passage in Ezekiel is — 

* Behold, I will open your graves and cause you to come up out 
of your graves, Ο my people.' 

The passage in 4 Esdras is — 

'Those that be dead will I raise up again from their places, 
and bring them out of the graves : for I have known my name in 

The passage in Clement is — 

elaiXBere €ΐς τα rn/xeia μικροί/ όσον όσον €ως ου TrapeXOij η οργή κα\ ό 
θυμός μου' και μνησθησομαι ημίρας άγαβης κα\ αναστήσω υμάς ck των θηκών 


(5) C. LVI. 
In c. 56 there is a passage which is composed of Ps. 117 
(118). 18, Prov. 3. 12, and Ps. 140 (141). 5: 

J^S. 117 (118) παώ^υων ίττα'ώΐυσίν μβ ό κύριος, 

καϊ τω θανάτω οΰ τταρί^ωκίν μ€' 
Prov. 3 ^ν γαρ άγαπα κύριος παώεύζΐ [sO Codd. AS in LXX., 

Cod. Β eXeyxeij 

μαστιγοΊ. be ττάντα νίον ov παραΒ^χζται. 
Ps. 140 (141) τταώζύσΐΐ μ€ γάρ {φη<τι) Βίκαιος iv eXeei και eXey^ei /ze, 

eXaiov de αμαρτωλών μη Χιπανάτω την κίφαΧην μην. 

But the want of cohesion between the third quotation 
and the two first makes it probable that this is rather a 
series of quotations on a cognate subject than a single 
quotation from a composite poem. 

2. Barnabas. 

(l) C. V. 

In c. 5 there is a passage which is composed of Ps. 118 
(119). 120: 21 {12). 17 : 

Ps. 118 (119) καθηΧωσόν μου τας σάρκας, 

Ps. 21 (22) δτι πονηρενομίνων συναγωγαΐ ίττανίστησάν μοι. 

It is immediately preceded by the quotation of Ps. 21 
{22). 21, but the καί which (in Codd. Sin. Const.) immediately 
precedes seems to mark it as a separate quotation. 

Neither of the quotations corresponds exactly to the 
text of the LXX. : (i) in Ps. 118 (119) the LXX. text is 
καθηλωσον εκ του φόβου σου ras σάρκας μου : (2) in Ps. 21 (22) 
it is συναγωγή Έονηρ^υομίνων ττεριεσχομ /xe. In other words 
the quotation is not from the LXX. but from a psalm based 
upon the LXX. : but it possibly has a critical value in that 
it may help to solve the difficulty which the words καθήλωσαν 
μου τ as σάρκας present in Ps. 118(119). These words are 
not in any sense a translation of the Hebrew, which means 


' My flesh trembleth for fear of thee : ' and they have no 
appreciable bearing upon the context. They must have 
been in early MSS. of the LXX. because they are trans- 
lated in the Old Latin versions * Confige (infige) timore tuo 
carnes meas : ' and Hilary, Ambrose, and Augustine com- 
ment upon the unusual expression. A clue to the original 
reading is afforded by Aquila's translation ηλώθη . . . . η σαρξ 
μου: and it may be conjectured that the present reading is 
due to a scribe's recollection of the composite psalm which 
Barnabas here quotes, or possibly adapts. 

(2) c. XI. 

In c. II is a passage composed of Jerem. 2. 12, 13 and 
Is. 16. I, :^ : 

Xeyci yap 6 προφήτης (Jcr. 2. 12)• €κστηθι ovpape, και eVt τούτω πλείον 
φρι,ζάτω η γη οτι bvo και πονηρά ζποίησίν 6 Xaos ovtos' epe ΐγκατίΚιπον πη- 
γην ζωής κα\ iavTols ώρυζαν βόθρον θανάτου' (Is. 1 6. ΐ) μη πέτρα (ρημόί 
εστίν το opos το ayiov μου Σινά ^ εσεσθΐ yap oas πετεινού νοσσοϊ άνιπτάμενοι 
νοσσιάς αφηρημένης. 

The critical interest of the quotation is considerable : the 
text of the quotation from Jeremiah is in some points 
nearer to the Hebrev^ than the LXX. is, but the substitution 
of βόθρον Θανάτου, ' an empty pit into which they will fall and 
be killed,' is a complete change of the metaphor : the text 
of that from Isaiah is nearer to the LXX., and preserves the 
points in which the LXX. differs from the Hebrew : it may 
therefore be presumed to be quoted from the LXX. If so, 
it affords an important correction of the LXX. text : for 
whereas all the MSS. of the LXX. have Σιων, the context 
and the Hebrew require Σίνα, which is read in all MSS. of 

The quotation has the further interest of being also 
found, with some changes, in Justin M. Tryph. 114, where 
the whole of it is attributed to Jeremiah. Justin's quo- 
tation consists of Jer. ij. 13, Is. 16. i, Jer. 3. 8 : 


oval νμίν, (Jcr. 2. 1 3) ort eyKareXiTreTe πηγην ζώσαν και ώρύ^ατ€ 
eauTotff "KUkkovs σνντΐτριμμβνονς οι ον δννησονται σννίχ^ιν ύδωρ* ^Is. 16. ΐ) 
μη €ρημον fj ου βστι το opos Σιώι/ ort 'ίβρονσάΚημ βιβλίον άποστασίου έδωκα 
βμπροσθεν νμων j 

It may be noted, without discussing in full the critical 
points of the quotation, (i) that Justin's text follows the 
LXX. in having kaKKovs συντ€τρίμμ€νονί for the βόθρον 
θανάτου of Barnabas : (2) that it preserves the Σιων of the 
LXX. text as against the Σινά of Barnabas. 

(3) c. xvi. 
In c. 1 6 is a passage composed of Is. 40. 12: 66. i . 

(Is. 40. 1 2) τίς ΐμίτρησΐν τον ovpavov σπιθαμή η τις την yrjv δρακί ; 
ουκ βγω ; Xeyet κύριος (Is. 66. ΐ) 6 ουρανός μοι θρόνος η δε γη νποπόδιον 
των ΊΓοδων μου' ποίον οίκον οίκοδομησίτί μοι ] η τίς τόπος της καταπανσεως 

The text of the quotation from c. 40 nearly corresponds 
to the LXX., rfi χζφΐ το vhωp being omitted, as it is also 
in the quotation in Clem. Alex. Protrept. 8, which shows 
that a recension in which the words were omitted was 
current : that of the quotation from c. 66 agrees throughout 
with Codd. AS, except only ns roviO's for ttoios totios^ and 
with Cod. 16 except only in omitting Aeyet κνριο^ after 
οικοδο/Λ?ίσ€τ€ μοι. 

3. Justin Martyr. 

(i) Tryph. c. XXVII. 

The most interesting of the composite quotations in Justin 
is that of Tryph. 27. It forms part of the same cento which 
is quoted by St. Paul, Romans 3. 10-18, and is made up of 
passages from Ps. 13 (14). i, 2, 3 (or 52 {^z)- 2» 3) = 5- 9 = 
139 (140). 4 : 9. 28 (10. 7). Is. 59. 7, 8. 




Ps. 13(14). I h. 

ovK €στι ποιών χρηστό- 
τητα^οϋκ ΐστιν ea>S€v6s\, 

Ps. 52 (53). 2 h. 
ουκ €στι ποιών αγαθόν 

Ps. 13 (14)• 2, 3«• 
52 (53)• 3. 4- 

τον idelv ei €στι 

σννιών η (κζητών τον 

πάντα €^€K\ivaVf άμα 

ουκ €στι ποιών χρηστό- 
τητα [Ps. 52 αγαθόν^ 

Pom. 3• 
ν. ΙΟ. 
ουκ Ζστιν dUaios ουδΐ 


Try ph. 27. 

νν. Ιϊ, 12. 

OVK %στιν 6 σννιών, 
ονκ €στιν 6 €κζητών τον 

πάντ€ς i^iicKivav, άμα πάντες (γαρ) (ξβκΚιναν, 
ηχρ€ΐωθησαν, αμα [MS. apaj ηχρειώ- 


ονκ €στιν 6 ποιών χρη- ονκ %στιν 6 σννιών, 
στότητα, ονκ €στιν (ως ονκ %στιν €ω: ivos' 

P.f. [13(14) 3:] 5• 
ΙΟ δ. 

τάφος aveoiypivos 6 
λάρνγξ αντών, 

ταΐς γλώσσαΐί αντών 

Ρ.Γ. [ΐ3(ΐ4).3:] 139 
(ΐ4θ). 4. 

toff ασπίδων νπο τα 

χ^Ιλη αυτών' 

Ρ^. [13(14)3:] 9. 28 

ον apas το στόμα αντον 
γ€μ€ΐ κα\ πικρίας' 

[Λ. 13(14)• 3] Is. 59• 

7, 8. ^ 

οί de πόδβί αυτών τα- 
χινοί ίκχίαι αίμα [Ps. 
13 (14)• ^^^'■^ ο*' πόδίί 
αντών €κχ€αι αΐμα^. 

V. 13. 

τάφος άν^ωγμίνος 6 
Χάρνγ^ αντών, 

ταΐς γλώσσαις αντών 

ταΊς γΚώσσαις αντών 

τάφος άν^ωγμίνος ό 
λάρνγξ αντών 

Ιος ασπίδων νπο τα Ιος ασπίδων νπο τα 

χ^ίλη αντών 

V. 14- 

ων το στόμα άρας κα\ 
πικρίας ye/xet' 

νν. 15, ι6, 17- 

0^619 01 πόδες αντών 
ΐκχίαι αίμα' 

χίίί^η αντών 



σνντριμμα και σνντριμμα και ταΚαι- σίιντριμμα και τάΚαι- 

τάΚαιπωρία iv ταϊς odois πωρία iv reus odois ττωρία iv τοις oBois 

αυτών, αυτών, αυτών, 

κα\ Sdov ΐίρηνης ουκ καί ό8ον ΐίρηνης ουκ κα\ obov ζΐρψης ουκ 

οϊδασι' (γνωσαν έγνωσαν' 

Ps. 35 (36). Ι h. ν. 1 8. 

ουκ Ζστι φόβος $eov ουκ ΐστι φόβος θίου 
άπίναντι των όφβαΚμών απέναντι τών οφθαλμών 
αύτοΰ. αυτών. 

There can be no reasonable doubt that the text of 
Ps. 13 (14) has been tampered with to make it agree with 
the quotation by St. Paul. The verses and words inserted 
above in square brackets are not found either in the Hebrew 
or in the majority of MSS. of the LXX. : they are found in 
BS\ but omitted by AS^ and 94 cursives. Jerome, Praef. 
m Isai. S7^ toin. iv. 66"]^ writes on the subject of their in- 
sertion, and says that all Greek commentators obelized 
them, and so admitted that they were not in the original 
text of the LXX. but in the Κοινή. 

{i) Tryph. c. xxiv. 

In Tryph. 24 are two quotations which might be con- 
sidered to be one, except that the introduction of the 
phrase βοα δια *Ησαιου appears to make a distinction be- 
tween them. 

The second quotation is from Is. 6^. 1, 'Z, ^ a. 

The first quotation is composite and is drawn partly from 
Is. 3. 5, 6, 9 and partly from unknown sources : 

dexJTe συν ίμο\ πάντες οι φοβούμενοι τον θεόν, 
οι θελοντες τα άγαθα ΊερονσάΚημ ΙΒεΊν' 

^ευτε πορενθώμεν τω φωτΐ κυρίου' 

άνηκε yap τον Χαον αυτού τον οίκον Ιακώβ* 

^εντε πάντα τα έθνη σνναχθώμεν εΙς *Ιερονσα\ημ 
την μηκετι ποΧεμονμενην bia τας ανομίας τών λαών, 
Ρ 2 


The source of the first strophe is unknown. The second 
strophe is from Is. 2,, ^b^6 «, with Ιακώβ, as in many cursives, 
instead of Ίσραηλ which is read by Codd. ABS. It is also 
evident that άνηκ€ is used by Justin in the sense of ' par- 
doned,' as in Is. l. 14 ονκίτι άνήσω ras αμαρτίας νμων : but 
that is clearly not the sense in which it is used by the 
LXX. here, or in which Justin himself uses it in a more 
exact quotation of the passage in Tryph. 135 : the Hebrew 
tLtoj and the context require it to mean ' forsook.' The 
source of the third strophe is also unknown. 

The three strophes evidently form part of a fine poem, 
a relic probably of the Judaeo-Christian poetry, of which 
the Sibylline Books are almost the only other remaining 

(3) ApoL I. c. Lii. 

In the First Apology c. 52 is a passage which, though 
assigned to Zechariah, differs so widely from the text of 
Zechariah as to be in reality a composite quotation, into 
which some passages of Zechariah enter. 

I ΙντΐΚονμαι rots reaaapaiv avepois 
σννάζαι τα €σκορπΐ(τμ€να τίκνα^ 
ivreKovpai τω βορρά φζρίΐν 
κα\ τω νότω μη προσκσητΐΐν' 
5 καΧ τότ€ iv Ιερουσαλήμ KoneTos μίγας, 
ου κοπετός στομάτων η χεΐλεων, 
αλλά κοπετοί Kapbias' 
κα\ ου μη σχ^ίσωσιν αυτών τα Ιμάτια^ 
αλλά τας diavoias' 
ΙΟ κόψονται φυΧη προς φυλην' 

κα\ τότε οψονταί εΙς ον εξεκεντησαν 

και εροΰσι' τι κύριε επΧάνησας ημάς άπο της όδοΟ σοι» * 

ή δόζα ην ευλόγησαν οι πατέρες ημών 

€γενηθη ήμΐν εΙς ονει8ος. 

11. Ι, 2 are a reminiscence, but not a quotation, of LXX. Zech. 
2. 6 εκ Ίων τεσσάρων άνεμων του ουρανού συνάξω υμάς, Χεγει κύριος. 


11. 3, 4 are a similar reminiscence of LXX. Is. 43. 6 epS) τω βορρά 

"Aye, και τω Αφ\ Μη κώλν€. 

I. 5 resembles Zech. 12. 11 쀕γαΚννθησ€ται 6 κοπετός iv Ιερουσαλήμ. 

II. 6, 7 cannot be traced. 

11. 8, 9 resemble Joel 2. 13 διαρρήξατε τάς καρδίας υμών κα\ μη τα 
ιμάτια υμών. 

1. ΙΟ expresses the same idea as Zech. 12. 12 κα\ κόψεται η γη κατά 

φυ\α9 φυΧάς. 

1. II is a translation of Zech. 12. 10: whether it is that of the 
LXX. is uncertain: the majority of the MSS. in that passage have 
the singular reading ίπιβλίψονται προς pe άν& hv κατωρχήσαΐ'το, which 
Jerome notes as having arisen from a mistake of the Seventy, who 
confounded llpl from li^*], 'to pierce/ with ΠΡ"Ι from ΠΡ"^, 'to 
dance': but (i) Codd. 22, 23, 26, 36, 57, 62, 68, 86, 87, 95, 97, 
ii4j 157^ 185, 228, 238, 240, some of which, e.g. 26, 86, are of 
authority, read εξεκβντησαν ; (2) εξεκεντησαν was read by the Greek 
Fathers, e.g. Clem. Alex. p. 984, and hence also in ps.-Ignat. ad 
Trail. 10 ; (3) it was read in the recension which underlies the Latin 
version used by Tertullian, who Vi?>ts pupugerunl or compugerunl in 
contexts which show clearly that he is quoting Zecharias, e.g. 
c. Judaeos c. 14, p. 148, c. Marc. 3, p. 671, by Cyprian Testim. 2, 
p. 294, and by Lactantius Instil. 4. 18. It may reasonably be 
supposed that St. John's quotation, c. 18. 37, is from the same 
recension : it may also not unreasonably be supposed, from the use 
which was made of the quotation in the Judaeo-Christian contro- 
versy, that the alteration in the text of the LXX. was from εξεκεντησαν 
to κατωρχησαντο, and not the reverse, and that it w^as made by Jews 
and not by Christians. This hypothesis will be still more probable 
if it be true that the LXX. text has been handed down by a Jewish 
rather than by a Christian tradition. 

I. 12 is a quotation of LXX. Is. 63. 17. 

II. 13, 14 are a quotation of LXX. Is. 64. 11 with the exception 
of the substitution of eh όνειδος for πνρίκαυστος : the LXX. text of the 
passage is quoted exactly in Apol. i. 47, which is one of many 
indications that this cento was a separate poem. 

It may be noted as a common feature of all these quota- 
tions, v^hether from Clement, Barnabas, or Justin, that they 
are introduced by the same formulae which are used for 
quotations of single passages of the canonical books. The 


formulae are, in Clement, (i) Aeyet [sc. το ayiov τν^υμα], 
(2) δια τον ττνζύματοί του αγίου οΰτωί τταρακαλξΐταυ ημάί, 
is) ^.eyet γαρ η γραφή, (4) γίγρατΐται γαρ, (5) οΰτω? φησίν δ 
άγιοί λόγο^. In Barnabas, (ι) Xeyet ό 'προφητ€ύων €π αυτω, 
(2) Xey€6 6 Ίτροφητη^, (3) ττώς Aeyet κύριοι καταργών αυτόν ; 
In Justin Μ., (ι) βοα [^sc. το άγων ττνξυμα], (2) δια Ζαχαριου 
του προφήτου ττροφητξνθέντα ζλ^χθη οντωί. 


There is ample evidence that the original LXX. text of 
the book of Job was much shorter than that which has 
come down to us in existing MSS. ; that the original text 
was revised by Origen in order to bring it into conformity 
with the Hebrew; that the passages which were absent 
from the LXX. text, but present in the Hebrew, were 
supplied by him from the version of Theodotion ; and that 
the text of all existing Greek MSS. is the revised and 
composite text which Origen thus formed. 

The divergences between the earlier and the later texts 
are indicated by Origen himself [Epist. ad African.^ Op. 
ed. Delarue, vol. i. p. 15) as consisting in the omission in 
the Greek of ' frequently three or four, sometimes fourteen or 
nineteen verses': the total amount of such omissions is said 
by Jerome to have been 700 or 800 verses {Praef, in Hiob^ 
tom. ix. 1097). 

The passages which were absent from the original LXX, 
text, and which were supplied by Origen from Theodotion, 
were marked by him in his text of the Hexapla with an 

* The author thinks it due both to himself and to Professor G, Bickell to say 
that although he had read his dissertation De indole ac ratione Versionis 
Alexandrinae in interpretatido libro Jobi (Marburg, 1862) before delivering the 
lecture on which the present essay is based, and derived from it, as he has since 
derived from his papers in the Zeitschrift fiir katholische Theologie, some 
valuable hints, the views which he here sets forth were suggested to him in- 
dependently, in the course of his examination of early quotations from the 
LXX., by the fact that Clement of Alexandria {Strom. 4. 26, p. 641) quotes, or 
appears to quote, c. xxxvi. 10-12 in the form which it had before Origen's 
revision : that is to say vv. \ob, ii are omitted• 

2i6 ON origen's revision of 

asterisk : and these asterisks have been preserved in three 
distinct groups of authorities : 

(i) They are found in two Greek MSS. of the LXX., 
the Colbert MS. 195^^ in the Bibliothkque Nationale at Paris, 
and the Vatican MS. 346 (which was collated for Holmes 
and Parsons, and is numbered ^^48 in their list). 

(2) They are also found in at least two Latin MSS., viz. 
the Bodleian MS. (God. Lat. 2426, which contains the Old 
Latin version, and Jerome's version separately) ; and a 
MS. which was formerly in the monastery of Marmoutiers 
(God. Majoris Monasterii), and which was published by 
Martianay in his edition of Jerome, vol. i, and reprinted by 
Sabatier in his Biblioriim Sacrorum Latinae Ver stones 

(3) They are also found in the Syro-Hexaplar version, 
i, e. the Syriac version which the monophysite bishop, 
Paulus Telensis, made in A. D. 617, from one of Eusebius's 
copies of Origen's Hexapla. The book of Job in this 
version exists only in one MS., now in the Ambrosian 
Library at Milan, which has been published (i) by Middle- 
dorp in the Codex Syriaco-hexaplaris (Berlin, 1835), (2) more 
recently in facsimile by Geriani (Milan, 1876). 

To these three texts and versions which preserve Origen's 
asterisks has recently been made the important addition of 
a version of the text itself as it existed before Origen's 
time. It is the Sahidic ( = Thebaic) version, which is (with 
the exception of the last leaves, which are at Naples) con- 
tained in a MS. in the Museum Borgianum at Rome : its 
only lacuna, c. xxxix. 9-xl. 7, can be supplied from a 
Sahidic MS. at Paris ^. 

It is of importance to note that these several sources of 

' The only information which I possess of this version is contained in a letter 
of Bishop Agapios Bsciai to the Monitetir de Rome of October 26, 1883, quoted 
at length by Lagarde Mittheilungen, No. 21, p. 203. The letter is sufficient for 
the present purpose inasmuch as it contains a list of the passages which the 
Sahidic version omits. 


evidence in the main agree : they differ, as must be 
expected when critical marks are transferred from one 
MS. to another at wide intervals of time, in the length of 
the obelized passages : but they agree in all important 
instances, and there is an especial agreement between the 
Syro-Hexaplar and the Sahidic versions. 

The question to the consideration of which the present 
essay is designed to be a contribution is. How are we to 
account for these wide divergences between the original 
and the later texts of the LXX. ? 

i. It seems probable that some of them are due to a care- 
less or unintelligent correction of the text by Origen or his 
scribe : of this the following four passages are examples : 

In c. ix. 3 there is a double version of ^33JJ^ δ^7, (i) ov μη νπακονση 
αυτω, {2) ϊνα μή avTeiirr]. The former of thesc IS due to Symmachus 
and Theodotion : the latter is probably a modification of an original 
LXX. reading ov μη άντ^ίπη, which has survived in the readings ουδέ 
μη avreinjj in Cod. 254, and ούδ* ου μή άντίίπϊ] in the margin of 
Cod. 250. 

In c. xxiii. 14, 15 the translation of the Hebrew of v. 14 is omitted, 
and V. 15 is translated twice, 

(1) V. 14 δια τοΰτο iir αυτω ίσπούδακα' 

νονθ€Τονμ€νος 5e (φρόντισα αντον. 

(2) V. 15 fVi τούτω άπο προσώπου αυτόν κατασπουδασθώ* 

κατανοήσω κα\ πτοηθήσομαι e^ αύτου. 

Of these two versions the first is that of the LXX., the second 
that of Theodotion. That is to say, Origen substituted the more 
accurate version of Theodotion for that of the LXX., but either he 
or his scribe erased v. 14 by mistake for v. 15. 

In c. xxviii. 26, 27 there is apparently a double rendering 

of nnaO^^^ ΠΝΊ TX, viz. (l) όντως Ιδών ήρίθμησ€, (2) τότί eidev αντήν 

κα\ €ξηγήσατο αυτήν. The first of these renderings is probably the 
translation of the LXX., since άριθμ^Ίν is used to translate "iSp in 
xiv. 16, xxxviii. 37, xxxix. 2 : the second is that of Theododon. 
But the translation of ΡΠ "it3K)p is omitted : and the first of the 
above translations takes its place, so that the passage gives no 

2i8 ON origen's revision of 

intelligible sense. The explanation is probably to be found in the 
fact that according to Codd. Marm. Bodl. and the Syr.-Hex. and 
Sahid. the words κα\ όδον . . . ίξηγησατο αντην Were inserted from 
Theodotion : when this was done the words οΰτως ldu>u ηρίθμησε of 
the original translation should have been erased : when they were 
left in by the negligence or ignorance of a scribe, the object of 
ore €ποίησ€ν, i.e. vera πρόσταγμα (or equivalent words), was omitted 
as destroying the symmetry of the στίχοι. 

The original form of the LXX. translation of vv. 24-28 may be 
supposed to have been as follows : 

23 θ(6ς €v συνβστησβν αντης την όδόν, 

avTos δε οι'δε τον τόπον αντης' 
2 4 avTos yap την νπ* ουρανον πασαν ζφορα^ 

ilbois τα iv τγι yfj πάντα' 
2 5 [Ρ"^^] €ποίησ€ν άνίμων σταθμόν, 

vdaTos re μ€τρα [jjrot/iaae] 
26 6τ€ (ποίησαν [ίζτω πρόσταγμα^ 

[όδόι/ re κυ8οιμων1' 
2*] ^τότί^ Ι^ων ηρίθμησ€, 

€Τθΐμάσας βζιχνίασξν' 
28 eiTre δε άνθρωπω. Ιδού ή θ€οσ€β(ίά €στι σοφία, 

το δε άπ€χ€σθαι άπο κακών €στ\ν επιστήμη. 

The words in brackets are conjectural : the reason for each of 
them is as follows : in vv. 24, 25 Cod. Β reads πάντα Ιποίησ^ν, 

Codd. AC^ 254 πάντα a (ποίησαν ξποίησ^ν δε, Codd. 23, 55, 68, 1 57, 
160, 161, 250, 252, 255, 256, 257, 260, 26 r πάντα a εποίησαν, 
Codd. 106, 1 10, 137, 139, 147, 248, 249, 255, 258, 259 πάντα re 
α βποίησ^ν, Codd. I38, 251, 254 πάντα οσα €ποίησ€ν : since οτβ follows 
in the next verse, and since the Hebrew TN requires τότ€ (which 
Theodotion has) in v. 27, it may be conjectured, in face of the 
great variety of readings, and not out of harmony with it, that otc 
was read here. In v. 25 the missing translation of ]'?^ may be 
supplied by ήτοίμασ^, since the same Hebrew verb is translated by 
€τοιμάζ€ΐν in the song of Hannah, i Sam. 2. 3. In v. 26 the missing 
translation of ""^W is clearly, as elsewhere, Ιίτω and that of ΡΠ may 
be πρόσταγμα, as in c. xxvi. 10: the translation of Hipp ΓίΠ? "ϊρΊ) 
was probably 686v re κυδοιμών as in c. xxxviii. 25. 

In c. xxix. 10, II the words "'^"}.?ϊ?ί!ΐ1 ^νΌψ ]]}< "Ί are translated, 
(l) ot δε άκονσαντ€ς ίμακάρισάν μ€, (2) more literally, on ovs ηκονσ€ καί 


€μακάρισ€ μ€ : the first of these translations takes the place of the 
translation of ^^^^nj ΰ'•1''3ί"")^ρ^ « the voice of the nobles was hid ' : 
and it, rather than the second, is likely to have been the LXX. 
translation because the noun ]ψ (in the dual) is translated by the 
verb oKoveiv elsewhere, viz. c. xiii. 17, Ezek. ix. 5 : x. 13. Cod. 248 
obelizes v. 1 1, the Syr. Hex. and Sahid. obelize vv. 10 i5, 11 a. These 
facts taken together seem to point to the existence of an earlier 
text, and the simplest hypothesis as to its form is that v. 1 1 in the 
Hebrew is a duplication of v. 10, and that vv. 10 <5, 11 α in the 
Greek are a duplication of vv. go, 10 a. 

ii. It is conceivable that some of the divergences are due to 
the circumstances under w^hich the translation v^^as originally 
made. It was made after Judaism had come into contact 
with Greek philosophy. It may be presumed to have been 
intended not only for Greek speaking Jews but also for 
aliens. The tendency, which found its highest Hterary 
expression in Philo, to show that Judaism was in harmony 
with Greek culture, may have influenced the mind of the 
translator, and led him to soften down some of the vivid 
Semitic anthropomorphisms, and throw a veil over some of 
the terrors of the law. Even in the Pentateuch which from 
its greater sacredness, and from its liturgical use, was 
translated with especial fidelity, a paraphrase or circum- 
locution sometimes takes the place of the literal expression 
of an idea which a philosopher would have found difficult 
to assimilate : and it is natural to expect that a poetical 
book, to which no idea of special sanctity was attached, 
and which had no liturgical use, should be translated with 
some freedom. 

But the hypothesis of the intentional omission of passages 
which were out of harmony with the Hellenized theology 
of Alexandria, though it may in some cases be true, is 
inadequate, because, in the first place, it would account for 
only a small proportion of the passages which were absent 
from the original version : and because, in the second place, 


many passages which remain have the same theological 
character as those which are omitted. 

The same remarks would apply to the hypothesis that 
the omissions are due to the difficulty of the language in 
certain passages : it would account for only a few of the 
obelized passages : it would not explain the fact that many 
passages are omitted of which the translation is easy, and 
that many remain of which the translation is difficult. 

Two other hypotheses remain : the one is that the book 
was more or less arbitrarily curtailed by the translator : the 
other is that at a time subsequent to its first translation the 
original Hebrew text was amplified, and that the original 
LXX. text represents, in the main, this original Hebrew. 

The first of these hypotheses is improbable, nor does it 
admit of either proof or disproof. The second is not without 
its difficulties, but it at least bears examination. I propose 
in the following pages to test its truth, and its sufficiency 
as an explanation of the facts, by enquiring how far the 
passages which Origen inserted can be omitted without 
detriment to the argument of the poem. 

The passages to which the hypothesis is chiefly applicable 
occur in the third (c. xxii-xxxi) and fourth (c. xxxii-xxxvii) 
groups of speeches : but there are also some passages in the 
second group (c. xiv-xxi) and in the fifth (c. xxxviii-xlii. 6). 
I propose to give some examples from the second and third 
groups, but to deal mainly with the fourth, the speeches of 
Elihu : there is the more reason for doing this because the 
speeches of Elihu are, from the point of view of a critic, 
the most interesting portion of the book, and because it 
is hoped that the hypothesis which is here adduced may 
help to solve some of the more difficult problems which 
the criticism of those speeches involves. 


i. The second group of Speeches : c. xiv-xxi. 
c. xvii. ^-S. 

vv. ^-^ a are obelized in Cod. Colb. and in the Sahid. : 
vv. '>^-^ in Cod. Marm. : wv. '^b, 4.b^ ^a in Syr.-Hex. 

The obelized words are difficult of explanation in both 
the Hebrew and the Greek : their omission gives a con- 
secutive sense which is even clearer in the Greek than in 
the Hebrew. It may be noted that the Greek and Hebrew 
of v. 2 are quite different : but since the Greek is in harmony 
with the sense of the non-obelized verses i, 6, 7, 8 it may 
be supposed that it represents a lost Hebrew verse, which 
was displaced when vv. '^-^ were inserted : in other words 
V. 'i in the Hebrew belongs to the added portion, but in 
the Greek belongs to the original. 

1 οΚξκομαι "^ ιτνΐνματι φ^μόμ^νος, I am consumed, being agitated in 

spirit (?) : 
Βΐομαι Se ταφής και ov τυγχάνω' I pray for the grave, and obtain 

it not. 

2 Χίσσομαι κάμνων, I am weary with entreating. 
και τι ποίησας ; And what hast thou done ? 

3 (κλίψαν Se μου τα υπάρχοντα And strangers have stolen my 

άλΧότριοι' goods, 

Tis koTiv cvTos ; Trj χβιρί μου Who is this one ? let him strike 

συνδίθήτω' hands with me : 

4 oTi Kapbiav αυτών (κρυψα^ άττό For thou hast hid their heart 

φρονήσ€ω5, from understanding : 

Sia τούτο ου /χτ) υψώσχίζ αυτούί' Therefore shalt thou not exalt 


^ In this, as in the other quotations in this chapter which are arranged in 
parallel columns, inasmuch as neither a critical discussion of the meaning of 
the variants of the Greek text nor a philological discussion of the meaning of 
the Hebrew would be pertinent to its main point, (i) the LXX. is quoted, 
except where otherwise specified, from the Sixtine text, (2) the Revised English 
Version has been followed wherever the meaning of the Hebrew approximates 
to that of the Greek. Where the Hebrew text varies to any great extent 
from the Greek, an independent translation of the latter has been given. 
The Roman type indicates the Revised Version, the Italic type indicates an 
independent translation of the Greek : the larger type indicates what the author 
believes to have been the original text of the book, the smaller type the passages 
which he believes to have been added. 



5 T^ yiteptSt άΐ'α77€λ€Γ κακίας, ? 

οφθαλμοί δΐ Ιφ' vlois Ιτάκησαν Even the eyes of his children 

failed : 

6 %βον be με θρύλλημα iv (θνεσι, Thou didst make me also a by- 

word among the people : 
yi\(cs he avrois άπ^βψ' And I became a laughing-stock to 


7 πιπώρωνται yap απ οργής oi Mine eye also is dim by reason 

οφθαλμοί μου, of Wratk, 

π(πολιόρκημαι μ€γάλω5 νπο πάν- I am besieged greatly by all men. 

c. xxi. 38-33. 

These verses are obelized in all the authorities : and 
Cod. 248 adds to them v. 27 b. 

The sense will be found to run on, and even more clearly 
in the Greek than in the Hebrew, from v. 27 to v. 34. The 
obelized section may be regarded as a poetical expansion 
of either v. 27 or v. 34 a. 

27 ώστε οίδα υμάς, 

on τόλμη €πικ€ΐσθ€ μοι. 

28 ώστε epHve, Τΐοΰ Ιστιν oJkos 

άρχοντοί ; 
και τΓον Ιστιν ή σκίιτη των 
σκηνωμάτων των άσεβων ; 

29 (ρωτη(Χατ€ ΐΓαραττορ€υομίνον$ 

καΐ τα σημίΐα αύτων ουκ άπαλ- 

3θ ΟΤΙ ds ήμίραν άιτωλύα5 κουφί- 

ζ€ται δ πονηροί 
els -ήμίραν ορηψ αύτοΰ άπα- 


31 τίί άπαγ^ίΚζΐ Ιττί προαώπον 

αντον TTjv δδόν αύτοΰ ; 
καΐ avTOs Ιττοίησί, tis ανταπο- 
δώσει αύτω ; 

32 και avTos els τάφουί άπη- 

καΐ avTos eni σωρών η^ρύννη- 

So that I know you, 

That with boldness ye set upon me : 

So that ye will say, Where is the 

house of the prince ? 
And where is the shelter of the 

tents of the wicked ? 
Ye asked them that go by the 

And their tokens ye shall not 

That the evil man is reserved to 

the day of calamity. 
That they shall he led forth to the 

day of wrath. 
Who shall declare his way to his 

And who shall repay him what 

he hath done ? 
Yet hath he been borne to the 

And hath kept watch over the 



33 ί-γλυκάνθησαν αντω χάλίκ€$ The cuj>s of the brook have been 

χαμάρρον sweet unto him, 

και δηίσω αντοΰ was άνΘρωπο$ And all men shall draw after 

aireXevaerai, him, 

καΐ €μτΓροσθ€ν αντου άναριθμη- As there were innumerable before 

Toi' him : 

34 πως be παρακάλ€Ϊτβ μ€ K€va 'j How then comfort ye me in 

TO be €μ€ καταπαύσασθαι αφ* And rest for me from you is there 

νμων ovdev, none. 

ii. TAe third group of Speeches : c. xxii-xxxi. 

c. xxiv. J4c-iSa. 

These verses are obelized in Codd. Colb. Marm., and in 
the Syr.-Hex. and Sahidic : so also in Cod. Vat. except 
V. 14 c, and in Cod. Bodl. except vv. 14 c, J 5 a, δ. 

The omission of the obelized verses gives an intelli- 
gible sequence of ideas. In LXX. v. 13 Job enquires why 
God does not visit the wicked who oppress the poor and 
know not the way of righteousness. The answer is at 
once given in LXX. v. 14 a, b, that when He takes cogni- 
zance of their deeds He delivers them over to darkness : 
and this idea of punishment is continued in v. i8<5, ' may 
their portion be cursed upon earth, and their fruits be 

The insertion of the obelized section, on the contrary, 
interrupts the sequence, and appears almost like a digres- 
sion leading off from the double sense of σκ6το<ί. In v. 14 <^ 
it is used in the sense of ' Sheol,' but in v. 14^ it is ap- 
parently taken in the sense of ' night,' and this leads to the 
thought of the thief and the adulterer. 

The entire absence of correspondence between the Greek 
and the Hebrew in vv. 33^, 14^, b, 18 i:, 19, 10 a, b makes 
it possible to suppose that the introduction of the obelized 


section led to changes in the verses immediately preceding 
and following it. 

1 3 avTos be Bia τι τοντων Ιτησκοττην Why has he not made a visitation 

ου ττίποίηται ; /or these things Ρ 

erri yrjs όντων αντων και ουκ Upon earth they Were, and they 

βτΓβγι/ωσαΐ', acknowledged him not, 

14 obov he δικαιοσύνης ουκ fjdeiaav But the way of righteousness they 

knew not, 
oide ατραπούς αίιτης €πορεύβη- Neither walked they in the paths 

σαι/. thereof. 

ψους de αυτών τα epya, But ivhen he took knowledge of 

their works 
παρί^ωκίν αυτούς eh σκότος. He delivered them over to darkness. 

και vvKTos 'έσται ws κλ€πτη5• And at night he slmll be as a 


15 Kox δφθαλμυε μοιχού efvXa^e The eye also of the adulterer 

aKOTos, waiteth for the darkness, 

Xeywv, Ού irpovorjaei μe δψθαλ- Saying, No eye shall see me, 

και άιτοκρνβ^ν -ηροσώττου eOero• And he putteth a covering on his 

face : 

16 διώρυξ€ν Iv cKOrei οίκΐαί, In the dark they dig through 

ήμίραί Ισφρά^ισαν eavTous, They shut themselves up in the 

ουκ ίττί^νωσαν <ρω$. They know not the light. 

1 7 oTi ομοθυμαδόν avTois το πρωϊ For the morning is to all of them 

σκιά θανάτου, as the shadow of death, 

oTi Ιτη^νώσίται ταράχαί CKids For he shall know the terrors of 

θανάτου. the shadow of death. 

18 (λαφρ65 koTiv eirl πρόσωπον He is swift upon the face of the 

ϋδατο5• waters : 

καταραθύη fj μίρις αυτών em γης, May their portion be cursed upon 


19 άναφαν€ίη 8e τα φυτά αυτών επι May their trees appear barren 

γης ξηρά' upon earth. 

άγκαλίΒα γαρ ορφανών ηρπασαν' For they plundered the armful 

[gleanings ?) of orphans. 

20 eir' άν€μνησθη αυτού ή αμαρτία' Then his sin Was remembered, 
ωσπΐρ be ομίχλη δρόσου αφανής And as the mist of deW he 

cycj/cro* Vanished ; 


συντρίβζίη 8e ttus άδικος Ισα' And may every unrighteous man 
ξνλω άνιάτγ. be broken like a tree that cannot 

be healed. 

c. xxvi. 5-11. 

The following verses are obelized : 

vv. 5-10 in Codd. Colb. Marm., vv. 5-1 1 in the Syr.-Hex. and 
Sahid., vv. 6-10 in the Cod. Vat. In Cod. Bodl. c. xxvi forms a 
continuation of the speech of Bildad in c. 25 : there are five asterisks, 
but it is not clear where they are meant to begin and end. 

The omission would make the description of the power 
of God shorter, but not less emphatic : the obelized verses 
give a poetical expansion of the main idea, but do not 
materially add to it. 

It may be noted that v. 14 a, b, also is obelized in the 
Syr.-Hex. As that verse stands (i) its first two clauses 
Ihov . , . . h αυτω would be less intelligible if it had been 
preceded by only the short enumeration of God's ways 
which the omission of vv. 5-1 1 would leave, (2) its last 
clause is in intelligible sequence with vv. 12, 13, and it may 
possibly have been immediately preceded by a clause 
which was omitted when vv. 5-1 1, 14 a, b^ were inserted. 

c. xxviii. 13-22. 
The following verses are obelized : 
vv. 13-19 in Cod. Vat. 

vv. 14-19 in Codd. Colb. Marm., and in the Syr.-Hex. and Sahid. 
V. 21 in Codd. Colb. Vat. Marm. : v. 21 3 in Codd. Bodl. and in 
the Syr.-Hex. and Sahid. 

V. 22a\Ti the Syr.-Hex. and Sahid. 

The sequence of ideas is not in any way disturbed by 
the omission of the section vv. 14-19, which amplify the 
main thought of the passage with singular poetical beauty, 
but do not add to its substance. 




It will be noted that v. 20 is a repetition in both form 
and substance of v. 12, and v. 21 <2, in substance though not 
in form, of v. 13 ; and also that v. 22 is in substance ana- 
logous to vv. 14 sqq. Consequently v. 23 begins an answer 
which is common to both the sections vv. 11-19 and 

There is another fact which enters into the consideration 
of the original form of the passage, viz. that Clement of 
Alexandria {Strom. 6. 6^ p. 763) possibly, or probably, 
quotes vv. 20, 21 in a form which does not survive in any 
existing MS. of the LXX. : Aeyet 6 αδτ]? rr^ άττωλεια* etdos 
μ\ν avTov ουκ €lbo^€V, φωνην 6e αντου ηκονσαμ^ν. If these 
words be a quotation from this passage, they may be taken 
to be a relic either of the original form of the passage, 
which was modified when vv. 14-19 were inserted, or of 
the poem which was incorporated with it. 

12 ή δε σοφία noOev (νρίθη ; Where shall wisdom be found? 

TTolos δε τόπος earl της άπιστη- And where is the place of under- 
standing ? 
Man knoweth not the way 

thereof : 
Neither is it found a?fiong men : 

The deep saith, It is not in me : 

And the sea saith, It is not with 

He shall not give . . .for it : 

Neither shall silver be weighed 
for the price thereof. 
* * * 

* * 5f! 

[Whence then cometh wisdom ? 
And where is the place of under- 
standing ? 
Seeing it is hid from the eyes of 
all living,] 
And kept close from the fowls of 
the air. 

13 ovK oice βροτος Sdop αντης, 

ovBe μην ίνρίθη iv άνθρώποις. 

14 άβυσσο$ (Ίπ^ν Ουκ eveariv kv 

και ή θάλασσα etnev Ουκ ev- 
ecTiv μ€τ' (μου. 

15 ου δώσα συ-^κΧπσμον αντ' 

κα\ ου σταθήσ^ται άρ-γύριον 
άντάλλαΎμα avrijs. 

VV. 16, 17, 18, iQ * * * 

* * * 

20 Γί? δε σοφία ττόθ^ν ζίιρίθη j 
Tccios he τόπος ear) της σννίσζως] 

21 ΧζΚηθΐ πάντα ανθρωπονΛ 

Hal άπ6 ΊΓ^τ^ινόύν του ουρανού 


22 jy άττώλΗα καΐ δ θάνατοί ΐΐτταν Destruction and death say 
άκηκόαμίν δέ αύτψ το KXios We have heard a rumour thereof 

with our ears : 

23 δ θώς €v συν^στησΐν αντης την God understandeth the way 

όδόν, thereof, 

avTos δε olde τον τόπον αντης. And he knoweth the place 


c. xxxi. 1-4. 

These verses are obelized in Cod. 248, and in the Syr.- 
Hex. and Sahid. : parts of vv. 1-3 are obelized in Codd. 
Marm. Bodl. 

The verses are in no way necessary to the general argu- 
ment ; the section which begins with c. xxxi. 6 is in a 
more natural sequence with c. xxx. than c. xxxi. i. 

iii. T/ie Speeches of Elihu. 

I. The first speech, c. xxxii. 6-xxxiii. 

In the first speech of Elihu there are two groups of 
obelized passages, (i) xxxii. 11-17, (2) xxxiii. 28-33. 

(i) xxxii. 11-17. 
The following verses are obelized : 

V. 1 1 in Cod. Marm. : 1 1 3 in Codd. Colb. Vat., and in Syr.-Hex. 

V. 1 2 in Codd. Colb. Vat. Marm., in Syr.-Hex., and Sahid. 

V. 13 in Codd. Colb. Marm. : 13 iz in Sahid. 

V. 14 in Cod. Marm. 

V. 15 in Codd. Colb. Marm., in Syr.-Hex. and Sahid. 

V. 16 in Codd. Colb. Vat. Marm., in Syr.-Hex. and Sahid. 

V. 17 in Cod. Marm. 

It is probable that vv. 11-17 were all absent from the 
original text. It will be noted that the Hebrew has the 
same clause at the end of v. 10 and at the end of v. 16, 
''^^'Π^ ^'yi Ti^yy^ : the intervening words form a separable 
section : and the connexion of ideas between v. 10 and the 



beginning of v. 17 is close and natural, 'I said, Hearken 
to me ; I also will shew mine opinion. For I am full of 

6 ν€ωτ€ρος μ€ν et/xt τω χρονω νμ€ΐς 

de iare πρεσβύτεροι 
dio ησύχασα φοβηθείς τον νμίν 
άναγγείλαί την εμαντον (πί- 
στη μην. 

7 (ΐπα δβ "Οτι ουχ ό χρόνος [Cod. 

Α €Ϊπον δε οτι χρόνος^ εστίν 

6 Χολών, 
iv πολλοΓί δε ετεσιν οΐ'δασι 

9 ονχ οΐ πολυχρόνιοι είσι σοφοί, 
ουδ' οι yipovTCs ο'ίδασι κρίμα. 

Ι Ο διό είπα, ακούσατε μου, κα\ αναγ- 
γελία ύμιν α oiba. 

11 ενωτίζεσθε μου τα ρήματα, ε ρω 

yap υμών άκονόντων, 
ό,χρίί ου ετάσητε Koyovs. 

1 2 και μέχρι υμών σννησω, 

καΐ Ιδού OV/C ην εν Ίώβ εΚε^χων, 

άναποκρινόμενο5 βήματα αύτοΰ 
Ι£ υμών 

13 tJ'tt p-rj εϊπητε Ένρομεν σοφίαν 

κνρίφ -προσθεμένοι' 

14 άνθρώπω δε επετ ρίχνατε Χαλησαι 

τοιαύτα ρήματα. 

15 επτοήθησαν, ουκ άιτεκρίθησαν 

ετταΚαίωσαν εζ αυτών λόγου?• 
ΐ6 υπέμεινα ου yap ελάλησα, 

οτι έστησαν ουκ άπεκρίθησαν. 

1 7 (υπολαβων δε Έλιους λέγει, 
πάλιν λαλήσω^ 
πλήρης yap ειμί ρημάτων 
ωλεκει γάρ με το πνεύμα της 

Ι am young, and ye are very old : 

Wherefore I held back, and durst 
not shew you mine opinion. 

I said, Days should speak, 

And multitude of years should 

teach wisdom. 
It is not the ancients that are wise, 
Nor the aged that understand 

Therefore I said, Hearken to me, 

I also will shew mine opinion. 

Give ear tinto fny words, 

For I will speak while ye listen, 

Until ye have searched out what to 

Yea I attended unto you, 
And behold there was none that 

convinced Job, 
Or that answered his words among 

Beware lest ye say, We have found 

y^'vsAovci^being joined to the Lord. 
But it was a ?nan that ye permitted 

to speak such words : 
They are amazed, they answer no 

more : 
They have not a word to say. 
/ waited, for I spake not, 
Because they stood still, and an- 
swered no more. 

For I am full of words 
The spirit of my belly 
straineth me. 



i8 ^ Sc γαστηρ μου ωσπ^ρ άσκος Behold my belly is as wine that 
yXevKOvs ζίων debephos, hath no vent ; 

η ωσπ€ρ φνσητηρ χαλκ€ω5 ^ρρη- Or like a smitK s hellows hurst- 
^ώί. ing : 

19 λαλήσω ΐνα άναπανσωμαι, I will speak that I may be re- 

άνυίξας τα χ^ίλη' I will open my lips and answer. 

There are two other points, besides the fact of their 
being obelized, which give an exceptional character to 
vv. 11-17. 

(i) With the exception of v. iS δ (where the LXX. prob- 
ably read νΤΊΠ, ' a smith,' instead of ^70' * i^^w') the trans- 
lation of the rest of the speech follows the Hebrew closely, 
whereas that of vv. 11-17 in several instances varies widely 
from it. 

(2) The obelized verses are characterized by great 
varieties of reading, especially in vv. 11, 16, which, on the 
hypothesis which has been offered, form the points of 
junction between the original and the added portions. 

The more noteworthy of these variants are the follow- 
ing : 

In V. II Codd. BS^ and the Syr.-Hex. omit βρώ γάρ, w^hich makes 
the sentence unintelligible ; Cod. A, and other Codd. which are 
mentioned by Olympiodorus (ap. Field's Hexapla z'n loc.) add after 
άκονόντων the duplicate, and more accurate, translation l8ov ήκονσα 

τους λόγους νρων' (νωτισάμην ρίχρι σννεσζως υρων '. SO Cod. 23, with 

the addition of yap after ιδού, and with a further duplication of κα\ 

€ως υρων συνησω after συνέσεως υμών. It must be SUppOSed that there 

were several concurrent versions of the passage, and that the reading 
of the Sixtine text, which is that of the majority of MSS., is a scribe's 

In V. 16 Cod. A has €λάλησαν : Cod. 254 has ^σίγησαν for ίστησαν I 

Codd. 106, 110, 137, 138, 139, 147, 161, 249, 251, 255, 256, 258, 
260, 261, Colb., and the Syr.-Hex. add on αποκριθώ κάγω pepos after 
άπ€κρίθησαν, SO, without οτι, 259 : of these words Cod. Colb. men- 
tions that pepos {to μέρος μου) IS due to Symmachus. It may be noted 


that although the words represent the Hebrew ^ϊ^^'Π ""JN-PiN njJfX they 
leave the following half of the verse, ιβδ, which is a repetition of 
V. 10 b, untranslated. This is entirely in harmony with the hypo- 
thesis that ιβΒ was only needed to serve as a point of junction 
between the added section and the following words of the original 
text . . . . ' For I am full of words.' It may be further noted, as a 
mark pointing in the same direction, that the want of such words in 
the current text of the LXX. probably accounts for the interpolation, 
which has no equivalent in the Hebrew, -noKiv λαλήσω. 

{2>) xxxiii. 27-33. 

Three sets of facts must be considered in relation to this 

(i) The follow^ing verses are obelized : 

vv. 28-29 ^^ Codd. Colb. Vat. Marm. Bodl., in the Syr.-Hex. and 

vv. 31-33 in Codd. Colb. Bodl., in the Syr.-Hex. and Sahid. 
vv. 32-33 in Codd. Vat. Marm. 

In other words vv. 27, 30 are the only verses of the 
section which remain in the Colbert text of the Greek, in 
the Bodleian text of the Latin, or in the Syriac and Sahidic 

(ii) After v. 30 Codd. A, 23, and the margin of the Syr.- 
Hex., insert the following words : 

ντΓολαβων δε Έλιούί Xeyei, 

άκονσατ€ μου σόφοί, (πιστάμΐνοι (Ρωτιζ^σθαι το καλόν' 

δη €φηκ€ν Ίώ/3 [23 OmitS Ίώ/3] Ίδον ταϋτα πάντα ipyarai 6 Ισχυρός 

όδους τρίίς μ€τα avdpos, 

του €πιστρ€ψαι ψυχην αυτοΰ βκ διαφθοράς, 

του φωτίσαι αυτω iv φωτΐ ζώντων. 

Of these words, lines ι, 2 are the beginning of c. xxxiv, 
as it stands in most MSS. : the Sixtine text omits το καλόν. 
It will be noted below that vv. 3, 4 of c. xxxiv are obelized, 
so that not only lines i, 2, but also the words otl ζϊρηκ^ν 


^1(ύβ, belong to that chapter. This fact is a strong cor- 
roboration of the hypothesis that at any rate vv. 31-33 did 
not form part of the original text. The words that follow, 
ibov ταντα . . . ζώντων, are a duplicate, and more exact, trans- 
lation of vv. 29, 30. They are altogether out of place in 
the mouth of Job, and do not contain the opinions which 
Elihu proceeds to answer. 

(iii) Neither the text nor the meaning of the Greek of 
V. 2y is certain : but no meaning can be attached to any 
form of the text which will bring it into harmony with the 
Hebrew : and neither the Greek nor the Hebrew is in 
intelligible sequence with the context. 

The general result is that, in the original text of the 
speech, vv. 28, 31, 32, ^^ were certainly omitted, and that 
the speech ended with v. 30, which is not obelized in any 
of the MSS. or versions, and the true form of \vhich is 
preserved in the duplicate translation in Codd. A, 23. To 
these omissions that of v. 27 should probably be added : 
but although v. 29 is obelized by all the authorities, 
the fact that it is preserved with v. 30 in the duplicate 
translation, and that it coheres well with the general 
sense of the passage, raises a presumption in favour of 
its retention. 

The following is suggested as having been probably the 
original form of the passage, the inserted portions being 
printed in smaller type : 

26 €υξάμ€νο5 δε προς κνριον κα\ He prayeth unto God and he is 

δεκτά αυτω €σται, favourable unto him, 

elaeXevaerni προσώπω ΐΚαρω σνν So that he Secth his face with 

(ξηγορία- joy, 

άποΒά)σ€ΐ δε άνθρώποις δίκαιο- And he restoreth unto man his 

σύνην righteousness : 

27 εΓτα τότε άττομίμχρίται άνθρωττοί 

Κίηων Οία συν^τίΚουν, 


Kal ουκ άξια ητασί μι ων And it was not requited unto me : 

28 σωσον φυχην μου τον μτ) €\θ(ΐν He hath redeemed my soul from 

ds διαφθοράν, going into the pit, 

και ή ζούή μου φώ$ οψ^ται. And my life shall behold the light. 

29 Ihov ταντα πάντα epyarai 6 Lo, all these things doth God 

Ισχυροί work, 

Sdovs τ pels μ€τα avbpos' Twice, yea thrice, with a man, 

30 [Codd. A, 23.] 

του imaTpe^ai ψνχψ αντον eK To bring back his soul from the 

διαφθοράς^ pit 

τον φωτισαι αντω iv φωτι ζών- That he may be enlightened with 

των the light of the living. 

[Codd. BCS cett.] 

αλλ' €ρρύσατο την ψνχην μου €κ But he rescued viy soul J'rom 

θανάτου, death, 

ίνα η ζωη μου iv φωτ\ αΐν^ That my life might praise him in 

αυτόν. the light. 

3 1 (νωτίζον ^Ια}β και ακουί μου, Mark well, Ο Job, hearken unto me : 
κώψ€νσον και hyai (Ίμι λαλήσω. Hold thy peace and I will speak. 

32 ti (Ισί σοι λοΎοι, άποκρίθητί If thou hast anything to say answer 

μοί' me : 

λάλησον, Θέλω yap ζικαιωθήναί Speak for I desire to justify thee. 

33 d μή, συ άκουσον μου' If not, hearken thou unto me : 
κωφίνσον καΧ δι5άξω σ€. Hold thy peace, and I will teach 

thee wisdom. 

2. T/ie second speech of Elihu, c. xxxiv. 

In the second speech of Elihu there are two groups of 
obelized passages, (i) vv. ^-j, (2) vv. 2,3-^3» 

(i) vv. ^-7. 
The following verses are obelized : 

vv. 3, 4 in Codd. Colb. Vat. Marm. Bodl., and in the Syr.-Hex. 
and Sahid. 

vv. 6 3, 7 in Codd. Colb. Marm. Bodl,, and in the Syr.-Hex. and 

v. 8 Λ in Cod. Bodl. and in the Syr.-Hex. 


The variants are not important except in v. 8, where the 
most noteworthy are the following : 

Codd. 139, 147, 256 omit ουχ άμαρτων ούδε άσφησας'. Codd. A, 23 

read ούδβ, Codd. CS^ 106, no, 137, 138, 139, 147, 157, 160, 161, 
248, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 261, read 
ονδ* δλωί, Cod. 249 reads η 6^ov, Cod. 260 reads η old' 6λως, for rj 
οι5δ' ου of Cod. Β and the Sixtine text : Cod. A adds 6dov after 


The omission of vv. 3, 4 is supported, as mentioned 
above, by the readings of Codd. A, 23 in v. 30 of the pre- 
ceding chapter : and it helps rather than hurts the sense of 
the passage. The main difficulty is that of v. Sa which 
has no equivalent in the Hebrew, and which, as the passage 
stands, affords no intelligible sense : this may account for 
its being obelized in Cod. Bodl. and the Syr.-Hex. The 
difficulty may perhaps be solved by noting that if v. 6 δ be 
rightly obelized, v. 6 is left without a second member, and 
by conjecturing that 8 ^ is that second member. On this 
hypothesis the whole passage originally read as follows : 
the added portions are printed, as before, in smaller type. 

2 άκονσατ€ μου σοφοί, Hear my words, ye wise men ; 
(πιστάμ^νοι €νωτίζ€σθ€. And give ear unto me ye that 

have knowledge. 

3 oTt CVS Xo^ovs δοκιμάζίΐ For the ear trieth words 
καΐ λάρνγξ yeverai βρώσιν. As the palate tasteth meat. 

4 κρίσιν ίΚώμίθα kavTois, Let us choose for us that which is 

γ/ώμίν avh μέσον ίαυτων 6 Tt Let us know among ourselves what 

καλόν. is good. 

5 on €Ϊρηκ€ν Ίωβ, Δίκαιος €ΐμι, For Job hath Said, lam righteous, 
6 Kvpios άπηλλαξ€ μου το κρίμα And God hath taken away my 

right : 

6 (ψ^νσατο be τω κρίματι μου' And hath been false in my judg- 


βίαιον TO Pikos μου aviv aZiKias. My wound is incurable, though I am 

without transgression. 

234 ON origen's revision of 

7 Tts άνηρ ωσττ€ρ Ίώβ What man is like Job 

ηίρων μυκτήρισμον ωσπ^ρ νδωρ Who drinketh up scorning like 


8 ονχ άμαρτων oide άσφησας, Though I have not Sinned nor 

dealt wickedly 
ovbe [Codd. A, 23, or ούδ' όλως Nor gone in company with the 
as in CS^ and most cursives] workers of iniquity, 

κοινωνησας μβτά ποωνντων τα 

τοί' πορενθηναι μίτα άσ^^ων So as to Walk with wicked men. 

(2) vv. 23-33. 

The following verses are obelized : 
V. 22 b m Codd. Colb. 255. 

V. 23 in Codd. Colb. Bodl. Marm., and in the Syr.-Hex. and Sahid. : 
it is omitted in the early Latin. 

V. 25 <5 in the Syr.-Hex. and Sahid. 

vv. 25-34 in Codd. Colb. Marm. Bodl. 

vv. 28-33 ill Cod. Vat. and in the Syr.-Hex. and Sahid. 

The omission of the section vv. 23 (or 22)-33 would in 
no way affect the argument of the speech; the answer of 
Elihu in vindication of God against Job is fitly concluded 
with either v. 21 or v. 22, and in v. 34 he turns again to the 
' men of understanding/ in the full assurance that they will 
say that Job has spoken without knowledge. 

3. The third speech of Elihu ^ c. xxxv. 

In the third speech of Elihu there are two obelized 
passages, (1) vv. "j b-ioa, (2) vv. 15-16. 

(i) vv. jb-ioa. 

These verses are obelized in Codd. Colb. Marm., in the 
Syr.-Hex. and Sahid. : vv. %-ioa in Cod. Bodl. 

The argument is made clearer and more pointed by the 
omission of the passage, which has no necessary connexion 
with the rest of the speech. 



(2) vv. 15-16. 

These verses are obelized in Codd. Colb. Marm. Bodl., 
and in the Syr.-Hex. and Sahid. 

The passage, like the preceding, is in no way necessary 
to the argument : and the hypothesis that it is an addition 
to the original text is supported by the fact that the LXX. 
has a different ending to the speech, viz. the clause of v. 14 
κρίθητί . . . ώ? €στϋ, which is no less difficult than the 
Hebrew, but which is both more appropriate and more 
emphatic than vv. 15, 16. 

The connexion of ideas in the speech, from v. 5, will be 
seen from the following reprint of it : 

5 άνάβ\€\ΐΑον (Is τον ovpavov κα\ iSe, 
κατάμαθζ be νίφη ως υψηΧα αττό 


6 el ημαρτίς, τι npa^eis j 

€1 δε κάί τΓολλα ηνόμησας, τι 
δννασαι ποιησαι J 

7 και el [Codd. A, 23, 249 ; 

Codd. Β cett. eVel de ovv] 
dUaios el, ri δώσείί αύτώ 

ή τί eK xeipos σου ληψ^ται ; 

8 άνδρι τω δμοίω σοι η άοίβϋά 

καΧ υΐω άνθρώττου ή δικαιοσύνη 

9 από πληθουί συκψαντουμίνοι «e- 

βοήσονται από βραχίονοί πολ- 
ΙΟ καΐ ουκ €?π6 Τΐοΰ ίστιν δ Qios 
υ ποιήσα5 μ€, 

6 κατατάσσων φύλακας νυκτ€- 
1 1 ό διορίζων pe άπο τ€τραηόδωρ 

Look unto the heavens and see, 
And behold the skies which are 

higher than thou. 
If thou hast sinned, what doest 

thou against him ? 
And if thy transgressions be 

multiplied, what doest thou 

unto him ? 
If thou be righteous, what givest 

thou him ? 

Or what receiveth he of thine hand ? 
Thy wickedness may hurt a man 

as thou art ; 
And thy righteousness may profit 

a son of man. 
By reason of the multitude of 

oppressions they cry out, 
They cry for help by reason of the 

arm of the mighty. 
But none saith, Where is God my 


W/io ordereth the watches of the 

Who separateth me from the beasts 

of the earthy 



αττό Se 7Γ€τ€ΐνων ουρανού [Codd. 
23, 253 add σοφίζβι ήμάδ], 

1 2 (KU Κ€κράξονται καί ου μη 

και [Codd. Α, 23, ι6ι omit] 

αττό ύβρεως πονηρών 

13 άτοπα γαρ ου βουλ€ται ide7v ό 

αυτός γαρ 6 παντοκράτωρ ορατής 

14 των συντίΚούντων τα άνομα 

κα\ σώσει μ€. 

κρίθητι δε ivavTiov αυτού 

f ι δυνασαι αυτόν αΐνίσαι ως €στι 

1 5 /fat νυν ΟΤΙ ουκ ΐστιν Ιτησκ^ιττό- 

μ€νο5 δρΎ^ν αυτού, 
και ουκ ^γνω παράπτωμα τι 
ΐ6 καί Ίώβ ματαίωί avoiyu το 
στόμα αυτού, 
hv άηνωσ'ια ρήματα βαρύνει. 

And from the fowls of heaven ? 

There they cry, but none giveth 

Because of the pride of evil men. 

Surely God will not hear vanity, 

For the Almighty himself is an 

Of those who commit unrighteous- 

And he will save me. 

Plead thou in his sight 

If thou canst praise him as he is. 

But now, because he hath not 

visited in his anger, 
Neither doth he greatly regard 

Therefore doth Job open his mouth 

in vanity, 
He multiplieth words without 


4. The fourth speech of Elihu, c. xxxvi-xxxvii. 

So large a part of this speech is obelized, that it w^ill be 
most conveniently considered as a whole. The antiquity 
of the shorter form is shown by the fact, which has been 
mentioned above, that Clement of Alexandria [Strom. 4. 
26, p. 641) quotes it: i.e. in quoting c. xxxvi. 10-12 he 
omits the obelized portions. 

The following are the obelized passages : 

c. xxxvi. 

V. 5 in Cod. Colb. : 5 ^ in Codd. Vat. Marm., and in the Syr.- 
Hex. and Sahid. 

vv. 6, 7 in Codd. Colb. Vat. Marm., and in the Syr.-Hex. and 
Sahid. : v. 7 in Cod. Bodl. 


vv. 8, 9 in Codd. Vat. Marm. Bodl., and in the Syr.-Hex. and 

vv. 10, II in Codd. Vat. Marm. Bodl. : vv. io<5, 11 in Cod. Colb. 
and in the Syr.-Hex. and Sahid. 

v. 13 in Codd. Vat. Marm. Bodl., and in the Syr.-Hex. and Sahid. 

V. 16 in Codd. Colb. Vat. Marm. Bodl., and in the Syr.-Hex. and 

V. 19 3 in Cod. Marm. 

V. 20 in Codd. Colb. Vat. Marm., and in the Syr.-Hex. and Sahid : 
V. 20 b in Cod. Bodl. 

V. 2 1 in Codd. Vat. Marm. Bodl. : v. 2 1 ^ in Cod. Colb. and in 
the Syr.-Hex. and Sahid. 

V. 22 to c. xxxvii. 6 in Cod. Vat. 

vv. 22 (2, 23 α in the Sahid. 

V. 2\h, 2ζα in Codd. Colb. Marm. Bodl., and in the Syr.-Hex. 
and Sahid. 

V. 26 in Codd. Colb. Vat. Marm. Bodl., and in the Syr.-Hex. and 

V. 27 in the Codd. Vat. Marm. Bodl.: v. 27 ^ in the Syr.-Hex. 
and Sahid. 

v. 2Sa in Codd. Vat. Marm. Bodl., and in the Syr.-Hex. and 

V. 29 in Codd. Colb. Vat. Marm. Bodl., and in the Syr.-Hex. and 

V. 30 in Codd. Vat. Marm. Bodl. and in the Syr.-Hex. : v. 30 α 
in Cod. Colb. 

C. XXXVll. 

v. I in Codd. Colb. Vat. Marm. Bodl. : y. τ a in the Syr.-Hex. 

vv. 2-5 a in Codd. Colb. Vat. Marm. Bodl. and 2 δ-ζ a in the 

V. 5 ^ in the Sahid. 

vv. 6 <5, 7 iz in Codd. Colb. Bodl., and in the Syr.-Hex. and Sahid. 

V. 9 ^ in Codd. Colb. Marm. 

V. 10 Cod. Vat. : v. \oa Codd. Colb. Marm. Bodl. and in the 

V. II in the Syr.-Hex. and Sahid. 

V. 1 2 in Cod. Colb. and in the Syr.-Hex. and Sahid. : v. 1 2 λ in 
Cod. Marm. 



V. 13 in Cod. Vat. : v. 13 ^, <r in Cod. Bodl. and in the Syr.-Hex. 
and Sahid. 

V. 18 in Codd. Marm. Bodl. and in the Sahid.: v. 18 3 in the 

(i) c. XXX vi. 5-21. 

5 γίνωσκζ on 6 κύριος ου μη άττο- Know that God Will not cast aw ay 

ποιήσηται top ακακον, 
^vvarbs Ιαχνί καρδία$' 

6 άσ^βή ου μτ) ζωοποιηστ} 
και κρίμα πτωχών δώσα. 

7 ουκ άφ^λίΐ άττο δικαίου δψθαλ- 

μου$ αύτου 
καΐ μίτά βασιλέων els θρυνον 
κ(ύ καθία avToiis els vinos και 


8 καΐ οι πeπeδημevoι kv xeiponedais 
ο'υσχεθησονται ev σχοινίοΐ5 ne- 


9 κάΙ avayye\ei alTOvs τά epya 

καϊ τταρατττώματα αυτών οτι 

10 αλλά του ΒικαΙον €ΐσακούσ€ται' 

the guiltless man, 

He is mighty in strength of under- 
He preserveth not the life of the 

But giveth to the afflicted their 

He withdraweth not his eyes from 

the righteous, 
But with kings upon the throne 
He setteth them for ever and they 

are exalted. 
And those that are bound in fetters, 
Shall be taken in the cords of 

affliction ; 
And he shall shew them their 

And their transgressions, that they 

have behaved themselves proudly. 

But he will give ear unto the 
righteous : 

And commandeth that they return 
from iniquity. 

If they hearken and serve him. 

They shall spend their days in 

And their years in pleasures. 

But the ungodly will he not pre- 
πάρα το μη βοΰλ^σθαί αυτούς For that they were not willing to 

elhevai τον κνριον know the Lord. 

και hioTi νουθ€τούμ€νοι άνήκοοι And because when admonished 
ήσαν' they hearkened not. 

και eiirev οτι ίτηστραφήσονται 
1^ άδικίas• 

1 1 eav άκούσωοΊ και δoυλeύσ^ωσι, 
συντ€λ4σουσι toLs "^pepas αύτων 

ev ayaeo7s, 
και τά ίτη αύτων ev eύ^tτpe^πeίaιs' 

12 άσΐβίΐ,ς δε ου διασωθεί, 

13 κα\ ύποκριταΐ καρδ'κ:^ τάζουσι 

But they that are godless in heart 
lay up anger, 


ov βοήσονται on ΐ^ησ^ν avTovs' They cry not for help when he 

bindeth them. 

1 4 άττοθάνοι τοίννν iv νίότητι η Their soul dieth in youth, 

"ψυχή αυτών, 
η δε ζωη αυτών τιτρωσκομίνη νπο And their life wounded by angels, 

1 5 avff hv Ζβλι-^αν άσθ^νη κα\ Because they afflicted the weak a?id 

αδύνατον' helpless, 

κρίμα δ€ πράτων ίκθησα. And he will execute judgment for 

the meek. 

16 και προσατιηπάτησύν ff€ €Κ στό- 

ματο5 Ιχθρον, 
α,βνσσο5 κατάχνσΐ5 υττοκάτω αν- 

και κατάβη τράιτ€ζά σον ττλήρψ 


1 7 ουκ υστ€ρησ(ΐ δε άπο δικαίων Judgment shall not fail from the 

κρίμα, righteous, 

1 8 βυμοί δε in άσφείς εσται. But Wrath shall be Upon the 

hi άσίββιαν δώρων ούν εδίχοντο For the wickedness of the gifts 
iiT άδικίαις' which they received for un- 

ig μή σ€ (κκλινάτω ίκών 6 vox)s Let not thy mind willingly turn 
δεησεως thee aside froiu entreaty, 

iv άνάγκτ} όντων αδυνάτων' When the helpless are in distress. 

20 Κ(ύ πάντα5 Toiis κραταιονντα$ 

μ^ k^e\Kvar)s r^v νύκτα, 
του άναβήναι Xaovs άντ' αυτών 

2 1 άλλα φύλαξαι μη πράξης άτοπα' But take heed that thou do not 


ΐπΐ τούτων yoip k^ei\ov and 

If the non-obelized verses ^a, το a, 12>, 14, 15, 17, τ8λ, 
be read consecutively it will be found that they give a 
consecutive and appropriate sense. They are a contrast, 
in clearly defined antithesis, of God's dealings with the 
righteous and the wicked. 


In the same way if vv. 5^, 6, 7, 8, 9, τοΰ, ii, 13, be read 
consecutively they also give a consecutive and intelligible 
sense. They form two connected sections : in vv. 6, 7 
there is a contrast between God's dealings with the righteous 
and the wicked : in the other verses there is a contrast 
between the effects of God's discipline upon the righteous 
whom he has afflicted for their transgressions, and the 
godless who *cry not for help when he bindeth them.' 
The only verse from which some words seem to have 
fallen away is 10 b, which requires an additional member 
to connect it, without harshness, with v. 9, and to explain 
its initial και. 

So far as these verses of the LXX. are concerned they 
form two interwoven but separable poems. 

The main difficulties of the passage lie (i) in the non- 
obelized verse 19, and {2) in the obelized verses 16, 30, 21 ^. 

In regard to (i) there is almost certainly a corruption of 
the text. The note of the wickedness of bribed judgments 
having been struck in v. 1 8 ^ it is natural to expect by way 
of antithesis an exhortation against receiving bribes in 
v. 19 : the words as they stand are barely intelligible, and 
it may be inferred from the fact that μη €κκλίνάτω σ€ is a 

good translation of ^tS^"^i?, that the other words represent 

a lost translation of Ί5^"^ΐ1? 'the greatness of the ransom.' 
If this be so, the next non-obelized words, v. 31 'But take 
heed that thou do not iniquity ' will follow in natural 

In regard to (2) vv. 16, 20 are altogether unintelligible 
as they stand : the varieties of reading in v. 16 point to a 
corruption of the text : and both verses, as also 21 <^, appear 
to be fragments of other translations of the Hebrew, since 
single phrases in each of them correspond to single phrases 
of the Hebrew, which were worked into an early text of 
the LXX. by an unintelligent scribe. 



(%) xxxvi. 2^2-xxxvn. 13. 
22 I80V 6 ισχυρός κραταιώσα iv Behold, God doeth loftily in his 

Ισχΰί αυτού' pOWer, 

TtV yap eVri κατ αυτόν δυνάστης ; Who is a mighiy one like unto 

2 3 τίς 8e €στιν 6 ΐτάζων αυτού τα Who enquireth into his works ? 
epya ; 

η τίς 6 (Ιττών, "Εττραξζν άδικα. Or who can say, Thou hast 

wrought unrighteousness ? 

24 μνησθητι οτι μ€γάλα €στ\ν αυτού Remember that thou magnify his 

ret epya 

Sjv ηρζαν dvSpes, 

25 was άνθρωτΓ05 eldev kv kavr^. 

Every man hath seen in himself^ 

όσοι τιτρωσκόμζνοί €ΐσι βροτοί. 

26 Ιδον δ Ισχυροί ττολυ?, καΐ ού 

άριθμοζ €των αυτού και άπίραν- 


27 άριθμηται δε αυτω OTayovis 

και Ιττιχνθησονται υίτω ds vi- 

28 ρυήσονται παλαιώματα 
Ισκίασί δί ν^φη Ιπί αμύθητων 


Behold, God is great, and we 
know him not : 

The number of his years is un- 

Numbered by him are the drops 
of rain, 

And they shall be poured forth in 
rain into cloud : 

Aitd he hath made the clouds over- 
shadow the countless race of 

ωραν Ζθζτο κτηνζσιν, 
οΊδασι δε κοίτης τάζιν' 

He hath set a season to the beasts 
And they know the order of their 
lying down, 
iiii τούτοις πάσιν οίκ βζίσταταί At all these things thy mind is not 

σου ή διάνοια, astonished, 

ουδέ διαλλάσσξταί σου η καρδία Nor is thy heart parted from thy 
άττό σώματος. body. 


καΧ kav αυντι α-ηίκταοιν [Cod. 

Β άπε/ίτασίί] ν(φέλη$, 
Ισότητα σκηνηί αυτού' 
3θ ιδού €ΚΤ€ν€Ϊ Ιτγ' αυτόν ήδώ \ 

And if thou dost understand the 

spreading of the clouds, 
The .... of his pavilion : 
Behold, he will stretch his bow 

^ For this, which is the reading of almost all MSS., Codd. A, 23 read to to^ov, 
which is the correct translation of the Hebrew i-ii« : here, as in some other 
passages, i and η were confused, so that ^δώ is a transliteration of ii>M. 


/foi ριζώματα θαλάσσης €κάλν- 
31 €V 'yap avToTs Kpivei λαοΰί, 

δώσίΐ τροφην τω Ισχύοντι 
[Cod. Β άκούοντι]. 

32 6771 χπρών €κά\υφ€ (pCjs 

και kv€Teikaro π€ρΙ αύτψ hv 

33 avayyeXei τηρΐ αύτου <}>ί\ον 

αύτοΰ Kvptos, 
κτήσι$ καΐ τηρΐ aSiKias. 
C. xxxvii. I και avb ταύτψ ΙταράχΘη 

ή καρδία μου 
και άπ€ρρνη ίκ του τόπον. 

2 aKove [Codd. A, 23, 254» add 

Ίώβ~] άκοτ^ν kv opyrj θυμού 
και μίλίτη Ικ στόματος αύτοΰ 

3 ύποκάτω ttuvt^s του ουρανού ή 

άρχη αύτοΰ, 
και το (pu>5 αύτοΰ kni τΐτίρύ^ων 
τψ yrjs. 

4 οπίσω αύτοΰ βοήσ^ται φωντ}, 

βρόντησα kv φωντ} νβρ€ω5 αύ- 
καΐ ούκ άνταλλά^€ΐ αύτοΰ$, 
ΟΤΙ άκούσ(ΐ φωνην αύτοΰ. 

5 βροντήσει δ ισχυροί kv ψωντ) 

αύτοΰ θαυμάσια' 

€ποίησ€ yap μεγάλα α ουκ f;Set- 

6 συντάσσων χιόνι Τίνον €π\ γης, 

καϊ χαμών υ€τυς 

καΐ χίΐμών ύβτων δνναστίίαί 

7 kv χαρί παντόί ανθρώπου κατα- 


Ινα γνώ πάς άνθρωπος την iavTOv 

8 βΙσηΚθΐ 5e θηρία νπο την σκβπην 

And he covereth the bottom of 

the sea : 
For by these he judgeth the 

He giveth meat ίο him that is 

He covereth his hands with the 

And giveth it a charge that it 

strike the mark : 

At this also my heart was 

And is moved out of its place. 

And meditation shall go forth 

from his mouth. 
Beneath the whole heaven is his 

And his light unto the ends of 

the earth. 
Behind him shall he shout with 

a voice, 
He shall thunder with the voice 

of his majesty. 

For thou shall hear his voice. 
God shall thunder marvellously 

with his voice, 

Great things doeth he, which we 

cannot comprehend. 
For he saith to the snow, Fall 

thou on the earth ; 
Likewise to the shower of rain 

And to the showers of his mighty 

He sealeth up the hand of every 


That all men may know iheir 

weakness : 
Then the beasts go into their 




ησύχασαν oe ewi κοίτης. 
9 €K ταμκίων ίπερχονται obvvai, 

ά -nb he ακρωτηρίων ψνχοί 
Ο και από irvoijs Ισχυροί) δώσβι 

οίακίζα δέ το νδωρ ώ? kav βού- 
Ι καϊ (κ\€Κτον καταπλάσσ^ι νε- 

διασκορίΓκΐ νβφοί φω5 αυτόν, 

2 και αυτο3 κυκλώματα διαστρέ- 


εν θεεβουλαθώθ, εΙ$ tpya αυ- 

πάντα οσα αν εντείληται αυ- 


3 ταντα συντετακται παρ' αύτοΰ 

επι τψ "yrjs, 
εάν τε ets παιδείαν εάν εls ττ^ν 

yrjv αύτοΰ 
εάν els ελεο5 ενρήσει αυτόν. 

And remain in their dens. 

Out of the chambers come forth 

And from the extremities cold, 
By the breath of God ice is 

And he steereth the water as he 



He spreadeth abroad the cloud of 

his light, 
And he himself %vill turn about 

its circuits : 

All things whatsoever he com- 

mandeth them : 
These things are ordered by him 

upon the earth, 
Whether it be for correction or 

for his earth 
Or for mercy, he shall find him. 

It will probably be found, after a more minute com- 
parison of the Greek text with both the Hebrew and the 
other versions, that, in this section, four poems, two of them 
original and two added, have been fused together. Each of 
the poems has the same theme, the greatness of God as 
seen in nature, and its effect on the mind of man. 

The first of the non-obelized, and therefore presumably 
original, poems seems to consist of c. xxxvi. 22, 23, ΐΔ^α, 
and the section ώραζ; iQero κτηνζσιν which is in some MSS. 
placed at the end of c. xxxvi. 28 and in others in the 
middle of c. xxxvii. 5. It may reasonably be supposed 
that this section forms the end of an enumeration of some 
of the works of God, which has been replaced by the added 
verses 26, 27, 28. 

The second of the non-obelized poems seems to consist 
of the fragments c. xxxvii. ^d, 6 a, y δ (?), 8, g a. It begins 
with the second half of a verse of which the first half 

R 2 


probably resembled the beginning of two other poems, viz. 
xxxvi. I'Xa^ 16 a. The poem, like the preceding, enu- 
merates some of the works of God ; (compare the mention 
of the beasts in xxxvi. :ϊ8 and xxxvii. 8). 

The third poem seems to consist of the obelized passages 
c. xxxvi. ^6, 27, 28 Λ, b, 29, 30, 31, 32, 0,% 34 ( = xxxvii. i). 
It begins, like the first poem, with a declaration of the 
greatness of God, and proceeds to an enumeration of his 
works ; and it concludes with a description of the effect of 
the consideration of those works upon the mind of Elihu 
(και άττο ταύτης €ταράχθη η καρδία μον, καϊ ά'π€ρρνη €κ του 
τόττον avTTJs) which is in apparent contrast with the effect on 
the mind of Job (c. xxxvi. 28 [xxxvii. 5] e^rt tovtols ττασιν 
ουκ ^ζίσταταί σου η biavoLa, ουδέ διαλλάσσεται σου η KapbCa 
άττο σώματοί). 

The fourth poem seems to consist of the obelized 
passages c. xxxvii. 2-5 a, 6 3, J a (and δ ?), g δ, 10-13. This 
poem is more fragmentary than the others, and contains at 
least two verses, 11, 12, which in their existing form are 
not intelligible. 

It is probable that the remainder of the chapter, vv. 
14-24, forms another poem : it contains many philological 
di faculties, but only one obelized verse, v. 18, and therefore 
it comes less than the preceding parts of the speech within 
the scope of this chapter. 

The result of the enquiry is that the hypothesis which 
was advanced at the outset explains satisfactorily the 
majority of the passages which Origen supplied from Theo- 
dotion. In other words it seems probable that the book of 
Job originally existed in a shorter form than at present ; and 
that in the interval between the time of the original transla- 
tion and that of Theodotion large additions were made to 
the text by a poet whose imaginative power was at least not 
inferior to that of the original writer. The additions are in 


general harmony with the existing text, though they do not 
always exactly fit in to their place : nor is it likely that the 
difficulties will be solved until the ten factors which are 
necessary to their solution have each engaged the attention 
of skilled specialists, namely, the philology and the textual 
criticism not only of the Hebrew, but also of the Greek, the 
Syro-Hexaplar, the Sahidic, and the Latin versions. Of 
these ten factors, only the first two, namely the philology 
and the textual criticism of the Hebrew, have as yet been 
dealt with by competent scholars. 



The text of Ecclesiasticus has come down to us in a form 
which, as it is frequently unintelligible, must be presumed 
to be corrupt : but since it is a translation of which the 
original is lost, and since, consequently, its textual diffi- 
culties cannot be explained by reference to that original, 
we cannot, in all cases, know for certain whether they are 
due to imperfections in the translation itself or to an im- 
perfect tradition of it. It has the further element of un- 
certainty that, like all paroemiastic literature, it was altered 
from time to time. The wisdom of the fathers gave place 
to the wisdom of the children : one generation had little 
scruple in correcting, amplifying, and supplementing the 
proverbial sayings of its predecessors. And since there 
are some parts of the book in which the Latin and Syriac 
texts differ not only from the Greek text but also from 
one another, it must be presumed that the original text 
was not only altered but altered in different ways, in dif- 
ferent countries, or at different times. 

The probability of recovering the original text of the 
whole book is consequently small. But for the greater 
part of it we have the same means of determining the text 
that we have in the case of the New Testament ; that is 
to say, we have not only the Greek MSS. but also early 
versions which point to a text that is probably earlier than 
that of the earliest existing MSS. It is remarkable, con- 
sidering the great intrinsic interest of the book, its impor- 
tance in the history of ethics, and the place which it has 


occupied in Christian theology, that so few attempts have 
been made to apply these means to the determination of 
the text where it is doubtful, and to the recovery of it 
where it is at present corrupt and unintelligible. The 
present essay is a study in that direction : its object is to 
show both how much remains to be done and how far the 
existing materials help us to do it. It will begin by a short 
survey of those materials, and proceed to apply them to 
the criticism of some passages. 

1. Greek MSS. 

The Greek MSS. which contain Ecclesiasticus, and of 
which collations have been published, are the following : — 

Uncial MSS. : Codices Alexandrinus A, Vaticanus B, Sinaiticus 
S, Ephraemi rescriptus C (in T\?>c\iQnaon Monumenia Sacra, vol. i), 
Codex Venelus, a MS. of the 8th or 9th century. No. i in the Ducal 
Library (Holmes and Parsons, No. 23). 

Cursive MSS. : No. 55', a Vatican MS. (No. i of Queen Chris- 
tina's MSS.) probably of the twelfth century: No. 68, a Venice 
MS. (No. 5 in the Ducal Library) probably compiled from earlier 
MSS. by order of Cardinal Bessarion, very partially collated for 
Holmes and Parsons : No. 70, a MS. of the 15th century in the 
Library of St. Anne at Augsburg, probably the same as that which 
was collated by D. Hoeschel (see below); only c. i was collated for 
Holmes and Parsons : No. 106, a Ferrara MS. described as being 
apparently written ' in charta papyracea Aegyptiaca,' and dated 
A.D. 734 .'' (yThe First Annual Account of the Collation of the MSS. 
Oxford, 1789, p. 64) : No. 155, a MS. of the i ith century, formerly 
in the Meerman Collection at the Hague, and now in the Bodleian 
Library (Auct. T. Π. 4) : No. 157, a Basle MS. : No. 248, a Vatican 
MS. (346) of about the fourteenth century : No. 253, a Vatican MS. 

^ The numbers are those of Holmes and Parsons : the references in the fol- 
lowing pages to the cursive MSS., with the exception of No. 155, which has 
been collated independently, are made from the MS. collations, now in the 
Bodleian Library, and not from the printed edition. The numbers which are 
placed in brackets, e.g. (ΐίι7)> are those in which the collator has made no note 
of variation from the printed text which he used, and in which, consequently, the 
reading of the MS. is inferred, more or less uncertainly, e silentio. 


(336) also of about the fourteenth century : No. 254, a Vatican MS. 

(337) of about the thirteenth century : No. 296, a Vatican MS. 
(Codex Palatinus, No. 337) probably of the eleventh century : 
No. 307, an incomplete Munich MS. (129, formerly 276) of the 
fourteenth century : No. 308, a Vatican MS., described by Holmes 
and Parsons (Praef. ad Hbr. Ecclesiastici) as Codex Palatinus Vindo- 
bonensis : but the MS. collation was made at Rome, and describes it 
simply as * MS. Palatinus,' without further identification : (there is 
no trace of it in Stevenson's catalogue of the Codices Graeci Palatini). 
In 1604 D. Hoeschel published an edition of Ecclesiasticus with 
variants from a MS. in the Library of St. Anne at Augsburg, which 
he does not further identify, but which is probably of the fifteenth 
century (Holmes, IVm/k Annual Account, Oxford, 1797, p. 25). 

In addition to these there are many MSS. of which no 
published collations exist : of these probably the most im- 
portant are the palimpsests of the 6th or 7th century 
at St. Petersburg, which Tischendorf promised to publish in 
his Monumejtta Sacra, vol. viii. Two Vienna MSS., Cod. 
Theol. Gr. xi (quoted below as Vienna 1) and Cod. Theol. 
Gr. cxlvii ( = Vienna 2), both of which were brought by 
Busbecq from Constantinople, have been partially collated 
for this work. 

It is desirable in the first instance to form a working 
conception of the character and relations of the chief MSS., 
in order to ascertain what kind of presumption for or 
against a reading is afforded by the fact of its occurring 
in a particular MS. or group of MSS. Such a conception 
may to some extent be derived from an examination of 
other books of the Bible in the same MSS. But there are 
two considerations which limit that extent : the first, which 
is the less important one, is that the MSS. of the whole 
Bible were written by different hands, and that no two 
scribes can be assumed to have copied with precisely the 
same degree of accuracy : the second, which is the more 
important consideration, is that different books or groups 
of books may be supposed to have been copied from dif- 


ferent originals. The main ground for this supposition in 
the case of the two books of Wisdom is that though they 
are always placed together, their place, like that of other 
books which were probably circulated separately, is dif- 
ferent in different MSS., for example. 

In the Sinaitic MS. the order (omitting the earlier books) is . . . 
Major Prophets, Minor Prophets, Psalms, Proverbs, Ecclesiastes, 
Canticles, Wisdom, Ecclesiasticus, Job. 

In the Alexandrian MS. the order is . . . Minor Prophets, Major 
Prophets, Esther, Tobit, Judith, Esdras, Maccabees, Psalms, Job, 
Proverbs, Ecclesiastes, Canticles, Wisdom, Ecclesiasticus. 

In the Vatican MS. the order is . . . Psalms, Proverbs, Ecclesiastes, 
Canticles, Job, Wisdom, Ecclesiasticus, Esther, Judith, Tobit, Minor 
Prophets, Major Prophets. 

In the Ferrara MS. (Holmes and Parsons, No. 106) the order is 
. . . Job, Proverbs, Wisdom, Ecclesiasticus, Major Prophets, Minor 
Prophets, i and 2 Maccabees, Psalms. 

These differences of position seem to be best explained 
by the hypothesis that, although at the time when the 
MSS. were written there had come to be a general agree- 
ment as to the books which should be included, the books, 
or small groups of them, existed in separate MSS. 

It is consequently possible that the original MS. from 
which the scribe of e.g. the Vatican MS. copied Ecclesias- 
ticus may have been different from that from which he, or 
his earlier colleague, copied the Pentateuch. So that no 
inference lies from the accuracy or inaccuracy of the one 
text to the accuracy or inaccuracy of the other. Hence 
the MSS. of each book must be separately considered in 
relation to the book : and a general estimate, or working 
conception, of their value, and , of their relation to each 
other, must be formed before the text of the book can be 

The following is an endeavour to show the way in which 
such an examination may be made upon the comparatively 


neutral ground of grammatical forms and usages, i.e. upon 
ground on which the scribe was not led to vary the reading by 
a desire to harmonize, or to interpret, or to paraphrase it. 

I. Forms of Words. 

In I. 3 : 18. 6 all MS S., without a variant, have a form of the 
Hellenistic φχνιάζω: in 42. 18 they have, also without a variant, a 
form of the Classical ^ξιχνίνω : in 6. 27 all MSS. except Codd. 253, 
307 have a form of εξιχνβύω, but in 18. 4 Codd. 253, 307 agree with 
Codd. ACS, 155, against Cod. Β and the rest, in having a form of 


1.6: Codd. ACS, 23 have the classical form πανουργ-ηματα, Codd. 
B, cett. the Hellenistic πανονργ^ύματα : so also in 42. 18 Codd. AS\ 

307 τνανονρ-γημασιν, Codd. B, cett. navovpyevpaaiv. 

I. 27: Codd. ACS, 55, 70, 106, 157, 254, 296, 307 πραντης: 
Codd. B, (23), (155), (248), (253) πραότης. But in 3. 17 : 4. 8 : 
10. 28 : 36. 28 all important MSS. read πραντης : and in 45. 4 Cod. 
A reads πραότης, against the πραύτης of all other MSS. 

27. 13 : Codd. AS προσώχθίσμα : Codd. BC προσόχθισμα, 

40. 5: Codd. AS, 106, 157, 253, 307 μψιμα: Codd. 55, 155, 

254 μήνις: Cod. 308 μψισμα'. Cod. 248 μίμημα: Codd. BC μηνίαμα, 

a word which is not elsewhere found. 

2. Inflexions. 

4. 3 : Codd. AS παρωργισμ^νην : Codd. BC 7Γαροργισμ€νην. 

8. 6 : Codd. AS, 23, 106, 157, 248 ev yrjpa: Codd. BC, cett. eV 


14. 14: Codd. AS, 55, 106, 155, 157, 248, 253, 254, 296 παρ^λ- 

θίτω : Codd. BC, (23) παρζλθάτω. 

14. l8 : Codd. AS dev^pov δασβω? : Codd. BC devdpov daaeos. 

15. 2: Codd. AS, 55, 106, 155, 157, 248, νπαντησα: Codd. BC, 
(254), (296) ντταντησζται '. Codd. 23, 253 άπαντησ^ταί. The future 

of υπαντάω in late Greek seems to have been νπαντησομαι : Sext. 
Emp. adv. Phys. 10. 60, p. 644, probably after the analogy of 
άτταντάϋ). (But the future active of απαι/τάω is found, without variant, 
in Mark 14. 13). 

15. 3: Codd. ACS, 155, 157,254, 296, 307 ποτίσω: Codd. Β, (55), 
{lo6), (248), (253) ποηα. So C. 24. 31. 


15. 4 ^ Codd. ACS, 23 στηρισθησζται : Codd. B, Cett. στηριχθησβται : 

but elsewhere in the book, viz. 24. 10: 29. 32 : 42. 17, the form 
with X is found without any important variant. 

17. 27: Codd. ACS, cett. eV αδον : Cod. Β eV adovs: Cod. S^ 

61/ αδη. 

28. 26 : Codd. AS^ όλισθ^ί : Codd. BCS' οΧισθησ^ς [S^ -σι?]. 
All the other aorist forms of the word in the book are, as usual in 
Hellenistic Greek, first aorist forms, viz. 3. 24 : 9. 9 : 14. i : 25. 8, 
without important variant except Cod. C in 9. 9 όλισθϋς for ολισθησυς. 

3 . Use of the paroemiastic fu ture. 

3. 3: Codd. ACS\ 106, (157), 253, 254, 296, 397 ε|ιλάσίί€ταί : 
Codd. B, 23, (55), (155), (248). (308) eltXfWrm. 

4. 13 : Codd. ACS, 23, 55, 106, 157, 248, 253, 254, 307 ευλογεί : 

Cod. Β evXoyrjaei. 

4. 17: Codd. ACS^ 55, 157, 248,2^4, zgonnpevaerai : Codd. BS^ 

(23), 70, 106, (155), (253), (308) πορεύεται. 

11. i: Codd. ACS, 23, 55, 106, 155, (157), 248, 254, 307 
άννψώσα [307? νψωσ^ι^ : Codd. Β, 296, 308 άννψωσί. 

12. 3 : Codd. AS, 23, 155, (157), 248, 253, 254 ουκ εσται : Codd. 
BC, 55, (106), 296, (308) ουκ €στι. 

i6. 25: Codd. AS, 23, 55, 106, 157, 248, 253, 254 βκψανώ : 

Codd. BC €κφαίνω. 

19. 30 : Codd. ACS, 55, 106, 155, 157, 248, 253, 254 avayyeXer. 
Cod. Β auayyeWei. 

4. Omission or insertion of the Article. 

(a) Instances of insertion in Cod. A and other MSS.., and of 
omission in Cod. Β : 

6. 23 : Codd. AS, 155, 157, 307 τψ yvtii^r]v μην: Codd. Β cett. 

yuωμηv μου. 

7. 19: Codd. AS, 23, 55, 106, 155, 157, 248, 296, 307 ^7 yhp 

χάρις avrrjs '. Codd. B, 253, (254)5 (S^S) και yap χάρις. 

7. 20: Codd. AS, 55, 106, (157), 248, 253, 296, 307 δώόντα την 
ψνχην αντον: Codd. Β, (23), (l55)j (S•^^) δίδοι/τα ψνχην αντυυ. 

ΙΟ. 4 • Codd. AS, 23, 55> ιο6, ΐ55) ΐ57 ^ εξουσία τψ yrjs : Codd. 

BC, 248, (253), {254) ^ονσία της y^. 


12. 2: Codd. AS, 23, 55, 106, 155, 157, 248, 253, 254, 307 
πάρα τον ν^\τίστον '. Codd. BC, (296) τταρα υψίστου. 

15• 5 : Codd. ACS, 55, 106, 157, 307, 308 τό στόμα: Codd. Β, 

(23)5 (l55), (248), {253)5 (254) <^^όμα. 

21. 20 : Codd. ACS, 55, 155, 157, 254, 308 την φωνψ : Codd. 

Β, (23), (lo6), (248) φωνην. 

40. g: Codd. ACS, 55, 106, 155, 248, 254 eVi τό ϋψος της γης'. 

Codd. Β, (23), (ι 57), (253) «τ^'^ Η''^ ^^^ 7^^• 

(β) Instances of omission in Cod. A and others, and insertion in 
Cod. Β : 

4. 28: Codd. ACS, 55, 106, 155, 157, 248, 253, 254, 296, 307 

ecos θανάτου : Codd. B, 23, (308) €ως τυυ θανάτου. 

7. 8 : Codd. AS, 23, 55, 106, 155, 157, 248, 253, 254, 307 eV 

yap μια : Codd. BC ev γαρ tjj μια. 

12. 5: Codd. AS, 55, 106, 155, 157, 248, 253, 254, 296, 307 

ταπ€ΐνω: Codd. BC, (23) τω ταπ€ΐνω. 

12. 7: Codd. AS, 248, 253, 307 αμαρτωλού: Codd. BC, (23), 

(55). (106), (155), (157), (296) Tod αμαρτωλού. 

5. Syntactical usages, 

4. 17: Codd. B, (55), 157, (254), 296, (308) have 5e in apodosi^ 
φόβον δε και δ^ιλίαν €πάξ€ΐ : Codd. ACS, 23, 106, ΐ55? 248, 2533 3°7 
omit be. This use of Se is so rare in BibHcal Greek that it is more 
likely to have been added by Cod. Β than omitted by the other 
MSS. : and it is noteworthy that in one of the two instances, both 
of which are disputed, of the same usage in the N. T., viz. i Pet. 
4. 18, it is Cod. Β which, against almost all other MSS., both uncial 
and cursive, inserts 8e in the quotation from Prov. 11. 31. 

9. 12 : Cod. A μη €υδοκηστ)ς evboKia άσεβων: Codd. CS, 157? 248 
. . . €υ8οκίαις : Codd. B, (55), (155) ... iv €vdo<ia. I Codd. 23, I Ο 6, 

254, 296, 307. There is a similar variation elsewhere in the con- 
struction of evBoKelv : it is found with iv in 2 Kings 22. 20 ; i Chron. 
29. 3 ; Ps. 43 (44)• 3 ; 48 (49)• 13 ; 67 (68). 16 ; 146 (147). 10 ; 
Hab. 2. 4 ; Mai. 2. 17 ; i Mace. 10. 47 : without eV in i Esdr. 4. 
39; Sir. 18. 31; I Mace. i. 43. 

II. 7: Codd. AS, 23, 55, 248, 254, 307 have πρ\ν η c. subj. 
followed in v. 8 hynp\v c. infin., in both cases with a negative main 
sentence : in 18. 19: 19. 17: 23. 20 they have nph η c. infin. with 


an affirmative main sentence. In 11. 7 Cod. Β has τΐρίν c. subj. 
followed in v. 8 by πρΙν η c. infin. There are similar variations in 
the construction of π piv fj in the N. T. : (i) when used with the 
infinitive, there was a tendency to drop ή\ which is found without 
variant only in Matt. i. 18, Acts 7. 2, whereas it is omitted in Matt. 
26. 34 by all good MSS. except L, in Matt. 26. 75 by all except A, 
in Mark 14. 30 by ^ζD, and in Acts 2. 20 by ^ζACD : (2) its use 
with the subjunctive tended to disappear, for in Luke 2. 26 Codd. 
^?^L and others add ap to 7, Cod. Β omits ή and inserts av, and in 
Luke 22. 34 Codd. nBL substitute εω? for πρ\ν η, which is read by 
A only of the greater uncials. 

41. 2 : Codd. AS, 55, 155, (157), 307, 308 ^Χασσουμίνω iv Ισχνϊ: 
Codd. BC, (23), (106), (248), (253), (254) iXaaaovpivco Ισχνί. 

44. 5 : Codd. AS, 55, 106, 155, (157), 248, 254, 296 κ€χορηγη^ 
μίνοι iv Ισχνϊ: Codd. Β, 23, 253, 308 κ(χορηγημ€νοι Ισχνϊ, 

45• 2 : Codd. AS ωμοίωσ€ν αντον iv 8όξτ] ά-γίων : Codd. BC cett. 

. , . do^T] άγιων. 

45. 15 : Codd. A, 25, 106, 155, 157, 248, 254 βγ^νηθη αυτώ eh 
δίαθηκην αΐώνιον και τω σπίρματί αντον iv ημύραις ονρανον '. Codd. BC, 
cett. . . . κα\ iv τω σττ/ρ/χαη αντον . . . 

46. 5 • Codd. AS, 155 iπηκovσ€v αντών μ€γας κύριος λίθοις χαΧάζηα Ι 
Codd. BC, cett. . . . iv λίθοις χάλάζης. 

It will be noted that although, as is usually the case, no 
MS. is uniform in either its forms or its syntax, the Hel- 
lenistic forms and constructions preponderate in the Vatican 
Codex. It will also be noted that in almost all cases the 
majority of MSS. are against that Codex in these respects. 
The more difficult question remains undecided, whether the 
Hellenisms or the Classicalisms belong to the original text : 
in other words whether a Hellenistic text was purged of 
some of its Hellenisms by purist scribes with the view of 
rendering the work more acceptable to educated persons, 
or whether a Classical text was altered by Hellenistic 
scribes who substituted a more familiar for a less familiar 
form or phrase. 

2 54 on the text of ecclesiasticus. 

2. Latin and Syriac Versions. 

I. The Latin Versions. The old Latin version, which 
was left untouched by Jerome, has come down to us in 
the following MSS. 

(i) The Toledo MS., the collation of which was first published 
by Bianchini in his Vindiciae Biblicae,\eTon2L, 1748, from which 
work it was reprinted by Vallars in the Benedictine edition of St. 
Jerome, vol. χ : (2) the Codex Amiatinus, the text of which is 
printed at length by Lagarde, Miiiheilungen, p. 283 : (3) the MSS. 
collated by Sabatier, viz. two Corbey MSS., one St. Germain MS., 
and one MS. of St. Theodoric of Reims. 

But it is probable that the large quotations from the book in 
St. Augustine's Speculum (last edited by Weihrich in the Vienna 
Corpus Scripiorum Ecclesiasticorum, vol. xiii) represent a more 
current form of the text than any of the above MSS. 

%, The Syriac Versions. There are two Syriac versions, 
the Peschitta and the Syro-Hexaplar. 

{a) The Peschitta, or current Syriac version, was first printed, 
with a Latin translation, in Walton's Polyglott, vol. iv : it has more 
recently been edited, with the help of six MSS. in the British 
Museum, by Lagarde (Lihri Veteris Testamenti Apocryphi Syriace, 
1861): the photographic reproduction of the oldest MS., that of 
the Ambrosian Library, has not yet been completed, {h) The 
Syro-Hexaplar version has been published for the first time, from 
an Ambrosian MS., in photographic facsimile by Ceriani in his 
Monumenta sacra et prof ana ^ vol. vii, Milan, 1874. 

There are some parts of the book in which the Latin and 
Syriac differ so widely from both the Greek and one an- 
other as to force upon us the hypothesis that the original 
text underwent in very early times different recensions. 
But for the greater part of the book the Latin and the 
two forms of the Syriac clearly point, with whatever dif- 
ferences in detail, to the same original as the Greek. The 
relation of the Latin and the Syro-Hexaplar to the Greek 
is clearly one of derivation. The relation of the Peschitta 
to the Greek must be considered to be still sub judice : nor 


can it be determined with any approach to scientific com- 
pleteness until after the exact study of the Greek text itself, 
to which the present essay is designed to be a contribution. 

The question of this relation of the Peschitta to the Greek is 
extremely complex. There are some passages in which the Syriac 
appears either to be based on an earlier Greek text than that which 
has come down to us, or to have been revised by reference to the 
Hebrew. There are, on the other hand, passages in which both 
the Greek and the Syriac have an unintelligible phrase which points 
to a mistranslation of the same Hebrew original. For example, in 
25. 15 the Greek ουκ eVn κεφαλή vnep κζφαλην οφεως, and the Syriac 
equivalent ' No head is more bitter than the head of a serpent,' 
point to a mistranslation of tJ^NT, viz. 'head' for 'venom': but 
there is nothing to determine whether the mistranslation is common 
to the two versions, or was derived by one from the other. The 
question of derivation will be positively determined by the examina-' 
tion of the passages, some of which are mentioned below, in which 
an error which has grown up inside the Greek text, is copied by the 
Syriac : for example, if it be true that in 5. 6 the Greek originally 
read παρ' αυτού, with a verb such as iXevaerat in place of eXeos, the 
Syriac, which is a translation of παρ' αντώ without an expressed 
verb, must be presumed to be derived from a Greek text in which 
Trap* αυτω was read, and from which the verb had already disappeared. 
So also, if it be true that in 25. 17 the reading αρκος is a mistake for 
apKvs, and that σάκκον {σάκκος) was a gloss upon αρκυς, even if it be 
not an equivalent early reading, with the same signification, the Syriac 
' sackcloth ' can only be a misinterpretation of the Greek σάκκον. 

But a more important question than that of the relation 
of the Peschitta to the Greek is that of the contributions 
which both the Latin and the Syriac make to the deter- 
mination of the original text. It will be found that all 
three versions are more or less corrupt, that they also have 
a common tendency to paraphrase, and that in a large 
proportion of passages each of them supplements the other. 
The justification of this remark can of course only be found 
in the examination of a considerable number of passages : 
the two following are taken, almost at random, as examples : 



δ σοψος kv Xoyois npoa^ei 

καΐ άνθρωποί ψρόνιμοί 
apiffei μ€'<(ΐστασιν 

δ €pyaζόμ€vos yrjv ανυ- 
ψώσει θημων'ιαν αύτοΰ 

και δ άρ^ΰκων μΐ^ιστάσιν 
€ζιλάσ€ται άμαρτίαν 

(ΐ) XX. 27, 28. 
Cod. Amiat. 

sapiens |in verbis pro- 
ducet seipsum 

et homo prudens place- 
bit magnatis : 

quioperatur terram suam 
exaltabit acervum 

et qui operatur iustitiam 
ipse exaltabitur : 

qui vero placet magnatis 
effugiet iniquitatem 


He who is full of the 
sayings of wisdom, 
how shall he show 
himself small ? 

And a wise servant shall 
be lord over princes. 

The first four lines of the Latin give two well-balanced 
couplets : 

A man who is clever in speech will advance himself, 
And a man of understanding will be pleasing to princes : 

He who works his land will raise a high heap of corn. 
And he who works justice will himself be raised. 

The fifth line of the Latin, 

He who is pleasing to princes will escape injustice, 

is out of harmony with the context, and is easily under- 
stood as a gloss upon the second line. But it is a trans- 
lation of the fourth line of the Greek, where it is equally 
out of place. It seems probable that the fourth line of 
the Greek was originally a gloss upon the second line, that 
the original fourth line should be restored from the Latin 
fourth line, and that the Latin fifth line was added when 
the present fourth line of the Greek had superseded the 
original fourth line. 

The Syriac seems to paraphrase the first couplet and 
to omit the second : its diminished paroemiastic force 
makes it difficult to take it as the original form. 



άνθρωποι άνθρωηω συντη- 
ρεί οργην, 

και παρά κυρίου ζητ€Ϊ 
ϊασιν ; 

Ιτγ' ανθρωττον ομοιον αυτω 
ουκ έ'χβί eXeos, 

και nepi των αμαρτιών 
αύτου Seirai ; 

αΰτόί σαρξ ων διατηρβΐ 

μην IV 
Tis (ξιλάσεται tcLs άμαρ- 

Tias αύτοΰ ; 

μνησθητι τα. ΐσχατα και 
■πανσαι Ιχθραίνων, 

καταφθοράν και θάνατον 
καΐ (μμενε evToXais' 

μνησθητι εντολών και μη 
μηνίσχι^ τω πλησίον 

καΐ διαθήκην υψίστου και 
πάριδε dyvoiav. 

(2) xxviii. 3-7• 
S. Aug. S^ec. p. 142. 

homo homini servat 

et a Deo quaerit medel- 


in hominem similem sibi 
non habet misericor- 

et de peccatis suis de- 
precatur ? 

ipse dum caro sit ser- 
vat iram, 

et propitiationem petit 
a Deo? 

quis exorabit pro delictis 

memento novissimorum 
et desine inimicari, 

tabitudo enim et mors 
imminent in mandatis : 

memorare timorem Dei 
et non irascaris prox- 

memorare testamenti al- 
tissimi et despice ig- 
norantiam proximi. 


A man who cherishes 
wrath against a man, 

How should he ask for 
healing from God ? 

He who is himself a man 
is not willing to for- 

shall any one forgive that 
man's sins ? 

Remember death, and 
lay aside enmities, 

the grave and destruc- 
tion, and abstain from 
sinning : 

Remember the com- 
mandment and hate 
not thy neighbour be- 
fore God : 

nay, give him that of 
which he is in want. 

Each of the first three couplets of the passage in the 
Greek and Latin appears to express the same idea in 
a sh'ghtly altered form. But while the duplication of an 
idea is common, the triplication of it is so unusual as to 
suggest the hypothesis that one of the forms is a gloss. 
The hypothesis is supported by the fact that the sixth line 
of the Latin is clearly another form of the second, and that 
it is introduced out of place between the two lines of the 
third couplet, so that the six lines of the Greek are repre- 
sented by seven lines in Latin. It is even more strongly 



supported by the fact that the third couplet is altogether 
omitted from the Peschitta. 

In the fourth couplet of the Latin 'tabitudo enim et 
mors imminent ' clearly show a corruption of ' imminent ' 
for ' immane ' = €μμ€ν€, and a consequent corruption of the 
nominatives ' tabitudo ' and ' mors ' for the genitives ' tabi- 
tudinis ' and ' mortis.' 

The last line of the Syriac is also clearly corrupt. The 
exhortation of the Greek and Latin ' overlook the ignor- 
ance (transgression) of thy neighbour ' is in entire harmony 
with the drift of the passage : the exhortation to almsgiving 
is a commonplace which gives no suitable antithesis to the 
preceding half of the couplet. 

The whole passage consists, in other words, of two 
quatrains which are best represented by the first two and 
the last two couplets of the Greek text : but the third 
couplet of the Greek text is an intrusive gloss. 

3. Examination of some important instances 
of variation. 

I now proceed from the short survey of the materials 
to the examination of some passages in which the variants 
are important, and in which the text can only be deter- 
mined by the help of whatever critical aids we possess. 

i. 13. 
Codd. ACS, 23, 70, 155, 157, 248, 253, 296, 307,Vienna i iv ημη.α 

Τίλΐντης αύτου €νλογηθησ€ται : Codd. Β, (55)> (ΐ^^)? (3°^)> (254)? 

Vienna 2 . . . €νρησ€ΐ χάριν. 
Latin : ' in die defunctionis suae benedicetur.' 
Syriac : Pesch. ' in the end of his days he shall be blessed.' 
It seems clear that ^υλογηθησεται is the correct reading : the 
diplomatic evidence against evprjaei χάριν is supported by the fact 
that that phrase does not appear to be used absolutely in the LXX., 

but always with the addition eu όφθαλμοίς {βναντι, ενώπιον) αυτού 

{κνρίσυ\ e.g. infra, iii. 18. 


i. 23. 
Codd. ACS^, 23, 157, 253, Vienna I varepov αΙτω dvaboiaei €υφρο- 

σννην. Codd.B,(55), 106, i55,(248),(254), 296,(3o8),Vienna2 

.... ευφροσύνη '. Cod. 70 άναδωσει els (ύφροσννηρ. 

Latin : ' et postea redditio jucunditatis.' 

Neither ευφροσύνη nor (νφροσύνην seems to be grammatically 
possible : the former because it involves a neuter sense for άναδωσ^ι, 
the latter because αι/αδώσ^ι has no subject. The Latin suggests 
the conjecture that the original reading was avadoais ευφροσύνης: 
the substitution of άνάΒωσις for dvaBoais by an early scribe would be 
a not uncommon change, and would sufficiently account for the 

iii. 10. 

Codd. ABCS, 106, 157, 254, 296, 308, Vienna i ov yap eVrt σοι 

δόξα προί άτιμίαν '. Cod. 253 • • • • δό^α ώί ατιμία: Cod. 1 55 
.... Βόξα άτιμίαρ : Vienna 2 nps ατιμία. 
Codd. (23), (55)» (248) .... δο£α πατρός ατιμία, 

Latin: 'ηοη enim est tibi [Cod. Am. omitsj gloria sed confusio.' 
Syriac : Pesch. ' for it will not be a glory to thee : Syr. -Hex. ' for 
it will not be an honour as a disgrace to thee' : (the subject ' the 
shame of thy father,' is continued from the preceding clause). 
The difficulties in the way of accepting πατρός ατιμία as the ori- 
ginal reading are mainly (i) the difficulty of accounting for the 
corruption of so simple and obvious a phrase into προς άτιμίαν in 
the majority of MSS., (2) the absence of an equivalent phrase in 
both the Latin and the Syriac. If προς άτιμίαν were the reading of 
only a small group of MSS., it might have been supposed that 
some one scribe had written πατρός in the contracted form πρς, and 
that the copyists of this MS., mistaking the contraction, had adapted 
ατιμία to the Supposed preposition. But this hypothesis hardly ac- 
counts for the facts (i) that προς άτιμίαν is read by MSS. of such 
different character as those enumerated above, (2) that the Syro- 
Hexaplar supports the reading ως ατιμία of Cod. 253. 

iii. 26. 
Codd. ACS, 23, 55, 106, 155, 157, 248, 253, 254, 296, 307, 

Vienna I ό αγαπών κίνδυνον iv αντω άπολύται : Codd. Β, (308) 
. . . . iv αυτω 4μπ€σύται. 

Latin : ' qui amat periculum in illo [Cod. Tolet. ' ipso '] peribit.' 



It may be noted that although Β probably stands alone, the 
quotation in S. Aug. de civit. Dei τ, 27 'qui amat periculum incidit 
in illud ' shows that it preserves an ancient variant. 

iv. II. 

Codd. ACS, 23, 55, 157, 248, 253, 296, 307 r] σοφία vlovs αντης 

[55> i57> 248, 296 (αντης'] άννψωσί: Codd. Β, 155, (254), 

(308) .... νίονς iavTTJ άνυψωσ^ρ : Cod. 106 avrrj νΐονς νψωσε. 

Latin : the MSS. agree in reading ' sapientia filiis suis vitam : ' 

they differ in regard to the verb, Cod. Tolet. ' inspirabit,' 

Cod. Amiat. ' spirat,' Cod. S. Germ. ' inspiravit,' Codd. cett. 

' inspirat.' 

The Latin seems to show that the Greek verb was originally 

€\1τνχωσβ or (ν^ψνχωσβ : and this hypothesis is confirmed by what 

appears to be a reference to this passage in Clem. -Alex. Strom. 7. 

16, p. 896 η σοφία, φησ\ν 6 Σολομών, ΐν^φνσίωσζ [€ν€φνσησ€? cp. SUpra, 
p. 148] τα iavTTJs τίκνα. 

iv. 15. 
Codd. ACS, 23, 55, 106, 155, 157, 248, 253, 296, 307, Vienna 2 

6 προσίχων avrfj κατασκηνώσει πεποιβώς '. Codd. Β, (254), (308) 6 


Latin : ' qui intuetur illam permanebit [Cod. Amiat. ' permanet '] 

There is a similar variation of readings in i Tim. 6. 3, where 

Cod. S^ reads κα\ μη προσίχ^ται νγιαίνονσιν 'Χόγοις, which is supported 
by the uniform translation of the Latin ' acquiescit, (-cet) ' whereas 
all the other Greek MSS. read προσίρχΐται. 

v. 6. 

Codd. ACS, 55, 106, 155, 253, 254, 296, 307, Vienna 2 eXeo? yap 
κα\ opyt) παρ* αυτω : Codd. Β, 23, (308) .... τταρ αυτόν : Codd. 
i57j 248 . . . . παρ' αντω ταχννα,. 
Latin : ' misericordia enim et ira ab illo cito proximat ' [so Codd. 

Tolet. Amiat. : Codd. cett. ' proximant.'] 

Syriac : Pesch. ' for mercy and wrath are with him.' 

The Latin confirms the reading of Codd. 157, 248 in respect of 

ταχνι /fi, but suggests that παρ αυτόν was read rather than παρ' αντω. 

The Syriac on the other hand is in harmony with the majority of 

Greek MSS. The absence of a verb would be out of harmony 


with the verses which precede and follow : whereas the introduction 
of ταχυνΰ makes the verse closely parallel to v. 7 3 i^amva yap c'^e- 

Xfvaerai οργή κυρίου. 

The exegetical difficulty of the verse lies in eXeos: for the 
whole of V. 6 <5 seems to be an answer to the sinner's plea ' His 
compassion is great, he will make propitiation for the multitude of 
my sins : ' and it is conceivable that the corruption of the text is 
greater than either the MSS. or the versions show. The exegesis 

seems to point to an original reading [e^JeXeuaerai yap opyfj παρ' αυτού 

' for wTath shall come forth from him, and his anger shall abide 
upon sinners/ The next verse, assuming that the sinner will ac- 
cept this assurance, and repent, urges him to do so speedily : on 
the ground that not only will wrath come forth but that it will do 
so speedily: hence ίξά-κινα e^eXeiaerai would be not a repetition 
but a natural expansion of the supposed i^ikeiaerai in v. 6 b. 

The clause TKcos yap κα\ opyr) ivap αυτού is found also in 16. 12 

where the mention of mercy as well as wrath is quite appropriate, 

and is amplified in the following clause δυνάστης εξιλασμών κα\ εκχίων 

vii. 18. 

Codd. AS, 23, 1 55) ΐ57? Vienna i μη άΧλάξης φίλον αδιάφορου: 
Codd. BC, (55), (253), (254), 296, 308, Vienna 2 μη άλλάξτ}: 

φιλον eveKfV (elveKcv^ αδιάφορου '. Cod. I06 μη aXka^rjs φίλον αδιά- 
φορου κατά μηδέν : Cod. 248 fu) άλλάζης φίλον αδιάφορου μηδέ ev '. 
Cod. 307 Μ f^ey^jys φίλον €V€K€V αδιάφορου. 

Latin : Codd. Am., S. Theod. ' Noli praevaricari [Cod. Am. -re] 
in amicum pecunia differenti : ' (' praevaricari in ... . '=παρα' 
βαίνειν, e.g. Is. 66. 24 'qui praevaricati sunt in me :' cf. Rom. 
4. 15 'ubi enim non est lex nee praevaricatio.') 

Syria c : Pesch. ' Barter not a friend for money.' 

It must be gathered both from the Latin and the Syriac that the 
word in the genitive, whether αδιάφορου or another word, was taken 
to mean * money ' : but (i) διάφορον, not άδιάφορον, is the Hellenistic 
word which has this sense : e.g. Corpus Inscr. Graec. 2347 r, 56 
TO aiTOTeTaypivov eis τον στίφανον €Κ του νόμου διάφορον ' the money as- 
signed for the crown in accordance with the law: ' 2 Mace. i. 35 
τΓολλά διάφορα ελάμβανε κα\ μετεδίδου * he took and distributed many 
sums of money : ' (2) the Latin ' differenti ' points to a reading 
διαφόρου in the text which the Latin translator used : the addition 


'pecunia' maybe regarded as having been added either by the 
translator to define the uncertain meaning of ' different!/ or as a 
gloss at a subsequent time. 

The original text of the LXX. was thus, in all probability, μη 
άλλαξες φίλον διαφόρου : the Other readings are attempts to explain 
αδιάφορου, as is most clearly seen in Cod. 307, which changes the 
meaning to ' Do not rebuke a friend for a trifling cause.' 

X. 17. 
Codd. ACS, 23, 106, 155, (157), 248, 254, 296, 307 €ξηρ€Ρ 

αυτούς [C, αυτάς, S^, 23, 296, e^ αύτών^ κα\ anaXcaev αυτούς 
[C, αυτάί] : Codd. Β, (308) €ξηραν€ν i^ αυτών: Cod. 55 e^- 
r)pavev αυτούς. 

Latin : ' arefecit ex ipsis et disperdidit illos [eos].' 
Syriac : Pesch. * he destroyed them, and overthrew them.' 
The reading i^rjpavev is supported by the Latin : but it has (i) 
the exegetical difficulty that it would be a mild word inserted 
among strong ones, (2) the critical difficulty that it does not ac- 
count for the reading εξ αυτών, with which it is incompatible. On 
the other hand e^rjpev, which is always elsewhere in the Apocryphal 
books constructed with an accusative followed by i^, e.g. i Mace. 
12. 53: 14. 7, 36, not only gives a congruous meaning, but also 
accounts for both αυτούς and εξ αυτών. It may be conjectured that 
the latter phrase was in the original text εξ ανθρώπων [i.e. ΕΞΑΥΤί2Ν 
=ΕΗΑΝί2Ν] : the words ' he put them away from among men and 
destroyed them ' would thus find a natural balance in the following 
clause, ' he caused their memorial to cease from off the earth.' 

X. 27. 

Codd. A, 106, 157? 296, Vienna I κρείσσων εργαζόμενος κα\ περισ- 
σεύων [i57j "fi^o"] CI' τασιι/ ή περίπατων δοξαζόμενος κα\ υστερών 

[ιο6, 296, Vienna ι άπορων^ άρτων [ιο6, Vienna ι άρτου]. 

Cod. Β κρείσσων εργαζόμενος εν πάσιν η περίπατων η δοξαζόμενος κα\ 

άπορων άρτων. 
Cod. 155 κρείσσων εργαζόμενος εν πάσιν η περίπατων δοξαζόμενος και 

άπορων άρτου. 
Cod. S κρείσσων εργαζόμενος η [S^ omitS η and adds εν πάσιν] κα\ 

περισσεύων εν πάσιν [S^ OmitS εν π.] η περίπατων δοξαζόμενος κα\ 

άπορων άρτων. 


Codd. 23, 248 κρζίσσων yap 6 εργαζόμενος καί περισσεύων iv πάσιν η 

ό δοξαζόμενοί καϊ άπορων αρτον. 
' Codd. 55? 254? Vienna 2 κρείσσων εργαζόμενος εν πόνοις η περίπατων 

δοζαζόμενος κα\ άπορων άρτων. 
Cod. 3^7 κρείσσον εργαζόμενος εν πάσιν η περίπατων εργαζόμενος και 

άπορων άρτων. 

Latin : ' melior est qui operatur et abundat in omnibus quam qui 
gloriatur et eget pane.' 

Syriac : Pesch. : ' better is one who works and abounds in riches, 
than one who boasts and wants food.' 

The Latin and Syriac show that Codd. 23, 248 have preserved 
the original text. The variants from that text may probably be 
accounted for thus : — the earliest variant may have been that which 
is found in Cod. A, and which added περίπατων as a gloss to δοξαζό- 
μένος : a later scribe finding ή περίπατων in some copies took it to 
be a correction for και περισσεύων, and omitted the latter [hence 
Cod. B], and since εν πάσιν was difficult to explain after εργαζόμενος 
it was altered to εν πόνοις [so Cod. 55] : a later scribe restored και 
περισσεύων but retained the η [so Cod. S^] which was further cor- 
rected by omitting the ή, and placing the restored κα\ περισσεύων 
after instead of before εν πάσιν [so Cod. S^]. 

xi. 9. 

Codd. ACS, 23, 248, 296, 307, Vienna I περί πράγματος ου ουκ 

ran σοι μ^ ?ριζε : Codd. Β, (55), (ιο6), ΐ55, (ι57), (254), (3θ8), 

Vienna 2 .... ου ούκ εστί σοι χρεία 

Latin : ' de ea re quae te non molestat ne certeris : ' [but the 

original scribe of Cod. Tolet. omitted ' re.'] 
Syriac : Pesch. ' if it be in thy power do not contend : ' Syr. -Hex. 
' about a matter which is not a trouble to thee do not contend.' 
It seems probable that the MSS. from which χρεΊα is absent pre- 
serve the original reading, and that ου is to be explained as an or- 
dinary instance of inverse attraction. If ερΊζειν be used here in its 
sense of a legal contest, the meaning will be ' contend not (at law) 
about a matter which is not thine.' 

xii. 12. 

The following is the text of Cod. A : — μη στηστ]ς αύτον παρά σεαυτω 
μη άναστρε-ψας σ€ στ^ επϊ τον τόπον σου' μη καθιστάς αυτόν εκ δεξιών 
'σου μήτΓΟτε ζήτηση την καθεΒραν σου. 


The variants on this text are Codd. B, 23, 106, 155, 308 παρά 

aeavTov : Cod. 106 omitS μη άναστρ^ψας .... τόπον σον : Codd. 
BC, 55., 253, (254), 296, (307) avarp^^as: Codd. 23, 248* 
καταστρίψας {2 4S μη ποτ€ κ.): Cod. 155 ^Ο'τητζζσβ arfj : Cod. 
253 ί*'α μη avaTpeyj/as (Ις τον τόπον σον orrj : Codd. 296, 308 eVl 
τον τόπον σον: Codd. 1 06, 248 add ΧαβΐΊν after καθίδραν σον. 
Latin : (see below). 
Syriac : Pesch. ' set him not near thee, 

lest, turning round, he stand i"n thy place : 
set him not at thy right hand, 
lest he desire to take thy seat.' 

It is obvious that the two pairs of phrases are in effect duplicates 
of each other : but it is not clear whether or not the duplication 
be intended by the writer. The Greek of all MSS. except Cod. 
106, and also the Syriac, would be quite intelligible on the hypo- 
thesis of an intentional duplication : and some analogies could be 
found for it elsewhere in the book. 

But the Latin suggests the hypothesis that one of the two pairs 
of phrases is a gloss of the other, since it arranges them in the 
order in which they would occur if a gloss had been incorporated 
into the text. 

The earliest text is probably that of S. August. Speculum^ p. 130, 
which agrees with Codd. Amiat., S. Germ., S. Theod. : (the sup- 
posed glosses are here printed in italics) : 
' non statuat ilium penes te 
nee sedeat ad dexieram tuam 
ne conversus stet in loco tuo 
ne forte conversus in locum iuum inquirat caihedram iuam! 

The Toledo MS. has— 

' non statuas ilium penes te in loco tuo 
. nee sedeat ad dexteram tuam 
ne forte conversus in locum tuum inquirat cathedram tuam^ 

The later MSS. and the Vulgate are based upon this, and 
have — 

' non statuas ilium penes te in loco tuo 
nee sedeat ad dexteram tuam 
ne forte conversus in locum tuum inquirat cathedram tuam.' 

If the words printed in italics be omitted from the oldest of 


these texts, the remainder will suggest that the original Greek 
text was — 

μη στηστ]5 αντον πάρα σ^αυτω 
μη άνατρίψας σ€ στγι inX τον τόπον σον. 
The only important variants in the Greek are άναστρ^^ας and 
άνατρ€ψας : the uniform translation ' conversus ' in all the Latin 
MSS. indicates that the former is the older reading. It may be 
supposed that the common use of the verb in the LXX. as a neuter 
was unknown to some of the Greek scribes, and that (i) they 
added ae to it, (2) substituted άνατρί^ας for it : the interchange of 
άναστρίφω ανατρέπω is not infrequent : there is an instance of it 
below, V. 16, where Codd. S, 22, read άναστρεψαι^ Codd. AB, άνα- 


xiv. 20. 

Codd. S^, 106, 248, 253 μακάριος άνηρ 6s iv σοφία μ€λ(τησ€ί καλά 

[S^ omits καλά] : Codd. AB, (23), (55), 155, 157, (254), (296), 

308, Vienna I τίλίυτησα : Cod. 307 reXevra. 

Latin : S. August. Speculum, p. 468 ' Felix sapiens qui in sa- 
pientia sua veritatem et justitiam meditatur : ' Cod. Amtai. 
' beatus vir qui in sapientia sua morietur et qui in justitia sua 
meditatur : ' Codd. cett. and Vulg. * beatus vir qui in sapientia 
morabitur et qui in justitia sua meditabitur.' 
Syriac : Pesch. * Blessed is the man who thinks upon wisdom, 
and meditates upon understanding : ' Syr. -Hex. ' Blessedness 
is for the man who in wisdom meditates well.' 
The original reading was clearly μ€λ€Τ77σ€ΐ=' meditabitur:' the 
Latin duplicates ' morietur ' ' meditabitur ' show the combination of 
two Greek texts, and the antiquity of both of them : the later 
' morabitur ' is possibly an emendation of ' morietur.* 

XV. 6. 

Codd. ASS 106, 248— 

(νφροσύνην κα\ στβφανον άγαΧΚιάματος βνρήσα, 

κα\ όνομα αΙωνος [ΐθ6, Vienna Ι, αΐώνιον] κατακΚηρονομησα [ΐθ6, 
Vienna ι, κληρονομήσει, 248 adds αυτόν] 

Codd. BC, (23), (55), 155, ΐ57> 253, (254), 296, 307, 3ο8— 

€νφροσύνην κάί στβφανον άγαΚΧιάματος Γΐ55) 3^7 "yαλλtάσeωsJ 
κα\ όνομα αΐωνιον [23, Ι55> Ι57> 253 at^j/os] κατακληρονυμησει. 

Latin : ' jucunditatem et exultationem thesaurizabit super ilium, 
et nomine aeterno hereditabit ilium/ 


Syriac : Pesch. ' With joy and gladness will he fill him, and he 

will cause him to possess an everlasting name/ 
The diificulty as to ευρησ^ι is that the preceding verses seem to 
require the subject Kvpios to be continued ; hence most Greek MSS. 

omitted evprjad. 

The key to the original text is supplied by the Latin ' thesauri- 
zabit : ' the original text may be supposed to have been (reading 

άγαλλιάσβως with Codd. 1 55, 307) — 
hrh\A\iKCEOJCQ\\ChY?\CE\, i.e. άyaλ\Lάσeωs θησαυρίσει: but 

a careless scribe passed from one C to another and wrote 

^vFAAAI^CELUCAYPICCI, i.e. αγαλλίασεως ανρισ€ΐ : and since αυ 
was a not uncommon error for ευ, and ι for η, the word ανρισα 
which followed α-γαΚΧιασ^ωί was interpreted as (ίρησ^ι. 

xvi. 3. 

Codd. AS, 23, 155, (157), 248, 253, 254, 296 μη 6πεχ€ eVt τό 
πλήθος αυτών : Codd. BC, 308 .... eVi τον τόπον αυτών '. Codd. 

106, 307 omit the clause. 
The Latin ' ne respexeris in labores eorum ' points to a reading 
κόπον or πόνον: but the context makes τ6 πλήθος almost certain, 
since the following clause is κρείσσων γαρ eh η χίλιοι. 

xvi. 17. 

Codd. AS, 23, 106, 155, 157, 248, 253, 307 μη etnrjS 6τι [248 
omits], από κυρίου άποκρυβησομαι, κα\ εξ ΰψους [S^ υψίστου j τις μου 
μνησθησεται ; Codd. BC, 55j (254)? 296, {308) μη e| ύψους .... 

Latin : ' non dicas a deo [Cod. Tolet. ' ab eo '] abscondar, et ex 

summo quis mei memorabitur ? ' 
Syriac : Pesch. ' Say not, I shall be hidden from the sight of the 

Lord, and in the height of heaven who will remember me ? ' 
The Latin and Syriac confirm the reading of Codd. AS. 

xvi. 18. 
Codd. AS, 23, 155, 157, 253, 254, 296, 307, Vienna i 

ίδού ό [l55 Ol^litS o] ουρανός καΐ δ ουρανός του ουρανού 
άβυσσος κα\ yrj [S, 2g6 η γτβ iv τη επισκοπή αυτοΰ σαΚευθησονται 
[23, 253 σαλεύοι^αί, 155 σαλευθησεταϊ^ 

Codd. Β, (55), (3ο8)— 

έδού 6 ουρανός κα\ 6 ουρανός του ουρανού του θεοΰ, 
άβυσσος κα\ yrj σαλευθησονται εν τη επισκοπή αυτοΰ. 


Cod. 106 — 

Ιδού 6 ovpavos τον ουρανού 

άβυσσος και γη καΐ τα. iv αντόΐς iv τη ΐττισκοπη αντον σαΚΐνθησονται. 
So Cod. 248, except that κα\ 6 ovpavos is retained. 
Latin : ' Ecce caelum et caeli caelorum, abyssus et universa 

terra, et quae in eis sunt in conspectu illius commovebuntur ' 

[in Cod. Tolet. * commovebuntur ' is added by a later hand]. 
Syriac : Pesch. ' Behold the heaven and the heaven of heavens, 

the deep, and the earth, stand by his manifestation upon 

them : ' Syr, -Hex. ' . . . . are trembUng at his visitation of 


It is probable that τον 6fov has come into the Greek text as an 
alternative translation of an original Hebrew ^^, as in Is. 14. 13. 
But the insertion seems to make του θ^οϋ a predicate, ' the heaven 
and the heaven of heaven is God's : ' which destroys the parallelism 
with the following verse. 

xvii. 27. 

Codd. ACS, 106, 155, 157, 248, 296, 307 avTi ζώντων κα\ διδόντων 
άνθομολόγησιν : Codd. Β, (23), (55). (253). (254), (3θ8) άντΙ 
ζώντων καϊ ζώντων καΐ διδόντων άνθομολόγησιν. Latin : ' CUm 

vivis et dantibus confessionem Deo.' 
It is only an inference from the silence of the collators to 
suppose that any MS. supports Β in the addition καΐ ζώντων : the 
addition is most like only the error of a scribe who wrote the 
words for καϊ διδόντων, and afterwise corrected them. But the fact 
of the words occurring, if they do occur, in other MSS. would be 
an important contribution to the genealogy of those MSS. 

xviii. 32. 
Codd. ACS, 155, 157, 248, 254— 

μη ίνφραίνου eVi πόλΧτ} τρυφη [248 adds σου] 

μη [Codd. C, (157), 248, 254, Vienna i, μηδβ, Cod. 155 κα\ 

μηδί\ προσδεηθηε συμβολί} [248 συμβουλής, Vienna I συμ- 
βουλή^ αντης. 

Cod. Β, (55), (253), 307 /^'/δε [307 μ] προσδ^θης. 

Cod. 106 μηδΐ συνδίθης .... 

Cod. 23 καΙ (υφραίνου .... και προσδίθης. 

Latin : Codd. Am. Corb. 

' ne oblecteris in turbis nee inmodicis, 
ad duas est enim commissio illorum : ' 


Cod. Tol. 

' ne oblecteris in turbis nee inmodieis deleeteris, 
ad duas est enim commissio illorum : ' 
S. August. Specul. 134-5 

'ne oblecteris in turbis 
nee inmodieis deleeteris : * 
Codd. cett., and Vulg. 

' ne oblecteris in turbis nee inmodieis : 
assidua enim est commissio eorum.' 
Syriac : Pesch. ' Delight not in a multitude of delights, lest at 
length thou become poor : ' Syr. -Hex. ' Delight not in a multi- 
tude of delights, and do not tie thyself to a portion of them.' 
The Latin ' commissio ' (probably = ' comissatio/ for which 
* comissa' is found, cf. Ducange s. v.) points to συμβολή having been 
in the nominative case in the text which it translated. Assidua also 
points to the possibility of the difficult variants προσδεθης, προσδ€ηθί}ί 
being the representatives of a lost adjective. But there is no apparent 
clue to the original reading. 

xix. 22. 
Codd. ACS^, 106, 155, (157), 254, 308 Koi ουκ €στι βονλη αμαρ- 
τωλών φρόνησις \ Codd. Β, (23), (55); (248), (253)» (296) f^ca 
ουκ €στιν οπού βουλή αμαρτωλών φρόνησις. 

Latin : ' et non est cogitatus peccatorum prudentia.' 
The use of the classical ουκ εστίν οπού { = ουδαμου) in Cod. Β, 
which is possibly not supported by any other MS., is improbable. 

xxi. 17. 
Codd. ACS, 23, 155? 157? 253, 254 στόμα φρονίμου ζητηθησ^ται iv 
εκκλησία, καΐ τους λόγους αυτού διανοηθησοναι iv καρδία : Cod. Β, 

(ιο6), (248), (296) . . . διανοηθήσεται. Latin: 'verba ejus 
cogitabunt in cordibus suis.' 
The singular διανοηθησεται is unintelligible on account of the 

accusative τους λόγους : the subject of the plural διανοηθησονται is 

clearly implied in the preceding clause. 

xxii. 27. 

Codd. AS, 155? 296, 308 eVi τών χζίλεων μου σφραγίδα πανούργων : 

Codd. BC, (23), (55), (106), 157, (248), (253), (254) .... 


Latin : ' super labia mea signaculum certum.' 


It is probable that navovpyov is correct : it is found in a good 
sense elsewhere in the book, =* clever/ e.g. 6. 32 : but a doubt 
arises from the fact that it is always used in the LXX. of persons 
and not of things: hence possibly here σφρ. πανούργων =^ ά seal of 
clever men/ i. e. cunningly devised : cf. βούλας πανούργων Job 

xxiii. 10. 

Codd. AS, 55, 157, 254 ό ομνύων κα\ [Codd. AS και ο] ονομάζων 
δια παντός το όνομα κυρίου άπο αμαρτίας ου μη καθαρισθί) ; Codd. 
BC, 23, (106), Ι55> (248), (253) omit το όνομα κυρίου. 

Latin : ' omnis jurans et nominans in toto a peccato non purga- 

Syriac : Pesch. ' Whoever swears on any (slight) occasion, it is 
an abominable thing, nor will he be guiltless : ' Syr. -Hex. * He 
who swears, and names Him, on any (slight) occasion will 
not be guiltless.' 
The antithetical clause οΙκ€της ζξ^ταζόμ^νος seems to require a 
single participle here : and the variants are best explained by the 
hypothesis that ό ονομάζων το όνομα κυρίου was added in early times 
as a gloss of 6 ομνύων : the phrase apparently comes from Lev. 
24. 16, and the separation of it into two parts by the insertion of 
δια παντός probably accounts for the loss of the words to όνομα 
κυρίου in most MSS., including those from which the Latin transla- 
tion was made. 

xxiv. 17. 

Codd. AS, 23, 55, 106, 155, (157), 248, 253, 254, 296 βγώ ώ? 

αμπβλος (βλάστησα χάριν (248 ευωδίαν) : Codd. BC, (s^^) 

Latin: 'ego quasi vitis fructificavi suavitatem [Cod. Amiat. 'in 

suavitate '] odoris.' 
Syriac : Pesch., Syr. -Hex. ' I am like unto a vine of fairest beauty.' 
The Latin is remarkable as supporting not only Codd. AS, cett. 
against BC, but also the reading εύωΒίαν of Cod. 248 against 
all the other MSS. 

XXV. 15. 

Codd. A, Vienna 2 συνοικησαι [Cod. A συι/οικτ^σ?] λίοντι κα\ δράκοντι 
ξύδόκησβ, η συνοικησαι μζτα γυναικός πονηρας : Codd. BCS\ 253 
συνοικησαι λίοντι και δράκοντι [253 δράκοντι κα\ Xeoj/riJ (υδοκησω η 
(ροικησαι μ€τα γυναικός πονηράς: Codd. S^, 23, 55^ ^55? 296, 


Vienna I . . . evdoKrjaai η συνοικησαι . . .: Codd. ΐοό, 254 . . . 
(Ιδοκησαι η οίκησ-αι . . . : Cod. 248 .. . ευδοκώ η συροίκησαι: Vienna 
2 . . . €ν8όκησ€ η συνοικησαι . . . 

Latin : ' commorari leoni et draconi placebit quam habitare cum 

muliere nequam.' 
Syriac : Syr. -Hex. ' I prefer to live with a serpent and with a lion, 

than to dwell in the house with a wicked woman.' 

The Syriac supports the personal ευδοκήσω or ευδοκώ against the 
impersonal (νΒόκησβ, and the Latin supports the future (ν8υκησω 
against the present evdoKOo. It seems probable that the reading 
evSoKTjaai has arisen from the influence of the following hoLKrjaai, and 
that the impersonal €νδόκησ€ of Cod. A is only a scribe's error for 
fvboKTjaai. It is probable that ^νοικησαι is correct rather than σννοικησαι 
in the second clause, because the meaning of the former • to live in 
the house ' is more suitable to the passage than the meaning of the 
latter, which in relation to a woman is almost always ' to cohabit.' 

XXV. 17. 
Codd. AS, 23, 55, 106, 155, 157, 248, 253, 254, 296, Vienna i, 2 

(πονηρία yvvaiKosj σκοτοί το πρόσωπον [254, 308 την ορασιιΛ αυτής 
ως αρκος : Codd. BC, (308) . ... ως σάκκον. 

Latin: 'obcaecat [obcaecavit, obcaecabit] vultum suum tanquam 

ursus, et quasi saccum ostendit/ 
Syriac : Pesc/i., Syr. -Hex. ' it makes her face dark as the colour 
of sackcloth.' 

The Latin shows the antiquity of both the Greek readings, 

αρκος and σάκκον. 

αρκος {=αρκτος) is Unintelligible: it can hardly be doubted that 
the original reading was αρκυς in the sense of a net for the hair : so 
Hesychius αρκυς' γυναικύον κ€κρνφαλον. For headdresses of this 
kind, see Baumeister, Denkmdler des klassischen Alter turns, fig. 81 
(a Pompeian wall-picture, from Mus. Borhon. vi. 18) and fig. 392 
(a Herculanean picture from Antic, di Ercol. i. 79). 

σάκκον has probably the same sense as αρκυς: it was a cloth 
like that of the terra-cotta which is pictured in Baumeister, fig. 850 
(from Stackelberg's Grader der Helleneri). The neuter form of the 
word does not occur elsewhere. 

It may be conjectured that each of the two words αρκυς and σάκκον 
(σάκκος) had a local or restricted use, and the one was substituted 


for the other by the scribe of a different locaHty. The Latin trans- 
lator, finding the corrupt reading αρκος translated it ' ursus/ and 
not understanding σάκκον, but taking it for an accusative, con- 
structed the new clause ' et quasi saccum ostendit.' 

The meaning of the passage, whether apKvs or σάκκον be read, is 
' the wickedness of a woman changes her appearance, and darkens 
her countenance as when a wimple is drawn over it.' 

XXV. 21. 

Codd. AS, 106, 155, (157)5 3°^ γυναίκα iv KoKkei μη €πίποθηστ}ς : 
Codd. 55j 254? 296 γυναϊκα iv κάλλβι μη €πιθυμήστ]ς I Codd. BC, 
(23), (253) γυναίκα μη €πιποθηστ]9 : Cod. 248 γυναϊκα μη (πιποθησης 
fis τρυφην. 

Latin : ' non concupiscas mulierem in specie.' 

The first clause of the verse, μη προσπ€στ)ς em κάλλος γυναικός, is 
inadequately balanced by the reading of Codd. BC, and although 
the reading of the majority of MSS. iv κάλ\€ί is supported by the 
Latin, 'in specie,' yet it is too nearly a repetition of eVl κάλλος to be 
quite satisfactory. Hence there is a probability that the true reading 
is preserved in Cod. 248 eh τρυφην, in the sense of the Latin 
' luxuria.' 

XXV. 25. 

Codd. AS, 23, 106, 155, (157), 253, 254 {μη 8ως) . . . μη^e 
γυναικί πονηρά παρρησίαν : Codd. BC, (55)» 296, 3^^ • • • H-V^^ 
γυναικί πονηρά €ξουσιαν : Cod. 248 .. . παρρησίαν εξόδου. 
Latin : ' nee mulieri nequam veniam prodeundi.' 
Syriac : Syr. -Hex. ' nor to a wicked woman liberty.' 
The antithetical clause μη δως ΰδατι dU^odov seems to favour the 
reading παρρησίαν in the sense of 'freedom of speech,' in which 
sense it is used in Job 27. 10, Prov. i. 20. But the Latin shows 
that iξoυσίav, in the sense of ' liberty to go out of doors/ was 
an early variant, to which e^obov was probably added as a gloss. 

xxvi. 5. 

Codd. AS^, 55, 106, 155, 157, 248, 253, 296 eVi τω τΐτάρτω 
προσωπω €φοβηθην : Codd. BC, (23), (254) . . . ibeηθηv. Latin : 

' et in quarto facies mea metuit.' 
The variation of reading is probably due to the unusual con- 
struction of φοβεϊσθαι with iπί: but ebeηθηv gives no intelligible 


sense. The Latin connects προσώπω (ψοβηθην, * I was afraid in 

xxvii. 27. 

Codd. AS^ 55, 106, 155, 157, 253, 254, 296, 307, 308 6 ποιων 
πονηρά etff αυτά κυΚισθησ^ται [lo6, 254 βγ^υλίσ^ι^σεταί] : Codd. 
Β, (23) . . . us αυτόν κυ\ισθησ€ται : Cod. 248 ποιουντι πονηρά. 
eV αυτόν κν\ισ6ησ(ται. 

Latin: S. Aug. Speculum, p. 142, Cod. S. Theod. ' facienti 
nequissimum consilium super ilium devolvetur: ' Codd. Tolet. 
Amiat. ' facienti nequissimum super ipsum devolvetur.' 
Syriac : Pesch., Syr. -Hex. ' he who devises evil will fall into it.' 
The most noteworthy point is the agreement of the Latin with 
Cod. 248 in the possible but harsh construction 'to him that doeth 
mischief, it will roll upon him :' the reading of Cod. Β is gram- 
matically impossible, but critically interesting because it preserves 
in αυτόν the middle link between the reading of Cod. 248 and that 
of the majority of MSS., i. e. it may be supposed that when the 
dative ποιουντι was changed into the nominative, αυτόν was in some 
cases retained by an unintelligent scribe from an earlier MS. 

xxviii. I. 

Codd. ABCS, 68, 157, 253, 296, 307, Vienna I ray αμαρτίας 
αυτού (157, 253 αυτών) διαστηριων διαστηρίξΐ : Codd. 23, (106), 
(248), 254? Vienna 2 τάς αμαρτίας αυτού (254 αυτών) διατηρών 
διατηρήσει: Cod. 55 '""^ αμαρτίας αυτών διατηρήσει: Cod. 1 55 
^ιατηριών 8ιατηρίσ€ΐ : Cod. 308 (apparently) διαστηριών διατηρήσει. 

Latin : ' et peccata illius servans servabit.' 

Syriac : Fesc/i., Syr. -Hex. ' for all his sins will be carefully pre- 
served for him,' i. e. for God. 

The reading διατηρών διατηρήσει is Confirmed not only by the 
versions but also by the context. The purport of the context is 
evidently that a man should not avenge himself upon one who has 
wronged him, but wait for the vengeance of God. The Pauline 
* I will recompense, saith the Lord ' is here expressed as ' their 
sins he will surely keep (in remembrance).' In the reading διαστηριών 
διαστηριεί there is (i) the grammatical difficulty that the use of the 
participle in the future would probably be without a parallel, 
(2) that the meaning ' their sins he will surely confirm ' is not 
relevant to the context. 


xxix. 4. 
Codd. AS, 23, 106, 155, 157, 248, 253, 296, 307 πάρζσ-χον κόπον 
[307 Koknovj Tois βοηθησασιν αύτοίς : Codd. BC, (55/5 (2 54)» 

(308) . . . πόνον. Latin : ' praestiterunt moiestiam his qui se 
adiuvaverunt (adiuverunt). ' 
κόπος and πόνος are similarly interchanged elsewhere, e. g. Job 3. 
10; Ps. 9, 35 (10. 14): 54 (55). 10, II ; Wisd. 10. 10. 

xxix. 7. 
(i) Codd. AS^ 55, 155, 157, 248, 254, 296, Vienna i πολλοί 

ουν χάριν πονηρίας άπ€στρ€ψαν (Codd. 55) 106, Ι57> ^54 ^^^ 
χεφα, 248 adds τον ανθρωπον) : Codd. S^ 23, 253^ 3^7 ""^^^^^ 
ου χάριν πονηρίας άπ€στρζ'\Ιταν ΐ Cod. Β, (308) πολλο) χάριν πονηρίας 
άπ€στρ€\Ιταν : Cod. Ι06 ττολλοί χάριν πονηρίας άπ€στρ€ψαν χ^'ιρα. 
(2) Codd. ABS, 106, 155, 157» 254, 296, (307), 308 άποστ€ρη- 
θηναι 8ωρ(άν (νλαβηθησαν : Codd. 23, ^^^, 248, 2 53) ^ί^^Ι^^ 2 
άποστερηθηναι δε . . . : Cod. 248 omitS δωρβα'ι/. 

Ladn : ^ multi non causa nequitiae non fenerati sunt sed fraudari 

gratis timuerunt.' 
Syriac : Pesch. ' many turn away from lending, by no means 
on account of wickedness, but because they are afraid of 
an empty quarrel : ' Syr. -Hex. (the last clause) ' . . . but they 
shall be deprived because they feared without cause.' 
In the first clause it is possible that both ουν and ου may be 
correct. The latter word is required by the whoh structure of the 
passage, and is supported both by good Greek MSS. and by the 
versions. The former is possible, because the verse is of the nature 
of an inference from v. 6. 

The verb απόστρεψαν requires an object, and the analogy of v. 9 
leads us to expect a personal object : hence the τον ανθρωπον of 
Cod. 248 seems preferable to the χ^φα of other MSS. 

In the second clause 6e is clearly necessary, and the retention of 
it in Cod. 248 shows that that MS. is based upon one which read 
ου in the first clause. 

xxix. 13. 

Codd. AS, 23, 55, 106, 155, 157, 254, 296, 307, 308 υπίρ άσπί8α 
κράτους (l57 κράνους) κα\ υπέρ [55 Omits] δόρυ όλκης : Codd.BC, 
(248), (253) • • • ^^Φ ^^Ρ^ άλκης. 



Latin : ' super scutum potentis et super lanceam.' 

Syriac : Pesch. ' a strong shield, and a spear, and a wall will it 

be for war.' 
The reading ολκής is not only better attested, but is also a more 
common word in later prose and Hellenistic Greek than the 
poetical άλκψ: 'it (sc. almsgiving) will fight for him in the face 
of the enemy better than a strong shield or a heavy spear/ 

XXX. II, 12, 13. 

Cod. 248 

I μη δως αυτω i^ovaiav iv ν^ότητι 
και μη παρ'ώΎ]ς τας αγνοίας αντου 
κάμψον τον τράχηλον αυτού iv ν€Οτητι 
κα\ θλάσον τας πλζνρας αντυΰ ως Ζστι νηπιος 

5 μη 7Γοτ€ σκληρυνθείς άπαθηση σοι' 
κα\ Ζσται σοι οδύνη ψνχης. 
παίδβυσον τον νΐόν σου κα\ 'έργασαι iv αυτώ 
ινα μη iv τη άσχημοσύνη αυτοΰ προσκόψτ)ς. 

Codd. ABCS, 23, 55, 68, 155, ι57, 253, 296, 308 omit νν. 2, 3, 
6 : Cod. 106 omits νν. 2, 3 ' Cod. (254) places νν. 2, 3 after 
V. 8. 

The variants are: v. i, Cod. 307 δός: v. 4, Codd. A, 106, 155 

€ως iστί: V. 5, Codd. ACS, 157, 307, 308 άπ€ΐθησ€ΐ, Cod. 155 

€πιθησ€ΐ : v. 6, Cod. 106 adds e^ αυτού after σοι : V. 7, Cod. C 

has ως €στι νηπιος for ipyaaai iv αυτω '. V. 8, Cod. 296 ... eV τΐ] 
αισχύνη αυτοΰ ττροσκό-^ης, Cod. 55 • • • ^^ "^Ό αίο'χημοσύνη σου 
προσκόψτ/ς^ Cod. 3^^ . . . iv τη αισχημοσύνη σου προσκόψη. 

Latin: 'ηοη des ilH potestatem in juventute 
et ne despicias cogitatus illius : 
curva cervicem ejus in juventute 
et tunde latera illius dum infans est, 
ne forte induret et non credat tibi 
et erit tibi dolor animi : 
doce filium tuum et operare in ilium 
ne in turpitudinem illius oifendas.' 
Syriac: Syr. -Hex. 

'Give him not power in his youth. 
Nor forgive him all his transgressions : 
Keep low his heart while he is young, 
And break his back while he is little : 


Lest when he is grown strong he rebel against thee. 
Teach thy son grief of mind, 
And show thyself rough towards him : 
Lest he cause thee to stumble by his foolishness.' 
Both the Latin and the Syriac confirm the general reading of 
Cod. 248 against all the other MSS. But the original of the Syriac 
translation of vv. 6, 7 was evidently different from any Greek text 
which has survived. 

XXX. 39 (xxxiii. 31). 
Codd. ACS, 23, 55, 157, 253, 254, 296, 307, Vienna 2 

cl eaTL σοι οίκβτης έστω cos συ 

δνι iv αίματι €κτησω αυτόν' 

€1 [S^ om.] €στί σοι οΙκ€τη5 aye αυτόν ως αδίλφόΐ', 

ΟΤΙ ως η ψνχη (του ίττώ^ησ^ις αντ^. 

Codd. Β, (3θ8) 

€ί eWt σοι οίκζ'της έστω ως συ 

ΟΤΙ iv αιματι ΐκτησω αυτόν' 

et €στι σοι οϊκετης aye αυτόν ως σ^αντόν, 

ΟΤΙ ως ή ψνχη σου €πώ€ησ€ΐς αυτω. 

Cod. 106 

61 €στι σοι οΐκίτης [marg. add. πιστοί] έστω ως συ 

ΟΤΙ iv αιματι iκτησω αυτόν' 

aye αυτόν ως άδβΧφόν, 

ΟΤΙ ως η ψυ\η σου iπώeησeις αύτώ. 

Cod. 155 

el €στιν σοι οίκ^της ayaye αυτον ως άbe\φόv, 
οτι ως η ψυχή σου βττιδβϊ/σίί αυτω. 

Cod. 248 

et eari σοι οΐκίτης, Ζστω σοι ως ή ψυχή σου 
οτι iv αιματι iκτήσω αυτόν' 
el ίστι σοι οΐκίτης aye αυτον ως άδεΧφον 
οτι ως ή ψνχη σου iπιbeήσeις αυτω. 
Latin : 

' Si est tibi servus fidelis, sit tibi quasi anima tua : 
quasi [Cod. Tol. ' et sicut '] fratrem sic eum tracta, 
quoniam in sanguine animae comparasti eum.' 
[Cod. Tol. * . . . animae tuae': * parasti ' in the margin.] 
Syriac: Pesch. 

' If thou hast one bond-servant, let him be to thee as thyself. 
Because like thyself will be the loss : 
Τ 2 


If thou hast one bond-servant, treat him as thy brother ; 

Fight not against the blood of thy soul.' 
The passage is one of the most difificult in the book : it seems 
evident, both from the Greek MSS. and from the Latin, that part 
of it has been duplicated. The key to the diversities of the Greek 
MSS. seems to be afforded by the Latin, which makes it probable 
(i) that 61 eavL σοι oiVeV»;? should be read only once (as in Codd. 106, 
155) • (2) that ως η ψνχη σον is an epexegesis, or the original form, 
of ως σν \ (3) that άδίλψον is the correct reading, if the whole clause 

aye avrov cos αδελφοί/ be not an added paraphrase of έστω ως σν (ως η 
ψνχη σον). 

It seems also probable that the unintelligible clause δη ως ή ψνχη 

σον €πώ€ησείς αντω veils a paraphrase of ev αίματι βκτησω αυτόν. 

xxxii. 22. 
Codd. ASS 55ί Io6, 155, 157, 253, 254, 307 κα\ κρινύ δικαίοις 
κα\ ποιήσει κρίσιν : Codd. Β, (23), (296) . . • . δικαίως'. Cod. 248 
.... δίκαιον ς. 

Latin : ' sed judicabit justos et faciet justitiam.' 
The context clearly requires δικαίοις : cf. Is. 11. 4 κρινεΙ ταπεινω 

xxxvi. (xxxiii.) 3. 

Codd. AS, 23, 55, 106, 155, (157), 248, 253, 254, 296, 307, 

308 άνθρωπος σννετος €μπιστ€νσ€ΐ νόμω κα\ δ νόμος αυτω πιστός ως 
ερώτημα δηλών [ΐθ6, 3^7 δί/λοι/, 248 δ^^^ωι^] ετοίμασον Χόγον κα\ 
όντως άκονσβηστ] : Codd. BC . . . . ώς ερώτημα δίκαιων [accent 

Latin : 'homo sensatus credit legi dei et lex illi fidelis : qui in- 
terrogationem manifestat parabit [Cod. Amiat. ' paravit '] 
verbum et sic deprecatus exaudietur.' 
The ordinary punctuation of the passage connects ώς ερώτημα 
δηλών with the preceding words : and it is possible that this punc- 
tuation is anterior to Cod. B, and accounts for the reading δικαίων 
(if δικαίων and not δίκαιων be intended). 

But the Latin helps to make it probable that the clauses properly 
run as follows : — 

άνθρωπος σννετος εμπιστενσει νόμω, 

κα\ νόμος αυτω πιστό?* 

ώς ερώτημα δηλών, ετοίμασον λόγον, 

και όντως .... [? = ' dcprecatUS '] άκονσθηστι. 


* A man of understanding will put his trust in the law, 
And the law will be to him trustworthy : 
Fashion thy speech, as one who states a question 
And so ... . shalt thou be listened to.' 

The use of ερώτημα in the philosophical sense of a formal ques- 
tion or problem is not out of harmony with the character of the 

xxxvi. 18. 

Codd. AS, 55j ^55) 253, 254 πάλιν ά-γιάσματός σου .... τόπον 
καταπανματός σον : Codd. 23, 106, 1 57? 248, 296, 3^7 '^όλιν 
άγιάσματόί σου .... τόπον καταπαύσβώς σου Ι Codd. Β πόλιν 
αγιάσματος σου .... πόλιν καταπανματός σου. The Latin sup- 
ports Cod. Β : * civitati sanctificationis tuae .... civitati re- 
quiei tuae.* 

xxxvi. 22. 

Codd. AS, 155 ^Ισάκουσορ Kvpie 8€ησ€ως των οΙκ€τών σον'. Codd. 

BC, 23, 55, (106), (157), (248), (253), (254), (296), (307), 
(308) .... ίκίτων σον. The Latin supports Codd AS : ' exaudi 
orationes servorum tuorum:' but in Ps. 73 (74). 23 Cod. S 
agrees with Cod. Β in reading ίκ€των : (Cod. A is there defi- 
cient : and neither word is a correct translation of the Hebrew 

xxxvi. 31 (28). ' 

Codd. AS, 23, 55, 157, 253, 254, 296, 307 τίς yap ττιστβύσβι 
€νζώνω ληστ^ άφαλλομ^νω εκ πόλίως eh πόλιν [296 πεδίον: SO 
308] : Codd. BC .... σφαλλομενω . . . .: Codd. 106, I55j 
248 .... εφαλλομενω .... 

Latin : ' . . . . quasi succinctus lateo exsiliens de civitate in civi- 

Syriac : Pesch. ' who would trust a youth like a goat leaping 

from city to city ? ' 

The Syriac appears to supply the missing element in the meta- 
phor : the wifeless and homeless man, wandering from city to city 
is like a goat leaping from rock to rock. 

xxxviii. 27. 
Codd. AS, 55, 106, 155, (157), 253, 296, 307 Kdi 17 [55, 106 


omit η] €πιμονη αυτού άλλοιωσαι ποικίλίαν : Codd. BC, 23, (248), 
(254)5 {308) ' ' ' ' V νπομονη .... 

Latin : ' assiduitas ejus variat picturam.' 
The Latin confirms επίμονη, ' assiduity ' or ' perseverance ' as 
distinguished from υπομονή, ' moral endurance.' 

xxxviii. 28. 

Codd. A, (157)» 3^7 '^"^ καταμανθάνων epyov σώηρου : Codd. S, 55> 
106, 254, 308 .... epya σώηρου : Cod. 296 epyots σώηρου: 
Cod. 155 • • • • φγασ"/αΐ' σώηρου : Cod. 23 ... . ^ργω σώηρου: 
Cod. 248 eV €ργω σώηρου: Codd. BC .... άρ-γώ σώηρω: Cod. 
253 €ργου σώηρου. 

Latin : ' considerans opus ferri.' 

The reading άργω σώηρω 'unwrought iron' {apyos is used of metal 

in this sense in Joseph. B.J. 7. 8. 4 apyos re σώηροε κα\ χαλκός eTL δε 
και μόλιβδος, SO Pausan. 3• 12. 3) is in itself possible : the smith is 
sitting at the anvil and looking at the glowing unwrought mass on 
which he is about to work : but the difficulty of the use of the 
dative case with καταμανθάνων seems insuperable. If the reading of 
Cod. A, epyov σώηρου, be correct, there does not appear to be any 
adequate reason for the numerous variations : the Syriac translation 
' implements of weight ' suggests that the original reading was the 
comparatively rare word cpyaXela (σώψου), which is found only in 
Ex. 27. 19: 39. 21 (40). The picture would thus be that of a 
smith sitting at the anvil, and scanning his implements : very soon 
Kapbiav 8ώσ€ΐ els συντίλααν epyωv, ' he will give his mind to the com- 
pleting of the works.' 

xxxix. 13. 

Codd. ACS, 23, 106, (157)5 248, 253, 296, 307, 308 β\αστησατ€ 
a>s podov φυόμ^νον eVi ρεύματος lypox) : Codd. B, (55)) 1 55» (254), 
.... eVi peΰμaτos aypoO. 

Latin : ' quasi rosa plantata super rivos [Cod. Amiat. ' rivum '] 

The quotation of the passage in Clem. Alex. Paed. 2. 8, p. 216, 

ώρ ρό8ον Ίΐ€φυτ€νμίνον em. ρ€υμάτων υδάτων βλαστήσατε, is remarkable 
as giving the Greek original of the Latin, and thereby showing 
that a recension existed which does not survive in any MS. 


xlii. 5. 

Codd. ACS, 155, 1575 253> 307 Trepl διαφόρου πράσ^ως ψπόρων : 
Codd. 23, 106, 248, 254? 296 περί άδίαφόρον πράσ^ως εμπόρων'. 
Codd. Β, (55)5 (3°^) "^^Ρ*- αδιάφορου τιράσ^ως κα\ εμπόρων. 

The Latin, ' de corruptione emptionis et negotiatorum/ points 
to a reading διαφθοράς for διαφόρου : probably through a mis- 
understanding of the meaning of διαφόρου, ' purchase-money/ 

xliii. 9. 
Codd. ACS-, 55, 106, 155, (157), 248, 253, 254, 307 KOa-pos 

φωτίζων ev υψίσ-τοις κυρίου: Cod. 23 κόσμον φωτίζων iv υ-^ίστοις 
κύριος : Codd. Β, (296), (3^^) ι^όσμος φωτίζων ev υ>\τίστοις 
Latin : ' mundum illuminans in excelsis dominus.' 
It seems probable that Cod. 23 has preserved the right reading, 
and that there are four parallel clauses, each referring to the moon : 
that is to say, the moon is described as 

κάλλος ουρανού, 
δόξα άστρων, 
κόσμον φωτίζων, 
ev ύψίστοις κύριος. 

xliii. 25. 

Codd. ACS κτησις κτηνών: Cod. 248 κρίσις κητών: Codd. 106, 

157 ι^τησις κητών: Codd. 254? 3^7 '^Tto'iff κτηνών: Codd. Β, 

(23). (55), (ΐ55), (253), (296) ι^τίσις (308 πτίσις) κητών. 

The Latin, ' creatura belluarum/ makes it probable that κτίσις 

κτηνών is the true reading. But itacisms are so frequent that nothing 

certain can be determined from the Greek MSS. 

xliv. 17. 
Codd. AS^, 55, 106, 155, 157, 254, 308— 

Νώε €ΰρίθη rcXetos δίκαιος' 

ev [106, 157 και eV] καιρώ ορ-γης eyeveTO άντάλλα-γμα• 

δια τοΰτο ζγρνηθη κατάΚίίμμα Trj yrj, 

οτ€ eyeveTO κατακΧυσμός [ΐθ6, 1 55» ^57 ^ icciT.j. 

Codd. 23, 248 — 

Νώβ €υρ€θη reXeios δίκαιος' 

ev καιρώ 6pyr]s eyeveTO άvτάλ\ayμa' 

δια τούτο iyeveTO κατακλυσμός Γ 2 48 ό κατ.Ί. 


Codd. B, 253 — 

Νώβ €νρ€θη TeXetoi dUaios' 
iv καιρώ opyrjs iyeutro αντάλλαγμα' 
bia τούτο ΐγ^νηθη [253 iyiv€To\ κατάλεψμα Trf yff 
^Lo. TOVTO iyevero κατακλυσμός, 
Latin : 

'Noe inventus est perfectus Justus 
et in tempore iracundiae factus est reconciliatio/ 
Syriac : Pesch. 

' Noah was found just, a peacemaker in his time : 
At the time of the flood he was appointed a ransom for 

the world, 
And for his sake was salvation made/ 
It seems probable that ore iyev^To is the true reading, and that 
the phrase ore iyevtTo κατακλυσμός balances and explains iv καφω 
opyrjs. But it is also possible that the Latin preserves the original 

form of the passage, and that ^γενηθη κατάλειμμα Trj yfi and OTe iyevcTO 
κατακλυσμός are glosses respectively of eyei/ero αντάλλαγμα and ev καφω 
οργής : this hypothesis would account for the shortened form which 
is found in Codd. 23, 248. 

xlv. 20. 

Codd. AS, 55^ 253 άπαρχας πρωτογενημάτων €μ€ρισ€ν αντω αρτον 
πρώτοις ητοίμασεν iv πλησμοντ}. The variants οη this text are 
Cod. 248 άπαρχην, Codd. 68 αυτοίς, Cod. 23 αρτοις πρώτοις, 
Codd. 106, 157? (254) ^^ πρώτοις, Cod. S^ πρώτον γενηματος, 
Cod. Β αυτοΐς and πλησμονην, Codd. 106, 1 57 eh πλησμονην, 
Cod. 155 ττλησμονη. 

Latin : ' primitias frugum [Cod. Amiat. ' fructuum'] terrae divisit 
ilH : panem ipsis in primis paravit in satietatem.' 

Syriac : Pesch. ' he made the firstfruits of the sanctuary his in- 
heritance, and the order of the bread, for himself and for his 

The Latin suggests that the original text was .... (μβρισεν αντω, 

ItpTov αυτοΊς iv πρώτοις ητοίμασ^ν ets πλησμονην : this hypothesis will 

account for the variants of Cod. B, 23, 106, 157. 

xlvi. 15. 
Codd. ACS, 23, 55, 106, 155, 157, 248, 253, 254, 296, 308 κα\ 

iyvώσθη [l55 iπeyvώσθη'] iv ρήμασιν [23, 55, 248, 253, 254, 296 


ρηματι] αυτόν πιστός [23 πίστα, 253 TTiVrts] 6ράσ(ως [248 omitS 
πιστός όράσίω?] : Cod. Β πίστ€ΐ for ρημασιν (^ρηματι). 

Latin : * et cognitus est in verbis suis iidelis quia vidit Deum 

The Latin confirms the reading of the majority of MSS., and 
gives a remarkable gloss of δράσεως : ' his words showed that he 
was trustworthy in respect of his vision,' i.e. 'that he was to be be- 
lieved when he said that he had seen the God of light.' But the 

phrase in C. xlviii. 22 is πιστός iv όράσ€ΐ αύτον. 

Such an examination as the preceding, since it is limited 
to a small number of passages, does not warrant a final in- 
duction. But inasmuch as the passages have not been 
chosen with a view to support any previously formed 
opinion, they may be taken as typical, and consequently 
as both suggesting provisional results and indicating the 
lines which further research may profitably pursue. 

The points which will probably be most generally allowed 
to be established by the preceding examination are these : 

(i) The great value of the versions in regard to the 
restoration of the text. The glosses and double versions 
which they embody frequently point to readings which 
have not survived in any Greek MS., but which carry with 
them a clear conviction of their truth. 

{2) The inferior value of some of the more famous uncial 
MSS. as compared with some cursives. Of the uncial MSS. 
the Venetian MS. (H. and P. No. 23) is clearly the most 
trustworthy : whereas the Vatican MS. Β preserves in many 
cases a text which is neither probable in itself nor supported 
by other evidence. The book affords in this respect a cor- 
roboration of the opinion that the same MSS. have different 
values for different books. 

(3) The field which is open to conjectural emendation. 
There are cases in which neither MSS. nor versions have 
preserved an intelligible text : and since it is clear that the 
book has existed in more than one form, that it has passed 


through the hands of scribes who did not understand it, and 
that there was no such reverence for it as would preserve 
its text from corruption, the same process may legitimately 
be applied to it which is applied to the fragments of Greek 
philosophers. In some cases such conjectures have a degree 
of probability which closely approximates to certainty. 


Passages treated at length or explained are marked with an asterisk 
after the page. 


1. I, 2 

1. 10 
1. 24 
1. 26 
1. 27 

1. 3o 


2. I 
2. 16 
2. 19 


6. 17 

7. 22 


8. 21 , 

9. 25 , 
9. 27 

12. I, ; 


12. 9 

13. 6 
13. 9 

13. 16 . . 

14. 14 (17 

14. 2 

15. 5, 6 
15. 13, 
18. 1-3 
18. 10 . 
18. 14. 
18. 20, 23 



8, etc 



























18. 27 . . 

19. 17, 22 

19. 19. . 

20. 6 . . 

21. 10 . . 

22. I, 2, II. 

22.3,4 . 
22. 6 . . 
22.7,8 . 
22. 16, 17 
22. 17 . . 

2. 6 . , 
2. 13, 14 
2. 19 

2. 22 

3. 2 
3. 16 
3. 19 
3. 24 

5. 23 

6. 2-. 
6. 12 


8. 19 

9. 13 
12. 27 
12. 43, 

15. 17 

16. 4 

17. 2, 

18. 25 
20. 9 


















i6, 26 












21. 6 
21. 19 
23. 6 
23. II 

23. 16 

24. II 

25. 2 

25. 10 

26. 24 

30. 19 

34. 22 

35. 9 
35. 21 

35. 22 

36. I 
36. 2 
38. (37) 
40. 2 

i9 (35) 

. 26 

• 19 

• 75 

• 75 
. 102 

• 75 
. 28 

• 17 
. 104 
. 42 

• 63 
. 19 

104, 108 
. 19 
. 104 
. 48 
. 28 
. 104 

100, lOI 

103, 104 

. 104 

. 108 
. 42 



1. 9, etc. 

2. 6 . . 

2.15. . 

3.9 . . 

4, 2, 22 . 

5.15. . 

6. 2 (5. 22) 

17.9 . . 

19. 4 . . 

19. 10 . . 

19. II . . 

19.14. . 

19. 15 . . 

19.17. . 

23. 15 . . 


28, 90 










23. 22 . . 


25.25. . 


26. I . . 



26. 6 . . 


26. 13 . , 

• • 



5. 9 ... . 16 

5. 18, etc 

. 16 

6.4 . . 

. 16 

8. 10,11, 


25 28 

9. 6 . . 

loi, 102 

10.32. . 


12.3 . . 


14. 22 . . 


15.30. . 


21. 5 • • 


21. 5 • • 


2L 27 . . 


22. 22 . . 


22. 29 . . 


22.32. . 


23. 21 . 


24. 24 . . 


25.4 . 


31.5 . 


32. 9 . 


35. 19 . 



1. 28 . . . . 107 


103, 107 



5. 22 . 


6.5 ' 



3, 104 

6.7 . 



. 50* 

7. 2 . 

. 29 


. 78 


. 71* 


. 18 

9.5 . 

. / 

^7, 107 

10. 12 . 

. 104 

10. 16 . 

. 27 

11. 13 . 

. 104 

n. 18 . 

. 103 

12. 21 . 

. ,107 

13.3 • 

. 104 

13. 8, 9 

. 26 

13. 15 . 

. 29 

14. 26 . 

. 107 

15. 10 . 

. 106 

15. II . 

• 74 

15. 17 . 

. 19 

17. 20 . 

• ^°5^ 

19.3 • 

. 51* 

19. 14. 

. 26 

20. 8 

20. 17 

21. 8 

22. 10 
24. 13 . . 
24. 13 (15) 
24.14. • 
24. 14 (16) 
24. 15 (17) 
26. 6 
28. 60 

28. 67 

29. 3 

29. 4 

30. 2, 6 

32. 10 

33. 29 

34. 9 













2. II 
5. I 

9. 2 
9. II 
10. 28 
14. 7 
14. 12 

22. 22 

23. 14 


. 103 
. 100 
. 106 
. 63 
. 25 
. 102 
. 99 
. 18 
98, 104 
. 23 
103, 108 
• 78 

1. 12 

2. 8 

3. 10 

3. 12 


4. 12 


















2. 22 . . . . 


6.9 . . . 


8. 3 . . . 


8. 19 . . . 


8. 21 . . . 


9.17. . . 


9. 28 . . . 


15. 19 . . . 


16. 15 . • . 


16. 25 . . . 


16. 26, 30 . 


17.13. • . 


18. 16, 17 . 


18. 20 . . . 


18. 28 . . . 


19. 5 . . .2 


21. 14 . . . 


6. 19 
9. 19 , 
9. 22 

10. 12 


12. 24 

13. 20 

16. 23 

17. 16 

18. I, ; 

18. 23 

19. II, 

19. 14 

20. 4 

20. 30 

21. 4 (δ) 

22. 2 
22. 8 

22. 15 

23. 21 


24. ΙΟ 
24. 14 
24. 16 

27. I 

28. 5 
30. 22 
30. 24 















66, 70 












• 29 



66, 70 





2 Samuel. 

2. 26 

3. 21 









4.9 ... . 31 1 

11. 18 . . . . 179 1 

1. 20 ... . 19 

f). 12 . . . 


12. 9, 10 . 


2. I . . 


6. II . . . 

. 4^* 

16. 14 . . 


2. 3 . . 


7. 3 . • • 



17.10. . 


3. II . . 


7. 12 . . . 

52, 53 

19. 6, 22 . 


4.17. . 


7.13. . . 



19.37• • 


7.4 . . 


7. 13, 16, 26 . 


22. 20 . . 


7. 6 . . 


7. 13, H, 26 . 


24.14. . 


8. I . . 


7.24. . . 


8.3,5 • 


7.27. . . 


I Chronicles. 

9. 22 . . 


12. I, 3» 4 • 
12.7 .. . 
14. I . . . 


4.38. ... 167 



. 63 



14. 16 . . . 



. lOI 

1. 6 . . 

46*, 103 

15. 8 . . . 


14. 2 

• 54 


• 99 

17. 8 . . 


15. 29 

. lOI 

2.4 . . 


17. 10 . . 


16. 10 

. . 106 

2.8 . 


18. 9 . . 


16. 14 

. . 19 

2. II . 


19.5,9 • 

22, 23 

16. 23- 


192, 195* 

3. 18 . 



19. 9 . . 


16. 30 

• ^L 

4. 16. 


20. 20 . . . 


21. I 

. . 46* 



22. 2 . . 


22. 19 

ΙΟΙ, I03 

5. 20 . 

22, 23 

22. 18 . . 



• 179 

6.4 . 


22. 28 . . 


28. 2 

. . I07 

6.8 . 


22.44. • 

23, 33 

28. 12 

. . io8 

6. II . 


29. 3 

. . 252 


22, 23 

29. 18 

. . 104 

7. II . 


I Kings. 



1. 12 . . . 22, 23 

2 Chronicles. 

9.3 . 


2.4 . . 
2.44• . 
2.45. . 


6. 7, 8 . . 

6. 30, 33, 39 

7. II . . . 

. 107 

. 52 

. lOI 


9. 18. 

10. I . 

. 151 
. 106 

3. 12 . . 

4. 29 (3λ) 



7. 20 . . . 
9 I . . . 

. 66 


10. 3 . 
13. 12 . 

• 65 

7. 3 (15) 


9:23. . . 
13.7 . . . 
15. 12 . . . 

. 108 

13. 17 . 

. 219 

8.17. . 


. 106 

13. 27 . 

• 17 

8. 39, 43, 4< 

. 52 

. 104 

. 101 

. 19 

: > 

. 179 

14. 16 . 

. 217 

8. 66 . . 
9.4 . . 

9. 7 . • 
10. 24 . . 

11.14. . 

. 106 
. 107 

. 46* 

15. 15 . . . 

16. 10 . . . 
20. 24 . . . 
24. 8, 10, 11 
24. II . . . 

14. 19 . 

14. 22 . 

15. 2 . 
15.8 . 
15. 13 . 
15. 34 . 

. io6 

. 108 

• 57 
. 100 

11.37. • 

. 107 

31. 21 . . . 

. lOl 

. 222* 
. 221* 
. 221* 

17.17. . 

17. 21 . . 

18. 40 . . 

. 99 
. 104 


32. 25 . . . 
36. 22 . . . 

. 105 
. 107 

17. 1-7 
17.2 . 
17. 3, 5 

19. 17 . . 


17.4 • 

. 108 

20 (21). 5 

. 105 


17. 15 . 

• 29 

21 (20). 20 



20.5 . 

. 92 

21 (20). 39 

. 17 

21. 25 . 

. 106 


21. 27 . 

• 25 

1. ί . . . . 23 
5. 7 . . . . io8 
7. 4 ... • 32 

21. 28^33 

. 222* 

2 Kings 
2. 19 . . . 

. 78 

21.34 a 
22. .^0 . 

. 223 
• 23 

4. 16, 17 

. 158 

23. 14, 15 

. 217* 

4.27. . 
5. 7 . . 

. 106 

24. 4 . . 

• 74 



24. 13 . . 


3*, 224*^- 

6. II . . 

. loi, 103 

1.6 ... . 27 

24. 14 a, b 

• 223 

10. 20 . . 

. 107* 

1 1. 16 

. 27 

24. 14b, 


. 223* 





24. I4c-i8d 


36. 5a-i8 

a . . 239 

24. 14 c, 15 a, b 


37. 5-21 . 

238*, 239* 

24. iSb . . 


36. 5-30 . 

. • 237 

26. 5 . . . 


36. sb-io 

b . . 240 

26. 5-11 . 22: 

*, 225 

36. 6 . . 

• . 74 

26. 12, J^ . 


36. 10-12 

. . 236 

26. 14 a, b . 


36. 11, 13 

. . 240 

27. I . . . 


36.13. . 

. . 92 

27. 10 . . . 


36. 16, 20 

. . 240 

28. 11-19 • 


36. 16, 20, 

21 b 240 

28. 12 . . . 


36. 16, 21 

b . . 240 

28. 13 . . . 

. 226 

36. 18 b . 

. . 240 

28. 13-22 . 


36.19. • 

. . 240 

28. 14 . . . 

. 226 

36. 21 . . 

. . 240 

28. 14, 19 . 22 

5, 226 

36. 22-30 

. . 241 

28. 20 . . . 

. 226 

36. 31-33 

. . 242 

28. 20-22 . 


36. 34 (37 

i) . 105 

28. 21 a . . 


37. 1-8 . 

. . 242 

28.23. . . 


37. 1-18 . 

. . 237* 

28. 24-28 . 

. 218* 

37. 9-13 . 

• • 243 

28. 26, 27 . 

• ''K 

37. 14-24 

• • 244 

28. 27 . . . 


38.37. • 

. . 217 

29. I . . . 


38. 39 . • 

. . lOI 

29. 7 . . . 

. 51* 

39. 2 . . 

. . 217 

29. 10, II . 

. 218 

39. 7 . . 

. . 179 

29. 10 b, 11 a 

. 219 

29. 12 . . . 
30.15. . . 

• 74 
. 99 



31. 1-4 . . 

. 227* 

1. I . . 

. . 180 

3L6 . . . 

. 227 

1.5 . 

. . 180 

3L ^2 . . . 

• 17 

2. 2 . . 

. . 63 

32 6-19 . . 

. 228* 

3. 4 . . 

• • 17 

32. 11 . . . 

. 229* 

3.6 . 

. 176, 190 

32. II-I7. 22J 

*, 229 

5. 6 . . 

. . 4 

32. 17 . . . 

. 228 

5. 10. . 

. . 98 

32. i8b . . 

. 229 

5. lob . 

. . 210 

32. 28-33 • 

. 227 


• • 17 

33. 20 . . . 

. 107 

6.4 . . 

• . 105 

33. 27 . . . 

. 231* 

6.9 . 

• • i 

33. 27-33 . 

. 230 

7. 2 . 

. . 176 

33. 28 . . . 


7. 12 . 

. . 25 

33. 30 . . .2. 

μ. 233 

8.4 . 

• 52, 54 

33. 28-33 • 

. 231 

9.8 . 

. . 53 

34.3,4 . -K 

J2, 233 

9.13. . 

. . 74 

34.3-7 • • 

. 232 


• 74, 75 

34.6b. . . 

• 233 

9. 23 (10 

. 2) . 74 

34. 6 b, 7. . 

. 232 

9. 28 (10 

. 7) . 210 

34.8 a . .2 


9. 30 (10 

•9) • 74 

34. 10, 34 . 

. 108 

9. 31 (10 

.9) . 76* 

34. 23-33 . 2 


9. 38 (10 

• 17) . 74 

34. 28 . . . 


9. 39 (10 

. 18) . 53 

34.29. . . 

• 63 

10 (11). 2 

• • 53 

34. 30 . . . 

. 91 

10 (11). 4 

, 5 . 29 

35. 7b-ioa . 

. 234 

11 (12). 4 

a . . 205* 

35. 8-Joa . 

. 234 

11 (12) 4 

b-5 . 204 

35. 9 . . . 


11 (12). 5 

. . 75* 

35. 10 . . . 

. 19 

11 (12). 6 

74, 75, 76 

35.15,16 .2 

34» 235 

12 (13). 2 

. . 108 

36. 3 . . . 

. 19 

12 (13). 6 

. . 106 

13 (14). I, 2, 

3 . 209 

13 (14). 3 . 

. 210 

13 (14). 4 


13 (14). 6 

• 74 

15 (16). 9 . 


16 (17). 3 

. 29 

IG (17). 13 

. 23 

17 (18). 3 

• 17 

17 (18). 28 

74, 75* 

17 (18). 44 

23, 31 

17 (18). 45 

. 180 

17 (18). 49 

• 23 

18 (19). 6 

. 189 

18 (19). 15 


20 (21). 2 

loi, 103 

20 (21). 13 

• 53 

21 (22). 3 

. 190* 

21 (22). 5, 9 

. 23 

21 (22). 15 

. 106 

21 (22). 17 

27, 180, 

21 (22). 19 . 

. 180 

21 (22). 23 

. 180* 

21 (22). 25 

• 74 

21 (22). 27 

\ ^4 

22 (23). I I 

72^*, 174 

22 (23). 4 


23 (24). 2 


23 (24). 7 

. 191* 

24 (25). 9 

• 74 

24 (25). 10 

. 48 

24 (25). 14 

• 57 

24 (25). 16 

• 74 

24 (25). 17 

. 167* 

24 (25). 22 


25 (26). 2 

. 29 

25 (26). II 


26 (27). I 

. 173* 

26 (27). 3 

. 106 

27 (28). 7 

. 99,107 

28 (29). 2 

. 196 

30 (31). 2 

. 23, 49 

30 (31). 3 

. . 31 

30 (31). 6 


30 (31). II 

• 27 

30 (31). 18 I 

73**, 204 

31 (32). 1,2 

• 175 

31 (32). 5 

. 98 

31 (32). 6 

. . 27 

31 (32). 7 

• 23,31 

31 (32). 10 

. . 205 

32 (33). 5 

. . 49 

32 (33). 14 

. 52,53 

32 (33). 16 

• 31 

33 (34). 3 

. . 74 

33 (34). 4 

. . 63 

33 (34). 5 


33 (34). 6 

74, 76, 99 

33 (34). 11 

. . 17 



12 . 
1 2-1 

33 (34) 
33 (34) 
33 (34). 19 

33 (34). 23 

34 (35). 9 
34 (35). 10 
34 (35). 13 
34 (35). 14 
34 (35). 15 

34 (35). 24 

35 (36). lb 

36 (37). 4 172* 
36 (37). 7 • • 
36 (37). 10,12, 

14, 17, 18, 20, 
21,32,40; 28, 

36 (37). 10,11 . 

36 (37). II . 

36 (37). 14 • 

36 (37). 15 

36 (37). 35-37 

36 (37). 40 

37 (38). 4 
37 (38). 10 

37 (38). 21 

38 (39). 2 
38 (39). 8 

38 (39). 12 

39 (40). 6 

39 (40). 18 

40 (41). 2 
40 (41). 3 

40 (41). 9 

41 (42). 4 

41 (42). 7 

42 (43). I 

43 (44). 3 
43 (44). 12 
43 (44). 15 
43 (44). 20 
43 (44). 27 
45 (46). 5 
47 (48). 9 

47 (48). 14 

48 (49). 2 
48 (49). 8, 16 
48 (49). 9 

48 (49). 13 

49 (50). 16- 
49 (50). 21 

49 (50). 22 

50 (51). 3-19 
50 (51). 8 
50 (51). I 



103, 107 

• 17 





52 (53). 
52 (53) 
54 (55). 
54 (55). 

54 (55). 

55 (56). 

56 (57). 

60 (61). 

61 (62). 

62 (63). 

63 (64\ 

64 (65). 

64 (65). 

65 (66). 
67 (68). 

67 (68). 

68 (69). 
68 (69). 
68 (69). 
68 (69). 
68 (69). 

68 (69). 

69 (70). 

70 (71). 
70 (71). 

70 (71). 

71 (72). 
71 (72). 
71 (72). 
71 (72). 

71 (72). 

72 (73). 

73 (74). 
73 (74). 

73 (74). 

74 (75). 

75 (76). 

75 (76). 

76 (77). 

76 (77). 

77 (78). 
77 (78). 
77 (78). 
77 (78). 
77 (78). 
77 (78). 
77 (78). 

79 (80). 

80 (81). 

81 (82). 
81 (82). 
81 (82). 
83 (84). 

83 (84). 

84 (85). 

85 (86). 

85 (86). 

86 (87). 
















































. 105 

. lOI 



. 105 


. 108 

. lOI 


• 53 

• 29 

. 74 
. 252 

103. 107 
. 66 

. lOI 

. 74 

. 74 

• 75 

• 74 
. 23 

• 23 




. 190 

99, 103 

• 74 

. lOI 

. 172** 
. 25 
104, 106 

103. 108 


. 103 


• 17 


. 204 

. 172** 
• 29 


. IQ2* 

. 108 

. 173* 
. 98 

. 106 







87 (88). 15 

87 (88). 16 

88 (89). 3 
88 (86). 4 
88 (89). 15 

88 (89). 49 

89 (90). 4 

90 (91). 14 

91 (92). 5 

92 (93). 2 

92 (93). 3 

93 (94). 9 
93 (94). 19 

93 (94). 22 

94 (95). 9 

95 (96). 10 

96 (97). I 
100 (101). I 
101. iii. . 
103 (104) 
103 (104) 
105 (106) 
105 (106) 

105 (106) 

106 (107) 
106 (107) 
106 (107). 20 
106 (107). 26 

106 (107). 41 

107 (108). I. 

108 (109). 5. 
108 (109). 16 
108 (109). 22 
108 (109). 31 
111 (112). 7 

111 (112). 9 

112 (113). 6 
112 (113). 7 
113.9. . 
114 (116). 4 
117 (118). 6 
117(118). 12,22 

117 (118). 18 . 

118 (119). 10, 15 

118 (119). II . 
118 (119). 20 . 
118 (119). 28 . 
118 (119). 90 . 
118 (119). 114 . 
118 (119). 120 . 
118 (119). 122 . 
118 (119). 134 
118 (119). 143 . 
118 (119). 153 , 
123 (124). 7 . . 

126(127). 5. . 
129 (130). 6 . . 
129 (130). 8 . . 


2, 53 




!, 189 








75, 105 

• 75 

• 75 
. 75 

• 75 
. 174 

22, 31 

• 17 












130 (131). I . 

130 (131). 2. 

131 (132). 15 
132. II . . 
138 (139). lo 

138 (139). 14 

139 (140). 13 

140 (141). 5. 

142 (143). 4 

142 (143). 7 . 

145 (146). 4. 

146 (147). 3. 
146 (147). 6. 
146 (147). 10 

148. 8 . . . 

149. 4 . . . 






. 207 

99> 103, 

104, 105 




74. 76* 


1. I . 
1. 20 
L 23 
3. 12 
3. 19 

3. 34 

4. 18 

6. 21 

7. 16 

8. 20 

8. 27 

9. 18 

10. 10 

10. 28 


11. 13 
IL 17 

11. 31 


12. 20 

13. 8 . 

14. 20 
14. 21 
14. 29 
14. 33 

16. I (15 

16. 12 
16. 26 

16. 32 

17. δ- 
ι?. 23 

18. 14 

19. I, 7, 




; 103 


2, 92 
102*, 107 











100, lOI 

. 75 


20. 19 
2L 16 

21. 26 

22. 2, 7 
22. 9, 22 
22. II 
22. 16, 2 
22. 16 
22. 17 

22. 18 . 

23. 6 . 
23. II . 

23. 24 . 

24. 3 . 
24. II . 
24. 12 . 
24. 14 . . 
24. 37 (30. 

24. 77 (31. 

25. I . 
25. 5 . 
25. 25 . 

25. 28 . 

26. 7, 9 
26. II . 
26. 24 . 

26. 25 . 

27. 19. 

27. 23 . 

28. 3, 8 
28. 3, 16 
28. 6,27 
28. II . 

28. 26. 

29. II. 
29. 14. . 

29. 38 (31 

30. 4 . 


2. 14,15 

3. 19 . 

4. I . 

5.7 . 

6.8 . 

6.9 . 

7.3 . 

7.8 . 
7. 8 (9) 

10.3 • 
12.7 . 
12. 9 . 




















loi, 107 
. I04 
99, lor 

75, 90* 








. 100* 
. 29 
29, 100 
. 90 

• 75 

• 74 
. 100 

• 99 








II . . . 


1. 2 . . 
Lio. . 
1. 11-14 
1. 16-20 

1.16. . 

1. 27 . . 

2. 5, 6, 9 
2. 5 b, 6 a 
2. 8, 20 . 

2. 18, etc. 

3. 10 

5. 21 

5. 29 


6. 6 

6. 10 

7. 2 
7. 2, 4 
7. 10-17 
7. 16 

10. I 
10. 2 
10. 7 
10. 7,12 
10. 20 
14. 32 
16. I 
16. I, 2 

19. 20 

20. 6 

21. 4 
24. 7 

24. 16 

25. 4 

25. 9 

26. 6 
26. 20 

28. 17 

29. 8 
29. 13 



































208, 209* 
204, 208 





177, 205 



29. 14 . . . 


47. 14 . 

29. 19 . . . . 


48. 17,2 

30. 15 ... . 


48. 22 . 

30. 27 . . . . 


49.6 . 

3L5 . . . . 


49. 7, 26 

32. 2 . . . 



32. 6 . 92, ic 



32. 7 . . . 


49. 25 . 

32. 15 . . . 


49. 26 . 

33.13. . . 


50. 2 (3) 

33. 14 . . . 


50. 6, 7 

33. 16, 17 . 


50. 7 . 

33. 18 . . . 


50. 8, 9 

33.20. . . 


51. 10 . 

33. 22 . . . 



34. 10 . . . 



35. 4 . • .2 


52.9 . 

35. 9 . 22,31,: 


52. 14 . 

19, 20 


53. 2 

37. II, 12. . 


53.8 b 

37. 20, 35 . 



37.23. . . 



37. 32 . . . 


54. I 

38. 6 . . . 


54. 5, 8 

38. 12 . . . 



38.15. . . 


54. 11 

38.16. . . 


55.9 . 

40. 3 . . . 


56. I 

40.12. . . 



40.13. . . 


57. 16 

41.14. . • 



41.15,16 . 



4L17. . . 



41. 22 . . . 



42. 1-4 . . 

. 199* 


42.6,7 • • 

. 183 


42. 8 . . . . 

[o*, 41 


42. 12. . .4 


58. 10 

42. 22 . . . 


58. 11 

42. 25 . . . 

. lOI 

59. I 

43. I, 14 . . 


59. 7 

43. 3,11,12. 


59. 7, 8 

43.4 . . . 

. 102 

59. 16 

43. 6 . . . 

. 213 

59. 19 

43. 9 . . . 



43. 13 . . . 


60. 16 

43. 21 . . . 

. 40* 

60. 17 

44. 6 . . . 


60. 19 

44. 17, 20 . 



44.19. • • IC 

)i, 108 

61. 3 

44. 22, 23. 24 


6L 10 

45. I . . . 

. 183* 

62. 12 

45. 2 . . . 

. 183* 


45.7,9 . . 

. 153 

63. 7 

45. 17, 20, 22 



45. 20 . . . 

. 23 

63. 16 

46. 7 . . . 


63. 17 

47.4 .. . 


64. 11 






















22, 32 






















22, 32 



65. I, 2, 3 a 

65. 14. 

66. I . 
66. 2 . 
66.3 . 
66. 14 . 
66. 19 . 
66, 24 . 

. 211 
105, 106 
. 209* 


. 106 
• 23 
. 261 


2. 12,13 


2. 24 . 
3.8 . 

3. 10. 

3. 19 . 

4. 10 . 

4. 18. 
5.4 . 

5. 17 . 
5. 21 . 
6.6 . 


9. 13. 

10. 12 . 

15. 7 . 

16. II . 
18. 12 . 
19.5 . 
20. 13 . 

22. 16 . 

23. 9 . 

24. 2 . 
24.9 . 
28 (51). 1 
31 (48). 29 
32. 17 . . 
38 (31). 25 

38 (31). 33 

39 (32). 41 

40 (33). 2 

41 (34). 18 
44 (37). 15 
46 (39). 18 
49 (42). II 
51 (44). 28 































22, 23 
• 23 


2. II ... . 98 
2. 12 . . . . 104 


1. 20, 21 

2. II . . 

5. 17 . . 

9.5 • . 
10. 13 . . 





10. 17 

11. 19 

12. 19 
12. 22 

14. 21 
16. 44 
16. 49 
18. 2 
18. 3 
18. 7, 

18. 12 


19. 14 

20. 6, I 
20. 47- 
21 (31) 
22. 12 
22. 21 
22. 29 
28. 2, J 

35. 12 

36. 26 

37. 12 
37. 23 
39. 29 
44. 25 
47. 12 






9. 16. 

, 103* 



18, 100 










75» 90 












7. II 
13. 14 







2.6 . 

. . 75 

2.7 . 

28, 74, 75 

4. I . 

. . 75 

5. II . 

• . 75 

5. 12. 

. . 75 


. . 17 

8.4 • 

' 74, 75 

8.6 . 

• . 75 


2.4 ... • 
3. 2 . . . . 


2.4 ... . 
3.3 . 

3. 14 . 


. 252 
40*, 41 

• 74 
. 98 


1. 12 


3. 12 





5. II 

6. 12 
9. 2 

IL 14 

n. 17 
12. 10 

12. II 
12. 12 





40*, 41 








2. 17 . . . . 252 


4. 39 . . . . 252 

8. 23 . . . . 170 


1. I .... 107 
2 62 ... . 31 

2. 68 ... . 52 


2. 16 . . . . 206 


3. 8 ... . 79=* 
6. 18 ... . 55 

12. 7, II . . . 58 


2. 2 ... . 58 

49.... 13 

8. 24, 27 . . 72 

9. 12 ... . 5 


1. II 

2. 22 

2. 24 


3. 10 

4. I 

5. 13 

5. 18 

6. 24 

11. 10 
14. 15. 23 

14. 18, 27 

15. II . . 

16. 24 . . 

18. 4 . . 

19. 6 . . 











1. 23 

2. I 


3. 10 

3. 26 

4. II 
4. 13 

4. 17 
4. 28 

6. 27 

7. 18 
7. 20 

9. 12 
10. 17 

10. 27 

11. I 












. 252 
255, 260* 

. 251 

• 250 

• 14 

. 252 
. 261* 

• 251 
. 251 

. 250 

. 14 

. 252 

. 251 

. 262* 

. 262* 

. 251 

252, 253 
252, 253 
. 263* 
. 251 
. 252 
. 252 



12. 10 . 

12. 12. 

13. 26 . 

14• 4, 5 

14. 8, 9, 10 
14. 14 . 
14. 18 . 

14. 20 . 

15. 2, 3 

15.4 . 

15.5 . 

15.6 . 

16. 3 • 
16. 17, 18 

16. 25 . 

17. 27 . 

18. 4, 6 
18. 17 . 
18. 19 . 
18. 31 . 

18. 32 . 

19. 17 . 
19. 22 . 

19. 30 . 

20. 27, 28 

21. 20 . 

22. 22 . 

22. 27 . 

23. 10 . 

23. 20 . 

24. 17 . 

25. 15 . 
25. 17 . 
25. 21, 2 
26-5 . 
27. 13 . 
27. 16, 1 

27. 27 . 

28. I . 
28. 3-7 

28. 26 . 

29. 4 . 
29. 7 . 

29. 13 . 

30. II, 12, 
30. 39 (33 
32. 22 . 
34 (31). 23 
36 (33). 3 
36. 18 . 
36. 31 . 
36. 22 . 
38. 27 . 

38. 28 . 

39. 2, 3 
39. 13 . 
39. 16 . 
40.5 . 




. 78 

4L 2 . . 

. . 253 

. 263* 

42.5 . . 

. • 279* 

. 67 

42. 18 . . 

. • 250 

. 80* 

43. 9 . . 

. . 279* 

. 80 

43. 25 . . 

• . 279* 

• 250 , 

44. 5 . . 

• • 253^ 

. 250 

44.17. . 

. . 279* 

. 265* 

45. 2 . . 

. . 253 

. 250 

45. 15 . . 

• • 253 

. 251 

45. 20 . . 

. . 280* 

. 252 

46. 5 . . 

• • 253 

. 265* 

46. 9 . . 

• • 252 

. 266* 

46. 15 . . 

. . 280* 

. 266* 

47. 18 . . 

. . 67* 

. 251 

48. 22 . . 

. . 281 


. 250 

I Maccabees. 

. 14 

. 252 
. 252 

. 267 

• 252 

1. 43 • • • • 252 

10. 47 . . . . 252 
12. 53. . . . 262 
14. 7, 36 . . • 262 

. 268* 

. 251 

2 Maccabees. 

. 256* 

L 35 . . . . 261 

. 268* 

2.28. . 


. 252 

5.6 . . 


. 57 

6. 8 . . 



. 268* 

13. 21 . . 

" ] 


. 269* 

14. 38 . . 


. 252 

. 269 

255, 269* 

S. Matthew. 

255, 270* 

1. 18 . . 

. . 


. 271* 

L19. . 



3.3 . • 



4. I . . 



5. II . . 



5.25. . 



5. 29 . . 



5. 39 . . 



5.41. . 


. 273* 

6. I . . 


. 273* 

6. 8 . . 


. 273* 

6.13. . 

• "4 

'3*, 79* 


6. 19-24. 

. 81* 


• 275* 

7. 1 1 . . 

. 81* 

. 276* 

7. 16 . ' . 


. 80* 

9.4 . . 

. 8,25 

. 276* 

9. 16 . . 

• 25 

• 277* 

9.30. . 

• 25 

. 277* 
. 277* 

12. 18-21. 

j i99*,2oo* 
( 201* 

. 277* 

12. 20 . . 

. . 29 

. 278* 

12. 25 . . 

• 25 

. 67* 

12.45• . 

• 79* 

. 278* 

13.3 • . 

. 69 

• 144 

13. 10 . . 

• 69^ 

. 250 

13. II . . 

. 58* 


13. 18 
13. 21 
13. 24 
13• 33 

13. 34, 35 

14. 8 

15. 18 
15. 19 

17. 12 

18. 20 



20. 23 




22. I 

22. 18 

23. 28 

25. 41 






















S. Mark 




2. 21 


3. 23 

4. 2 

4. ΙΟ 
4. II 
4. 13 
4. 17 

4. 3o 
4• 33: 

5. 30 

10. 7 
10. 40 
12. I 

12. 12 


13. 14 
13. 28 





58* 69 









79*» 93* 







S. Luke 



L 28 

2. 26 

3. 14 

4. 2 

4. 23 

5. 22 
5• 36 

6• 9. 35 



8. 2 



8. 10 

8. II 


11. 26 


12. 14 
12. 16, 

-12. 42 

12. 58 

13. 6 

14. 7 

16. I 
37. 33 

18. I, ( 

19. 11' 

20. 9 
20. 18 

20. 19 
2L 25 

21. 29 
24. 16 

S. John 
5. 21 . 
10. 6 . 

11• 33, 38 

12. 6 . 

13. 28 . 
13. 29 . 
16. 25 . 

18. 37• 
20. 20 . 















58*, 69 









, 62 















2. I 
2. 20 
2. 46 

4. 24 

5. 12 

5• 39 

7. 2 

7• 3 

7. 26-28 

7• 57 


12. 20 

14. 17 

15. 25 
17. 22 

17. 26 

18. 12 

19. 29 

20. 9 
20. 19 

25. 19 

26. 7 


2. 10 . 

3. 8 . 

3. 20 . 

4.3 • 

4. 17 . 
4. 18 . 
4. 20, 21 
8. i^sqq 

8. 20 

9. 12 
11. I 

11. 34 
14. I 
16. 23 
16. 25 
2L 9 














14, 72* 






204, 209, 
210*. 211'' 









2. 16 

6• 5 

I Corinthians 





13. 12 . 
15. 51 . 

2 Corinthians 
11. 17 

11. 26 

12. 20 






3. 6 .... 156 
4. 30 . . . 160 


L6 . 

3• 3, 4 

3. 18 

4. 13 

5. 32 

6. 19 


1. 20 



1. 26, 27 . . 

2. 2 . . . . 

2. 18 . . . . 
4. 3 ... . 







1 Thessalonians. 
3.9 ... . 59* 
3. 16. . . . 59* 

2 Thessalonians. 

2. 7 ... . 59 

1 Timothy. 

3. II ... . 47 

4. 13 . . . . 39* 
6. 3 . . . . 260 

2 Timothy. 

2. 3 ... • 47 

1. 20 ... . 25 



Hebrews. | 


2.18. . 

. 72* 

3.14. . 

. 89* 

4. 12 . . 

• 25 

4. 15 . . 

• 73* 

6. 13, 14 

. 162 

9.9 . . 

69. 70* 

0. 26 . . 



. 89* 

L4 . . 

. 152 

L 16 . . 

• 54 

L19. . 

69, 70* 


1. 26, 27 

2. 23 . . 

I Pet 
L6 . . 
2. I . . 
2.9 . . 
2. 15, 20 
2. 21 . . 

4. 4 . . 
4. 12 . . 
4. 18 . . 


2 Peter 




L2, 8 . . 



L3 . . . 


L5 . . . 


2.9 . . . 


2. 16 . . . 



2. 20 . . . 



2. 22 . . . 

69, 70* 





1. 20 . . . 



3. 10 . . . 



10. 7 . . . 

. 59* 


17.7 . . . 




The Holy Scriptures, Apocrypha, etc 

The Bible in English 
Reprintofthe Authorized Version of 1611. 4to. isss. £3 3s. net. 
The Authorized Aversion. 

Complete lists of Oxford Bibles, Oxford Bibles for Teachers, Oxford Prayer 
Books, Church Sets, etc, in all styles and bindings can be obtained from 
any bookseller, or from Mr. Henry Frowde, Oxford University Press, 
Amen Corner, London, E.C. 

The Revised Version. 

[The Revised Version is the joint property of the Universities of Oxford 

and Cambridge.] 
Edition for the Church Desk. 

Folio. In buckram, with the Apocrypha, £Ά 2s. net. Bound in 
Grained Cowhide, £2 10s.; with the Apocrypha, £3; in Turkey 
Morocco, £4 4s.; with the Apocrypha, £4 16s. 
Library editions, in large type (pica). Bound in cloth and greatly reduced 
in price. 

Royal 8vo, with large margins : with the Apocrypha, 6 vols., from £1 9s. ; 

the Apocrypha only, 10s. 6d. ; the other five volumes, 18s. 6d. 
Demy 8vo, with the Apocrypha, 6 vols., £1 Is. ; the Apocrypha only, 
7s. 6d. ; the other five volumes, 14s. 
Single volume editions : 

Royal 8vo, on Oxford India Paper, and in large type (pica), in Persian, 

£2 12s. 6d.; with the Apocrypha, in Turkey Morocco, £3 19s. 
Royal 8vo, small pica type. In cloth, 10s. 6d. ; in leather bindings, from 

19s. 6d. 
8vo, small pica type. In cloth, 7s. 6d. ; in leather bindings, from 10s. 6d. 
8vo, minion type. In cloth, 4s. ; with the Apocrypha, 6s. ; the Apocrypha 

only, 3s. 
16mo, ruby type. In cloth, 2s. 6d. ; with the Apocrypha, 4s. ; the 

Apocrypha only, 2s. ; with the Oxford Helps, in leather, from 7s. 
16mo, nonpareil type. In leather, from 3s. 

16mo, pearl type. In cloth boards, from 8d. ; with Oxford Helps, 
from 2s. 6d. 

The Revised Version with marginal references. 

8vo, in bourgeois type, from 6s. ; on Oxford India paper, bound in 

leather, from 15s. 
8vo, in minion type, from 5s.; with the Apocrypha, from 7s. 6d.; with Oxford 
Helps, from lis. ; on Oxford India paper, bound in leather, from 14s. 
16mo, in nonpareil type, in leather, from 6s. ; on Oxford India paper, 
bound in leather, from 10s. 6d. 
The Parallel Bible, containing the Authorized and Revised Versions in 
parallel columns. Small quarto, in minion type. On ordinary paper, from 
10s. 6d. ; on Oxford India paper, bound in leather, from 16s. 
The Two-Version Bible, being the Authorized Version with the 
differences of the Revised Version printed in the margins, so that both texts 
can be read from the same page. With references and maps. Bourgeois 
type. Cloth, 7s. 6d. ; leather, from 10s. 6d. ; on Oxford India paper— cloth, 
10s. 6d. ; leather, from 15s. ; interleaved with writing-papeir, and bound in 
leather, from £1 Is. With the Oxford Helps, leather, from 18s. 6d. Printed 
on writing-paper with wide margins for notes, from 10s. 6d. net. 


Elementary Helps to the Study of the Bible 

Oxford Bible for Teachers, containing the texts of the Old and 

New Testaments, with or without marginal references, and in either the 
Authorized or Revised Versions, together with the Oxford Helps to the Bible 
(see below), and 124 full-page plates. 

In many styles and bindings, from 3s. to £5. A complete Hst can be 
obtained from Mr. Henry Frowde, Amen Corner, London, E.G. 

The Oxford Helps to the Study of the Bible, containing 

Introductions to the several Books, the History and Antiquities of the Jews, 
the Natural History of Palestine, with copious tables, concordance and 
indices, and a series of maps. With 124 full-page plates. 

8vo, in long primer type— cloth, 5s. ; leather, from 6s. 6d. 

8vo, in nonpareil type — cloth, 2s. 6d. ; leather, 3s. 6d. 

16mo, in pearl type — stiff covers. Is. net ; cloth. Is. 6d. ; leather, from 2s. 

Bible Illustrations, being 124 full-page plates, forming an appendix to 

the above. Crown 8vo, cloth, 2s. 6d. 
Old Testament History narrated for the most part in the words of 

the Bible. By George Carter. With maps. Crown 8vo. 2s. 
An Annotated Psalter. Arranged byJ.M. Thompson. Fcap 8yo. 2s. 

Dr. Stokoe's Manuals. Crown 8vo 
Old Testament History for Schools. By τ. η. Stokoe. Part i. 

(Third edition.) From the Creation to the Settlement in Palestine. Part II. 
From the Settlement to the Disruption. Part III. From the Disruption to 
the Return from Captivity. Extra fcap 8vo. 2s. 6d. each, with maps. 
Manual of the Four Gospels. With Maps, Ss. ed. Or, separately, 
Part I, The Gospel Narrative, 2s. ; Part II, The Gospel Teaching, 2s. 

Manual of the Acts. ss. 

The Life and Letters of St. Paul. ss. ed. Or, separately. 

Part I, The Life of St. Paul, 2s. Part II, The Letters of St. Paul, 2s. 

First Days and Early Letters of the Church, ss. Or, Part i. 

First Days of the Church, Is. 6d. Part II, Early Letters of the Church, 2s. 

Graduated Lessons on the Old Testament. By u. z. Rule, 

edited by Ll. J. M. Bebb. Selected Readings from the O. T. with para- 
phrases and connecting paragraphs ; with notes for teachers and pupils. For 
use in lower forms, and in elementary and Sunday Schools. The text is that 
of the R.V., but the book may be used with the A.V. In three volumes. 
Extra fcap 8vo. Is. 6d. each in paper covers, or Is. 9d. each in cloth. Vol. I. 
Creation to Death of Moses. Vol. II. Conquest of Canaan, Psalms, etc. 
Vol. III. Israel and Judah, the Captivity, the Prophets. 

The Gospel according to St. Mark. Ed. by A. s. Walpole. is.ed. 

Notes on the Gospel of St. Luke, for Junior Classes. By Miss 
E. J. Moore Smith. Extra fcap 8vo, stiff covers. Is. 6d. 

A Greek Testament Primer, being an easy grammar and reading- 
book for the use of students beginning Greek. By E. Miller. Second 
edition. Extra fcap 8vo, paper, 2s. ; cloth, 3s. 6d. 

History of the Bible 
List of Editions of the Bible in English, by h. Coito». 

Second edition. 8vo. 8s. 6d. 

RhemeS and Doway ; showing what has been done by Roman Catholics 
for the diffusion of the Bible in English. By H. Cotton. Svo. 9s. 

The Part of Rheims in the Making of the English 

Bible, by J. G. Carleton, containing historical and critical introduction, 
tables and analyses thereof. Svo. 9s. 6d. net. 

WycliiFe's Bible, portion edited by W. W. Skeat. See p. 6. 

Studia Biblica et Eeclesiastica. Essays chiefly in BibUcal and 
Patristic criticism, and kindred subjects. Svo. 
Vol. I, 10s. 6d. Vol. II, 12s. 6d. Vol. Ill, 16s. Vol. IV, 12s. 6d. 
Vol. V, complete, 16s., or, in separate parts. Part I, Life of St. Nino, by 
M. and J. O. Wardrop. 3s. 6d. Part II, Texts from Mount Athos, by 
K. Lake. 3s. 6d. Part III, Place of the Peshitto Version in the Ap- 
paratus Criticus of the Greek New Testament. 2s. 6d. Part IV, Baptism 
and Christian Archaeology, by C. F. Rogers. 4s. 6d. 

The Gospel of Barnabas. Edited and translated from the Italian 
MS. in the Imperial Library, Vienna, by Lonsdale and Laura Ragg. Svo, 
with a facsimile. 16s. net. 

The Journal of Theological Studies. Edited by J. f. Bethune- 

Baker and F. E. Brightman. Published Quarterly, price 3s. 6d. net. 
Annual subscription 13s. net, post free. Many of the back numbers are still 

The Old Testament and Apociypha 

The Psalter, by Richard Rolle of Hampole. Edited by H. R. 
Bramley. With an introduction and glossary. Svo. £1 Is. 

The Parallel Psalter : being the Prayer-Book version of the Psahns 
and a new version arranged on opposite pages, with an introduction and 
glossaries by S. R. Driver. Second edition. Extra fcap Svo. 3s. 6d. net. 

The Book of Job in the Revised Aversion. Edited, with 

introductions and brief annotations, by S. R. Driver. Crown Svo. 
2s. 6d. net. 

The Hebrew Prophets in the Revised Version. Arranged 

and annotated by F. H. Woods and F. E. Powell. In four vols., crown Svo.. 
Vol. I (Amos, Hosea, Isaiah i-xxxix, Micah), 2s. 6d. net. 

Lectures on Ecclesiastes. By G. G. Bradley. Second edition. 

Crown Svo. 5s. 6d. 

Israel's Hope of Immortality. By c. F. Burney. svo. 2s. ed. net. 

EcclesiasticUS, translated from the Hebrew by A. E. Cowley and 
A. Neubauer. Crown Svo. 2s. 6d. 


The Place of Ecclesiasticus in Semitic Literature. An 

essay by D. S. Margoliouth. Small 4to.• 2s. 6d. 

The Five Books of Maccabees, with notes and iUustrations by 

H. Cotton. 8vo. 10s. 6d. 

The Book of Enoch, translated from Dillmann's Ethiopic text (emended 
and revised), and edited by R. H. Charles. 8vo. [Out of print.] 

The Book of the Secrets of Enoch, translated from the Slavonic 
by W. R. MoRFiLL, and edited, with introduction, notes, etc, by R. H. 
Charles. 8vo. 7s. 6d. 

History and Song of Deborah (Judges iv and V). By g. a. 

Cooke. 8vo. Paper covers. Is. 6d. (Published by Mr. Fro wde.) 

DeuterOgraphs. Duplicate passages in the Old Testament. Arranged 
by R. B. GiRDLESTONE. 8vo. 7s. 6d. 

Astronomy in the Old Testament. By g. Schiaparelli. 

Authorized translation. Crown 8vo. 3s. 6d. net. 

Libri Psalmorum Versio antiqua Latina, cum Paraphrasi Angio- 

Saxonica. Edidit B. Thorpe. 8vo. 10s. 6d. 

Libri Psalmorum Versio antiqua Gallica e Cod. MS in Bibi. 

Bodleiana adservato, una cum Versione Metrica aliisque Monumentis perve- 
tustis. Nunc primum descripsit et edidit F. Michel. 8vo. 10s. 6d. net. 

Chandler's Life of David. Svo. 7s. 6d. net. 
Faussett's Sacred Chronology. Svo. 10s. 6d. net. 

Hebrew and Chaldee 
Notes on the Hebrew Text of the Book of Genesis. 

By G. J. Spurrell. Second edition. Crown Svo. 13s. 68. 
Notes on Samuel. By S. R. Driver. [Out of print.] 

Notes on the Hebrew Text of the Books of Kings. By 

C. F. BuRNEY. Svo. 14s. net. 

A Hebrew and English Lexicon of the Old Testament, 

with an Appendix containing the Biblical Aramaic, based on the Thesaurus 
and Lexicon of Gesenius, by F. Brown, S. R. Driver, and C. A. Briggs. 
Small 4to. Cloth, 34s. net ; morocco back, 45s. net. The several Parts can 
be supplied to complete sets. 

Gesenius's Hebrew Grammar, as edited and enlarged by E. 
Kautzsch. New edition in preparation. 

A Treatise on the Use of the Tenses in Hebrew. By 

S. R. Driver. Third edition. Crown Svo. 7s. 6d. net. 

A Commentary on the Book of Proverbs, attributed to 

Abraham Ibx Ezra. Edited from a MS in the Bodleian Library by S. R. 
Driver. Crown Svo. Paper covers, 3s. 6d. 


The Book of Tobit. a Chaldee Text, from a unique MS in the 
Bodleian Library ; with other Rabbinical texts, English translations, and the 
Itala. Edited by A. Neubauer. Crown 8vo. 6s. 

EcclesiasticUS (xxxix. 15 — xlix. 11). The Hebrew, with early 
versions and English translation, etc, edited by A. E. Cowley and A. Neubauer. 
With 2 facsimiles. 4to. 10s. 6d. net. Translation, see p. 67. 

Facsimiles of the Fragments hitherto recovered of the 

Book of Ecclesiasticus in Hebrew. 60 leaves in Collotype. £1 Is. net. 
(Published jointly by the Oxford and Cambridge University Presses.) 

The Psahns in Hebrew without points, stiff covers. 2s. 
Accentuation of Psalms, Proverbs, and .Job. By w. Wickes. 

8vo. 5s. 

Hebrew Prose Accentuation. By the same. svo. los. 6d. 
Lightfoot's Horae Hebraicae et Talmudicae. a new 

Edition by R. Gandell. 4 vols. Svo. £1 Is. 


VetUS Testamentum ex Versione Septuaginta Interpretum secundum 
exemplar Vaticanum Romae editum. Accedit potior varietas Codicis Alex- 
andrini. Tomi III. 18mo. 6s. each volume. 

A Concordance to the Septuagint and other Greek Versions of 
the O. T. (including the Apocryphal Books) ; by the late Edwin Hatch and 
H. A. Redpath. With Supplement. By H. A. Redpath. Fasc. I, containing 
a Concordance to the proper names occurring in the Septuagint ; Fasc. II, 
containing a Concordance to Ecclesiasticus, other Addenda, and the Hebrew 
Index to the whole work. Imperial 4to. Cloth in 3 vols. £8 8s. net (or 
Concordance, 2 vols., £6 17s. 6d. net. Supplement, £2 net). Parts II-VI, 
21s. net each ; Supplements I and II, 16s. net each. 

Origenis Hexaplorum quae SUperSUnt sive Veterum inter- 
pretum Graecorum in totum Vetus Testamentum iFragmenta. Edidit F. Field. 
2 vols. 4to. £5 5s. net. 

Essays in Biblical Greek. By Edwin Hatch. Svo. lOs. 6d. net. 

The Book of Wisdom : the Greek Text, the Latin Vulgate, and the 
Authorized English Version ; with an introduction, critical apparatus, and 
a commentary. By W. J. Deane. 4to. 12s. 6d. net. 

The Greek Versions of the Testaments of the Twelve 

Patriarchs. Edited from nine MSS, with variants from other versions. 
By R. H. Charles. Svo. 18s. net. 

Tattam's Major and Minor Prophets. See p. 85. 

The New Testament 

The Parallel New Testament, Greek and English ; being 

the Authorized Version, 1611 ; the Revised Version, 1881 ; and the Greek 
Text followed in the Revised Version. 8vo. 9s. net. 

The New Testament in Greek and English. Edited by 

E. Card WELL. 2 vols. 1837. Crown 8vo. 6s. 


The Greek Testament, with the Readings adopted by the Revisers of 
the Authorized Version. (1) Pica type. Demy8vo. 10s. 6d. (2) Long Primer 
type. \Vith marginal references. Fcap 8vo. 4s. 6d. (3) The same, on 
writing-paper, with wide margin, 15s. (4) The same, on India paper. 6s. 

Novum Testamentum Graece. Accedunt paraUela S. Scripturae 

loca,etc. Ed. C.Lloyd. 18mo. 3s. On writing-paper, with wide margin, 7s. 6d. 

Critical Appendices to the above,byW. Sanday. Extra fcap 8vo. 3s. 6d. 

Novum Testamentum Graece (ed. Lloyd), with Sanday's Appen- 
dices. Cloth, 6s. ; paste grain, 7s. 6d. ; morocco, 10s. 6d. 

Novum Testamentum Graece juxta Exemplar Millianum. 

Fcap 8vo. 2s. 6d. On writing-paper, with wide margin, 7s. 6d. 

Evangelia Sacra Graece. Fcap 8vo, limp. is. ed. 

Novum Testamentuni Graece. Antiquissimomm Codicum Textus 
in ordine parallelo dispositi. Edidit E. H. Hansell. Tomi IIL 8vo. £1 4s. 

Athos Fragments of Codex Η of the Pauline Epistles. 

Photographed and deciphered by Kirsopp Lake. Full-size collotype 
facsimiles, large 4to, in an envelope. 21s. net. 

Athos Fragments of the Shepherd of Hermas. Photo- 
graphed and transcribed by Kirsopp Lake. Full-size collotype facsimiles, 
large 4to, in an envelope. 17s. 6d. net. 

Outlines of the Textual Criticism of the New Testament. 

By C. E. Hammond. Sixth edition. Crown 8vo. 4s. 6d. 

Greswell's Harmonia Evangelica. Fifthedition. svo. 9s.6d. Pro- 
legomena. Dissertations on the Harmony. 
Jones's Canon of the New Testament, svo. 3 vols. iss. net. 

DiateSSaron. Edited by J. White. 3s. 6d. 

Horae Synopticae. By Sir J. C. Hawkins, Bart. Svo. 10s. 6d. net. 

Dr. Sanday's Books 

Sacred Sites of the Gospels, with sixty-three fuU-page illustrations, 
maps, and plans. Svo. 13s. 6d. net. 

Criticism of the Fourth Gospel, svo. 7s. 6d. net. 

The Life of Christ in Recent Research, svo, with two 

illustrations. 7s. 6d. net. 
Presidential Address to the christian Section of the Congress for the 
History of Religions, 1908. Svo. Is. net. 


The Logia 
Two Lectures on the ' Sayings of Jesus', delivered at Oxford 

in 1897, by W. Lock and W. Sanday. 8vo. Is. 6d. net. 

The Oxyrhynchus Logia and the Apocryphal Gospels. 

By C. Taylor. 8vo, paper covers, 2s. 6d. net. 

The Oxyrhynchus Sayings of Jesus, found in 1903 ; with the 

Sayings called ' Logia ', found in 1897. By C. Taylor. 8vo. 2s. net. 
Published by Mr. Frowde for the Egypt Exploration Fund. 
By B. P. Grenfell and A. S. Hunt. 
ΛΟΓΙΑ ΙΗΣΟΥ, from an early Greek papyrus, with translation and com- 
mentary. 8vo, stiff boards, with two collotypes, 3s. net; with two tone 
blocks, 6d. net. 
New Sayings of Jesus and Fragment of a Lost Gospel. With one Plate. Is. net. 
Fragment of an Uncanonical Gospel from Oxyrhynchus. With one Plate. Is. net. 

Coptic and Syriac 
The Coptic Version of the New Testament, in the Northern 

Dialect, otherwise called Memphitic and Bohairic. With introduction, 
critical apparatus, and English translation. 8vo. Vols. land Π. The Gospels. 
£2 2s. net ; Vols. Ill and IV. The Epistles. £2 2s. net. 

Tetraeuangelium Sanctum iuxta simpUcem Syrorum versionem ad 
fidem codicum, Massorae, editionum denuo recognitum. Lectionum supel- 
lectilem quam conquisiverat P. E. Pusey auxit, digessit, edidit G. H. 
GwiLLiAM. Accedunt capitulorum notatio, concordiarum tabulae, translatio 
Latina, annotationes. Crown 4to. £2 2s. net. 

CoUatio Cod. Lewisiani Evangeliorum Syriacorum cum 

Cod. Curetoniano, auctore A. Bonus. Demy 4ito. 8s. 6d. net. 

Old-Latin Biblical Texts : smaU 4to, stiff covers. 

No. I. St. Matthew, from the St. Germain MS (gi). Edited by J. 
" " " ■ ' ' "^ " th 

Wordsworth. 6s. net. — No. II. Portions of St. Mark and St. Matthew, 
from the Bobbio MS (k), etc. Edited by J. Wordsworth, W. Sanday, and 
H. J. White. £1 Is. net. — No. III. The Four Gospels, from the Munich 

from the Bobbio MS (k), etc. Edited by J. Wordsworth, W. Sanday, and 
H. J. White. £1 Is. net. — No. III. The Four Gospels, from the Munich 
MS (q), now numbered Lat. 6224. Edited by H. J. White. 12s. 6d. net. 

No. IV. Portions of the Acts, of the Epistle of St. James, and of the 
First Epistle of St. Peter, from the Bobbio Palimpsest (s), now numbered 
Cod. 16 in the Imperial Library at Vienna. Edited by H. J. White. 
5s. net. — No. V. The Gospels from the Codex Corbeiensis (ffj or ffn) with 
Fragments of the Catholic Epistle, Acts, and Apocalypse from the 
Fleury Palimpsest (h). Edited by E. S. Buchanan. 12s. 6d. net. 

Nouum Testamentum Latlne, secundum Editionem Sancti Hiero- 
nymi. Recensuit I. Wordsworth, Episcopus Sarisburiensis ; in operis 
societatem adsumto H. I. WHI1Έ. 4to. Part I (Gospels), £2 12s. 6d. (Some 
of the fasciculi of Part I can still be obtained separately.) Part II, fasc. i 
(Romans), in the press. Part III, fasc. i (Acts), 12s. 6d. 

Notes on the Early History of the Vulgate Gospels. By 

IJom John Chapman, O.S.B. 8vo. 

Sacrorum Evangeliorum Versio Gothica cum interpretatione 

Latina et annotationibus Erici Benzelii edidit Edwardus Lye (1750). Large 
4to. £1 10s. net. 


The Fathers of the Church and 
Ecclesiastical History 

Editions with Latin Commentaries 

Catenae GraeCOrum Patrum in Novum Testamentum edidit J. A, 
Cramer. Tomi VIII. 8vo. £2 8s. net. 

Clementis Alexandrini Opera, ex rec. Guil. Dindorfii. Tomi IV. 8vo. 
£3 net. 

Cyrilll Archiepiscopi Alexandrini in XII Prophetas edidit P. E, Pusey. 
Tomi II. Bvo. £2 2s. 

In D. Joannis Evangelium. Accedunt Fragmenta Varia. Edidit post 

Aubertum P. E. Pusey. Tomi III. Bvo. £3 5s. 
Commentarii \ in Lucae Evangelium quae supersunt Syriace edidit 
R. Payne Smith. 4to. £1 2s. Translation, 2 vols. 8vo. 14s. 

Ephraemi Syri, Rabulae, Balaei aliorumque Opera Selecta. See p. 85. 

Eusebii Opera recensuit τ. Gaisford. 

Evangelicae Praeparationis Libri XV. Tomi IV. Svo. 42s. net. 
Evangelicae Demonstrationis Libri X. Tomi II. Svo. 15s. 
Contra Hieroclem et Marcellum Libri. Svo. 7s. 
Annotationes Variorum. Tomi II. Svo. 17s. 

Canon Muratorianus. Edited, with notes and facsimile, by S. P. 
Tregelles. 4to. 10s. 6d. 

Evagril Historia Ecclesiastica, ex rec. H. Valesii. Svo. 4s. 

n. Josephl de bello Judaic© Libri Septem recensuit E. Cardwell. 
Tomi II. Svo. 17s. 

OrigeniS Philosophumena; sive omnium Haeresium Refutatio e Codice 
Parisino nunc primum edidit Emmanuel Miller. Svo. 10s. 

Patrum ApOStolicorum Clementis Romani, Ignatii, Polycarpi, quae 
supersunt edidit G. Jacobson. Tomi II. Fourth edition. Svo. £1 Is. 

Reliquiae Sacrae secundi tertuque saeculi recensuit M. J. Routh. 
Tomi V. Second edition, 1846. Svo. £1 5s. net. 

Scriptorum Eeclesiasticorum Opuseula recensuit M. J. Routh. 

Tomi II. Third edition, 1S5S. Svo. 10s. 

Socratis Scholastici Historia Ecclesiastica Gr. et Lat. edidit R. Hussey. 
Tomi III. 1853. Svo. 15s. net. 

Sozomeni Historia Ecclesiastica edidit R. Hussey. Tomi III. Svo. 15s. net. 

Theodoreti Ecclesiasticae HistoriaeLibriVrec.T. Gaisford. Svo. 7s. 6d. net. 
Graecarum AiFectionum Curatio rec. T. Gaisford. Svo. 7s. 6d. 

Notitia Scriptorum SS. Patrum. By J. g. Dowling. svo. 

7s. 6d. net. 


Editions with English Commentaries 
or Introductions 

St. AthanaSlUS. Orations against the Arians. With an account of his 
Life by W. Bright. Crown 8vo. 9s. 

Historical Writings, according to the Benedictine Text. 
With an introduction by W. Bright. Crown 8vo, 
10s. 6d. 

St. Augustine. Select Anti-Pelagian Treatises, and the Acts of the 
Second Council of Orange. With introduction by W. Bright. Crown 8vo. 

[Out of print.] 

St. Basil .* on the Holy Spirit. Revised text, with notes and introduction, 
by C. F. H. Johnston. Crown Svo. 7s. 6d. 

Barnabas, Editio Princeps of the Epistle of, by Archbishop Ussher, as 

grinted at Oxford, a. d. 1643. With a dissertation by J. H. Backhouse. 
mall 4to. 3s. 6d. 

Canons of the First Four General Councils of Nicaea, Con- 
stantinople, Ephesus, and Chalcedon. With notes by W. Bright. Second 
edition. Crown Svo. 7s. 6d. 

Eusebius' Ecclesiastical History, according to Burton's text, with 

introduction by W. Bright. Second edition. Crown Svo. Ss. 6d. 

Eusebii Pamphili Evangelicae Praeparationis Libri XV, 

Revised text edited, with introduction, notes, English translation, and indices, 
by E. H. GiFFORD. 4 vols. [Vols. I, Π, text, with critical notes. Vol. HI, 
in two parts, translation. Vol. IV, notes and indices.] Svo. £5 5s. net. 
(Vol. in, divided into two parts, containing the translation, £\ 5s. net.) 

The Bodleian Manuscript of Jerome's Version of the 

Chronicles of Eusebius, reproduced in coUotype, with an intro- 
duction by J. K. FoTHERiNGHAM. 4to, buckram. £2 10s. net. 

John of Ephesus. See p. 85. Translation, by R. Payne Smith. Svo. 10s. 

Philo : about the Contemplative Life ; or, the Fourth Book of 

the Treatise concerning Virtues. Edited, with a defence of its genuineness, 
by F. C. CoNYBEAHE. With a facsimile. Svo. 

Socrates' Ecclesiastical History, according to Hussey's Text, with 

introduction by W. Bright. Second edition. Crown Svo. 7s. 6d. 

Tertulliani Apologeticus adversus Gentes pro Christianis. Edited by 
T, H. Bindley. Crown Svo. 7s. 6d. net. 

De Praescriptione Haereticorum : ad Martyras : ad Scapulam. 
Edited by T. H. Bindley. Crown Svo. 5s. net. 


Works of the English Divines. 8vo 

Sixteenth, seventeenth, eighteenth and early 

nineteenth centuries 

Editions of Hooker and Butler 

Hooker's Works, with Walton's Life, arranged by John Keble. Seventh 
edition, revised by R. W. Chuhch and F. Paget. 3 vols. 12s. each. [Vol. II 
contains the Fifth Book.] 

Introduction to Hooker's Ecclesiastical Polity, Book V. 

By F. Paget. Second edition. 5s. net. 
Ihe Text, as arranged by J. Keble. 2 vols. lis. 

The Works of Bishop Butler. By W. E. Gladstone. 2 vols. 
• 14s. each. Crown 8vo, Vol. I, Analogy, 5s. 6d. ; Vol. II, Sermons, 5s. 

Studies subsidiary to the A¥orks of Bishop Butler. 

Uniform with the above. 10s. 6d. Crown 8vo, 4s. 6d. 

Pearson's Exposition OF THE Creed. Revised by E. Burton. Sixth edition. 10s. 6d. 

Minor Theological Works. Edited by E. Churton. 2 vols. 10s. 
Enchiridion Theologicum Anti-Romanum. 

I. Jeremy Taylor's Dissuasive from Popery, and Real Presence. 8s. 
II. Barrow's Supremacy of the Pope, and Unity of the Church. 7s. 6d. 
III. Tracts by Wake, Patrick, Stillingfleet, Clagett, and others, lis. 
Addison's Evidences. 3s. 6d. net. 

Allix's Works. 4 vols. 5s. each. Bentley's Sermons. 4s. 

Biscoe's History of the Acts. 9s. 6d. 
Bragge's Works. 5 vols. £1 13s. 6d. 

Bull's Works, with Nelson's Life. Ed. by E. Burton. 8 vols. £2 9s. 
Burnet's Exposition of the XXXIX Articles. 7s. 
Butler's Works, 1849. 3 vols. Sermons. 5s. 6d. Analogy. 5s. 6a. 
Clergyman's Instructor. Sixth edition. 6s. 6d. 
Comber's Works. 7 vols. £1 lis. 6d. 
Fell's Paraphrase on St. Paul's Epistles. 7s. 
Fleetwood's Works. 3 vols. £1 Is. 6d. 
Hall's Works. Edited by P. Wynter. 10 vols. £3 3s. 
Hammond's Paraphrase on the New Testament. 4 vols. 20s. 

Paraphrase on the Psalms. 4 vols. 20s. 
Horbery's Works. 2 vols. 8s. Hooper's Works. 2 vols. 8s. 

Jackson's (Dr. Thomas) Works. 12 vols. £3 6s. 
Jewel's Works. Edited by R. W. Jelf. 8 vols. £1 10s. 
Leslie's Works. 7 vols. 40s. 

Lewis' (John) Life of Wiclif. 5s. 6d. Life of Pecock. 3s. 6d. 
Lewis' (Thomas) Origines Hebraicae. 3 vols. 16s. 6d. 
Newcombe's Observations. 6s. net. 
Patrick's Theological Works. 9 vols. £1 Is. 
Sanderson's Works. Edited by W. Jacobson. 6 vols. £1 10s. 
Scott's Works. 6 vols. £1 7s. Smalridge's Sermons. 2 vols. 8s. 

Sherlock's Discourses. 4 vols. 42s. net. 
Stillingfleet's Origines Sacrae. 2 vols. 9s. 

Grounds of Protestant Religion. 2 vols. 10s. 
Stanhope's Paraphrase. 2 vols. 10s. Taverner's Postils. 5s. 6d. 

Wall's History of Infant Baptism. By H. Cotton. 2 vols. £1 Is. 
Waterland's Works, with Life by Van Mildert. 6 vols. £2 lis. 

Doctrine of the Eucharist. 3rd ed. Cr. 8vo. 6s. 6d. 
Wheatley's Illustration of the Book of Common Prayer. 5s. 


Early Ecclesiastical Histoiy 

The New Testament in the Apostolic Fathers, By a Com- 
mittee of the Oxford Society of Historical Theology. 8vo. 6s. net. 

The Origins of Christianity. By C. Bigg. Edited by τ. β. Strong. 
8vo. 12s. 6d. net. 

The Church's Task under the Empire, with preface, notes, 

and excursus, by C. Bigg. 8vo. 5s. net. 
Bingham's Antiquities of the Christian Church, and other Works. 

10 vols. 8vo. ^£3 3s. 
The Church in the Apostolic Age. By W. W. Shirley. Second 

edition. Fcap 8vo. 3s. 6d. 
Harmonia Symbolical Creeds of the Western Church. By C. 

Heurtley. 8vo. 6s. 6d. 

A Critical Dissertation on the Athanasian Creed. By 

G. D. W. Ommanney. Svo. 16s. 

Ecclesiae Occidentalis Monumenta luris Antiquissima : 

Canonum et Conciliorum Graecorum Interpretationes Latinae. Edidit 
C. H. Turner. 4to, stiff covers. Tom. I, Fasc. I, pars I, 10s. 6d. net; pars 
II, 21s. net. Tom. II, pars I, 18s. net. 

The Key of Truth ; being a Manual of the Paullclan Church of 
Armenia. By F. C. Conybeare. Svo. 15s. net. 

Baptism and Christian Archaeology, being an offprint of studia 

Biblica, Vol. V. By C. F. Rogers. 8vo. 5s. net. 

Ecclesiastical History of Britain, etc 

Adamnani Vita S. Columbae. Edited, with introduction, notes, 

and glossary, by J. T. Fowler. Crown Svo, leather back. 8s. 6d. net. 
With translation, 9s. 6d. net. 

Baedae Opera Historica. Edited by C. Plummer. Two volumes. 

Crown 8vo. 21s. net. 

Councils and Ecclesiastical Documents relating to Great 

Britain and Ireland. Edited after Spelman and Wilkins, by A. W. Haddan 
and W. Stubbs. Medium Svo. Vols. I-III, £3 13s. 6d. net. Also Vol. II (Parts 
I and II) and Vol. Ill, separately, 14s. net per volume. 

Nova Legenda Angliae, as coUected by John of Tynemouth and 
others, and first printed 1516. Re-edited 1902 by C. Horstman. 2 vols. 
Svo. £1 16s. net 

Wyclif. A Catalogue of the Works. By W. W. Shirley. Svo. 3s. 6d. 
Select English Works. By T. Arnold. 3 vols. Svo. £1 Is. net. 
Trialogus. First edited by G. Lechler. Svo. 7s. 

Cranmer's AVorks. CoUected by H. Jenkyns. 4 vols. Svo, £1 lOs. 

* Cranmer's ' Catechism with the Latin Original. Edited by 
Edward Burton (1829). Svo, with woodcuts. 7s. 6d. net. 


Records of the Reformation. The Divorce, 1527-1533. Mostly 
now for the first time printed. Collected and arranged by N. Pocock. 2 vols. 
£1 16s. 

Primers put forth in the reign of Henry VIII. svo. 5s. 
The Reformation of Ecclesiastical Laws, as attempted in the 

reigns of Henry VIII, Edward VI, and Elizabeth. Edited by E. Cardwell. 
Svo. 6s. 6d. 

Conferences on the Book of Common Prayer from 1551 to 

1690. Edited by E. Cardwell. Third edition. 8vo. 7s. 6d. net. 

Documentary Annals of the Reformed Church of 

England ; Injunctions, Declarations, Orders, Articles of Inquiry, etc, from 
1546 to 1716. Collected by E. Cardwell. 2 vols. 8vo. 25s. net. 

Formularies of Faith set forth by the King's authority during 
Henry VIH's reign. Svo. 7s. 

Homilies appointed to be read in Churches. By J. Griffiths. Svo. 7s. 6d. 

Hamilton's Catechism, 1552. Edited, with introduction and 

glossary, by T. G. Law. With a Preface by W. E. Gladstone. Svo. 12s. 6d. 

Ιλ Oelll CatechismUS slve prima InstitutlodisclpllnaquePletatlsChristlanae 
Latine explicata. Editio nova cura G. Jacobson. Svo. 5s. 6d. 

Sylloge Confessionum sub tempus Reformandae Ecclesiae edit. Subjlc. 
Catechismus Heidelbergensis et Canones Synodi Dordrecht. Svo. Ss. 

Histories written in the seventeenth (or early 

eighteenth) and edited in the nineteenth century 

Stillingfleet's Origines Britannicae, with Lloyd's Historical 

Account of Church Government. Edited by T. P. Pantin. 2 vols. Svo. 10s. 

Inett's Origines Anglicanae (in continuation of Stmingfleet). Edited 
by J. Griffiths. 1S55. 3 vols. Svo. 15s. 

Fuller's Church History of Britain. Edited by J. s. Brewer, 

1845. 6 vols. Svo. £2 12s. 6d. net. 

Le Neve's Fasti Ecclesiae Anglicanae. Corrected and con- 
tinued from 1715 to 1S53 by T. D. Hardy. 3 vols. Svo. £1 10s. net. 

Strype's Memorials of Cranmer. 2 vols. svo. iis. net. Life of 

Aylmer. Svo. 5s. 6d. net. Life of Whitgift. 3 vols. Svo. 16s. 6d. net. 
General Index. 2 vols. Svo. 

Burnet's History of the Reformation. Revised by N. Pocock. 

7 vols. Svo. £1 10s. 

Prideaux's Connection of Sacred and Profane History. 2 vols. Svo. 
10s. Shuckford's Continuation, 10s. 

Gibson's Synodus Anglicana. Edited by E. Cardwell. 1854. 
Svo. 6s. 


Recent Works in English Ecclesiastical History 
History of the Church of England from the abolition of the 

Roman Jurisdiction. By W. R. Dixon. 3rd edition. 6 vols. 8vo. 16s. per vol. 

Chapters of Early English Church History. By w. Bright. 

Third edition. With a map. 8vo. 12s. 

Registrum Sacrum Anglicanum : an attempt to exhibit the course 

of Episcopal Succession in England. ByW. Stubbs. 2nd ed. 4to. 10s. 6d. 

The Elizabethan Clergy and the Settlement of Religion, 1558-1564. 
By Henry Gee. With illustrative documents and lists. 8vo. 10s. 6d. net. 


Liturgies, Eastern and Western. Vol. i. Eastern Liturgies. 

Edited, with introductions and appendices, by F. E. Brightman, on the basis 
of a work by C. E. Hammond. 8vo. £1 Is. net. 

Rituale Armenorum : the Administration of the Sacraments and the 
Breviary Rites of the Armenian Church, with the Greek Rites of Baptism 
and Epiphany. Edited by F. C. Conybeare; with the East Syrian Epiphany 
Rites, translated by A. J. Maclean. 8vo. 21s. net. 

Cardwell's Two Books of Common Prayer. Ed. 3. svo. 7s. net. 

Gelasian Sacramentary, Liber Sacramentomm Romanae Ecclesiae. 
Edited by H. A. Wilson. Medium Svo. 18s. net. 

Leofric Missal, with some account of the Red Book of Derby, the Missal 
of Robert of Jumieges, etc. Edited by F. E. Warren. 4to. £1 10s. net. 

Ancient Liturgy of the Church of England, according to the 

Uses of Sarum, York, Hereford, and Bangor, and the Roman Liturgy 
arranged in parallel columns. By W. Maskell. Svo. 15s. net. 

Monumenta Ritualia Ecclesiae Anghcanae : the occasional 

Offices of the Church of England according to the old Use of Salisbury, the 
Prymer in English, and other prayers and forms, with dissertations and 
notes. By the same. Second edition. Three volumes. Svo. £2 10s. net. 

The Liturgy and Ritual of the Celtic Church. By F. E. 

Warren. Svo. 16s. net. 

Sharp on the Rubric. 8vo. 6s. net. 

Helps to the Study of the Book of Common Prayer. By 

W. R. W. Stephens. Crown Svo. Cloth, 2s. 6d. net ; also in leather bindings. 
Printed on Oxford India paper and bound with the Prayer Book, from lis. 6d. 

The Oxford Hymn Book 

Music Edition. Crown Svo, Ss. 6d. net ; gilt edges, 4s. net ; India 

paper, 5s. net. W^Ords Only, Crown Svo, Is. 6d. net ; gilt edges, 
2s. net ; India paper, 3s. net. 32mo, cut flush, 6d. net ; cloth boards, 9d. net ; 
India paper, 2s. net. 



Ancient Philosophy 

(For editions, translations, &c., see pages 44-47) 

Greek Theories of Elementary Cognition from Aicmaeon to 

Aristotle. By J. I. Beare. 12s. 6d. net. 
Plato's Doctrine of Ideas. By J. a. Stewart. Svo. es. net. 

Mediaeval and Modern Philosophy 
Opera hactenus inedita Rogeri Baconi. Edited by Robert 

Steele. Svo. Fasc. I, De Viciis Contractis in Studio Theologie. 5s. net. 
Fasc. II, Communiura Naturalium Lib. I. 10s. 6d. net. 

loannis Saresberiensis Policratici Libri viii recognovit com- 

mentario etc instruxit C. C. J. Webb. Two volumes. Svo. 36s. net. 
Bacon's Novum Organum, edited, with introduction, notes, etc, 
by T. Fowler. Second edition. Svo. 15s. 

Novum Organum, edited by G. W. KrrcHiN. Svo. 9s. 6d. 

Bentham's Introduction to the Principles of Morals and 

Legislation. Crown Svo. 6s. 6d. 
The Works of George Berkeley, formerly Bishop of Cloyne. With 

Erefaces, annotations, appendices, and an account of his Life and Philosophy, 
y A. C. Fraser. New edition (1901) in crown Svo. Four volumes. £1 4s. 
Some copies of the Svo edition of the Life are still on sale, price 16s. 
Selections from Berkeley, with introduction and notes, for the use of 

Students. By the same Editor. Fifth edition. Crown Svo. 7s. 6d. 
The Cambridge PlatonistS : being selections from their Writings, 

with introduction by E. T. Caiupagnac. Crown Svo. 6s. 6d. net. 
Leibniz's Monadology and other philosophical Writings, translated, 

with introduction and notes, by R. Latta. Crown Svo. Ss. 6d. 
Locke's Essay. Collated and annotated with prolegomena, biographical, 
critical, and historical, by A. C. Eraser. Two volumes. Svo. £1 13s. 

Locke's Conduct of the Understanding. Edited by τ. Fowler. 

Extra fcap Svo. 2s. 6d. 

A Study in the Ethics of Spinoza. By h. h. Joachim, svo. 

lOs. 6d. net. 

Hume's Treatise of Human Nature. Edited by L. A. Selby- 

Bigge. Second edition. Crown Svo. 6s. net. 
Hume's Enquiries. Edited by L. A. Selby-Bigge. Crown Svo. Second 

edition. 6s. net. 
British Moralists, being Selections from writers principally of the 

eighteenth century. Edited by L. A. Selby-Bigge. Two volumes. Crown 

Svo. 13s. net. Uniform with Hume's and Berkeley's Works. 

Butler's Works, edited by W. E. Gladstone. Two volumes. Medium 
Svo or Crown Svo Vol. I (Analogy) 14s. or 5s. 6d, Vd. II (Sermons) 14s. or 5s. 

The Optimism of Butler's Analogy. The Romanes Lecture, 
190S. By Henry SceTT-HoLLANO. 2s. net. 


Recent Philosophy 
Kant's Theory of Knowledge. By h. a. Prichard. svo. es.ed. net. 

The Logic of Hegel, translated from the Encyclopaedia of the PhOo- 
sophical Sciences, with Prolegomena, by W. Wallace, Second edition. 
Two volumes. Crown 8vo. 10s. 6d. each. 

Hegel's Philosophy of Mind, translated, with five introductory essays, 

by W. Wallace. Crown 8 vo. 10s. 6d. 
Lotze's Logic, in Three Books— of Thought, of Investigation, and of 

Knowledge. Translated by B. Bosanquet. Seconded. 2 vols. Cr. Svo. 12s. 
Lotze's Metaphysic, in Three Books— Ontology, Cosmology, and 

Psychology. Translated by B. Bosanquet. Seconded. 2 vols. Cr. Svo. 12s. 

BluntSchU's Theory of the State. Translated from the sixth 
German edition. Third edition, 1901. Crown Svo, half-bound, 8s. 6d. 

Green's Prolegomena to Ethics. Edited byA.c. Bradley. Fifth 

edition, 1906. With a Preface by E. Caird. Crown Svo. 6s. net. 
Types of Ethical Theory, by J. Martineau. Third edition. Two 

volumes. Crown Svo. 15s. 
A Study of Religion : its Sources and Contents. By the same 

author. Second edition. Two volumes. Crown Svo. 15s. 

The Principles of Morals. By τ. Fowler and j. m. Wilsow. svo. 

14s. Also, separately— Part I, 3s. 6d. Part II, 10s. 6d. 

Logic ; or, The Morphology of Knowledge. ByB. Bosanquet. 

Two volumes. Svo. £1 Is. net. 

Lectures and Essays on Natural Theology and Ethics. 

By W. Wallace. Edited by E. Cairo. With portrait. Svo. 12s. 6d. 

Studies in History and Jurisprudence. By Rt. Hon. j. Bryce. 

1901. 2 vols. Svo. £1 5s. net. 

The Theory of Good and Evil. By h. Rashdall. bvo. 2 vols. 

14s. net. 
The Herbert Spencer Lectures. Svo. 1905, by Frederic Harrison. 1906. TheVoluntaryist Creed. By Auberon Herbert. 1907. 

Probability, the Foimdation of Eugenics. By Francis Galton. Is. net. 

1908. Individualism and After. By Benjamin Kidd. Is. net. 
An Introduction to Logic. By H. W. B. Joseph. Svo. 9s. 6d. net. 
Essay on Truth. By H. H. Joachim. Svo. 6s. net. 

The Ethical Aspects of Evolution. By w. Βενεπ. 6s. net. 

Elementary Logic 
The Elements of Deductive Logic. By τ. Fowler. Tenth 

edition, with a collection of examples. Extra fcap Svo. 3s. 6d. 

The Elements of Inductive Logic. By the same. Sbcth edition. 

Extra fc^p Svo. 6s. In one volume with Deductive Logic, 7s. 6d. 



See also Anecdota Oxoniensia, pp. 98, 99. 

Sacred Books of the East 

Translated by various Scholars, and edited by the late 
Right Hon. F. Max Muller. Forty-nine volumes 

An Index Volume (Vol. L) is in the press. 

Sacred Books of India. Brahmanism 
Twenty-one volumes 

Vedic Hymns, Part I, translated by F. Max Muller. Part II, translated 
by H. Oldenberg. Two volumes (XXXII, XL VI). 18s. 6d. net and 14s. net. 
Hymns of the Atharva-Veda, translated by M. Bloomfield. 

One volume (XLII). 31s. net. 
The AS'atapatha-BrahmaTia, translated by Julius Eggeling. 
Five volumes (XII, XXVI, XLI, XLIII), 12s. 6d. net each; 
(XLIV), 18s. 6d. net. 

The Gnhya- Sutras, translated by H. Oldenberg. 

^ Two volumes (XXIX, XXX), each 12s. ed. net. 
The Upanishads, translated by F. Max Muller. 

* Two volumes (I, XV Second edition), each 10s. 6d. net. 
The Bhagavadglta, translated by Kashinaih Trimbak Telang. 

One volume (VIII), with the Sanatsu^atiya and Anugita. 10s. 6d. net. 

The Vedanta- Sutras, with >Sankara's Commentary, by G. Thibaut. 
Two volumes (XXXIV, XXXVIII), each 12s. 6d. net. 
The third volume (XLVIII) with Ramanu^ra's /S^ribhashya. 25s. net. 
Vol. XXXI V— Part I of the Vedanta-Sutras— is temporarily out of print. 

The Sacred Laws of the Aryas, translated by G. Buhler. 

Two volumes (II (Second edition) and XIV), each 10s. 6d. net. 
The Institutes of VishTlU, translated by Julius Jolly. 

One volume (VII). 10s. 6d. net. 
Manu, translated by Georg Buhler. One volume (XXV). 21s. net. 

The Minor Law-books, translated by Juuus Jolly. 

One volume (XXXIII, Narada, BWhaspati). 10s. 6d. net. 

^ Jainism and Buddhism. Twelve volumes 

The Cr aina- Sutras, translated from Prakrit by H. Jacobi. 

Two volumes (XXII, XLV). 10s. 6d. net and 12s. 6d. net. 
The Saddharma-pUTli/arika, translated from Sanskrit by H. Kerk. 

One volume (XXI). 12s. 6d. net. 
Mahay ana Texts, by E. B. Cowell, F. Max muller, and J. Takakusu. 
One volume (XLIX). From the Sanskrit. 12s. 6d. net. 

The Dhammapada and Sutta-Nipata, translated by F. Max 

Muller and V. Fausboll. One vol. (X, Ed. 2). From the Pali. 10s. 6d. net. 
Buddhist Suttas, translated from the Pali by T. W. Rhys Davids. 
One volume (XI). 10s. 6d. net. 



This book is due on the last date stamped below, 
or on the date to which renewed. Renewals only: 

Tel. No. 642-3405 
Renewals may be made 4 days prior to date due. 
Renewed books are subject to immediate recall. 

jHK ' 

1 Q r 1071 Ο r? 


BtC'O to 


APR: ΰι973 5 7 

vote cm jm.*»'H 


I '^'S 8 4 

hUU i i ix-.- 

py^a ^^'* 


(P2001slU)476— A-32 

— iic¥iini*«n 

General Library 

University of California 


LD 21-100m-7,'33