The Journal of philology

Google

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project
to make the world’s books discoverable online.

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books
are our gateways to the past, representing a wealth of history, culture and knowledge that’s often difficult to discover.

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book’s long journey from the
publisher to a library and finally to you.

Usage guidelines
Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the

public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to
prevent abuse by commercial parties, including placing technical restrictions on automated querying.

We also ask that you:

+ Make non-commercial use of the files We designed Google Book Search for use by individual
personal, non-commercial purposes.

and we request that you use these files for

+ Refrain from automated querying Do not send automated queries of any sort to Google’s system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.

+ Maintain attribution The Google “watermark” you see on each file is essential for informing people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other
countries. Whether a book is still in copyright varies from country to country, and we can’t offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book’s appearance in Google Book Search means it can be used in any manner
anywhere in the world. Copyright infringement liability can be quite severe.

About Google Book Search

Google’s mission is to organize the world’s information and to make it universally accessible and useful. Google Book Search helps readers
discover the world’s books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web
ai[http: //books . google. com/|

The Hournal

OF

PHILOLOGY.

EDITED BY

W. G. CLARK, M.A. FELLOW OF TRINITY COLLEGE,

J. E B. MAYOR, MA. FELLOW OF ST JOHN'S COLLEGE,

AND

W. A. WRIGHT, M.A. TRINITY COLLEGE, CAMBRIDGE.

VOL. III.

London and Cambrider ;
MACMILLAN AND CO.
DEIGHTON, BELL AND CO. CAMBRIDGE.
1871.

SOS
S86

Cambridee :
PRINTED BY C. J. CLAY, M.A
AT THE PRESS

CONTENTS.

No. V.

On an Accadian Seal. A. H. Sayce ‘
On the End of the Epistle to the Romans. F. J. A. ‘Hort .
On the Enneakrunos at Athens. Thos. H. Dyer .

On the Lengthening of Short Final Byline | in Vergil H. N ettle-
ship

On neas’ Voyage Round Sicily. W. Everett

On the Chronology of St John v. and vi. J. P. Norris
Note on the ‘ Arzareth’ of 4 Esdr. xm. 45. W. A. Wright
On Lucretius, Book VI. H.A.J. Munro . .

A Theory of Job xrx. 25-=27, C. Taylor

On the History of tho Ravenna Manuerpt of Aristophanes
W. G. Clark . ‘ .

Notes on Thucydides and the Acharnaas of Aristophanes
W. C. Green ° ° ‘

Notes on the Supplices of Eechylun A. H. Wratislaw . ‘
On the Athenian Proedri. EL. Hicks . . ° .
On the Sixth Satire of Persius. Thomas Maguire. . ;
On a Theban Inscription at the Fountain of Dirce. W. E. Currey

161

189

iv CONTENTS.

No. VL

The Epistle to the Romans, J. B. Lightfoot ; ‘
Thought, Word, and Deed. E. B. Cowell . ‘

The Eastern Origin of the Christian Pseudepigraphic Weng
K.H. Palmer. ‘ ‘ ‘

Prof. Munro’s Notes on Juvenal I, 13, and on Aetna 590,
Thomas Maguire

The Roman Capitol, as laid down in 1 Mr Burn’s “Rome and the
Campagna.” Thos. H. Dyer . . ‘

Acts xx1. 37, 38. A. H. Wratislaw e
On Lucretius, Book VI. R. Ellis . .
A Passage in Hidipus Rex. Thomas Maguire
Two Passages in Vergil. Thomas Maguire .

Westphal’s Methodische Grammatik der Griochischen Spruce
Evelyn Abbott

Notes on the Translation of Genesia. Charlee ‘Taylor
Note on Gen. v1. 16. F. Field

An Introduction to Greek and Latin Btymology. By J ohn
Peile, M.A. J. B. Mayor

‘Decadence. J.B. Mayor

Horatians. Horace, Carm. I. 20. H. A. J. Munro 2
Carm. 2.13,14. Panus perhorrescit. J. E. Yonge ‘

On Two Triple Readings in the New Testament. A. A. Vansittart

236

277
278

282

291
327

328
347
349
353

357

THE JOURNAL

OF

PHILOLOGY.

age.

few

ERBaTa her

are NZ ters

Pare 215, > Maniekacrn ves4 Wins. lest

for Weprncich rsqd wer: Poa Tia,

Pare 233, ter Ashiva rad Nike - 1ese

een

em-

the

land of Accad” only. ...-. ruld
seem to refer to the mountainous country to cue ... the

Euphrates. Sumiri (also called Cassi) is apparently “the
plains”; and Dr Haigh has suggested with great probability
that it is the original of "YIY. Dingir, the ancient form of the
Accadian word for “god”, became afterwards Dimir; and Gin-
gir, the Accadian Astarte, is perhaps identical with Gimir “a
foreigner’. The cuneiform system of writing was an Accadian
Invention, each sign being a hieroglyphic representation of the
object for which it stood, gradually corrupted, as is the case
with Chinese, into the forms which we meet with on existing
monuments. Hence, without some knowledge of the language
of the inventors, a full acquaintance with Assyrian, which en-
Journal of Philology. vou. m1. 1

2 THE JOURNAL OF PHILOLOGY.

deavoured to express a Semitic grammar and vocabulary in a
foreign syllabary, is out of the question. Considering the paucity
and scantiness of any Accadian inscriptions, such a knowledge
might have seemed impossible. Fortunately however the bi-
lingual tablets of Assyrian and Accadian, drawn up by the
order of Assur-bani-pal, the son of Essarhaddon, have been pre-
served in a more or less mutilated condition. Had they come
down to us perfect, we should have had, without doubt, a com-
plete grammar of the ancient tongue of Chaldea. As it is,
however, the fragments frequently are broken off just where
their preservation would have been of most importance. It is
necessary to proceed in great measure by the help of induction
and comparison. Hence I have been compelled to relinquish
the design I had originally formed of drawing up a complete
Accadian grammar. The gaps and imperfections would have
been so numerous that I have judged it my best course to take
& single inscription, and to make the philological analysis which
I have attached to it the means of setting forth all the facts of
Accadian grammar which I have been able to get together.
I shall conclude by endeavouring to fix the position of the
Accadian among the recognised families of speech, and so to
Justify such analogies from other languages as I have brought
forward in the commentary. The inscription which I have
selected is one which, so far as I know, has never yet been
published. It differs, also, from the chief part of those with
which we are acquainted, in its not being royal. With the
exception of the tenth line the characters are very legible.
The inscription is as follows :—

(1) (an) mi’s-'su-ta ud-du-[a]

(2) (%sar) id zi-da

(3) na-pal-la-ci-ge

(4) (nam) tsil-lil

(5) (an) il-zi us cal-ga

(6) (?’sar) (Huru)-ci-ma-ca-cu

(7) ci-lum-la gu-za-lal

(8) tur lik-ba-bi-ge

(9) mu-na-(? ban)-sab-ba

(10) (?’sar)-mu......... di ga ca ni

ON AN ACCADIAN SEAL. 3

(11) ga-an-tsil-lil
(12) mu-bi
TRANSLATION :—

“To the god who issues forth in power (Nergal), king of the
right hand in the city of Napalla, for the life of Ilzi the strong
male, king of all the land of the city of (Huru), I, Cilumla, the
throne-supporter, the son of Lig-babi, perform-sacrifice. To
my king, to the end of his days(?) may his name give life.”

PHILOLOGICAL COMMENTARY :—

(1) The first character has the usual phonetic value of an.
The full form is annap, which in the Elamite inscriptions sig-
nifies “god”. The word would connect itself with Zyrianian
yen, Zakute yenem (“god”) and Wotiak tn (“heaven”), inmar
(“ god”, lit. “he who (is of) heaven”). The final -p is appa-
rently a formative, which shows itself in khtlt-p, another Acca-
dian word for the deity. The temple of Anu or Anna (= “the
god’’) at Huruk is called indifferently Bit-Anna, and Bit-khili-
Anna, where the final -p is dropped. Comp. the emphatic
affix -pd of the Finnish (Wot. -pa, Esth. -p, Ost. ap’). The ordi-
nary Accadian word for “god”, however was dingir, which in
later times became dimir. The root is a wide-spread one:
Tartar tengri (“god”), Turk. tangri (“heaven”), Jakut. tan-
gara (“heaven”, “holy”), Fin. taimas, perhaps even Samoiedian
adjaan. It has been borrowed by the Chinese under the form
tien. It is to be noticed that the same law of corruption seems
to have been at work in the Accadian and the Finnish. Before
the plural affix -ene, Dimir becomes dimirri. Sometimes the
+-e is contracted; kharra (“prince”) becomes karrine (“princes”).
(a2) The oldest mode of forming the plural seems to have
been by the repetition of the word (see § 7): thus khar-khar
is interchangeable with kharrine, and dimtr-gal-gal-ene is
the customary way of expressing “the great gods.” The last
example shows clearly the primitive nature of the Accadian.
Each agglutinative affix preserved its full force as an indepen-
dent word. How far the corrupting influences which have made
the Ugrian dialects tend towards the phenomena of an inflec-

1 Cf. Castrén, Ostjakische Sprachlehre, p. 25.
1—2

4 THE JOURNAL OF PHILOLOGY.

tional language prevailed, will be stated in the course of the
paper. The repetition of the root to signify the plural is com-
mon to all barbarous tongues. Traces are even to be found in such
Greek words as da:dadXo, or such Latin words as gurgulio. The
system of “ pair-words” is characteristic of the Ugrian idioms.
Thus “from year to year” would be eszenddrél eszendére in
Hung., jepest japar in Lapp. In Canarese collective nouns are
formed in this way. In Malay again the use of pair words is
very extensive. Often the root is merely repeated, as in the
Accadian plurals of which we are speaking: e.g. api-apt = “a.
scout”, bdésar-bdsar = “very great”. So in Basque we find
traces of a plural in the verbs and in the postpositions. In the
verbs and most of the postpositions the plural is 2¢ or eta, with
which compare the Finnic plural below: in the postposition -z,
however, the plural is formed by reduplication, zaz. It is
noticeable that the plurals are prefixed, like the Accadian ene
in ene-mun. (8) A second and most common method of form-
ing the Accadian plural was by means of the suffix ene. I
cannot discover that this word had a separate meaning of its
own. Ina contracted form -ne it was used to mark the plural
of the present tense. I would explain its origin in the following
manner. In, or rather inni, the 3rd pers. pron., became nt or
ne before a vowel. To form the plural it was reduplicated—
nene, “they”. This was divided, as was the case with the
future of Semitic verbs, and the aorist of Aryan verbs, to make
the 3rd pers. pl. of the present tense. Hence ene or ne came to
be regarded as a plural formative. In one instance it seems
changed to nu according to the law of harmony; ct-nu-cu is
given as signifying “to the places”; and once I have found it
prefixed to the noun; “bricks” being rendered by ene-mun.
The Wotiak uno (“many”) can hardly be connected. We may
compare the formation of the Basque plural by suffixing the
demonstrative (z.e. the 3rd pers.) pron. (y) Another way of
forming the plural is by adding mes, “many”, to the singular,
The same affix makes the plural in Elamite (Third Acheme-
nian). Comp. Jakute myz “collect together”; Wot. myzon “an-
other's”, myd “in several ways”. It is possible that the Ugrian
plurals yas (Zyr.), -yos (Wotiak), -¢ (Fin., Lapp., Tcherem.,

ON AN ACCADIAN SEAL. 5

Mordv., Hung.), -& (Lapp., Mordv., Hung., Ostiak), and -n or -la
(Ost.) have the same origin. According to Prof. Max Miller
“the old Ugric termination of the plural is -as’.”. Analogous is
the Turkish -s or -k, Mandschu se, Buriatian -da(?). Initial m
or 6 (first changed to w) is constantly dropped. In Hungarian,
mek forms the plural of nouns of relationship. I am not sure
whether or not the Accadian possessed a dual. In the Assyrian
inscriptions words expressing dual ideas, such as “hands” or
“feet”, had the numeral 2 affixed. If this were of Accadian
origin, it would have been pronounced kats. This might pos-
sibly be compared with the dual termination possessed by the
Ostiak, Lapponian, and Samoiedian, kan in the former, ga in
the latter. Castrén, however, derives these from the enclitic At
“also”, which bears a strong resemblance to the Accadian cs
“with”, and hence “also”, and to a form urv-ct, which must
mean “cities” in a phrase quoted below (§ 4), and may be a
dual referring to Huru and Lar'sa, the only cities mentioned in
the inscription. In the Taic Kassia the plural is expressed by
the preposition ht. The word for “god” was also often used
with the signification of “prince”, and in this sense placed
before royal names. So in Basque, jauna=“lord”, and jainco
= “God”.

(2) Aft's-'su may be regarded either as a single or as acom-
pound word. It is used by Assur-izir-pal (B. Jf. S. Vol. I. pl.
28), who speaks of “the gods Ussur and Nergal”, (Si-dun “he
who marches before”) “who have exalted” (or “chosen” ac-
cording to Dr Hincks) “power”. The first character, with the
value miz means “strong”; if sounded sit it =“to measure”,
if cistp it = “a foot”. The second character usually signifies
“to magnify”. The number of compound words in Accadian is
considerable, as is the case with Zyrianian, according to Castrén.
The predicate generally follows the subject, not only in the case
of pronouns (as in the Ugric and Tartaric languages, which
herein differ from the Bhotiya, Lohitic, Tamulic, Chinese, &c.),
but also in the case of adjectives which are conceived as de-

1 Vast Results of the Turanian Researches” in Bunsen’s Phil. of Univ.
Hist. Vol. 1. p. 460.

6 THE JOURNAL OF PHILOLOGY.

pendent genitives. This rule of position is found always in
Basque and occasionally in a few dialects of the Ugrian, Cauca-
sian, and Taic families, though the general order of the sentence
in these languages is the reverse. It is, however, only the
natural order to be followed by such idioms as affix the pro-
nouns and use postpositions. And Accadian adopts the uni-
versal principle of the Allophylian dialects of placing the object
(and generally the subject also) before the verb. A familiar
example of what I have said above is to be met with in ct-a “a
ford,” literally “a place of water”. Without doubt many of the
words which at first sight appear to be compounded really are
not so; the combination has merely taken place in the group
of ideographic characters which were used to express the vocable.
Thus aral: “death” was expressed by ideographs which re-
spectively denoted the ideas of “house,” “land” and “corpse”.
Still the order of the signs would follow the customary order of
thought. In thus placing adjectives and genitives after the

a)

subject the Accadian 1 18 imitated by the Basque. Here qgun on
(2) (8) (1)

would be “ good day”, eched gizon onen would be “house of this

man”. I find some instances, bowever, in Accadian in which
the converse arrangement of the sentence, found among the
Altaic dialects, occurs. Thus kha-luba or kha-dibba is “ fish-
pond ”, kha being “fish”. So again the words for “country” are
regularly affixed to the proper name, as will be seen in § 8.
But in this case it is possible that the proper name was not
regarded as a dependent genitive, the defining words being
rather co-ordinate. And when united with the suffix ta so as
to form a compound postposition, ct by tmesis precedes the
personal pronouns (see § 3 (1)). On the other hand, the Acca-
dian, like all the so-called Turanian languages, makes large use
of postpositions. It is not until we reach the boundary of the
Taic family that we meet with prepositions. In another cha-
racteristic point, also, the Accadian has the same usage as the
Altaic and Tamulic languages, though herein it differs from the
Basque. The relative clause, or rather the governing verb-
participle, is always post-fixed to the noun. Thus Rim-(‘Sin)
says “dimir-gal-gal-ene (Huruk)-ci-ma - - - kat-mu-cu_ banin-

ON AN ACCADIAN SEAL. 7

“ses-a”, “the great gods who have filled Erech into my hand”,
and a common formulary with which Burna-buriyas and his son
Curigal’su end their accounts of temple-building is “which to its
place was given back”, It will be noticed that in this last
instance the Accadian departs from the usage of the Altaic lan-
guages, in which the relative clause precedes the word or words
on which it depends. The Accadian arrangement is, however,
more in keeping with the general order of words. A great pro-
portion of the compounds are formed by thus post-fixing the
participle: e.g. na-zika (literally “mark-making”) is “ memo-
nial”, gis-pa for “ gis-pa-a (literally “speaking-wood ”) is “scep-
tre”, tr-gar (lit. “judgement-making”) is “ruler”, ca-ztk (lit.
“ speech-making”) is “seal”. In short, the determined element
in &@ compound word, if viewed either as a participle or as a
genitive, is regularly post-fixed. In the case of the participle,
the Accadian, though agreeing, as I have said, with the Altaic
languages, differs from the Basque, which prefixes the participle,
e.g. u-argia “the moon” (lit. “destroying-light”). However,
- as we have seen, Accadian usage is already beginning to waver.
On these grounds, therefore, miz-zu may be considered as a
compound, zu being the participle, for zu-a. The final vowel,
however, may belong to the simple word. Accadian roots are
often lengthened in this way. Thus- gurus =“a valiant one”,
guruse =“ valour”; uddun “go forth”, uddune “the not-going
forth”. So we find gic “setting”, but cuga or gissu “sunset”;
babbar “rising”, but babbara “sunrise”, “silver”. In many
cases, however, the abstract noun is marked by the feminino
prefix «': e.g. cus “resting”, tt-cusu “rest”; dur “fortress”,
vd-dur “dwelling”; mar “abode”, tt-mar “ brick-house”. This
sf seems originally to have been the ideograph of a “comb”,
and so to have stood for “woman”; it also signified “prince”
and “hand”. A feminine noun was also expressed by the prefix
ca: as in the word “loving”, which is written man, god, making
for the masculine, and this ca followed by the monograms of
god and making for the feminine. For the primitive significa-
tion of ca see § 13. In Basque, as in the Taic and Lohitic

1 It, however, may here mean simply “‘one”’; see § 6.

8 THE JOURNAL OF PHILOLOGY.

dialects, gender is marked by sugixes which denote respectively
male (arra, cp. Accadian kharra) and female (emea, cp. Accadian
um). The same want of the distinction of gender characterises
the Ugric and Tartaric languages. Cw is another preformative
of the same kind. Thus dara and cu-dara signify “name”;
sag is “head”; cu-sag “supremacy”.

(3) Za is a common postposition. It is generally used
in the sense of “from’’; e.g. é-ta 1b-tan-uddu “he drove from
the house”. Its original meaning, however, was “in”. Thus
Khammurabi says of himself: banuv é-par é Parra Lar'sa-ma-
ta, “he raised Bit-Parra, the Temple of the Sun, at Senkereh”.
The postpositions mark out the Aceadian as a member of the
Turanian family of speech. In modern Allophylian dialects
these postpositions are naturally the most conservative part
of the language. They are often almost the only words which
do not convey a distinct and independent meaning to the mind
of the speaker. Hence in comparing the Accadian vocabulary
with those of existing idioms, we ought first to take the post-
positions, So far as I have been able to discover, the Accadian
possessed the following :—

(1) Cit or cota “with”. With the personal pronouns cif
suffered tmesis; e.g. ci-mu-ta “with me”, ct-zu-ta “with thee”.
Sometimes the final ta was dropped altogether, and then ct
preceded the word it governed, as though it were a nominative:
ct Huru-ci-ma-ca stands for “with all the land of the city
Huru”. Dr Hincks has well explained the origin of this word.
It is a compound of ct (for cings or gin “land”) and the post-
position ta, and would literally be “in the place of”. With
true Turanian instinct the Accadians never lost sight of the
independent signification of ci; hence its employment in some
cases before its case. (tt must be distinguished from ge, which
is often expressed by the same character. Comp. Zakute khonu
“field”, “broad place”, and kidng “broad”. So in Wotiak intyin
(from tnty “place”)=“pro”; and the Basque alde-an “near”
is from alde “place”.

(2) Cu “for”. Cu is very common: e.g. garnam-bi in-
nun-cu in-sem “his soul for the child he gave” ; aria-da-cu ban-
semmu “into the river he throws her”. Before the plural

ON AN ACCADIAN SEAL. 9

affix, final -u is dropped, and an euphonic a inserted between
the last consonant of a word and the c-sound ; e.g. snnunac-ene
‘for children” or “things”. Cu often stands for the infinitive
of the substantive verb (cp. Mong. bii-kii “to be”): “to (be)
with him’”’, for instance, is ct-cical-bi-cu or ci-sium-bt-cu ; where
the final ta of cit is dropped before another postposition, and
the paraphrase for ‘‘self’’ is to be noticed. Crcal is literally
“strong place”, t.e. “assistance” and siwm (?) 1s “service”.
These paraphrases for the personal pronouns are common
throughout all the Allophylian tongues, and give rise to an
infinite number of pronominal forms. Jn Japanese there are
no personal pronouns properly so-called; a number of words,
originally meaning “body”, and the like, are used for all the
persons indiscriminately. The employment of cu to denote the
infinitive perhaps explains the Turkic infinitive affix mek, for
which the Mandschu has simply me. The future participle in
Basque is formed by the suffix co or go, which is properly a
postposition signifying “belonging to” (eg. Burgosecoa “of
Burgos”, nongo “where?”). Cu originally meant “to be near”,
“to establish”, in Accadian, hence “to serve”, “bea slave”,
“to capture”. Ci-cu “a seat” is literally “a place firmly-
established”: cicu-garra (lit. seat-making or existing) signifies
“one who is close by”. Postpositions of similar sound are
widely found. The dative is expréssed in Tamil by ku, in
Telugu by ki, ku, or ko; and the Basque ca “to”, “on”, or co,
go “in”, are naturally referable to the same root.

(3) Ga “in”, “having”. This is a word of extensive use,
as it is the only mark whereby adjectives can be distinguished
from substantives. Thus cal is “strong” or “strength” or “to
be strong”, with ga added it becomes “powerful”. Enwu ci-ga
is “lord of countries”, enw huru-ga is “lord of cities”. So us-ga
is “sea” ; cp. Mag. viz, Fin. vest “water”. Ga primarily signified
“to bind”, and is used for the “ yoke of a chariot”. It seems
to claim kindred with the Tartaric ga (Yakute gha, ga, ka), the
dative-suffix. In Basque gana is “belonging to”, “among”,
“with”, gattc “for”, “on account of”.

(+) Gab (2) “over-against”. I am not sure what was the
phonetic value of this character when it was used as a post-

10 THE JOURNAL OF PHILOLOGY.

position. It was sounded also khus, and in Assyrian was fsat
In many cases the Assyrians adopted the words employed by
their Turanian predecessors; and even, it would seem, several
of their prepositions had an Accadian origin. If tsat were the
pronunciation of the Accadian postposition, we might compare
Basque tzat “for”. Jb certainly was the usual word for “op-
ponent” or “opposition”, and fic was used in the same sense.
Mun-tic or mun-ib, for instance, equally meant “front of brick ”.
Saggut, again, had a similar signification, as well as urugal
or uragal.

(5) Ge, “over”, “taking”. Ge like gab denotes “ battle”,
“opposition”. Gt also signifies “foundation”, “strong”, and
with » added becomes the root which means “to dwell”, “es-
tablish ”, as in in-gin “he placed”, in-gine “he places”. Lastly,
the root = “to take”, and this, I fancy, is the origin of the
postposition rather than the other meanings. Many examples
of its use are found, e. g. enu cingt Accad-ge “lord of the land of
Accad”’, saggadhu imtete-na-ge “on the top of his person”. In
some cases I have found it replaced by cu: thus Gungunuy
calls himself “king ” Huru-ma-ca-cu “ of all the land of Huru.”

(6) Gim “like”. With this word we may perhaps com-
pare the Basque cintzoa “suitable”, or the postposition kin
“with”. An-gim is frequently met with, explained to be “like
the god Anu”.

(7) La “among”, “for”. This postposition is of rare oc-
currence; indeed I am inclined to think that it is merely an
euphonic alteration of ta. It is certainly interchanged with ta in
the title of Nergal with which our inscription begins. We have
it in ucu-mal-ene “among my men ”, though perhaps the middle
character here is to be read mur: certainly in the 3rd s. present
of the verb which signifies to “dwell” it is followed by the
syllable ra. The postposition occurs again in an inscription of
Burna-buriyas, who calls the Sun “the lord in the land of
Sippara”, Larsa-ct-mati-la ; and in the legends of Amar-(? Sin)
li follows the name of Mul, where other inscriptions would
have lal-ge. So, also Nit-(Sin?) says that he is “the power-
ful male, the shepherd...created by Mul”, IJnucié-ls garrd.
In Wotiak and Zyrianian ly forms the dative, len (T'cheremis.

ON AN ACCADIAN SEAL. II

lan) the genitive, les (Teherem. letz) the ablative. Jakute ly,
ny, dy (Tartar lat, changed to das after a consonant) builds
the adverbial case, while lyn (or dyn, Turk. atlah) signifies
“with ”.

(8) Lal “under”. JZal is sufficiently common in Acca-
dian inscriptions : e. g. Burna-buriyas calls the sun enu gal ancia
ci-lal “great lord of heaven and earth”. al is primarily “fill-
ing”, hence “deep”. Cognate words are the Ost. tel, (Mag-
yar tele (“full”), Wot. tyro (“fill ’’).

(9) Inbis “the midst”, bis-ta “in the midst”. The
compound a-lb, “ water-surrounded ”, is the usual word for a
“piece of ground”’, formed in accordance with the same con-
ception that has made a-caga or a-dega (lit. “ water-on-the-top-
of”) “the surface of the ground”. Immine Anna an-libista
translated “the flood of Anu in the midst of heaven”; and
libis-ga is as common a compound as /tis-ta. Libis seems
originally to have signified “near”; cp. Wotiak tupato “to
make suitable”.

(10) Na “of”,“on”. This postposition is found in such
expressions as si-ni-na “in his sight”, cicu-ani-nam “on his
seat”, where the na is lengthened according to an euphonic
law before a following n. The origin of na is to be sought in
the demonstrative na, the formative element of the personal
pronouns in Accadian as well as in the Ugrian and Tartaric
languages, and which appears in the Jakute mmnd@ “there” or
“thence”, described by Bohtlingk as the locative of a lost
pronominal stem tn. The postposition na has many analogies.
In Elamite (Third Achemenian) the genitive is formed by the
affix na (or inna after the plural ending -tp), the same element
being repeated in the genitive of the lst pers. pron. hu-ni-na.
The genitive, again, is made by en in Mordvin. and Lapponic,
by -n in Finnish, by -tn in Turk., by ns in Mands. and by yin
in Mongol., while in Jakute (which has no genitive) na or yna
is the mark of the locative, nan of the instrumental (like Wot.
yn, Zyr. én), and na, da or la of the acc. indef. The acc. def.
is made by affixing + or y, which 1s preceded by n after vowels.
This n Bohtlingk traces to the pronominal in, just as in Basque
the nasal of the dative (ons) and of one form of the nom. and

12 THE JOURNAL OF PHILOLOGY.

gen. (onéc, and onen) of the demonst. pron. is not an euphonic
interpolation, but a veritable part of the old pronoun. In
Tamulic adjectives are formed by the affixes ana, na, ni, in;
and Tschuvashian gives us from man “I”, man-yng “mine”,
and man-yng-yng “of me”. Basque forms its genitive by
suffixing -en, and n (pl. etan, compounded with ga in gan “in”,
and with ki in kin “with ”) is the locative postposition. After
a vowel of the t-order na in the Accadian became tm: hence
the character which had this value is used to express the
Assyrian preposition adi “to”.

(11) Ra “to”. We meet with many examples of this
postposition in the inscriptions, After an t-vowel ra becomes
tr (e.g. Dumugu ‘sar (?)-anir “to the moon his king”). In
the same way after an u-vowel, it becomes ur; thus Rim-Sin (?)
has Nintp ‘sar (?)-mur “to Ussur my king”. Analogously,
after an a-vowel the form of the postposition would be ar.
This euphonic law seems to apply to all monosyllabic affixes
which terminate in d short: it is not applicable, however, to
such as end in 4 long, like ta; though even this, as we have
seen, sometimes loses its final vowel when compounded with
ct. One of the meanings of the root ra is “to inundate”.
This may be the origin of its use as a postposition. Compare
the Basque ra (pl. etara) “to”; ronz (pl. etaronz) “towards”.

(12) Ruv (2) “according to”. I am not sure what was
the proper pronunciation of this postposition. The character
had the further values of as and ina, and both these values
represented Assyrian prepositions. One or other of these latter,
if not both, were in all probability derived from the Accadian.
If the first, we may compare the Wot. oz, Zyr. ddz, the ter-
minative affix (e.g. ta dyroz “up to this time”), or the Basq.
# (pl. zaz), Mong. etze (?), which form the ablative. In Basque,
also, the affix tz forms several of the adverbs of time. In the
Accadian itself assan is “high”. If ina be the word, we may
compare the Basque adverbs which are built by the postposi-
tion n. Instances of the use of the postposition in question in
Accadian are to be found in gubtagubba-ruv “for the being
fortified”, and in pakh-ruv (?), which is translated “much” or
“strongly”. I am inclined to think that the correct pronun-

ON AN ACCADIAN SEAL. 13

ciation of the word is as, since this is the value which the
character bears in all the Accadian inscriptions in which I have
found it; ¢.g. in inaddunas “they have caused to go”.

(13) Ta, te, “to”, “at”, “in”, “from”. I have already
given examples of this word. It is also used to form the ad-
verbial case ; ama-ta is “like a whirlwind”. A preceding dental
is assimilated and dropped: thus mad-ta (for mada-ta “in the
land”) becomes simply ma-ta. I do not know whether ¢a or te.
was the original form of the word: in the old inscriptions it
generally occurs under the form of te, a root which signified
“a basement”. We must compare the Basque di, dic, or tic
(pl. etatic, or etaric with the euphonic r) which denotes “from”.
We have & in Wot. od in Zyr. for the penetrative case ; in
the Tartaric dialects dan, or tan, Osmanli den, after hard con-
sonants, forms the ablative. The locative is expressed in
Mands., Mong., and Turk. by da, de, and du.

(14) Zig “across” “in front”, “behind”, “over”, “on”.
These various meanings are all to be traced to the two primary
significations of the root “to cross” and “a front”. Thus cra-tig
= “across a place of water”, t.¢e. “a ford”, ru-tig “front of a
front”, ma@-tig “top of a ship”, mun-tig “before brick”, cicu-tig
“on a seat”. Comp. Jakute tyz “what is before one”, Turk.
tush “opposite” (like tash “rock” compared with Accadian tag).

(15) Tug “for”, “to”. This is the participle of tug “to
have”, apparently identical with the Basque dugqu “habere”,
whence comes the common verb ukhen or ucan, the initial
dental being dropped, as in many other cases. An example of
this postposition is kharra-tug, which is rendered “to a man”.

These postpositions may be compounded one with the other,
and so produce a new set of postpositions. We have already
mentioned libis-ga and libis-ta. One of the most common is
ge-lal “up from under”, as in ar-gelal “up from under the
district” te. ‘‘a generation”. We find also lal-ge: the Moon-
god, for instance, is called tu sag Mul-lalge “eldest son of
Mul” So, again, in the mythological tablets we have Maruduq
tur sak Inuci-ga-ge “Merodach, eldest son of Hea”, tur-mes
Inuci-ga-ge “the children of Hea”. This composition of post-
positions is common both in the Ugrie dialects and in Basque.

I4 THE JOURNAL OF PHILOLOGY.
According to Prof. Max Miiller it is also to be found in

Canarese.

Before concluding this account of the Accadian postpositions
I must not omit to mention an affix which, like ruv(?), is used
for the adverbial case. This is bt or khas, I do not know which
was the correct pronunciation, but I fancy the latter. Thus
makh-khas is “much” or “supremely”, gal-khas is “greatly”,
susa-pallal-khas is “by way of punishment” (=takma “a
penalty”), susapallal being translated zamaru. Comp. the
Tartaric kiintz “daily” from kun « day”, and such temporal
adverbs in Basque as notz, maiz’. One curious fact about these
postpositions is that in the earliest inscriptions they are wanting
almost entirely. The position of the words, as in Chinese, deter-
mines the grammatical relations. Indeed the age of an inscription
can in great measure be settled by the absence or the frequency
of these connecting suffixes: and their occurrence in the inscrip-
tion which we are at present considering is the reason that -in-
duces me to regard the king addressed as not identical with the
monarch of the same name whose brick-legends we possess (see
§ 10). Even when postpositions became plentiful, their primary
meaning was as little obscured as it is in the Taic dialects, which
have in like manner developed a set of prepositions. A post-
position in Accadian was nothing more than a participle: if
used as a substantive, it followed the rule of substantives and
stood before its case. Several of these prepositions are to be
found. They are not indeed prepositions properly so-called, but
nouns followed by a genitive which answer to the prepositions
of European languages. Thus s+ “the eye” or “sight” (like
Assyrian pan, 235) was used to express the idea of “ before,”
e.g. st-gut “before an ox,” st-dun “going before”; mukh-bi
would be “over him,” mukh-st or bar-st or anna-st is “above
the eye”, though pi-anna is “ above the ear”, where anna is used
as a participle (2.e. a postposition). Anna may be the fuller
form of the demonstrative, and hence identical with the post-
position na, as may be seen by comparing the last example with

2 If the character is to be read khas, as, the initial guttural] being dropped,
it may determine the value of ruv to be as is common.

ON AN ACCADIAN SEAL, — I5

tag-na “a high stone” (lit. “on the stone”), though I should
prefer to identify it with anna “high”, whence annap and the
god Anna or Na (M3). How natural this primitive form of the
sentence is may be seen from its being observed in Taic and
Malay, which employ prepositions not postpositions (except in the
Malay acc. which affixes the Sanskrit pdda) e.g. di ndgri “in
the country”; while the genitive, without any mark of case
being attached, stands after the governing word. The Accadian
order of words is also observed in these languages by the place
of the adjective, which follows the substantive, by the plural
being formed by an affixed substantive, by the possessive
pronominal suffixes being (in Malay) merely the personal
pronouns added to the noun, and above all by the personal
pronouns preceding the root in the conjugation of the verb, the
different tenses being distinguished by affixes or infixes. The
words, however, which denote a difference of gender are affixed
not prefixed.

(4) Uddu-a or udduna is the participle of uddu or uddun
“going out”. If the root is derived from dun “to go” (like
sidun), a nasal must be inserted: however as I do not know of
any similar sense in which ud was used, and as we find udda
“fire”, I should prefer to read uddu and not uddun. Uddu before
its case is a substantive, ¢.e. a preposition, eg. uddu guza
Lar’ sa-ci-ma “on the throne of the land of Lar’sa”. The use of
the participle is very extensive in Accadian, as it 1s in all the
Turanian languages. The Turkic present (in -er) is really a
participle, and relative sentences in Basque are formed by the
same means. The participial termination is one of the few
portions of Accadian grammar which has disguised its derivation.
Arguing from the analogy of other Turanian languages it ought
to be the demonstrative pronoun. In this case the long a
would be corrupted from an original an, like Basque verbal ad-
jectives in 1, a, u, primitively na, nu, n, du. In the latter
language, again, a the 3rd pers. pron. seems to have been
originally an, the source of the postfixed article a or ac. To
this, again, we must refer the Basque participial ending in -an,
-en, -n. In the Tartaric dialects the gerundive in -a seems to
ally itself with the participles in -at and -an (cf. Mordvinian part-

16 THE JOURNAL OF PHILOLOGY.

endings van, vat, vit). In Jakute, according to Béhtlingk’, many
adverbs and postpositions are nothing more than this gerundive;
e.g. yla, “of”, from yl “to take away,” cytta “like”, “with”,
from kytyn “to join oneself.” This is exactly analogous to what
I have shewn in the last section to have been the case in
Accadian. We find many instances of the use of the participle
in the inscriptions to denote a relative clause. Thus dda uru
kanig-tug nt-bat-e uru-ci-mada-nu-cu wmmingarra-d is “the
river of the city for a canal he opens, which for the cities of his
country was fully made”; and a common phrase is ct-bi-cu
nen-dib-a “which to its place was brought back”. Sometimes,
however, for the sake of greater clearness and emphasis, the
relative clause was expressed not by the participle alone, but by
the participle and the character which stands for “man’’. This
was pronounced gum (sometimes gumma), though ucu was the
generic term. (Uruci?), for instance, calls himself “king of the
land of Accad”, gum é Mul-lal in-zig-a “he who has built the
temple of Mul”; and another king has “gum inbisi-cu ci-bt ne-
dib-a ‘‘he who has restored his country to affluence” (?). This
use of the word “man” for the relative pron. is common in the
Turanian dialects. In Basque the demonstrative is often added
to the participial clause. With gum or gu (=cu, comp. cuga
and gic) we may compare the Basque relative cen-a (giz-on
“man”), Zyr. kod, Tcherem. kudy, kit, Wot. kud, kin, Mordv.
kon, Ost. khot, Fin. ku, Elamite akka, appa, Tartaric kha, khas,
khan, kim, (kvzt ““man”). Cf. § 16.

(5) The first character of line 2 denotes “a king.” Its
value is doubtful, but it seems to have ended in -m. Owing
to a fracture of the tablet, the first sign which represented its
Accadian pronunciation in one of the bilingual syllabaries is lost,
and only the last sign -m is left. In another place I have
found this character followed by ma and translated by the
Assyrian sarru “king”. Apparently, however, it was also pro-
nounced sar even in Accadian times, since it is the first compo-
nent of the name of ‘Sargina, an ancient mythological hero as
well as an early king of Babylonia. ‘Sar-gina would be “rex

1 Ueber die Sprache der Jakuten, p. 214.

ON AN ACCADIAN- SEAL. 17

primus,” though when the name was adopted by the Assyrians
they transformed gina into their own word cinu from f\5. ‘Sar,
changed into sarru, was probably one of those many mono-
syllabic vocables which the Semites borrowed from their
Accadian predecessors and ‘Semitised’ by investing them with
a triliteral form. The bilingual tablets afford us numerous
instances of this process; the loan-words appearing in many
cases to have been early adopted and so to be common to most
of the Semitic languages. In some instances these loan-words
have been made stems for further derivations; thus the Accadian
kharra “man”, under the form khirru has been made to yield,
in accordance with the genius of Semitic speech, khiratu or
khirtu “woman”. It is possible that most of the monosyllabic
roots found in: the Semitic languages came from a Turanian
source. Like semi-civilised peoples generally, the Accadians
had a great number of synonymes for “king”.

(6) Jit‘“hand”. This is another Accadian word which has
passed into the Semitic tongues. Equally common in Accadian
to express the same idea is kat (or as it seems primarily to have
been sounded kattakh). Compounded with ti “to raise”, this
becomes katts “to seize”, a verb in which, with true Turanian
desire to keep each root clear and distinct, the objective case is
separated from ti in the tenses, and the personal pronouns
placed between them; e.g. kat-nen-ti “he took”, kat-bab-ti-e
“he takes.” Kat and it or yat are seemingly identical, the
initial guttural becoming lost, through an intervening form in
kh, as in other Turanian dialects. Thus Tcheremiss kol (“die”)
is the Mag. hal, Ost. had, Basque tl; Zyrn. kul (“hear”), Ost.
hud, and (by an interchange of the guttural and labial) pet
(“ear”), Zyr. pely, Mag. ful, Mord. pile, Basque belarria, Accad.
ps. It connects itself with the Turk. 1, Jakute «li “ hand.”
It also signified “one”. This origin of the numeral “one” is
carious. It takes us back to a time when the savage signified
his first idea of number by holding up his hand. Jt or kat is
clearly allied to akat, the base for “one” in the Ural-Altaic
languages according to Professor Schott. Hence Lapp. akt,
Fin. yht (which resembles the Accadian form very closely),

Esth. uts, Basq. bat, Ost. ot or st (and 2), Zyr. dttk, m6. egy,
Journal of Philology. vou. 111.

18 THE JOURNAL OF PHILOLOGY.

Mord. vatke. The final guttural in the last three words may
claim kindred with Accadian gina “primus”, Mong. nege, nikka
(“one”). The Accadian word for “two” has similar Ugnriec
affinities. It is kats, Esth. kats, Fin. kaks, Zyr. kyk, Ost. kat,
Magy. ketté; Yak. ikki, Turk. (y)iki, Mong. kuyar. The Basque
has borrowed the Aryan numeral bt. Out of this Basque has
formed bide “a road’’; just as in Accadian kats or kharan was
employed to represent the same idea. From kharan comes the
name of the city which commanded the high-road to the West.
We do not, unfortunately, possess the names of the other
Accadian numerals. Si, however, seems to mean “five”, and
esa “fifteen”. Comp. Fin. visi, Esth. wivs, Tcherem. vis, Magy.
6t, Samoiedian sam-ltk, Tchuvashian pilik, Yak. bids, Turk. besh,
Basq. bortzi or bost, Mong. tha-ba and ta-bun, Tung. sunja.
Sanabi perhaps is “forty”, and us or sus “sixty”. The latter
would remind us of the Basq. set (“six”). Ordinals were
expressed by adding either nalla “being” or gan (also ganva)
with the same meaning. With the latter, originally the demon-
strative (which seems shortened to na in gina), comp. the Tartar
ordinal formative n or in.

(7) Zida “right”, opposed to gupu “left”, connected with
im zidi “the north wind.” This word possesses the formative
da which is used extensively in Accadian. Kudur-Mabug is
called es-da mada Martu “citizen of the West”, es being
explained “house”, one of the monosyllabic roots adopted by
the Assyrians under the form esu “a building.” The affix
appears, again, in ma-da “country”, more frequently written
ma simply, a root found in most of the Turanian dialects (Zyr.
and Wot. mu, or Esthon. ma for example). The Elamite has
murun, and to this Tcheremiss adds da (muldnda) as in Accadian.
So again khir “to repel” becomes khirda “an enclosure”. The
suffix occurs in the brick-legends chiefly in the group of charac-
ters which represents the Semitic kiprat arbat “the four
peoples”. First comes the determinative prefix of divinity, as
little sounded as in the Elamite (an) ctg “the sky” (literally
“the divine blue”), or in the Accadian (an)e “heaven” (lit.
“the divine hollow”). Then follows ar “a region”, then the
individualising complement da. This is succeeded by the

ON AN ACCADIAN SBAL. 19

monogram of “four”, with ba sometimes added. The latter
addition has induced cunealogers to regard this group of charac-
ters as pronounced in a Semitic manner. In the bi-lingual
tablets, however, the group is given as Accadian, without any
suffix ba, and translated tupukatu irbittu or ciprat irbittu, the
correct form, since trba would not be in accordance with the
rules of Semitic grammar. If therefore, the Semitic origin of
the title in question is still insisted upon, we must consider it as
one borrowed by the Accadians from their Arab neighbours’.
An early intercourse between the two races is evidenced by
their common stock of traditions (of the Flood, the Garden of
Eden, the Tower of Babel, etc.), which seem to have had an
Accadian origin; not to speak of the Arab dynasty, which
according to Berosus held sway in Chaldea*. The termination
ba in the numeral “four” does not seem to connect itself with
Turanian analogies, although we have in the Mongolic dialects
tirba, durban and dorban. Another word which exemplifies
the use of da is a-da or aria-da “a river” from a or aria (comp.
Basq. ura) “water”. The suffix has an individualising, demon-
strative force; which reminds us of the Samoiedian affix da as
ldta-da “the board” from ldta “board”, which Castrén has
shown to be the possessive 3rd personal suffix’. Da or ta and
so (Tcherem. ty or tyda and seda) is the demonstrative pron. in
the allied dialects, like tt in Jakute; which reappears in
Ost. teu “he”, and Sokpa tha. The Sokpa would have the
same origin as the Buriat ene, which again refers us to the
Tataric ol.

In Basque d and t represent the 3rd pers. pr. sing. and
pl. in the verbal forms, and the article suffix is nothing more
than the demonstrative, which has probably lost an initial
dental, as is the case with ukan originally dukan, or with aurra
“child”, which seems to belong to the same root as the Ac-
cadian tur. Upon such grounds, therefore, I conclude that

1 Naram-'Sin, who bears a Semitic
name, and who has ciprativ arbraiv
(pl. masc.), belonged to the Assyrian
dynasty.

8 Many words were doubtlessly bor-
rowed mutually. Gabdiri, for instance,

one of the many Accadian terms for
S$ - =
“mountain” seems clearly , \a».
> De Affixis Personalibus Linguarum
Altatcarum p. 11.

2—2

20 THE JOURNAL OF PHILOLOGY.

da in Accadian also was primitively a demonstrative, a bye-
form of na set apart for a special purpose. We find da in
one place translated “a male’, and da-ri “a child”, while
du-ri is rendered “before a man”. The frequently-occurring
adjective, again, which means “long”, is more often buda
than bu. When at “a father” is to be specialised, da is
affixed, the preceding dental being assimilated. Thus we
have tur ad-da-na-ra adda-mu nu-mia bannendug, “a son
has said to his father, thou art not my father”; and Kudur-
Mabug is called by his son addae-mu, “my father”, where ae
was probably pronounced as a diphthong. So again len (?) is
“a memorial”, len-da “the memorial’; and the collective
mulu-da “subjects”, is thus formed from mul (“lord”). The
last example shows the way in which this termination came
to represent the plural, as in (an) Arda. Comp. the Buriat.
pl. -da. Accada itself is an instance of the affix, being de-
rived from aca “exalted”, which is also used in the sense of
“weighing”, t.¢. “raising” the scales.

I will here give a list of the other formatives which are
possessed by the Accadian :—

Ba: “side” or “part”, e.g. ca-ba “side of the mouth’,
dur-ba “part of the fortress”.

B: e.g. gub (“fortify”), dub (“tablet” compared with du),
ab (“ month”, compared with ai ‘“ moon”).

C: eg. gic (“difficult”).

Ci: e.g. gusct (“red” cp. Basq. gor).

E, I, A, U: e.g. me (“battle”), sizse (“sacrifice”, cp. Zis
“excellent”), ge (“conflict”); gemt (“the sea”), tst and 48
(“hill”), arali (“death”); ma (“ship”), é-a (“house”, generally
é only), wmte-va (“self” compared with imtete and imtez), dara
(“name”), tura and tur (“little”); abu (“flood”), enu (“lord”),
usu (“body”). ;

Kh: e.g. dikh (“ stone” compared with dub and du).

La: e.g. galla, gula and gal (“great”), din-la and din (“a
family”), mal (“ abode”, compared with mar “ dwell”, “ reign”).

Ma, m: 6.9. (nam) din-ma (“a family”), lamma (“a colos-
sus”), seslam (“a race”), titnum-ci (“ back”, 3.e. “ west country”,
but titnu “ behind”), sem and se (“to give”).

ON AN ACCADIAN SEAL. 21

N, Na: eg. agan (“supreme” compared with agazt), gin-
gina (“earth” compared with cingt), cingt (“land” compared
with ct), cin (“a work”): un (“man”) added, as in um-un
(“ prince,” “son”), uk-un (“ offspring”).

P: eg. annap from anna “high”, khilip and khili (see § 1),
tssep and cip (“leader”, “prince”), the latter perhaps from ct.

R, ra: e.g. mar (“to inhabit”, cp. ma “country”), zicura
and sigaru (“below”, “prince”, compared with zicum), parra
and par (“the sun”, compared with pa “to shine”), zanaru
and zana (“high”), barra (“high”, compared with bar “ top”).
So dingir (“god”), Jakute tangara, if, under the form dimir
compared with Jak. tammakh, or tammala “a drop of water”,
and the Accadian dim “water”, would show the final -7r to be
merely formative’. The word would then be derived from
the idea of ratn, just as Indra is from indu “drop”. So the air-
god, Wir or Aftr-mir, is drawn from mir “rain”: mir-mir
“ brightness” is a fresh derivative from the name of the god.

S: eg. libis and lb (“place”), amas (“nail”), sis (“bro-
ther”, whence Sisct, a name of the moon-god), zizse (“sacri-
fice”): us (“male”) added as in ucus (“man”, “soldier”), dhus
(“ soldier”), gurus (“hero”), cus (“a brother”).

T, ta: e.g. dugud (“heavy”), hurud (‘iron”, Wot. kort),
vara-ta and vara (“ancient ”); te (“raise”) added as in aganatets
“the raiser”.

Vowels are also prefixed: e.g. num (“high”, Wogul numan)
and enum or enuv (whence perhaps enu “lord”), egir (“after”
compared with gir “beginning”), wcu and guv and cus, ugu or
ugun (as in uguna-mi-cu “to future days”, ugunu-cu “for a
day”) “a day”, compared with Turk. kun (“day”), Jak. kun
(‘‘day’”, “sun”), Basq. egun; ? Ugrian nunaornunal. Gutturals
are inserted as in dingtr, gingir: and I have found cilam and
cebalam (“an opponent”).

I have found instances in Accadian of all the principal
consonantal changes common among the Turanian races. The
dental and the labial are changed in (an) gallam-la uddua,
a title of Negal, which is also written (an) gallam-ta uddua

' Cp. Turkic (Kazan) dengiz, deniz, ‘‘sca", Mong. denggis,

22 THE JOURNAL OF PHILOLOGY.

“he who goes forth in might.” The labial and the nasal,
again, are frequently interchanged: e.g. algubba for an-gubba
“he fortifies.” A and 7 take each other's place as in mal
and mar. The initial guttural undergoes the usual altera-
tions: c and g are constantly shifting, as in cuga and gic, gub
or guy and cu’u (“precious”), gum aud ucu. Still more common
is the interchange of c and k, e.g. ucu and uku (“army”);
and of all these with kh as in ga-gar “may he do”, kan-len
“may he proclaim”, kha-bara-uddu “may he go forth”, ukhbs
“clothing”, cuba “clothed”. Through the help of this strong
aspirate the guttural is lost altogether; e.g. kha-baran-duszu
“may he not take”, a-banin-duz “may it take him”. Hence
perhaps kan “to be”, and al (Turk. ol), have the same root.
The guttural and the labial also are interchanged, e.g. dtkh
and dip or dup “a stone”. Perhaps, too, dhus and cus imply
an interchange of the dental and guttural. Jf and 6 pass
into one another (ba being sometimes written for ma “country”,
and man-sem standing for ban-sem “he gave him”), and are
liable to be dropped altogether; thus mus is given in one place
for us. Z and g further are confounded (see § 10). Finaln
before m probably becomes m: thus ma is sometimes added
to kan, the mark of the ordinal numbers. 7’ is assimilated to a
following d.

The lengthening of words is a common feature of Accadian.
I shall speak presently of the use of this means to form the
present tense of verbs. The final consonant is doubled, and a
vowel affixed of the same class as those of the root. Thus bar
becomes barra, miz becomes mizzu, kur becomes kurra. As I
said in § 2, this is a very frequent mode of forming the abstract
substantive. It gives intensity to the idea by compelling the
mind and the voice to dwell longer upon it. But a further and
stronger way of producing the same result is to repeat the whole
word, the final consonant with its vowel being attached to the
second member. Thus bar-barra is “height.” This is pro-
perly, as we have seen, an emphatic plural like kur-kurru-tsy
“thy enemies”. It is by a similar, though inverse practice, that
the Tibetan and Lohitic idioms turn nouns into verbs; ¢.g. ndg
“black”, ndggo “it is black”: and the same means are used in

ON AN ACCADIAN SEAL. 23

Jakute and other Tataric languages to form diminutives (6g.
kiisl koluyd “a little sea”), adverbs, and gerundial expressions
(as izen ten “after long travels”). This alliteration has pro-
duced also many substantives and verbs in Mongolic and Ugric
dialects.

(8). Napalla-ci-ge “In the city of Napalla”. Napalla was
a Chaldean town. Ci when added to the name of a place in
writing was probably not sounded. The syllabaries translate it
by “place”, “land”, and “fortress”. The full form of the word
was cing, which is always written in Cong: Accad “the land of
Accad”, where the determinative preceded its genitive according
to rule, and was phonetic. Gin-gina, literally “the lands’ (see
§ 7), signified “the earth” generally. Besides ci, another
shortened form of the word, gi, was in use. Cv or git was
probably the original root, to which the formative n (the de-
monstrative) was added.

(9). (Nam)- tsillsl “life”. The usual form of the word is
tsilla or tstl, the vowel being affixed when the word is closely
attached to an enclitic, and the syllable lengthened by a redu-
plication of the consonants before the short vowel of the enclitic.
Thus Kudur-Mabug says (nam)-tsilla-ni-cu va (nam)-tsil * tur-
mu (sar?) Larsa-ct-ma-cu mu-naninzrg “for his life and for
the life of (? Nit-Tsin), my son, king of Larsa, I built them.”
Il is merely a formative, as in din-la, possibly connected with
the preposition, and probably a form of the demonstrative (see
§ 7). The word is another instance of the attempt to intensify
by increasing the final syllable. Followed by e, the affixed
vowel becomes i not d, as in ganamga tsilli nenu gisin ganelgar
“let the mouth, during the life of the king, act”. Nam is the
non-phonetic complement which generally precedes the word.
It is the determinative prefix, also, of nouns of relationship.
Nam inaa is rendered ‘‘crown of the lord”, and nam-nam “pro-
claimer”. Compounded with gar (“make”) it means “soul”.
Comp. Wot. nim “name”, nimo “renowned”; Mag. nem “a gene-
ration”, nemt “descended”; Elamite numan ‘“‘a race”. In the
sentence quoted above ganamga (perhaps for ca-namga “mouth-
having-speech”) seems to mean “mouth”; and under the signi-
fication of “crown” an original meaning of “enclosure” may lie

24 THE JOURNAL OF PHILOLOGY.

hid. Nam was also pronounced tsim, and this appears to suggest
& primitive relationship to tsil, which upon its side may be con-
nected with nal “esse”. Initial n is very liable to change or
loss in the Turanian dialects. In Turkish, when compared with
cognate dialects, it is either dropped altogether (as in eng “very”,
Mong. neng) or changed to j or d (as in jadi “seven”, Mand.
nadan, dil “tongue”, Mag. nyelv), while the Turk. 7 as often as
not expresses an original z (e.g. jaka “border”, Mong. Zacha.
jemek “eat”, Mand. tse-me). Another form of the verb in Ac-
cadian is al or alet (Turk. ol-mek, Mand. o-me, Fin. olla, Esth.
ollema, Basq. adi in such forms as nadin, nindeque, &c.) Ap-
parently its origin must be sought in the demonstrative.

(10). Jlzt or Ilgt. The characters which in Assyrian repre-
sent zi and gi have the same form in Accadian. The royal
name is preceded by the ideograph of “god”, here used as the
determinative prefix of a prince. This king can hardly be iden-
tified with an Ilzi, of whose brick-legends we possess a small
number, and who is mentioned by Nabonidus as the son of the
founder of the great Temple of the Moon at Huru, a monarch
whose antiquity is very great, and who is in fact the earliest
Chaldean sovereign with whom we are acquainted. He has been
called Urukh and compared with the Orchamus of Ovid, but
upon insufficient grounds. The first element in his name means
“lion”, pronounced in Accadian [ik or liccu. Ur (WS) was an
Assyrian value, the Accadians expressing ur by a different
character, as on the brick of Rim—(?Tsin), Col. 2.1.7. The
second element is the title of a god, the pronunciation of which
is unknown. In one place we find it sounded zicuy. Now the
legends of these early princes are marked by, the absence of
postpositions, a sign of antiquity which is not applicable to the
present inscription. For this reason I am disinclined to identify
the two Ilzis. The Chaldean kings were fond of bearing the
names of their predecessors: thus among the kings known to us
we have three Nimgirabis and two Kurgaltsus. The following
are the inscriptions of Ilgi the son of Liccu*:—(1) One found at
Tel-eid near Warka which runs; “(To) the lady of the land of
Mar, his Lady, Ilzi, the powerful male, king of the country of
Huru, king of the land of Accad, the Temple of Guk-é her high

ON AN ACCADIAN SEAL. 25

place, I built”. (2) Two from Mugheir:—‘Elzi, the powerful
male, king of Huru, king of the land of Accad”; and “Elzi, the
powerful male, king of Huru, king of the land of Accad, the
Temple of Cisaq, the temple of his high place, I built”. (3) On
a black stone:—“(To) Gingir, Lady of the Temple of Anna, his
Lady, Elzi, the powerful male, king of Huru, king of the land
of Accad, the Temple of Anna, her place, I founded; its great
fortification I built.” This Elzi and his father, however, were
not the most ancient sovereigns of Chaldea. They were pro-
bably the first who made Huru their capital city; but before
their time Hurug, “the city” as it 1s written in Accadian, must
have asserted its pre-eminence. A large number of these rulers
of Huru have preserved their names in brick-legends: besides
them we have a list of royal names belonging to one dynasty
which seems to have been Elamite. At all events the names,
which are translated into Assyrian, show a dialective distinction:
thus mili is “man” (in Accadian mulu and muluda), khali is
“great” (Accadian gal gula), cit is the “Sun”. Lastly, we come
to kings with Semitic names and in some cases with Semitic
inscriptions. I once endeavoured to show that these represented
the Assyrian dynasty of Berosus which began B.c. 1272 (cp.
Herod 1. 95), its leader being probably Khammurabi (an Elamite
name by the way) the Semiramis of Berosus. These Assyrians
will be the Casdim or Semitic “conquerors” of the Old Tes-
tament, who descended from Assyria and imposed a Semitic
domination upon the primitive Turanian population.

(11) Us,“amale”. Also mus and vus. Comp. Zyrianian
ydz “people ”.

(12) Cal-ga, “powerful”. The adjective formed from cal
or cala “strong”, by the postposition ga.

(13) Huru-ci-ma-ca-cu, “(for) all the land of the city of
Huru”. Huru was the name by which the city was called in
Semitic times: it does not follow that such was the Accadian
pronunciation. The name is written with the characters which
denote “name” and “house”, and the Accadian title may
have been as different from the Semitic one as Ca-dimirra or
Din-Tir (“ Homestead of the Tower”) was from Bab-tl. Huru
was the city of the Moon-god, in opposition to Larsa or Zi-par

26 THE JOURNAL OF PHILOLOGY.

“the abode of the Sun”. (Cui is added according to the rules
of Accadian writing, as the determinative: and it is in this
way that we can demonstrate the Accadian origin of Nineveh,
Asur, and the other great cities of Assyria. Ma follows to
express the whole country to which the city of Huru gave its
name; and then comes a character, which we find applied to
ma sometimes, though generally it is omitted. Ca signifies
“a mouth”, hence “a gate” (cf. Turk. cap), for which a
separate character has been set apart; it 1s also the determina-
tive prefix of “woman” thus early distinguished as “the
talker”. Compounded with zig for ztga it stands for “a seal” ;
and with ga postfixed forms an adjective of extensive use, ¢. g.
ar-caga “a people”. Cacaga (“mouth-speech-making”) is “a
command”; and the plural caca signifies “face” (like 6°35)".
Hence we get the word used for “in the face of”, “above”,
like khut and cun expressed by pa “speech”; as in ca-uzga
“top of the water”. This meaning becomes adjectival by the
addition of ga ; a-caga is “ water-above ”, 1. e. “the surface of the
ground”, Afa-ca would therefore be “the whole face” or “ sur-
face of the country”, in other words “all the country”. This
explains the employment of cag in the signification of “all”,
the guttural being reduplicated as in bab, sts, gic, &c. Cag
always preceded its noun, as in the longer form cagabi alsakh
“for all bliss”. So khirda is “an enclosure”, “a crown”, ca-
khirda “a circuit”. It is possible that the primary meaning
of ca itself was “enclosing”, “encompassing”, like the lips
This use of ca, however, in the signification of “all”, may have
a different origin. In Jakute didn, the participle of did “say”, |
Orenburgh dican (cp. Acc. dug), is affixed to substantives,
adjectives, and adverbs, to express that the whole thing as-
serted is absolutely the fact and nothing else, and may be
translated “namely”, “extremely”. At the end of dependent
sentences it stands in the sense of “that”, “to wit”. Similar
is the employment of dvb, div in other Tataric dialects, of kelan

1 Caca, the plural, may be inter- ation, ‘‘speech”, which is expressed
changed with ca, as in du-ca or du- by the character which has further va-
caca “a memorial”; caca itself alter- lues of mis, sit, rid, lag, and kal.
nating with papa, of the same signifi-

ON AN ACCADIAN SEAL. 27

(“ saying”, “ that”) in Mongolian, and of annu or yennu (“ say-
ing”) in Canarese. Cu concludes the whole sentence. It will
be noted that the postposition in Accadian follows the last
word of the sentence to which it belongs.

(14) Guza-lal “throne-supporter” or “councillor”. Guza
(whence the Assyrian cussu) means “a seat” (cp. Elamite kada,
Basque cudtra, coya ?)

(15) Tur Ltg-babt-ge “son of Liq-babi”. Tur, properly
“small”, is used in the general sense of “son”, like the Elam-
ite tur, Mord. tsur. Nam-tur has the general sense of “ child”
(see § 9), while gt signified “very small”. It is curious that
both tur (es in Tartan) and gi also meant “prince”. The first
syllable of the proper name denoted “a lion”: liccu is trans-
lated iibbu. The postposition attached to the genitive after
tur is not common: tur in Elamite is distinguished as a strong
word, being placed before the governed noun in opposition
to sakri which comes after it.

(16) Muna(nis ?)-sabba “I offer sacrifice to him”. Sab is
rendered by the Assyrian saramu (to burn in sacrifice) ; it also
signified “to heap” or “ fill”, as in sab-gal “a mound”, sab-tur
“a threshing-floor”. The third character in this word is un-
certain: I have not been able to identify it with any known
sign. Judging from analogy, however, it ought to contain a
nasal, probably also a sibilant; and it may be compared with
a character one of whose values is nts. It may, however,
be ban.

This verb introduces us to the most important and cha-
racteristic part of Accadian Grammar, the pronouns and the
verbs. It will first be necessary to treat of the pronouns. The
personal pronouns are: (1) Alu (“ Ego”) and tdbi-(duru ?), Gen.
mina (“mei”), Dat. dab (“mihi”), pl. ma (“nos”); (2) Zu
(“tu”), and tz, and (7) mun, pl. Zunene (“vos”); (3) Ne or ene
(“ille”) or tn or bs or abba, Acc. mt or min (“se”), pl. nene
or is (“illi”). “Most Turanian languages”, says Prof. Max
Mitller, “besides the usual personal pronouns, have produced
a large number of polite or conversational pronouns, such as
‘servant’, ‘Elder Brother’, ‘Sister’, ‘Blockhead’, &c. Their
pumber becomes smaller with the progress of civilisation and

28 THE JOURNAL OF PHILOLOGY.

literary culture. Hence but few traces of them remain in the
Tamulic, and hardly any in the Ugric branch’”, The same
may be said of the Accadian. Here we find a compound tdbi-
duru used instead of the first pers. pron. mu, from which the
dative is formed as in an-dab-site “he measures out to me”.
Owing to a defect in the tablet the reading duru is doubtful;
it is curiously like the Malay dirt “self”. We may compare
the Accadian dara “name”. Jdbi may be his “hand” or
“slave”. So, again, mun if used for the 2nd pers. must receive
a similar interpretation. M-n seems used for all the persons
alike: it forms the genitive of the first person, the postposition
na being affixed to mu which is shortened to mi, and hence in
the conjugation of the negative verb mal for man is employed
as the nominative, u being changed, according to rule, to a,
while mun is the prefixed dative; it appears as min in the
sense of the 2nd pers. after a negative, and perhaps as mun
in the instance quoted above; and in such cases as mt-nt-gtr
or min-ni-gir “he gave it”, it is used for the objective case of
the 3rd person. It would seem originally to have been merely
a demonstrative. In Japanese there are no words specially set
apart for the different pronouns; vocables expressive of the
ideas of “slave”, “body”, and so forth, being used for all the
persons indiscriminately. In Malay and Taic these represen-
tatives of the pronouns are very abundant for the first and
second persons, though they are not used indiscriminately. The
Basque preserves the same phenomenon in the various verbal
forms, distinguished by the difference of the incorporated pro-
nouns, which are employed according to the rank or age of the
person addressed. With mu, mina must be compared Fin. ma
(obj. mind), Esth. ma, minna, Lapp. and Wot. and Mordv.
mon, Zyr. me, Tcherem. mtn, Ost. ma (loc. mana), Magy. en,
Samoiedian man and modt, Elam. hu (Gen. mi), Basq. nt, Mong.
and Mands. 6: (gen. mint), Ouigur. man (gen. maning), Jakute
min or bin. The n would be the demonstrative. Idi, dab
may be compared with ¢ the incorporated first pers. pron. nom.,
and i or td the same pron. dat. of Basque verbs. Ms, the

1 In Bunsen’s Outlines of the Phil. of Univ. Hist., p. 465.

ON AN ACCADIAN SEAL. 29

plural, follows the usual Allophylian rule which makes the
pronouns alone form their pl. by a modification of the base,
instead of by an affix. This occurs even in the Taic dialects,
while the Malay pronouns afford the sole instances of a pl.
met with in the language. The Ugrian idioms give examples
of the same fact in the Ist and 2nd pers., most also in the 3rd:
thus for the plural of the lst pers. we find Fin. me, Esth. mete,
Lapp. and Wot. mt, Mord. min, Zyr. mt, Tcherem. md, Magy.
mi, Sam. me. Basque gives us a new root gu. In the Tataric
languages this rule does not hold, but it reappears in Mandschu
be (from bz), and sue (from sz); and in Mongolic (Buriatian),
which gives us in the 2nd pers. ta (from tscht).

Zu, or tz as it appears in verbs before a consonant, is the
Fin. sa, Esth. sa or sinna, Tcherem. tin, Wot. and Mordv.
ton, Zyr. and Magy. te, Sam. tan and tod: (pl. st), Basque zu
and ht (and c in verbs), Turk. sen, Jakute dn for zdén, Mands.
st, and Mong. zt. If mun be a genuine word, it would remind
us of the Basque incorporated dat. fem. in “tibi”, the only
trace of gender to be found in the language. Zunene “you
+ they ” is interesting, as finding so many analogies in Turanian
languages. The Basque zute, the nom. pl. of the incorporated
2nd pron. is compounded of zu (“tu”) and te (“illi”) exactly
as is the Accadian. Béhtlingk resolves the Jakute bis-tg (“vos”,
Turk. biz) into bin+zan (“ego+tu”) and dz-1gt (“vos”, Turk.
siz) into zan+zan (“tu+tu”)’. The Buriat. bida (“nos”),
seems to be “ego +ille”.

The 3rd pers. is properly the demonstrative, which originally
began with a guttural. Its usual forms are ene, wn or an (the
latter after an a-sound), which become nz or ne when preceded
by a vowel. The pl is another instance of the primitive mode
of forming the plural in this class of languages, nene is “ille
+ ille”. Comp. Basq. a (and the demonst. on-ec) and in verbs
the ace. d (“illum”) and nom. te or ate (“illi”), Fin. ne
(“illi”), Zyr. nya (“illi”), gen. ny-laén, Tcherem. nind (“illi”),
Esth. ta, temma (“ille”) and neet, nummad (“illi”), Jakute
kint (“ille”), Turk. ol (pl. an-lar), Buriat. ene, Tungusic -n.

¥ Ueber d. Sprache d. Jakuten, p. 168.

30 THE JOURNAL OF PHILOLOGY.

The second form bi which is equally: singular and pl. refers
us to the Basque be which forms the poss. bere (“his” and
“their”) as well as be-r-au (“himself”). Comp. too, Turk.
bu (“hic”), Jak. by and ba (“hic”), Samoi. pu-da (“his’,
=“he+his”), Fin. pt or vt in verbs. Bt becomes, accord-
ing to rule, 1b, wb, ab, and ba, as well as abba. We find
mt or min prefixed to the verbal nominative, sometimes, to
denote the accusative. It may be a bye-form of bi, but it is
more probably an independently developed demonstrative. In
the Taic idioms man “he” seems to have its source in annat
(“this”), annan (“that”). There is no distinction of gender
in the Accadian pronouns. This applies equally to the Basque
and the Ugric and Tataric languages, thereby distinguishing
them from the Tamulic.

The Pcss. Prons.:—These are postfixed, as in the Ugric,
the Tataric, the Mongolic, and the Tungusic, the Taic and the
Malay; the reverse being the case in the Tamulic, Lohitie,
Chinese, and Caucasian Bhotiya, as well as in Basque, except
in the case of the vocative. The Accadian possessives are
merely the personal pronouns placed after a noun, instead of
standing alone, or of being prefixed asin verbs. They are, (1)
mu, (2) zu, (3) nt or na or ant (after a vowel) and bz, (4)..., (5)...,
(6) nene and bt. The simple pronouns following their noun
denote the possessives in the same way, in the Taic, and Malay,
the Tungusic, Mongolic, Tataric, and Ugric, though the final
vowel is always dropped in the latter class of languages and
generally in the Mongolic and Tataric. In the 3rd pers. b¢
there is no more distinction of number than there is between
the sing. and pl. of the 3rd pers. in Basque, Finnish, Ostiak,
and Buriat.

The demonstrative pronouns :—These may all, as I have
already indicated, be traced back to gan or kan, which some-
times occurs instead of mi(n) as the prefix of the 3rd pers. acc.,
e.g. gannib-tugtug “he possessed it”, and which is probably the
origin of the substantive verb gan or kan (like Turk. ol-meg
and ol). The guttural is still found in the Jak. kint “he”,
with which we may compare the Elamite khi “this” and khe
“that”, as well as khir “him”. An inscription of Curi-galsu

ON AN ACCADIAN SEAL. 31

has khu-mun-z1q instead of the usual mu-na-zig or mu-nan-ztq,
and this khu for kha, according to the law of vowels, I should
be inclined to refer to the demonstrative, final n being omitted,
as in mt- for min-. Other forms of the demonstrative in Accadian
are na and nam, which at once take us to the 3rd pers. pron.
and the genitive postposition. (Cf. § 3 (10).) |

The relative pronoun :—This, as I have said above (§ 4), is
gum or cui (“aman”). Asin the Basque, the relative is used
only with the participle, the pronouns being pleonastically pre-
fixed to the verbal form, just as if no relative had been ex-
pressed. The Basque phrase cefiac min egin diden “which has
made me ill”, where cez-ac is the relative, d- the prefixed nom.
“jlle” and n the participial ending, is an exact parallel to an
Accadian sentence. Of the Tataric and Mongolic dialects Os-
manli alone has developed an independent prefixed relative
kim or ki, probably of the same origin as the Accadian. In the
other dialects this relative is still an affix, ki or gi (e.g. naghor-
de-ki “which is in the lake”), used like the Mandschu -ngge,
as in aracha-ngge (lit. “written-having”) “qui scripsit”, ms-
ningge (“ mine-what-is”) “das meinige”, ininge (“ his-what-is ”)
“das seinige”; the latter re-appear in Mongol. miniige or
ekoniige, Osmanli mininght. These terminations may, however,
go back to the Accadian postpositions ge or ga. At any rate
the interrogative, in Mong. kes, Esth. kes, ke, Hung. kz (and
kiht “quicunque”), Fin. kuka, Lapp. ku, gi, Wot. kin, has the
same root as guv.

The reciprocal pronoun :—This is barta-bi, barta-bi-cu being
“with” or “among one another”.

I should explain the word as compounded of barta the infin.
of bar “to bind”, and bi “their”, so that its literal meaning
would be “ their combination”.

The indefinite pronouns :—We find, first, udbab, “any one”
if that is the mght reading. Udbab-cu is “in any case”. The
derivation may be td “one”, and the reduplicated form of bs
which occurs in bab-ac “he has made”, su-bab-te-e “he takes”,
bab-zig-ine “they raise”. Another indefinite pron. 1s bamu, as
in a-bamu-ran-sem “let no one give”.

The pronouns introduce us to the verbs, which are little else

32 THE JOURNAL OF PHILOLOGY.

than the bare roots with the pronouns attached. These are
generally prefixed, in contradistinction to the use of the noun
with its possessive. Thus khir-mu would be “my writing”,
mu-khir “I wrote”. This is the simplest kind of grammatical
machinery, and is justified by logical relations which make the
person primary in thought in the verb, and secondary when
used as a possessive. Modern Turanian languages have ad-
vanced beyond this primitive stage of mere juxtaposition, and
the more polished tongues, those of the Ugrian group namely,
have corroded the pronouns almost to the form of inflectional
terminations, and have moreover affixed them not only in the
case of nouns but also in the case of verbs, with two important
exceptions. These are Basque and Tungusic. The Tungusic
idioms are the least developed of all the Altaic languages, and
are therefore likely to have best preserved the original forms
of agglutinative grammar. In Mandschu, as in Accadian, the
simple position of the pron. before the root creates a person of
the verb, bi-thege, si-thege are “I dwell”, “thou dwellest”, like
Accadian mu-tug, tz-tug. Among the tribes of Nyertshinsk,
however, Castrén found that affixes had been added even to
Tungusic verbs to distinguish the persons, and the Mongolic
shows no traces of prefixed pronouns. Here, as in other Tura-
nian languages, the possessive and verbal suffixes are kept
distinct, while an attempt is made to restrict roots to being
used either as verbs or as nouns alone. Basque is the only
advanced language of this family which has preserved the
original position of the pronouns. The auxiliary tzate has the
nom. pr. always prefixed, the other auxiliary observing the same
rule in the imperfects and the conditional. The present tenses
of this last auxiliary, however, have it postfixed, and the same
indecision is already marked in Accadian. Usually, as I have
said, the subject pronoun precedes, but we find not only in-
semmu “he gives”, twn-sem “he gave”, but also semmu-nin,
sem-nin, not only tn-gur “he restored”, but also gur-nin, not
only tn-male “he dwells”, but also malenin, not only ntn-segs
“he heaped up”, but also segi-nin; while inu-mu is regularly
“T am lord” as well as “my lord”. Dibdtb-ne, again, is “they
bring back”, and ztku-na seems to mean “he makes”. So, too,

ON AN ACCADIAN SEAL. 33

ne-garrinna is “he throws”, but garrina-zu “thou throwest”.
The Accadian verb thus shows the primitive mechanism of the
agglutinative languages, and marks out the stage of development
already attained. The Taic idioms (like the Chinese) prefix
the verbal pronouns ; so too do the Bhotiya and Lohitic, though
here the possessive is also prefixed asin Tamulic. The same is
the case with Malay.

The most distinguishing feature of the Accadian verb is the
incorporation of the pronouns. Thus tn-sem “he gave”, tn-nan-
sem “he gave him”. The pronouns used are those of which we
have already spoken, the two forms of the 3rd pers. ix and bz
being equally common. There are, besides, two sets employed,
one for preceding the nom. pron., and one for being inserted
between the nom. pron. and the root. The most frequent
are, necessarily, those of the third person. This is in full
nan in the sing. nantn in the pl. but nan often becomes
merely na, as in mu-na-zig “I built it”, or s.mple n as in
ban-sem “he gave it”; in some cases this n is even omitted
altogether, or rather assimilated to the following letter; thus
ba-bat “he slew him” stands for ban-bat. The plural xanin
occurs in the legends of Kudur-Mabuq and Khammurabi, refer-
ring to the temples those kings had built; but it may, after
ba, be used. as a lengthened form of the singular, e.g. ban-
nanin-khir “he wrote it”, where the first nasal represents the
long d. The second nasal in the form nen-sem is merely eu-
phonic, though it may denote the accusative (e.g. nen-khir “he
enclosed it”) if the verb be preveded by a word which ends in a
vowel other than 1, or in the semi-vowels m or v. The rule is
that any vowel (except 2) or semi-vowel m or v requires the
succeeding pronoun, if it be not the second form of the 3rd
person (51), to begin with a consonant, although we find mukh-
bs an-de-e “he strikes him”, but here, perhaps, a ba has
dropped out of the text (comp. tl-b% ban-tsiz “its foundation he
strengthened”). The second form of the 3rd pers, however,
preferably begins with the vowel if not immediately followed by
an incorporated pronoun. This must always happen before a
following ¢: thus we must write 1b-tan-uddu, tb-tugtug, 1b-turrt.
On the other hand, d may have the nasal before it. At the

Journal of Philology. vou. 111. 3

34 THE JOURNAL OF PHILOLOGY.

beginning of the sentence the vowel-forms, in, an, un are em-
ployed. These are regulated by the last vowel of the word
immediately preceding, unless this be a post-position, or a pos-
sessive pronoun. The second form of the 3rd pers. pron. is
preferably used in an intensive sense, translated by the Assyrian
pael and the secondary conjugations which insert & The full
form is abba, but this becomes bab, when followed by the verbal
root and preceded by a short vowel. If long ad, however, pre-
eedes, ab is used. Before an incorporated ptonoun the initial
vowel falls away altogether, or coalesces with the final vowel of
the preceding word if the last syllable of the latter be open.
So far as I have observed, 7b always occurs before t, whatever
the preceding vowel-sound may be except short d. In only one
instance is ¢ preceded by a nasal, and then the verb 1s passive
(an-ta-dudunmu"). Besides the incorporated accusatives (a)n-,
~nan, (a)nnan-, -nanin- which all require b(a), not ab, +b, or ub,
this secondary form of the 3rd pers. may be also used with the
incorporated pron. of the lst pers. In this case, however, the
latter pronoun will belong to the first of the two sets of pro-
nouns I ‘mentioned above, those namely which are prefixed, not
inserted. The secondary form of the third pers. pron. may be itself
incorporated. Thus while ga-gar for gan-gar is “let him do”,
gan-eb-gar is “let him do it”. We even find a combination of
the two forms in bannab-lal-e “he weighs out it", and in .gah-
tn-ban-tsil “let him give life to him”; though here I should
rather explain the form as the incorporation: of two pronouns,
one. for the accusative and the other for the dative, the dative,
as in dab “mihi”, preferring the form with b, Bannab-lal-e,
therefore, would not be exactly parallel to bannan-dug, but
rather banna-b-lal-e “he weighs it out for him”, while the 3rd
nasal in gan-in-ban-tsil would be euphonic. As in Basque,
there is no difference of number expressed by the pronouns,
except in that of the first person. Jn, nin, nanin, and bt are
all equally singular and plural, and when incorporated their
number can only be known by a reference to their object. But
this can never be obscure as the object is always expressed.

! The nom. pron. here is probably influenced by the double d of the root.

ON AN ACCADIAN SEAL. 35

When the number of the subject is to be signified, the plural
affixes are attached to the root. Of these I shall speak pre-
sently. The incorporated pronoun of the first person, which I
have as yet found only in the dative, is dab; e.g. an-dab-sit-e
“it measures” or “counts for me”, and Amar- (?Sin) says that
he is “king of the strong foe”, bab-dab-kurri-a, “who was
hostile to me”. I have not come across the pronoun of the
2nd person.

Besides these incorporated pronouns, there is another set
which is prefixed to the subject pronoun instead of following it.
These all belong to the form m-n, and thus seem, like the
demonstrative gan when used in the same way, to have retained
some consciousness of their originally independent signification
which may be referred to the root min, minna “size”, “great”
(cp. Jak. manga “ great”). The 3rd pers. is min, contracted to
mi; e.g. min-in-gub “he strengthened it”, min-in-zu “he added
it”, min-tn-sem “he gave him”, mi-nt-tt “he enslaved him”,
The first pers. is mun, as in mun-nab-zige “he strengthened
me” (where the n is doubled to express the length of the
preceding syllable), and khul-mun-s-ib “he greatly adored (?)
me”. The last example shows us how the pronouns were
incorporated when the nominative pronoun followed, instead of
preceding, the root. This system of incorporation, so character-
istic a feature of the Accadian, is not to be found in any of the
languages of the old world, with two marked exceptions. These
are the Basque, and more obscurely the Mordvinian. In the
latter language the verbal terminations are: Sing. Ist pers.
-m-ak (= “ me + tu”),-mam (=“me + ille”),-m-isk (=“ me+ vos”);
2nd pers. -t-an (=“te+ego”), -nz-at (=“ille+te”), -d-es
(= vobis + illud”) ; 3rd pers. -ze (= “ illud”), -n-k (“illud + vos”) ;
PL, Ist pers. -m-tsk (= “nos+ tu”), -m-ta (= “nos+illi”). In
Basque, the system of incorporation prevails even more exten-
sively than in Accadian. The numberless verbal forms are
distinguished from one another by the difference of form or of
position in the pronouns which they contain though now through
lapse of time greatly corrupted, and disfigured. Like the Accadian
the Basque requires the objective pronoun as well as the object
itself to be expressed. The incorporation of the pronouns,

38—2

36 THE JOURNAL OF PHILOLOGY.

however, only takes place in the first auxiliary, though tzak in
the imperfect seems to contain an accusative, ni-n-ze-n, ht-n-ze-
n, etc., the final nasal being the sign of the tense. The second
n may, however, be merely euphonic. Now it is in the first
auxiliary that the nom. pronouns are generally postfixed:
hence we must compare forms like n-a-zw (“ me-habes-tu”) and
min-du-cu-n (“me-habuisti-tu”) with such words as mun-nab-
giga or khul-mun-si-b. Except in forms like 2-td-d-a-n (“tu-
illum-miht-habuisti”), the Basque avoids bringing the nom. and
the acc. and dat. pronouns together, the nom. being generally
relegated to the end of the root. This is not the case, as we
have seen, in Mordvinian, where the forms bear a close resem-
blance to khul-mun-s-ib which without the inserted intensive
sibilant would be khul-mun-nib. In the ordinary Basque verbe,
other than the auxiliaries, the nom. pron. is postfixed, the acc.
being prefixed, e.g. d-aki-t “illud-scit-ille”, the plural being
denoted by tzt inserted between the root and the nominative.
Besides the machinery of the pronouns, the Accadian verb
employs a number of auxiliary words to denote what in other
languages would be expressed by tenses and conjugations. In
one instance, however, it has recourse to a genuine internal
modification of the root. As in most Turanian idioms, the only
radical distinction of time that is known is that between the
past and the present. The past tense has the bare root; while
the present is marked by a prolongation of the root, the last
consonant being doubled and a corresponding short vowel added.
This dwelling wpon the idea is the most natural way of express-
ing present time. We find the same contrivance in the
Tibetan and Bhotiya dialects; thus from jyed “‘to do” we get
nga jded-de “I am doing”. It is similar to the mode in which
abstracts are produced (see § 2). In the Ugric languages,
again, the difference between the two tenses is set forth by
a difference in the suffixed pronouns, which are shortened in
the preterite. So, too, in Tamulic the shorter personal termi-
nations are used for the past, the fuller for the present. The
infinitive is denoted, as in other Turanian languages, by a post-
position. This is ta with which the Basque verbal subst. in -te
may be compared. Thus the negative root mta becomes mia-ta

ON AN ACCADIAN SEAL. 37

“non-esse”. The postposition cu is also used for the same
purpose, e.g. ct-cidan-bt-cu “to (be) with him”. This is identical
with the Mong. ku as in bu-ku “to be”, and the Basque charac-
teristic of the destinative ko as in izaiteko “to be”.

Of the participle I have already spoken (§ 4). It is ex-
pressed by the affix 4, which is not to be confounded with d@ the
sign of the present sing. of a- verbs. This d@ is added after a
preceding short d, and it follows vowels of all kinds. Thus
garra-@ “which is made”, cus-va for cusu-d “rest”, cacava
*‘said”, (Le. “the end”) where the first vowels denote the pass,
The plural seems to be formed by areduplication of the ending;
e.g. gut dadunat “bulls who go frequently”. A participle, or
rather a nomen agentis, which cannot be used as a relative, may
also be formed by an external addition. This is the prefix ct,
probably for the postposition ctt, (see § 3, (1)). Thus from bal
“transgress”, we have ci-bal “‘transgressor” or “rebel” both
masc. and fem., and the common ct-dca “high place” or “ex-
alting” seems derived in the same way from aca “high”.

The plural of the two tenses is distinguished by the em-
ployment of the two words which represent plurality. Hne
denotes the present, (m)es the past. Ene becomes ine, ane, and
une when the 3rd sing. ends in t, a, and u; in some cases it is
contracted to ne. Mes, as I have remarked before, affords an
instance of the progress of phonetic corruption in Accadian,
being only found in the shortened forms -es, -as, and -us. When,
however, the root ended in m, this letter was doubled ; thus zn-
sem-mus “they gave”. The same careful distinction between
the vowels is exemplified in Elamite, where the vowels of the
personal terminations differ according to the vowel of the root
(e.g. turnas “they knew”, cusis “they built”, bitus “they
obtained”). In Basque the mark of the plural tzt is inserted
between the root and the postfixed nom. pron. in ordinary verbs,
e.g. d-aki-tzi-t “he knows them”; in the form of 2, it is
inserted into the root itself of the auxiliary. Other parts of
the verb are created by the means usually employed by the
Turanian languages. I have not found any instance of the
future. In the allied languages this is mostly marked by the
root ra or ar (as in Tatar, Mong, Mands and Elamite, the

38 THE JOURNAL OF PHILOLOGY.

Mag. fut. part. being expressed by the same syllable). In
Basque the fut. infin. is marked by the postposition ra, which
may explain the original sense of the particle. In Accadian,
however, ra gives an intensive force to the verb, translated by
the secondary conjugations of the Assyrian, just as in Tataric
and Mongolic ar, z, or ra form Inchoatives and Neuters. The
root ra primarily signifies to “inundate”, and is placed imme-
diately before the verbal base, and after the pronouns; eg.
ba-ra-tur “he altogether crossed over”, ba-ra-uddu “he went
fully out”, ba-ra-uddu-ne “they go fully out”. The 3rd pera
pron. has always the form ba before it. The negative is inserted
between ra and the verbal root, its final vowel being dropped,
e.g. ba-ra-n-tee-ene “they do not fully take”. The precative
prefix becomes kha, as kha-ba-ra-uddu “let him fully go forth”,
kha-ba-ra-n-male “let no one dwell”, kha-ba-ra-n-duzeu “let
him not seize”.

Another intensive form is that with su or st. This is
inserted between the nom. and acc. pronouns. Thus we find
in-s-in-sem “he gave them a price”, in-8-in-semmus “ they gave
them a price”, in-s-in-semmu “he gives them attestation”, sn-
8-in-semmune “they give them attestation”, mu-s-in-sem “I
gave it all”, in-s-in-zu “he despoiled him”, nu-ban-s-in-duz “ he
did not quite seize him”, khul-mun-s-tb “ he fully adored (?) me”,
When the accusative was not expressed, this intensive was
prefixed under the form of su; thus su-semmu “a gift” (unless
su here be for sev, the root being doubled to denote the passive),
su-nu-n-barra “he does not at all abandon”, su-nenin-ak-d
“who have made” in Khammurabi’s inscription. If ss were
the original form of the word, it might be connected with an old
root which meant “hand” and which is similarly cut off by the
pronouns from the verb with which it forms one word and pre-
fixed, as in su-nen-ti “he took” (see § 6). If st, it would signify
“presence”, “on the spot”. In the Turkish-Tatar dialects
s or tz, affixed to the root makes the cooperatives and re-
ciprocals.

The Accadian had yet another intensive form. This was
the prefix tmmi or tmma which originally signified “a flood”,
tm being “rain”. An ancient Accadian ritual speaks of wnme-

ON AN ACCADIAN SEAL. 39

ne (dimir) Na an-libista “the exceeding flood (lit. “ floods”) of
Anu (M3) in the midst of heaven”. The prefix several times
occurs in the brick-legends of the kings. In the bilingual
tablets we find tmmi-nin-durgas “they died of plague”, num-
ma-s-in-gi “he does not urge on” (where, however, the Assyrian
has the future). The latter example gives us the double
intensive si as well as ammi. Comp. m which builds factive
verbs in Zyr., Tcherem., and Wotiak.

The precative was formed by prefixing ga, which when
followed immediately by ba became kha and evena. Examples
are ga-n-dagga “may he restore”, ga-nin-ban-tsil “may he give
him life”, ga-neb-gar “may he make”, ga-gar (for ga-n-gar) the
same, ga-paga (for ga-n-paga) “may he fight” (whence ga-paga
durga, “may he die in battle”, lit. “may he battle death”),
kha-ba-ra-uddu “may he utterly go forth”, kha-bab-dibdib-ene
“may they bring back”, kha-ba-ra-n-male “let no one dwell”,
kha-ba-ra-n-duzzu “let him fully take it’”’, a-ba-nin-duz “let
him take it”, a-bamu-ra-n-sem “let no one at all give”. Kha
seems to be changed to @ after a preceding @ The Accadian
vowels were probably strongly aspirated at the beginning of a
word. <A guttural pronunciation is largely affected by all
primitive languages, more especially by the Allophylian in
which every word keeps intact its full sound, phonetic corruption
being contrary to the genius of the speech. Thus the river
Idiklat is written [pM in Hebrew, and the Elamite khapar
“high” is apparently connected with the Accadian bar. The
original form of this ga was, I believe, gan, the substantive verb;
the force of the prefix being, “since it is so, he”, etc. Comp.
the Basq. affix of the Conditional and Potential -ke. In the
latter language the characteristics of the various tenses are
postfixed to the root, with the exception of d, the mark of the
present in the Indic, Condit., and Pot., which is prefixed.
Era-(zo), also, the formative of the causative, may precede the
verbal-root, and ecin which denotes the Prohibitive always
stunds before the radix. The same is the case with a, the

1 In this instance it will be noticed the same as that of the inserted ne-.
that the form of the objective pron. is. gative,

46 THE JOURNAL OF PHILOLOGY.

out of the Ural-Altaic family, the Basque affords many ancl
striking similarities. At the same time, Basque has preservedle
beyond any other language of the Old World, distinct tracess
of those primitive contrivances of speech -which have so fax
become obliterated in the Finnic dialects as to make these
approach somewhat to the perfection of inflective languages —
Still, the Finnic dialects are stamped both in structure andl
in grammar with an unmistakeable Turanian imprint. Great&
as may be the distance between the Mandschu and the Wotiak,
it is one of degree only, not of kind, which is bridged over by”
the intervening idioms. At first sight, the Accadian and the>
Elamite seem widely separate, so much so as to justify”
the old title “king of tongues (uccu) and of Accad”, which»
alternated with “Sumir and Accad”, or the phrase perhapes
derived from Arab neighbours “the four peoples” (the D3 of
Gen. xiv. 1); but a close survey, while relegating the Elamitee
to the Ugric division and the Accadian to another and less
developed stock, will assert their primitive connexion. [
lay no stress here upon a community of words such as @
“water”, mil: or mulu “man”, tur “son”, &c., because these
might have been easily borrowed; but the spirit of the two
grammars, and the store of formal elements used by both, are
identical Granting, however, that each belongs to the same
Turanian family, we want to know more closely to what par-
ticular subdivision of that family does the Accadian belong?

Now it cannot but be observed that the analogies between
the latter and the Basque are peculiarly numerous and striking.
It 1s only in the most natural and necessary relations of gram-
mar, more especially the prefixing of the pronouns in the verb,
that the Accadian agrees with Taic or Malay: these languages
are built upon the isolated word, while Accadian takes its
start from the sentence. These natural relations of grammar
again, however contrary to the general principle of modern
Turanian speech, are to be found on the one side in Basque,
on the other in Tungusic, while traces of them may be detected
in Ugric. The postposition of the adjective, moreover, though
now opposed to the practice of Turanian dialects, is but the
earliest expression of Agglutinative grammar, and is still the

ON AN ACCADIAN SEAL. 4I

tion), cusva (for cusu-d) “rested”, dumu for dum-vu “made to
go”, which with the participial ending attached appears as
duma or duva, i.e. dunu-a, gubba (for gubva) “fortified”, gan-va
“been” &c. The addition of the pronouns seems to have been
felt as inconsistent with the impersonal non-active character of
the voice; though rarely the pronouns were omitted in the other
forms of the verb after the subject and object had been expressed.
So, too, in Basque incorpora‘ed pronouns might be used together
with an expressed subject or object. This cannot take place in
languages like the Taic and the Malay, where each word retains
its independent, isolated existence, and cannot in any way be
combined with another so as to form the general idea of a sen-
tence.
The Negative Verb is as marked a characteristic of the
Accadian, as it is of the Ugric, Tataric, and Basque, as well as
of the Tamulic. It appears in two forms, the first combining
the verb with the negative particle, the second being the regular
negative verb. The negative conjugation inserts or prefixes the
particle nii “not” according to circumstances. When the simple
root is used with the pronouns, the negative precedes the latter,
the initial vowel of the pronoun being lost (except in the case of
the second form of the 3rd pers.); e.g. nub-use “he does not
subdue’, nub-ziga “he does not found”, nun-zu-a “he who knows
not”. If, however, a formative is added to the root, the nega-
tive is placed between it and the root, losing its vowel in its
turn, e.g. ba-ra-n-uddu “he goes not forth at all”, ba-ra-n-tee-
ene “they do not take at all”; unless the formative be prefixed
to the pronouns, when, whether it may precede or follow the
latter, in either case it makes its vowel prevail; e.g. su-nu-
n-barra “they do not abandon”, num-ma-s-ingi “he doves not
urge on”, num-mu-n-s-in-male “he does not fortify it”. Before
ba, which seems to have along vowel, nu becomes nam or nab, as
in nam-ba-lale “he does not fill at all”. Mw is constantly used
with participles, as nu-cusva “‘unrested”, adjectives, as nu-sega
“anloving”, and even substantives, as nu dara “not a name’”.
1 Mr George Smith (North Brit. into the body of a word (like the plural

Rev. Jan. 1870), draws attention to the sign in Basque verbs). Thus ‘sidi is
fact that the negative may be inserted ‘‘ pure”, ‘si-nu-di ‘‘ impure.”

42 THE JOURNAL OF PHILOLOGY.

It is also found with the negative verb, intensifying the nega-
tive, e.g. nin-mu nu mia “my wife thou (art) not”, kharra nu
mia “a man not being”. In the Ugric dialects the negative is
combined with the pronoun which is inserted between it and
the verbal root, exactly as in Accadian. This negative is dn or
en as is clear from the imperative where it appears in full; with
which compare Zyr. en and Elam. anu used with imperatives
only. The negative in substantives and adjectives is an affix.
In Basque, the negative particle ez is prefixed to the root. The
negative me in Osmanli immediately follows the root or the
affixes which modify the root. The Tataric me, Mong. ume,
Ugric nem (which perhaps contains both negatives) conducts us
to the regular Accadian negative verb mia. The root mua,
which by the addition of the pronouns becomes a verb, signifies
“not being”. In the singular it is conjugated thus: mta za-e min,
“thou art not”, mia ene “he is not’, mia-ta “I am not”. Mia
za-e min is literally “a not-being (art) thou thou-there’”. Zu
becomes za before the connecting vowel e, a+e being probably
pronounced as a diphthong, while mn is the second form of the
2nd pers., primarily the demonstrative, like Mong. me, Tatar.
ma (from the dem. man) “there hast thou”. Its similarity of
sound to mia apparently causes it to be used with the latter.
Hence it comes to have a kind of negative force, like personne,
jamars. Still if used as a negative it seems to require a pre-
ceding nu; e.g. dam-mu nu min “my wife thou (art) not®”.
Mia-ta is properly the infinitive formed by the post-position ta;
the first person being understood in the speaker. Standing alone
it is the infinitive simply, and requires the addition of other
words to make it the first person. Besides this conjugation,
mia has also the participle of the substantive verb gan (for

1 We find similar repetitions of the
pronouns in tho allied languages. Thus
Jakute min-agha-bin “I am a father”,
literally ‘*I-a father-I"', or the Basque
ni hilsen niz (lit. ‘‘I dying I am”’).

* One example, however, apparently
uses it alone in a negative sense. This
is the phrase ai-'su cicu-ani-nav iz-da-
paggekha-min which is rendered ‘' thy

father on his seat thou doest not seat”,
the Assyrian verb being tugallat, the
pael pres. from a root which is pro-

- ~-

bably allied to As, however,
the first character has the usual value
of ne, and nav. ought not to be followed
by a vowel, it is possible that tugallat
is the 8rd pers. fem.

ON AN ACCADIAN SEAL. 43

ganda) attached. The infinitive has then to be used, preceding
the verb hke any other accusative. The absolute form is mia-
ta d-an, d being the connecting vowel after the preceding Za,
and an, probably pronounced han or rather khan, being the
substantive verb. This is another instance of phonetic decay in
Accadian. A-an is a common affix, added like gan or ganva,
with a connecting vowel. Owing to its use and the fact that its
vowels belong to the guttural class the hard g has become modi-.
fied. The singular of the new form of the negative verb is thus
conjugated; mia-ta mal-e-gan (for man-e-gan) “I am not”, mta-
ta 2d-a-gan “thou art not”, mia-ta en-e-gan “he is not”: and the
bilingual tablet adds further mta-ta-ta “from not-being”.

The Accadian sometimes attaches another particle in a -
curious way. This is va “and” (?comp. Wot. 7) which has often
the position which it requires in European languages; but it
may be prefixed to the verb which it couples with the next
sentence, as in va-ne-cu “he captured and”. Another word for
the conjunction was cama or cava. It may be the passive part.
of ca, “it being said”; but cp. Elamite aak, and cutta (“and”),
Mordw. yak.

The following will be the forms of the simple conjugation :—

PERFECT.

Mu-z1g, “I built” abba-gur, “he restored”

(on the bricks of Elzi tn-nin-zig, “he built it”

and Ismi-Dagon’). ni-nin-zu, “he added it”
Mu-na-zig, “I built it” tn-nan-gur, “he restored it”
Mu-n-zig, » ba-n-tsir, “he fortified it”
Mu-nan-219, » ban-nanin-khir, “he wrote it”
Mu-nanin-zig, “I built them” mi-ni-gir, “he gave it”
tn-zig, “he built” min-in-gub, “he strengthened it”
ftg-nin, in-gin-es, “they placed”
an-sem, “he gave” min-ak-es, “they made”
nin-khir “he surrounded” tb-zigis, “they raised”
ni-dun, “he went” &e.

ba-nuv, “he raised”

1 J should rather read the name Issep-Dagon, from issep ‘‘ a prince”.

Ad. THE JOURNAL OF PHILOLOGY.

PRESENT.
an-lal-e, “he fills” al-gubba, “he fortifies”
ba-lal e, “he fills” ba-nnab-lale, “he weighs it”
ni-lal-e, “he fills” ba-nnan-ca, “he addresses her”
anu-mu, “I am lord” (cp. Mord. an-dab-sit-e, “he measures for
paz-an, “I am lord”). me”
tz-dun-e, “thou goest” mun-nab-ziga, “he strengthens
nin-garrt, “he does” for me”
bab-gubba, “he fortifies” mun-giddhu, “thou spoilest”
1b-turrt, “he crosses” garrinna-zu, “thou throwest
male-nin, “he dwells” down”’
ub-use, “he subdues” in-semmune, “they give”
ba-n-khaa, “he slays him” wm-garrine, “they do”
ne-garrinna, “he throws down” in-lalene, “they weigh”
an-ake, “he weighs” &e.

Before concluding this part of the subject, I have to draw atten-
tion to a remarkable fact. Not only are the pronouns incorpo-
rated, but in some cases the system of incorporation is extended
even to substantives. In the other Turanian languages, the
Turkic for instance, as well as in Accadian, verbal roots are
incorporated, as I have already shown, whereby the various
conjuyations are formed. But in Accadian, a substantive con-
taining the same letters as the verb may be incorporated, an
agglutinative representative, as it may be described, of the
cognate accusative of other tongues. Thus we find ab-’sub-'subbi
“he builds a building”. This will show us more closely the
origin of the similar formation in Frequentatives, Causatives, and
Passives. The root it must be remembered, without suffixes or
without a position in a sentence, is not a part of speech at all.
It will now be necessary to state in general terms the
vowel-harmony of the Accadian. In the most primitive Tura-
nian idioms, the Mandschu for example, polysyllabic words
require the same vowels. This is also largely carried out in
Accadian, eg. gurus, amas. It does not prevail, however,
universally. But in combination, the vowels are always more
or less influenced by succeeding or preceding syllables, I
have already stated the rules which govern this in the case

ON AN ACCADIAN SEAL. 45

of the verbal pronoun suffixes, as well as in the case of the
lengthened root-forms. I have only to add that when the
root is reduplicated, the first (short) syllable is affected by the
final syllable if long; thus it is di@-dund, but diidumad. When
a modifying verbal root is inserted into the body of a verbal
form, should its vowel be a, @ or ~ must precede. In other cases
the vowel-harmony will be @ and 1 after u, % after % and ev, e
after ¢, and ¢, 3 or @ after a.

(17) (?Sar)-mu, “my king”. It may also be “I am king”.

(18) ... dt-ga-ca-ni, “all his final ...” The line is un-
fortunately mutilated, and the reading df? is not quite certain.
Di signifies “to end”, “to set”, also “to judge”, while didt is
“to possess”, “ conquer”.

(19) Ga-wtsillil, “may he give life”. The longer, em-
phatic form is used, though gan-teilli would have been suffi-
cient. The insertion of the predicate shortens the form; thus
“maay he give him life” is ga-nin-ban-tsil, not tsilli or tstllil.

{20) Muh, “his name” or “memorial”. This is the
erdinary meaning of mu, as in mu-khir (“name-writing ”), “a
tablet”. It is alse translated “very great”, and “prince”. It
further signifies “a year”, and “to give”.

The above contains all that I have been able to collect
upon the subject of Accadian Grammar. That it belongs to
@ primitive epoch in agglutinative speech is evident. It only
remaing to discuss the languages to which it is most nearly
related. Throughout this paper I have used “Allophylian” as
synonymous with “Agglutinative”, comprising the various
Asiatic families of speech known as Taic, Malay, Tamulic,
Bhotiya, Tibeto-Caucasian, &c. ; while I have confined the term
“Turanian” to a group of tongues—Mongolic, Tungusic, Ta-
taric, and Ugric—whose unity of origin has, I conceive, been
fully proven by German writers. In the last-named family
must be included the Basque, as has been shown by the labours
of Prince Lucien Bonaparte, Charencey, and others. A con-
tinuation of W. Yon Humboldt’s researches on local names has
extended the range of the Basque across the south of Europe
as far as Asia Minor, and the sub-family thus formed may be
conveniently named Iberian. To the Ugric idioms, specially,

52 THE JOURNAL OF PHILOLOGY. |

sight, we are met by a still worse confusion of incongruous tra-
ditions ; that is, the doubling of the Benediction (20 and 24). |
The great mass of early authorities of various groups con- .
cur in placing the Benediction at 20 only: so NABC 5 137
lat.vg(best MSS) memph aeth Orig.ruf. The pure ‘ Wester’
group D*FG (with Sedulius and perhaps the Gothic version)
places it only at 24°, evidently from the feeling that it
must be the close of the epistle. Minor shiftings and other
like freedoms taken by the same group of authorities occur
in almost every chapter of St Paul: two whole verses 1 Cor.
xiv 34f. are pushed 5 verses forward by DFG 93 and
some Latin Fatbers: compare 1 Cor. xv 26. The scribes of
the fourth century, bringing together MSS from differcat
regions, here as in countless other instances heaped up with-
out omission whatever they found, and so the Eenediction
was set down in both places. The compound reading appears
first in the Greek commentators of the fifth century fiom the
Syrian school, then in the Harclean Syriac (A.D. 50S—616) : in
extant MSS it is found only in L (=Jd) of the ninth ceatury
and the great mass of cursives. There is bowever a similar
combination in a few respectable authorities who retain the
Doxology and place the second Benediction after it (P 17, the
vulgar Syriac and the Armenian versions, and the Ambiosian
Hilary): and this implies the previous existence of MSS
which simply transposed the Benediction to their end of the
epistle, as (D*)FG transposed it to theirs*. Thus the historical

dam in fine adjiciunt epistole. Nos,
quoniam id non videbatur ad huno lo-
cum pertinere, semovimus in finem
hujus epistol#’’ (note on xiv 28 in ed.
princeps of 1516). ‘ Hee est pars qua
in plerisque Grecorum codicibus non
additur, in nonnullis alio additur loco,
sicut indicavimus, in quibusdam adji-
citur in fine. Id quod et nos fecimus,
presertim assentientibus Latinis ex-
emplaribus” (note on xvi 25 ff.).

1 D* and Sedulius add the Doxology
after the Benediction. The nature of
both authorities, as evinced by their

readings generally, explains this sis-
gular collocation. D is not so purely
Western as FG: Sedulius combines
the Old with the Hieronymic Latin.
In each case the Doxology must be a
later accretion. The Gothic has the
Benediction at 24 and (in xvi) no
Doxology: the extant fragments fail to
shew whether the Benediction was at
20 likewise.

* If, as is probable, the abifting of
the Benediction and the dropping of
the Doxology were simultaneous in the
common source of D*FG Sed., P17 é&e.

54 THE JOURNAL OF PHILOLOGY.

mutual peace through willing obedience to the common Lord.
As he had gone back to the perils and hopes of the Church
after the one set of individual greetings, so we can imagine him
joyfully returning to the yet higher sphere of God’s universal
purposes after the other set of individual greetings’. Nay the
parallelism between 17—20 and 25—27 is one of contrast as
well as likeness. The first passage gives vent to somewhat of
the anxious dread which lurks behind many a phrase of xv
1433, especially 30, 31. If these were St Paul’s last words
to the Romans except the two sets of greetings and the Bene-
diction of 20 b, the epistle might have appeared to end ina
note of discord: at all events its exulting comprehensiveness
would have died back into the rebuke and controversy proper
for the Galatians. The sudden upward flight of the Doxology
seems therefore to be almost demanded, to swallow up not only
trivial individualities of salutation but also the temporary strifes
of the Church.

But it is said that the Doxology differs too much in style
from the rest of the epistle to form part of it. I used to sus-
pect that it might be the ending to one of the forms of the
encyclical epistle to the Ephesians, which was preserved from
being lost to the Canon by being appended to St Paul’s longest
epistle. Dr Lightfoot (after Dean Alford) points out its resem-
blance to the Pastoral Epistles as well, and accordingly treats
it as marked by the Apostle’s later style generally. Before
scrutinizing words and phrases, let us look at the subject. The
starting-point is doubly personal; an anxiety about the stability
of the converts addressed, such as tinges the hopefulness of the
first and last words spoken to and about the Romans (i 11;
xvi 17—20); and a bold lifting up of what friend and foe knew
as the distinctive ‘Gospel’ of St Paul, (and that in its distine-
tive form of ‘preaching’, and with its distinctive appeal to
‘faith’,) such as marks the time of the conflict with Judaism
within the Church (i 1, 5, 9, 16; xv 16; x 8, 14, 15). Here

} Dr Lightfoot says (p. 292) that the type" 9 xdps «.7.A. But none of his
Doxology “has nothing incommonwith _ other epistles have a postscript, follow-
the usual endings of St Paul's Epistles, ing a benediction in that form already
which close with a benediction of the given.

END OF THE EPISTLE TO THE ROMANS. 55

the pronouns ‘ you’ and ‘my’ face each other with an emphasis
which in such a context is hard to explain till we remember the
preeaging instinct with which St Paul saw in the meeting
of himself and the Roman Christians, if indeed it was to be
vouchsafed, the pledge and turning-point of victory (i 10 ff;
rv 29—32; cf. Acts xix 21; xxviii 31). Then comes the idea
in which the Doxology culminates, the counsel of the far-seeing
God, the Ruler of ages or periods, by which the mystery kept
secret from ancient times is laid open in the Gospel for the
knowledge and faith of all nations. This idea no doubt per-
vades the Epistle to the Ephesians, though with considerable
ernchments. But is it foreign to St Paul’s earlier thought ?
The second chapter of 1 Corinthians at once shews that it was
hot and explains why the fact is not obvious. St Paul is deal-
ing there with converts who were in danger from pride of elo-
quence and wisdom (from i 5 onward). For fear of this danger,
he says (ii 1 f£.), he himself kept back all excellency of speech
or of wisdom when he came among them, and confined himself
to the bare preaching of the Cross as alone fitted to their im-
Perfect state. But for all that he desired them to know that
he too had in reserve a wisdom which he spoke among the.
Perfect. Its nature he briefly hints in words that closely re-
Semble our Doxology (“We speak a wisdom of God in a mys-
tery, that hidden wisdom which God fore-ordained before the
@ges unto the glory of us” &c. ii 7), and then hastens to ex-
Plain that, even after being laid open, it demands a spiritual
Power to discern it. The Churches to which he wrote about
this time, at Corinth, in Galatia, at Rome, were not in a state
to profit by an extended exposition of a belief which yet was
Strong in the Apostle’s own mind, and so the traces of it in
the early period are few. Later it filled a larger space in his
thoughts, it acquired new extensions and associations, and he
had occasion to write to Churches which by that time were
capable of receiving it. But it is not really absent even from
the Epistle to the Romans. Kindred thoughts find broken
and obscure utterance in viii 18—30. The belief itself is the
hidden foundation of the three chapters (ix—xi) in which God’s
dealings with Jew and Gentile are expounded, and comes per-

56 THE JOURNAL OF PHILOLOGY.

ceptibly to light in their conclusion (xi 833—36). Now it is
precisely in these chapters, as F. C. Baur (Paulus 341 f£) saw
long ago, that the main drift of the epistle is most distinctly
disclosed: all its various antitheses are so many subordinate
aspects of the relation of Jew and Gentile which in this seem-
ing episode is contemplated in its utmost generality as reaching
from the one end of history to the other. The whole epistle
could hardly have a fitter close than a Doxology embodying
the faith from which its central chapters proceed. Here at
last that faith might well be articulately expressed, though
& wise economy compelled it to be latent as long as the Apostle
was simply instructing the Romans. This Doxology is in fact
a connecting link between the epistle at large and the earlier
concentrated doxology of xi 36. In both alike human sin and
hindrance are triumphantly put out of sight’: but here the
eternal operation of Him ‘from Whom, through Whom, and
unto Whom are all things’ is translated into the language of
history.

An examination of single phrases is attempted in the fol-
lowing table, which includes some less obvious coincidences of
thought’.

Tg 3 duvaptry
buds ornpltac

Rom. xiv 4...crfxec 9 wlrrea’ crabjcera: 8, Svvara yap
6 xtpios orfica: atréy. Atwapuat, dvearés, dwardw with an in-
finitive are used of God Rom. itv 31; xi 28; 2 Cor. ix 8;
(xiii 8;) Gal. iii 21; [2 Tim. i 12: 7rQ..duvauéoy...Eph. iii
20.] Zrnpltw in St Paul is found elsewhere only Rom. i 11
(éwcwo0O ydp lseiy buads...els rd oryptxOfpas duds) and 4 times
in 1,2 Thess. ‘Standing fast’ is a common phrase in 1,
2 Thess., 1, 2 Cor., Gal., Rom.; though also found later:
** falling”’ is confined to 1 Cor., Rom.
xara 7d ebayyuby So Rom. ii 16; [2 Tim. ii8.] So also card 7rd ebayyé\uer
pou Rom. xi 28, for here as there the inclusion of the Gentiles
must be chiefly meant. (The ‘stablishment’ of the Romans
would presuppose the harmony of Jew and Gentile among
them.) In this light pov is illustrated by i 1—6, 9, 16;
xv 16.

1 They could not be left out in the
latter part of the Epistle,when St Paul's
own position and the dangers of the
Romans had to be spoken of (xv 14—
83; xvi 17- 20). But for this very rea-
son it was the more necessary that the

ground conquered at the end of xi
should be maintained at the final close
of the Epistle. See p. 54,

3 References to the later epistles are
in []: the chief passages are set out st
length by Dr Lightfoot, p. 298.

END OF THE EPISTLE TO THE ROMANS. 57

cal rd chpvypa ‘Ty-
oot Xpurros

Sed “I ye09 Xprorod
[¥]! 9 Séga els rods
alavas duty.

Compare Rom. ii 16; x 8—12; xv 5£; 1 Cor. i 21; xii
13f.; 2 Cor. i 19f.; Gal. iii 26—29; [2 Tim. iv 17; Tit. i 8:
also 1 Tim. ii 7; 2 Tim.i11.] The double name appears
to have special force in this connexion.

Rom. i 16 f£....els cwrnplay raryrl rg mioretorri, "lovdaly re
[spwrov] cal “EXXqm dixcasoodwn ydp Geot dv abrg [s0. r@ ed-
avyyeNly] dwoxaddwrera: éx xicrews els ricrw: here the histo-
fical ducatoodyn is a part of the pvorfpoy: and 80 iii 21 »urt
82 xwpls vbuou Sucatootvy Geod reparépwrar, waprupouyéryn bwd
Too wéuou cal roy rpodyrar, dixactvyn 8¢ Oeot did xlorews
[(Iycot] Xperrod els wdvras rods wicredovras: of. Gal. iii 22 f.
Rom. xi 25...7d pvorhpeow rovro...dre rdpwors dwd pépous rQ
"Topahy\ yéyover Exps oS rd wAhpwua tiv eva ecicéd\6y, xal
ores was "Iopahd cwhhoera. 1 Cor. ii 6, 7, 10 codlay 32
Aadovpen dy rots redelass...Oeov codlay év uvornply rh» daroxe-
Kpuypévny, Fy wpowpicey 6 Oeds wpe rev aldvwy...° uly ydp
dwrexdd\upyer 6 Geds ed rou wrvevuaros, (Eph. iii 8—11. IIlpd
Xpbvwy aluelaw 2 Tim. i 9; Tit. i 2.]

Rom. i 2...ebayyédcov Oeot 8 xpoerryyeldaro 3a tiv wpo-
Gyrev avrov dé» ypadais dylas; iii 21 (above); and ix—xi
passim.

{1 Tim.i1; Tit.i8.] But the meaning is given by Rom.
11, 5 &’ of [se. °I. X.] EkdBoper...dwrorrodiy els dxaxohy wl-
orews é» racw ras EOveow; x 15; and the mere formula
kar’ éxcrayhy 1 Cor. vii 6; 2 Cor. viii 8.

1 Cor. ii 7 (above); x 11; cf. Rom. xi 883—86. [1 Tim.
117 rg Bac roy aldévwr: also Eph. iii 9, 11; Col. i 26;
2 Tim. i 9; Tit. i 2.)

Verbatim in this connexion Rom. i 5 (above). This en-
larged sense of dwaxen, vraxovw, is confined to the early
epistles (Rom. vi 17; x 16; xv 18 els dwaxohy Ovary; ? xvi
19; 2 Thess. i 8; 2 Cor. vii 15; ?x 5f.)

Rom. i 5 above; xi passim; xv passim; xvi8f. Tvwpltw
is similarly used Rom. ix 22 f.; 1 Cor. xv 1; ?Gal.i11; as
well as (often) in the later period.

Rom. iii 29, 30 4 ‘Iovdalwy db beds pbvwr; ody! xal €Orcw; val
wal €0vue, efrep els 0 Oecs 8s x.r.X. [Mévy dep 1 Tim. i 17, a
kindred passage, which early caused riv aldewy to be in-
serted here after rods aldvas, and in its turn received od¢y
hence in the fourth century: cf. 1 Tim. vi 15; but also Jud.
4,25; John v 44 &€c.] Zodla is predicated of God by St
Paul with reference to the working out of a distant purpose
by unexpected means: so Rom. xi 33; 1 Cor. i 21, ? 80; ii 7;
[Eph. i 8; iii 10; Col. ii 3.]

Rom. v. 1f.; xv6f.; Gal. i4f.; [Eph.i5f., 11—14; iii
31; Col. i 27; 1 Tim. i 11, 17.]

1 gis probably an intrusion, notwithstanding the presumption in favour of an
irregular construction.

58 THE JOURNAL OF PHILOLOGY.

A minute examination of the passages briefly indicated in
this table will shew that the dominant thoughts of the Epistle,
—the thoughts which inspired its beginning (1 1—17), its
primary close (xv 6—33), and its three characteristic chapters
in which the old faith and revelation are invoked on behalf
of the new,—are precisely those expressed in the final Doxo-
logy ; and that the separate words and phrases of the Doxology
are for the most part what have already occurred in the
Epistle, while there are hardly any not to be found in epistles
of the same or an earlier period’. If this be so, the obvious
resemblances to parts of the later epistles lose all force as
evidence of date. The Doxology and 1 Cor. 11 6—10, a passage
absolutely inseparable from its context, support each other
in shewing that St Paul’s late teaching was his early belief;
while in each case there was an adequate motive for his ex-
ceptional transgression of the limits imposed on him by the
present imperfection of his converts. The condensed and cu-
mulative style, which he used more freely afterwards, arises
naturally from the compression of varied thoughts and facts
into a single idea in a single sentence under the impulse of
eager feeling. Rom. 1 1—7; ii 21—26; 2 Thess. i 3—10
offer a true analogy: what distinguishes them is their articula-
tion, which was hardly possible in a doxology. But we may go
further. As is the Epistle to the Romans itself in relation to
the monuments of St Paul’s early teaching, gathering up, har-
monizing, concluding, such is the Doxology in relation to the
Epistle. It looks at once backwards and forwards. Springing
from the keen sense of a present crisis, it gives old watchwords
of action a place in the dawning vision of thought which the
epistles from Rome were to expound, and anticipates in its
style as in its ideas the habitual mood of the time when the
crisis was victoriously ended, and the unity of the Church
secured,

II. The course thus far has been smooth, because the
chief textual difficulties have been out of sight. The end of

1 The only clear exception is xypdvoe + x11. On the other hand draned (at-
aluno: (2 Tim. i 9; Tit.i2), the idea © ¢rews), both phrase and sense, is pecu-
of which is preserved in 1 Cor. ii 7; liar to the early epistles.

END OF THE EPISTLE TO THE ROMANS. 59

the fourteenth chapter is a point at which various phenomena
present themselves which nothing in the context would have
led us to expect. Some of them (a) on the surface mark only
an interruption of the Epistle. The Doxology is inserted
either (1) here alone or (2) both here and in xvi. In (3) a
single MS G, one of the twin MSS which alone omit the
Doxolozy altogether, an empty space is left here, occupying
half a line at the bottom of an otherwise full page and 5 lines
of the next page. Secondly (8) the whole of the two following
chapters are supposed to have been omitted (1) by Marcion
(on the authority of Origen), (2) perhaps by Tertullian and
even Irenzus, and (3) in the capitulation of an unknown Latin
MS mentioned by Wetstein. The variety of this evidence, if
it stands proof, is a strong argument in favour of any theory
which will account for all the particulars.

The testimony of Origen requires consideration first. We
have it only in the greatly abridged version of Rufinus, th care-
less and licentious translator. This is not a passage with which
he is likely to have consciously tampered; but there is no
certainty that the language is Origen’s own. Characteristic
terms of expression as well as ideas may be recognized through
Rufinus’s Latin in almost every page; but none such are con-
spicuous here: rather the sentences are short and simple for
Origen. The comment on the Doxology (after xvi 23) begins
thus. “Caput hoc Marcion, a quo Scripturae Evangelicae
atque Apostolicae interpolatae sunt de hac epistola penitus
abstulit: et non solum hoc, sed et ab eo loco ubi scriptum est
‘Omne autem quod non ex fide peccatum est’ [xiv 23] usque
ad finem cuncta dissecuit. In aliis vero exemplaribus, id est
in his quae non sunt a Marcione temeruata, hoc ipsum caput
diverse positum invenimus, In nonnullis etenim codicibus post
eum locum quem supra diximus, statim cohaerens habetur ‘ Ei
autem qui potens est vos confirmare.’ Alii vero codices in fine
id ut hunc est positum continent. Sed jam veniamus ad capituli
ipsius explanationem.” As the text stands, it asserts plainly
that Marcion removed from the Epistle both the Doxology and
xv xvi; and that of the MSS unaffected by Marcion’s proceeding
some had the Doxology after xiv, some after xvi.

60 THE JOURNAL OF PHILOLOGY.

So the passage has been universally understood. On the
other hand for many years I have had a sirong impression
that the Benedictine text is wrong in three letters, and that
on the removal of this tiny corruption the whole interpretation
collapses. De la Rue’s notes on this book often mention the
readings of a certain Paris MS (Reg. 1639). Wherever I have
examined them, they have appeared usually to give the truest
text against all other known authorities, and very seldom to
be evidently wrong. In this place Reg. 1639 has tn instead
of ab. If the preceding hoc is likewise altered to hic, and so
small a variation may easily have escaped notice, we get an
entirely new and, I venture to think, more probable statement.
Origen begins by saying merely that “ Marcion, the falsifier'
of the Gospels and [St Paul’s] Epistles, removed this paragraph
completely from the Epistle.” Then it appears to strike him
that some reader might know the Epistle in a copy which
had the Doxology at the end of xiv (if not there alone), and
acquit Marcion as having at most only removed a superfluous
repetition®. He adds therefore explicitly “And not only bere
but also” at xiv 23 “he cut away’ everything quite to the
end.” Then, for fear the remark might not be understood
by those who knew the Doxology only in xvi, he explains

1 Interpolo in ancient Latin, it will
be remembered, does not mean to in-
terpolate, but properly to give a spuri-
ous look of newness to old things, and
so generally to falsify.

* Reasons will be given farther on
for suspecting that the MSS here no-
ticed by Origen had the Dorxology in
both places, At this point the differ-
ence is without importance,

* This is not, it must be confessed,
the natural meaning of the single word
dissecuit: but will the context on any
view tolerate another? As regards the
Doxology, abstulit is decisive. Is it
conceivable that Marcion only ‘separat-
ed’ xv xvi from the rest of the Epistle,
while still acknowledging their autho-
rity, whether he joined them to another

epistle or not? or that such an opera-
tion would be unrecorded? The diffi-
culty surely lies in the translation.
Dissecuit would not be an unnatural
rendering of repéxoyer or possibly zre-
piérexer, either of which would mean
simply ‘cut away.’ Compare Epiph.
Huer. 809 p od pdvew 82 rh» dpyhy dré-
renew [of St Luke’s Gospel]..., d\X\a cal
Tov rédous xal rw» wéowe woddAd wepré-
xowe ray ris dAnOelas Ad-yww «.7.d.: and
again dAX\d ria alray repréuwruwr, rod
&¢ d\Aouoas cepdrasa. In the first sen-
tence, so closely resembling Rufinus’s
in form, aroréyrw and repxowrw must
be practically synonymous, for the pre-
ceding sentence describes the Gospel as
wepixexoupévoy ard rhs doxis by Mar-

cion.

62 THE JOURNAL OF PHILOLOGY.

point in the Epistle. Certainly it might be so. But in that
case we should expect him to begin with the transposition of
his immediate text, and having so been carried to xiv 23 to
append by way of digression an account of Marcion’s proceeding.
The reverse order, which we actually find, has no logical justifi-
cation on the common interpretation, unless Origen himself
saw in Marcion’s supposed omission of xv xvi and in the trans-
position of the Doxology two facts connected by community of
origin. That however is a step in criticism which there is not
the slightest evidence that he took. He regarded Marcion’s
omission, whatever its extent, as an original and unprecedented
act; and he gives no hint that the transposition or repetition in
certain MSS was a consequence of Marcion’s mutilation: in
other words the two facts were in his eyes two independent
phenomena. How then came the one to suggest the other? If
Marcion omitted two chapters, the sole point of contact is xiv
23; and thus the transposition, which alone forms a bridge
from xvi 24 to xiv 23, must have preceded the omission in
Origen’s account. If on the other hand Marcion cut out only
what the scribes transposed, then no bridge is needed. The
first and the last sentences refer alike to the same subject, the
paragraph on which Origen is avowedly about to comment.
The second sentence refers partly to this place, partly to the
other; and likewise serves to anticipate an erroneous criticism
of the first statement, which might occur to Origen’s readers.
The commentary of Jerome on Eph. iii 5 explains diffusely
how St Paul could say that ‘the mystery of Christ in other
generations was not made known to the sons of men’ notwith-
standing the language of the Prophets. At the outset he
repudiates the doctrine juzta Montanum that the prophets
spoke in ecstasy, not knowing what they said. Three columns
further on he repeats “ Those who will have it that the prophets
understood not what they said, and spoke as it were in ecstasy,
bring to confirm their doctrine not only the present text, but
also that which is found [in the epistle] to the Romans in moet
MSS, reading Now to Him, &c.” The inference is obvious, that
the writer had seen or heard of MSS which did not contain the
Doxology. But who is the writer? Jerome in his preface

END OF THE EPISTLE TO THE ROMANS. 63

tells us that he had partly followed the three books of Origen
on this Epistle. Comparison of the Greek fragments proves
how freely he drew on his great predecessor’s ample stores ;
and any one familiar with Origen’s style will recognize it in
many places where the Greek is entirely lost. Throughout this
long disquisition Origen’s hand cannot be mistaken, though
Jerome may bave added or altered this or that sentence. The
controversy with Montanistic doctrine belongs moreover to the
third, not the fourth century’, The character of the MSS
hinted at as wanting the Doxology is sufficiently indicated
in the two sentences which follow the refutation ofthe Mon-
tanists. “And in like manner it is to be observed that the
mystery of our faith cannot be revealed except through the
Prophetic Scriptures and the coming of Christ. Let those
therefore know who understand not the Prophets, and desire
not to know, protesting that they are content with the Gospel
alone” &c. This evident allusion to the Marcionists, the other
great sect which threatened the Church in Origen’s days, sug-
gests the strong probability that the passages from his two
commentaries relate to the same subject. What he calls
“most MSS” here are identical with “those copies which have
not been corrupted by Marcion.” In the former case. the
Doxology is said to have been omitted*: may we not infer, in
the absence of evidence to the contrary, that this and this alone
constituted Marcion’s offence? Whatever the argument might
be worth taken independently, it appears to me a striking
corroboration of the result obtained thus far.

Tertullian’s language is ambiguous. After confuting Mar-
cion out of Galatians and 1, 2 Corinthians, he proceeds to
Romans (adv. Marc. Vv 13). Henceforth, he says, he will touch
but briefly on what has come before him already, and pass over

1 The dislike of the early Alexan-
drians to the Montanist theory of ‘pro-
phecy’ or inspiration is well known.

$ The words are “Qui volunt Pro-
phetas &c., cum praesenti testimonio
illud quoque quod ad Romanos in ple-
risque codicibus invenitur ad confirma-

tionem sui dogmatis trahunt, legentes
Ei autem” &. They do not formally
negative the omission of the two whole
chapters; but other language would
surely have been chosen had the Doxo-
logy been the mere conclusion of a
large section omitted.

64 THE JOURNAL OF PHILOLOGY.

altogether what has come before him frequently. He is tired
of arguing about the Law, and about God as a Judge, and 50
an Avenger, and so a Creator. Yet he must point out the
plain references to justice and judgement which meet him at the
beginning of the Epistle (i 16 ff; ii 2). It will be enough for
him, he declares, to prove his point from Marcion’s negligences
and blindnesses, from the sayings which he left undisturbed’.
He then runs over the Epistle in 5 pages, just half what he had
bestowed on the little epistle to the Galatians, passing over in
silence some long spaces of text containing appropriate matter,
as ili 1—20 and x 5—xi 32. The ethical paragraph xu 9—
xiii 10 tempts him to give examples of the anticipation of its
teaching in the Old Testament, and he concludes with insisting
on the harmony of Law and Gospel in inculcating love of neigh-
bours. There apparently he intended to stop, the doctrinal
part of the Epistle being ended, but his eye was caught by the
words “ judgement-seat of Christ” at xiv 10. He therefore adds
(14 8. f.) rather awkwardly, with evident reference to what he
had said on the beginning of the Epistle’, “Bene autem quod
et tn clausula tribunal Christi? comminatur, utique judicis 4
ultoris, utique creatoris, illum certe constituens promerendum
quem intentat timendum, etiamsi alium praedicaret.” And
then he proceeds to another epistle. The absence of allusions
to anything in xv xvi requires no explanation: it is hard to se
what could have been cited except xv 4, 8, 18, which are slight

1 He notices but one omission by
Marcion in this epistle, that of c. ix.
The limits are not given, but there is
little room for doubt. Eight other
(short) omissions are recorded by Epi-
phanius, who professes to furnish only
a selection (Haer. 817 f.). It is singu-
lar that Epiphanius should pass over
the loss of three consecutive verses:
but his silence would be far more
astounding if two whole chapters were
missing. Nothing could be safely in-
ferred in any case from his employ-
ment of the word dxpwrnpidiw, as ap-
plied to St Paul’s epistles (xal avrdr 52

hapurnpiacpérww curhOws 7G adrod pak
oupyia 3817p): his wide use of it is ma
nifest when he says (811 p) that the
Gospel, as jxpwryplacre: wire dex
Exov pire péoa phre réos, luariov pe
Spwyuérov twd wodAuw onraw éréxe Ti
Tpbmov.

4 So not long before he had said, not

~ it is not true of a book but of a passage

(1 Cor. ix 10—x 11), “‘ Denique et in
clausula praefationi [apostolus] re-
spondet” (c. 7).

3 The true reading is rod Geot, buat
confusion with 2 Cor. v 10 was easy.

66 THE JOURNAL OF PHILOLOGY.

to be contemned, but their value depends on the attendant cir-
cumstances. Seventeen verses only of the two chapters (xv 1—
13; xvi 17—20) were likely to be quoted. Of these Onga
once quotes one (setting aside the commentary), Clement three;
while of others it so happens that Origen quotes five, Clement
three, besides the Doxology.

Lastly Wetstein has a note at the end of xiv: “Coder
Latinus habet Capitula Epistolae ad Romanos 51, desinit autem
in Caput XIV; ex quo conficitur ista Capitula ad Editionem
Marcionis fuisse accommodata.” “Later critics,” says Dr
Lightfoot, “have nut been able to identify the MS and thus to
verify the statement.” Their failure however matters little.
The phenomenon here obscurely described is not peculiar tos
single MS: it belongs to what was probably a widely current
Latin capitulation, found e.g. in the earliest (540—550) MSS of
the Vulgate, the Amiatinus and the Fuldensis. The sections or
breves of Romans are 51, § 50 beginning at xiv 15, and § 51 at
xv 4. In the table of contents before the Epistle § 50 is headed
“De periculo contristante [sic] fratrem suum esca sua, et quod
non sit regnum Dei esca et potus sed justitia et pax et gaudium
in Spiritu Sancto,” a fair description of the section ; and § 51
“De mysterio Dumini ante passioneim in silentio habito post
passionem vero ipsius revelato,’ which in strictness applies
only to the Doxology'. If the marginal figures were lost, it
would be a natural inference that § 50 ended with xiv, that
§ 51 consisted of the Doxology, and that xv xvi were absent
from the MS on which the capitulation was originally formed.
But as on this view the table and the marginal figures con-
tradict each other, it seems hopeless to attempt to clear up the
confusion while the origin of the capitulation remains un-
known*., There is no Latin authority whatever for associating

1 Kither Wetstein examined only the
table of headings, or he overlooked the
inconspicuous figures li at xv 4, a place
where he would scarcely expect them.
This is the sole point of difference.

* Internal evidence proves that the
rections cannot, in their present furm,

answer to ecclesiastical lessons, Other-
wise one might have thought that the
Doxology was appended to xv 13 or 33
for public reading, and the rest of xv
xvi neglected. Some sections are de-
scribed only by their end, as others
only by their beginning.

68 THE JOURNAL OF PHILOLOGY.

was copied, must have preserved the abridged recension.” In
other words (1) the scribe of G copied i—xiv from one MS and
xv xvi from another; and (2) the scribe of F copied in like
manner from the same two MSS, though he left no mark of the
transition from the one to the other. If the first of these hypo-
thetical facts were true, we ought surely to find some evidence
of it in the respective texts; whereas the closest study fails to
detect a shadow of difference in the character of the readings

before and after the blank space.

F is free from no outrageous portent
found in G, but has to answer for many
of its own. No one can believe that
two scribes independently arrived at
e.g. Tlorurovow exar tyewceuvor Tw
Aoywy (both FG have w over -vov: F
further divides dyew. exvov) for vroru-
rwcw tye vytaworruw ASywv: and the
absence of division of words in the
archetype is proved by the numerous
self-corrections of the scribe of G,
where he has added to the end of one
word the first letters of the next, seen
his error, and begun the second word
afrosh with a space between. In these
cases he sometimes has forgotten to
put in the cancelling dots or line, and
then the writer of F confidingly tran-
scribes the whole. But usually he is
careful to follow only corrected read-
ings. In 1 Cor. xi 81f. dwd translated
by a happens to be under the end of
davrovs in G; and the stroke or ac-
cent which, as usual in G, caps a looks
like a cancelling line to the final s:
hence F reads éavrov though tho verb is
Scexplvouev. Other instances might be
given of the dependence of F on acci-
dents in G. The relations of the Latin
accompaniments (fg) are complicated,
but tend to the same result. The
body, so to speak, of g must have at
least a double origin, from a pure Old
Latin text and from one or more alter-
ed texts, either the true Vulgate or one

The partial adherence of D ex-

of the intermediate revised texts &
both. Where none of his materisls
represented the Greek literally enoug)»
the scribe evidently devised new rendet”
ings of words and still oftener chang~
ed their order. This is shown not onl ¥
negatively by comparison with tbh©
mixed and fragmentary yet frequentlY
copious evidence of all sorts as to varé.~
ations in Latin MSS and Fathers»
but also positively by mistakes arising=
from the wrongly divided Greek words®™
and the like. Sometimes g offers two~
or more alternative renderings, either”
all traditional or part traditional part
original. The body of f is tolerably
pure Vulgate, unequally but always
imperfectly assimilated to the Greek
with, I believe, the aid of no document
except g, all the elements of which may
be recognized. In 1 Cor. x, singled
out by Mr Scrivener for its frequent
departure from the Vulgate, out of the
46 variants 28 agree withd and 42 with
g, while the remaining 4 consist of 3
blunders, one correction of an obvious
blunder, and one interpretative change
of tense. The concordance of evidence
80 Various seems decisive against any
claim of F to represent the archetype
whero it differs from G. Nothing how-
ever in the text of this article is sub-
stantially affected by the result except
the sentences in brackets.

END OF THE EPISTLE TO THE ROMANS. 69

epted, this character is unique among existing Greek MSS: that
t should prevail equally in two MSS accessible to the scribe of
2 is possible certainly, but not likely; and the hypothesis in-
volves this further anomaly that the two originals, so singularly
hke in the main, must have differed on the capital point, the
Omission of xv xvi. [When F is taken into account, fresh embar-
rassments arise. Either the scribe of F copied one MS through-
out or he did not. If he did not, an exact repetition of the circum-
stances attending the writing of G is demanded, without such
evidence as the blank is said to afford. If he did, what becomes
of the primary original of G7] Theblank may, I believe, be
easily explained by a simple process. The Greek text of F and
G alike was copied from a single archetype wanting only the
Dorology. [The scribe of F wrote down exactly what lay
before him.] The scribe of G on arriving at xiv 23 remembered
the Doxology as occurring there in some other MS that he had
read (all extant MSS but 9 have it there, 4 older, 5 younger),
held faithfully to his archetype, but satisfied his conscience
by leaving a space which might be filled up hereafter if
heedful. He did in fact only what the scribe of B had done
four centuries before, when he left a blank column for the sup-
Plement to St Mark’s Gospel (xvi 9—20). It follows that FG
attest the omission of the Doxology alone, while the blank in G
vouches merely for the vulgar Greek text as it prevailed from
the fourth century onwards.

That reading of the vulgar text however remains to be ex-
Plained if possible, and remarkable without doubt it is. The
intrusion of the Doxology after xiv 23 appears in two forms:
Conjointly with its retention at the end in AP 5 17%, and some
Amenian MSS: in this place alone in L (= J) and all Greek
cursives but 8 (or 10), some MSS known to Origen (above,
Pp. 99), the Harclean Syriac and the Gothic’ (with, it is said,

? There is a doubt about 2 or 3 assimilation to the Greck.
others, and more will probably be found 2 The fragments of this version do
in due time: see also p. 70, notel. The not comprise xiv 20—xv 8. But the
introduction at xiv 23 by the second presence of the Doxology after xiv
hand of the Latin text in the trilingual would make the gap exactly equal in
109 is doubtless due to an imperfect length to the adjoining leaves of the

70 THE JOURNAL OF PHILOLOGY.

two other late and obscure versions), Chrysostom" and the
Greek commentators who follow him, and perhaps Cyril and
John of Damascus. Perplexities abound here. The first small
group is select* though not trustworthy: by the analogy of
other passages it indicates a reading of high antiquity, probably
current at Alexandria, but a correction. Origen’s MSS being
waived, the certain portion of the second group is practically
rubbish: that is, it contains no authority of the slightest value
hereabouts except as a rare adjunct to some primary authority
left nearly in solitude. That some MSS known to Ongen
should have attested a reading of the first group is exactly
what might have been expected: their association with the
second is passing strange. It suggests a doubt (more is not
permissible) whether Origen after all did not speak of thus
MSS which had the Doxology at xiv 23 as having it also at the
end. Rufinus’s clumsy scissors may easily have shorn off the
additional fact, especially as the antithesis became clearer iD
consequence: on this view the words about Marcion’s doings
‘not only here but also in that place &c’ would have increased
force, though it must be allowed they do not require it. Bu®
another difficulty remains. We might have supposed th®
double position of the Doxology to be owing to the combinatioe*
of texts from two sets of MSS, each of which had it in a differen™
place and there alone; yet the character of the authorities n—
verts this order. In cases like this it is ultimately found safer~
to trust to the historical relations of the evidence than to any
speculations about probability. But indeed here the only tole-
rable explanation that offers itself of the introduction of the
Doxology at xiv 23 in either group would point to the first
group as exhibiting the earlier form of corruption. Changes in
the Greck text of the New Testament, chiefly by interpolation,

Codex Carolinus, which alone has pre-
served the verses before and after.

text and commentary in both places,
and so might be added to the first

The 4 existing leaves of this MS shew
that xi 33—xv 13 was written on 8
leaves; and all the measures give the
same length to a leaf within a line.

1 One Vatican MS of Chrysostom
according to Mr Field (p. 647) has both

group. But internal evidence proves
that Chrysostom himself used only the
vulgar Greek text.

2 Though inferior to 17, 5 is a cur-
sive of the first rank.

END OF THE EPISTLE TO THE ROMANS.

71

arising from the modifications required for Church lessons are
common in MSS, though they have rarely found their way into
printed texts. The salutations in xvi might easily be thought
to disqualify the bulk of the chapter for public reading’, espe-
cially at a time when but a few select lessons were taken from
the whole Epistle*: and yet some church, for instance that of

1 The Greek ‘Euthalian’ capitula-

tion found in divers MSS (printed by
Mill N. T. 418 and elsewhere) has for
the heading of its § 18 wepl [ris] usu7-
Gews Tes Xpcorou dvetixcaxias, of § 19 wep
Tip Aaroupylas alrow ris dy dvarod\y Kal
8éea, and nothing after. These must
correspond to xv 1—138, 14—33._ It
follows that xvi (but not xv) is omitted,
evidently because not publicly read in
somechurch. The latest sectional num-
ber (24) in P stands at xv 14, doubtless
for a similar reason. By a singular
evincidence § 18 of the Vatican capitu-
lation begins with xv 1 as in the ‘Eu-
thalian’ capitulation: but they do not
coincide in the earlier chapters, and
the Vatican sections proceed to the
end, commencing § 19 at xv 25, § 20 at
xv30, and § 21 at xvi 17. Fritzsche
(Rom. i p. xlvii) pleads that on the
tame grounds we might argue the ex-
clusion of 1 Cor. xvi from public read-
ing, since no trace of its contents ap-
pears in the ‘Euthalian’ capitulation
for that epistle. Why not? The last
sectional numeral (20) in the margin
of Pin 1 Cor, is at xv 61. Thus again
both independent capitulations equally
agree with what the nature of the chap-
ter renders intrinsically likely. The
Capuan Lectionary in the Fulda MS of
the Latin Vulgate takes no lesson from
Bom. rv xvi except xv 8—14 (for the
Cireumcision), and none from 1 Cor.
xii—xvi.

3 Dr Lightfoot (287) refers to Reiche
as having shown that xv xvi were not
omitted in public reading. Reiche de-

pends on Fritzsche and after him Mey-
er, who argue (1) that the profound -
reverence of the early Christians must
have saved every letter of the N. T.
from being unheard in the churches ;
(2) that the lectionaries prove the whole
epistle to have been actually read. But
this continuous reading noted in the
lectionaries belongs only to the Daily
Lessons, which E. Ranke (Herzog R.
E. xi 376 ff.) shews to be of late date,
perhaps not earlier than the 12th cen-
tury. The ancient lessons for Sundays
and Saturdays are all more or less
selected, continuous only in certain
definite cases. The existing Synaxa-
ria, valeant quantum, give Rom. xiv
19—28 plus the Doxology as the lesson
(an appropriate one) for Saturday be-
fore ‘ Tyrophagus’ Sunday (Quinquage-
sima): see the tables in Scrivener In-
trod. 72; Scholz N. T. ii 459; Matthai
Rom. xxiv. They have but two other
lessons from this part of Romans, xv
1—7 for the 7th S. and xv 30—33 for
the Saturday before the 10th S. after
Pentecost (Scrivener 69 f.; Scholz 458;
Matthei ib.). All these arrangements
however are probably Constantinopo-
litan, and originally derived from the
‘use’ of Antioch. An Alexandrine
Table of Lessons is preserved in a
Vatican MS (46 Paul. of Wetstein),
and has been edited by Zacagni Coll.
Mon. 712— 722; but the first leaf, con-
taining from Easter to the 3rd S. after
Pentecost, is missing. In the part of
the year where Romans is chiefly read,
xiii 1—8, xv 1—6, 18—19, 80—83 oc.

72 THE JOURNAL OF PHILOLOGY.

Alexandria, may have been glad to rescue the striking Doxology
at the end for congregational use by adding it to some neigh-
bouring lesson’. It could not well be used by itself, even if it
were longer: it craved to follow some passage which in like
manner craved a close. Many would find in the benedictions
at xv 13, 33.a reason against appending the Doxology in either
place’, while it would make an impressive termination to a
lesson formed out of the latter verses of xiv which when alone
have both a harsh’ and an unfinished sound. Scribes accus-

cur consecutively; but no other lesson
from this Epistle after xiv 11 appears
anywhere. A few scattered lessons
agree with those in the common Syn-
axaria, but the coincidences are such
as might easily be accidental: the
systems are independent throughout,
though partly analogous. Saturday
lessons are wanting, according to the
custom of the early Alexandrine and
Roman Churches (Socrat. v 22), except
in Lent. But as it is the long eight-
week Lent of late Alexandrine usage,
compurison as to ‘Tyrophagus’ Satur-
day is out of the question. All the
Lenten Saturdays have in place of a
definite lesson the single obscure for-
mula 'Ex roi drogréddov els dylous: the
4 lessons els prelas dylwy, Rom. v 1—5;
viii 28—34; Heb. x 832—38; xi 88—xii
2, can hardly be meant, as Zacagni
seems to suppose; but the reference
may be to a Menologium, or Table of
Lessons for Holy-Days, not preserved
in the MS: the common Synaxaria
have lessons from Hebrews on the
Saturdays of their Lent. ‘Tyrophagus’
Sunday is one of the days of cvinci-
dence, the lesson being Rom. xiii 11—
xiv 4. In short nothing can be clearly
made out, except the prevalence of
Variety of usage and the utmost free-
dom in the selection of lessons; that
is, Fritzsche’s and Meyer's arguments
are found to have no support from facts.

1 The late Alexandrine lesson for St
Stephen’s Day begins Acts vi 8 and
ends vii 60, As the other lessons are all
short, this must have been made up of
two passages, the speech being omitted.
A similar Old Latin lesson for St Ste-
phen’s Day has been printed by Ceri-
ani (Mon. S, et P. im 127 f.), combin-
ing vi 8—vii 2 with vii 51—viii 4.
Ranke in Herzog R. E. x 81 notices
two Mozarabic lessons from Jeremiah,
one of which omits 18 verses in the
midst, and the other is a cento of 5
fragments.

2 Gabler in Griesbach Opusc. ii p.
XXVi.

3 This is the ground taken by J. A.
Bengel (App. Crit. 840 Burk), to whom
we owe the first suggestion about Church
Lessons, He says “Videntur Greci,
ne lectio publica in severam sententiam
Quicquid non est ex fide peccatum est
desineret, hanc ei clausulam attexuisse.
Conf. var. Matth. iii 11.” His note on
the omission of cal wvpl in this last
place is worth quoting. ‘Citra hme
verba finierunt Greci, v. gr. in Aug. 4
[the Lectionary numbered 24], lectio-
nem ecclesiasticam, ne tristis esset
clausula. Simili euphemismo et Ju-
dei post ultimum eumque severum
Tesais, Malachis, Threnorum, et Ko-
heleth versum rescribere penultimum
solent: et Grwmei nonnulli post ultimum
Malachia versum ponunt antepenulti-

END OF THE EPISTLE TO THE ROMANS. 73

tomed to hear it in that connexion in the public lessons would
half mechanically introduce it into the text of St Paul, just as
‘they seem to have introduced a liturgical doxology after the
Yord’s Prayer into the text of St Matthew (vi 13). Then in
the course of time it would be seen that St Paul was not likely
to have written the Doxology twice over in the same epistle,
and it would be struck out in one place or the other; while
familiar use would override any effort of critical judgement’,
and so the Doxology would vanish from the end of xvi, nothing
in the context seeming to demand its retention. Such I con-
ceive is the history of the position which the Doxology holds in
the vulgar Greek text, a position which it would probably
retain in the Received Text and in the popular versions of
Europe but for the confused impulse which led Erasmus in this
instance to adhere to the Latin tradition.

HII. In the two places of the first chapter (7, 15), where
the name of Rome is mentioned, it disappears in the single
MS G. Some leaves are wanting at the beginning of F;
doubtless if extant they would show the same omission. At
the first passage there is a note in the margin of 47 to the
effect that “he [or “it”: no nominative] mentions the phrase
év “Pen neither in the commentary nor in the text.” The
subject may be some unknown commentator, but is more
likely to be an “ancient copy” of St Paul’s Epistles which is
expressly cited in a similar marginal note on vi 24°, and which
like 47 itself may have been provided with a marginal catena

of ‘commentary ’®.

Bim. Etiam in Byz. [86] rédcs pri-
lm post hwo verba, deinde his erasis
ante, notatum est.”

‘Yet ancient criticism, finding the
Doxology between xiv and xv, would
Probably see nothing to object to ; while
it would readily stumble at the ap-
Parent violation of epistolary correct-
bes in xvi 25 ff. The influence of
MSS like FG may also have helped to
‘pel the final Doxology, while it would
be powerless to displace the same words

Dr Lightfoot thinks he sees a trace of the

where imbedded firmly in the text.

3 The reading there quoted from rd
wadatoy ayri-ypador is both rare and ex-
cellent: the other marginal readings of
47 are of no interest, nor is there I be-
lieve any other reference to another
authority. Cf. Griesbach Symb. Crit.
i 155 ff.

3 Anuncial MS with acatena, like Z
of St Luke, might be called “ the an-
cient copy” in the 11th or 12th cen-
tury.

74.—Ci«s THE JOURNAL OF PHILOLOGY.

same omission in Origen’s criticism as rendered by Rufinus,
notwithstanding the presence of Romae in the text. But the
context gives another turn to the language used. “Benedictio
autem pacis et gratiae, quam dat dtlectis Det ad quos scnbit
apostolus Paulus, puto quod non sit minor ea quae fuit bene-
dictio in Sem et in Japheth, quoniam per Spiritum impleta
est erga eos qui fuerant benedicti &c.” “Ad quos scribit” 1s
substituted for “qui erant Romae” because the point is that
St Paul’s benedictions had not less dignity and effect than the
sacred benedictions of the Old Testament; as Origen proceeds
“Non ergo his omnibus inferiorem duco hanc Apostoli bene-
dictionem, qua benedixit ecclesias Christi,” while any inference
from the generality of “ecclesias” is precluded by the further
remark that “haec Apostoli consuetudo scribendi non erga
omnes ab eo servatur ecclesias,” and by the classification which
follows. Still less can I recognize any sign of the omission in
the Ambrosian Hilary’s words “Quamvis Romanis scribat, illis
tamen scribere se significat, qui tn caritate Det sunt.” For he
goes on “Qui sunt hi nisi qui de Dei filio recte sentiunt? Ist
sancti sunt et vocatt dicuntur: sub lege enim agentes’ male
intelligunt Christum” &c. Every word becomes clear on com-
parison with a passage in the Prologue (25 AB) in which he
contrasts the “Romani” with the Judaizers who were equally
at Rome (év “Poyy): the meaning is that St Paul writes not
to all “at Rome” indiscriminately, but to those at Rome who
were “in caritate Dei.” The true text in full is maou rots
ovow év ‘Pan ayatrntots Geod xAnTois dyiows. A Western cor-
rection (D* lat. [the Greek lost] G, the 2 best MSS of the
Vulgate, apparently the Ambrosian Hilary, and perhaps Hilary
of Poitiers) substitutes év ayamn Oeod for ayanntots Geov, doubt-
less on account of the «Antois following (‘who...through the
love of God are called to be saints’). The result is that ENPQMH
and ENATATTHOY were left contiguous, each beginning with év.
The loss of one or other out of a pair of such groups of letters
is common in MSS of any form, and would be peculiarly liable
to occur in one written in columns of short lines, such as was

1 Not ‘they agentes’ but ‘they who agunt’.

76 THE JOURNAL OF PHILOLOGY.

Oeciav viii 28; &c. The omission in i 7 might therefore be
neglected without further thought but for the parallel omission
of trois dv ‘Poy ini 15, the name of Rome being confined to
these two passages in the Epistle. The coincidence would cer-
tainly be noteworthy if it were sustained by other documentary
evidence, or if there were independent reasons for believing a
recension of the Epistle to have existed in which the marks
of a special destination were purposely obliterated. There is
no such reason apart from the supposed removal of xv xvi:
the hypothesis is suggested by the reading of G ati7,15. We
may therefore be content to suspect that in these two verses
like causes produced like results.

All the phenomena of text alleged to prove a double re-
cension have now been examined. The enigmatical Latin
capitulation excepted, they have been found, if I mistake not,
to be more naturally explicable by other causes, This result
becomes clearer still when the hypothesis is examined as &
whole. The second recension, it will be remembered, was said
to consist of chapters 1 to xiv, with the Doxology, and without
the two namings of Rome. How is it then that every autho-
rity, which supports, or may be thought to support, some part
of this combination, contradicts some other part? For the
omission of xv xvi the one direct testimony, if such it be, 18
that of Marcion: and yet the one incontrovertible fact about
him is that he omitted the Doxology. If G is to be added ©o®
the strength of the blank space after xiv, yet again it leav@
out the Doxology. Once more there is no lack of authorit® ®
of a sort for subjoining the Doxology to xiv. We may was ~“
the fact that they all retain xv xvi. We cannot forget (1) th» ‘
they all make mention of Rome at i 7, 15; and (2) that tho™
have no sort of genealogical affinity with the MS that ignor ~
Rome, or with Marcion. In few words, the authorities, whic
as a matter of fact contain the rude outlines of the first recer=
sion, supply the main data for constructing the second. Mear=—
while neither recension is represented in the great mass Cc
good authorities, Greek, Latin, Syriac, Egyptian, or other, os*
which the text of St Paul stands in ordinary cases. Both rev-

END OF THE EPISTLE TO THE ROMANS. 77

censions, as wholes, are purely conjectural. If Rome and the
transposed Benediction are set aside, the first recension is
vouched for by FG (standing for a single archetype) alone of
extant documents and by some traditional evidence. The
second recension can be reached only through a hypothetical
text which Marcion altered, and a hypothetical duplicate ori-
ginal of G.

Such being the relations of the textual evidence, little re-
quires to be said on the intrinsic probability of the hypothesis.
There is nothing in it that we need hesitate to accept if only
the evidence were stronger. But it surely bas not that kind of
Yerisimilitude which would raise the feeling that it cannot but
be true. The only analogous instance known to us is the
encyclical epistle addressed to the Ephesians and other neigh-
bouring churches. But that letter appears (1) to have been sent
Simultaneously to its different recipients ; and (2) to have been
8eneral in form in the first instance, not a special appeal
trimmed for general use. Analogy apart, it is difficult to
imagine St Paul deliberately cutting out in after years the
Words that spoke of personal bonds to definite churches and
believers, and the passionate hopes and fears which they had
Nce called forth. If for any purpose he needed an impersonal
treatise on the old subjects, he would surely have written it
Mew. Indeed the fitness of our Epistle, however altered, may
Well be doubted. Its catholicity springs from the marvellous
balance that it holds between Jew and Gentile, which in its
turn rises historically out of the equal or almost equal combina-
tion of the two bodies in the metropolitan Church, as Dr Light-
foot has justly insisted (288 ff.). Is it probable that the same
characteristics would recur in the unlike “countries into which
he had not yet penetrated” (294)? Even that single point of
connexion disappears when we recall the pregnant paradox of
his relation to the Romans, that, though he had not seen them,
he knew them so well.

The inverse theory of several critics, that the original letter
to the Romans ended with xiv and, some add, with the Dox-
ology, and that St Paul afterwards appended xv xvi, escapes
these difficulties to plunge into worse. Paley proves con-

78 THE JOURNAL OF PHILOLOGY.

vincingly that xv can belong only to the time when the body o
the Epistle was written and can have been addressed only t
the Romans: and there is cogent evidence which he has over
looked. Dr Lightfoot has shown how much can fairly b
elicited from xvi to the same effect. The slight break more
over after xiv is onesided, and on the wrong side. The openings
words of xv furnish a tolerable beginning: the last words o:
xiv make a very bad end, even when the Doxology is allowed t
follow.

When all is said, two facts have to be explained, the inser.
tion of the Doxology after xiv, and its omission. The forme:
has occupied us enough already: the latter now claims a few
words. If the view taken in this paper be right, the omitting
authorities are FG, Marcion, and certain MSS twice noticed by
Origen, once distinctly and both times implicitly, as having
. been corrupted by Marcion. The readings of D* and Sedulius,
mixed authorities substantially akin to FG, likewise imply
omission as antecedent. Origen accuses Marcion of wilful
omission: is the charge just? There is analogy favourable to
either answer. It is now equally certain that Marcion some-
times mutilated the text of his favourite apostle, and that some
variations or omissions imputed to his pen were in fact simply the
readings which he found already in his MS. The reference to
‘prophetic Scriptures’ in v. 26 might conceivably annoy him,
though, as far as we know, he tolerated much of the same kind
that was less likely to please him. But the removal of four
words, an operation more in his manner, would have served
every purpose. Though copies of his Apostolicon were seemingly
current here and there in the Church, no extant document can
be shown to have been affected by any of his wilful alterations.
Indeed ‘ copies corrupted by Marcion’ need mean to us no more
than ‘copies agreeing in a certain reading with Marcion’s copy’:
and Marcion’s copy, prior to his own manipulations, appears by
various signs to have had much in common with the authorities
associated with him in the omission of the Doxology. On the
whole it is reasonably certain that the omission is his only as
having been transmitted by him, in other words that it is a
genuine ancient reading.

80

THE JOURNAL OF PHILOLOGY.

Henceforth conservatism and criticism would be on the same
side. Presently, when the Doxology had founda home after the
fourteenth chapter, every motive for replacing it at the end of
the Epistle was gone. We cannot wonder that the evidence for
retaining it there, and leaving inviolate the continuity of the
fourteenth and fifteenth chapters, is exclusively ancient and

good".

1 Since this article has been in type,
Dr Lightfoot has kindly pointed out to
me an oversight in pp. 66 f., 76. In
the Codex Fuldensis the table of head-
ings to Romans agrees with that in the
Codex Amiatinus &c. only in the latter
part, as Ranke himself observes, p.
xxiii The first 28 headings belong
to a totally different capitulation, and
exhaust the Epistle down to xiv 13.
Then follows No. 24 of the other table,
describing ix 1—5; and so on. The
previous or peculiar headings have no
marks or divisions answering to them
in the text itself. The scribe evidently
saw that his tale of 51 sections could
not be made up without borrowing
elsewhere, and he ventured to save ap-
pearances at the cost of sense. Whe-
ther he had actually reached the end
of the first table or only saw it near at
hand, is less clear. The headings are
not so exactly descriptive as to forbid
the inclusion of xiv 14—28 in § 23; and

F. J. A. HORT.

thus it is certainly possible that we
have two complete and independent
Latin capitulations in which xv xvi
are omitted. More cannot be said till
ancient capitulations generally have
been properly investigated, and this
demands a wide examination of MSS—
Meanwhile it should be observed thas
(1) the Fulda headings have no trace of
the Doxology; and (2) they are loaded
with Augustinian or Anti-Pelagian
phraseology, and cannot therefore be
dated much before 400 at earliest.

The sectional numerals in P, I now
likewise see, may possibly once have
been continued after Rom. xv 14; 1
Cor. xv 51: some numerals have faded
out of sight in almost every epistle,
and in Rom, i—x all have vanished;
cf. Tischendorf M.S.I. v p. xiv. Bat
as the §1 of each epistle (10) except
1 Cor. begins after the salutation, ana-
logy favours the view taken above (p.
71, n. 1).

ON THE ENNEAKRUNOS AT ATHENS.

PAvUSsANIAS, in his description of Athens, after conducting the
reader from the gate at which he entered to the western
foot of the Acropolis, mentions among other objects which he
saw at that spot a fountain called Enneakrunos. But modern
topographers, to a man, have asserted that he must assuredly
have been mistaken; that there cannot be any reasonable doubt
that Enneakrunos was really at the south-eastern extremity of
the city, near the Olympium; and that Pausanias, therefore, in
Mentioning it in this order, must have made an unaccountable
leap over half the diameter of the city, without notice, and
Without mentioning any intermediate object.

Leake (Vol. L p. 238 sq.) explains this extraordinary leap by
Supposing that Pausanias took it in order to connect his narra-
tive respecting the successors of Alexander the Great, which
begins at the eighth chapter and continues down to the four-
tenth; the statues of the Ptolemies before the Odeum, which
Sood near the fountain in question, affording an opportunity
for such connexion.

But, if this was the motive of Pausanias for disturbing the
lucid order of his narration, and puzzling his readers by so gross
& Piece of topographical blundering, we might at least suppose
that he would have confined himself to the Odeum, which was
the cause of his deviation ; instead of which we also find him
describing not only the fountain, but also the temple of De-
meter and Coré, of Triptolemus, of Eucleia, and other objects,
Which he had a better opportunity to do when he afterwards
atrived, in a proper order, at this part of the city. And after
this unseasonable episode, he as suddenly skips back again to
the Kerameikos (1. 14, 5).

Journal of Philology. vou. 111. 6

82 THE JOURNAL OF PHILOLOGY.

But, as Leake admits (p. 239), the narrative is not e
then consecutive; for it is interrupted in the eighth chapte:
describe several statues and other objects which were
doubtedly at the west end of the Acropolis; and indeed Le
himself has placed them there: as the statues of the epon:
of the tribes, those of Amphiaraus, Eirene, Lycurgus, Den
thenes, &c.; and the temple of Ares, with several neighbour
statues, and especially those of Harmodius and Aristogeiton.

Curtius (Attische Studien, No. 11. p. 15) explains the ori
of the supposed blunder of Pausanias as follows: “The m
extraordinary thing is the Enneakrunos episode, which is |
only at variance with any reasonable plan, but is also in it
difficult to comprehend as a separate part of the peregrinati
since a later tour brings us back to the same neighbourho
Nor can we assume a mutilation of the text, since we eviden
have two excursions. Wherefore, if we are not disposed
ascribe this irregularity to circumstances beyond all combi
tion, we are led to the following supposition. The places wh
Pausanias names in his first walk to the Ilissus, viz. the Oder
Enneakrunos, the temples of Demeter and Koré, as well
those of Triptolemus and Eukleia, all lie near the Itonian Gas
Pausanias first entered by this gate, and having afterwa
learnt better (eines Besseren belehrt) and begun a new s
more correct itinerary, which commenced at the princi
entrance on the west, it appears to me not impossible tl
as he had visited and described those points immediately
his first entrance, so that they formed a separate group
his journal, he afterwards inserted the description in anot
place, in order not to separate the remarkable objects in '
inner town. That he has not done this more cleverly n
not surprise us, seeing how little art and practice Pausan
displays in drawing up his description of Attica.”

This explanation seems a great deal more far-fetched s
unsatisfactory than that of Leake. For, first, if Pausanias }
entered the city, as Curtius says he did (Da nun Pausan
zuerst in dies Thor eingetreten ist), how could he have possi
seen these objects? For the Ilissus, on which Enneakrw
and the objects named, are supposed to have been, was 1

84 THE JOURNAL OF PHILOLOGY.

We shall first examine the authorities adduced by Leake
to the contrary (Vol I. p. 172 sqq.), which he considers s0
conclusive as not to leave any “reasonable doubt that Ennea-
krunos was really at the south-eastern extremity of the city.”

The first proofs adduced are the following passages from
Herodotus and Thucydides, which, as they serve for mutual
illustration, we put together.

‘Os &€ avroi AOnvatoe Néyouct, Sixaiws éFeAacas KaTonnpe-
vous yap Tous IleXacryous v6 TH ‘Tunoog, evOevrev oppempevorn,
adixéewy rade. ourav yap aiel tas odetépas Ovyarépas Te Kak
Tous mraidas em’ Vdwp eri tiv ’Evveaxpouvoy, ov yap elvas tovtor
Tov xpovoy ahiot Kw ovdé Toiat ddrAdrowoe “EAAHot oixéras. saws
Sé €rAOorey airat, Tors TleXacryots vad UBpws te Kal cdvyepits
BiacOat odéas.— HEROD, Vi. 137.

TO 5é mpd ToUTOU, 7) axpoTTONLs 7) Viv Ova, TrOALS HY, Kas TO
Um’ avrny pos voToy padioTa TeTpappevov. Texpnptoy Sé Ta yap
icpa év avtH TH axpoTrode: Kat ddXwv Oeay éott, nal Ta E&eo mpi
TOUTO TO pépos THS TOAEWS MAAXOV lOpuTat, TO Te TOU Avos TOU
"Orupriov, nat rd Tlv@tov, nai ro ths Tijs, wat ro ev Atpvats
Avwovicov, & ta apyaiotepa Atovicta tH SwoexaTn Toveitas &
pyvi ’AvOecrnpiar, @oTrep Kal of am’ ’AOnvaiwr “Twves ers Kal
viv vopitovow. ‘Ddputar dé Kai dAXa iepa Ta’Ty apyaia. «aitl
KpnVvn TH vov Lev TOV Tupavvey olTwW oKEevacayTw», ’Evveaxpovry
Kadoupévyn, TO 5é wadat, davepav tav myyav ovady, Kaddpoy
wvopacuevyn, exeivn’ Te eyyvs olan Ta TAEoTOU akia éxypawTo,
kai viv étt atrd Tov apyaiov po Te yautKay Kai és ddda THY
lepay vopiteras T@ VdaTt ypjoOat. xaretras 8é Sia Thy wadaw
TAUTH KATOiKNOLY Kal H axpoTrods méxpe ToddSE Ett Um "AOnvaiwr
mods.—THUCYD. I. 15.

Leake comments on the former of these passages as follows:

1 This is the reading of all the co-
dices, which Bekker has arbitrarily al-
tered to éxetvor. ’Exelyy is here used
pleonastically, as the demonstrative
often is both in Greek and Latin, as
76 yap TeXeuraidr coe riety dvexOev, éxeivo
Sevpl xaréwexyé oe. Lucian, Catapl.
T. 1. p. 638 (Amst. 1743). «atl Képo:Sov
$2 nal Marriddqe xal éxedvous dvotrous

gaoly, lian. V. H. xurr. 15: 'Tyeis &
r@ o¢ dxdnorlay re kal dxpaciay dpré-
forr:s wdvra roury pwdNeora éolxare. La-
cian, Cynic 8, T. m1. p. 543: pdeg 8
KadnXppon yevoudrn, Ade per’ dtovclas
Thy ldlay drwitpero réxyy. Chariton,
De Cher., &c. lib. 1. ¢. 14. In Latin:
Sall. Cat. 7: Cic. De Orat. tn. 18, &e.
See Drakenborch, ad Liv. xxv. 37, init.

ON THE ENNEAKRUNOS AT ATHENS. 85

“Herodotus relates on the authority of Athenian traditions

that the Pelasgi, to whom lands had been assigned at the
foot of Hymettus, as a reward for having fortified the acropo-
lis, were afterwards expelled from thence, because, among other
offences, they ill treated the sons and daughters of the Athe-
nians when the latter were sent (there being at that time no
servants in Greece) to draw water from Enneacrunos. The
fountain therefore was on the side of Athens towards Hymet-
tus, a position confirmed by Thucydides, who thus describes
Athens as it existed before the time of Theseus.”

And he then gives the following version of the passage

fom Thucydides: ‘“‘The city then consisted of that which is
how the citadel, together with that portion of the present city
Which lies below it towards the south. A proof of this fact is
afforded by the temples of the gods; for some of these are in
the citadel, and in the other situation are those of Jupiter
Olympius, of Apollo Pythius, of the Earth, and that of Bacchus
in the marshes, at which the more ancient Dionysiac festival
is celebrated at the twelfth of the month Anthesterion; a
custom still observed by the Ionians, who are descended from
the Athenians. There are other ancient sanctuaries in the
same quarter, as well as the fountain, which from having been
fitted with nine pipes by the tyrants [the Peisistratide], is
called Enneacrunos, but which when the natural sources were
open, was named Callirrhoé: this spring, being near the sanc-
tuaries, was resorted to for all the most important offices of
religion, and still continues to be emploped by women prior to
their nuptials, as well as for other sacred purposes in the
temples. It is in memory of this ancient condition of the city
that the Acropolis is even to this day called Polis by the
Athenians.”

Had Leake extended his quotation a little further, he
would have come upon a flat contradiction between the two
great historians. Herodotus, in the passage just quoted, says
that the ground assigned to the Pelasgi was under Hymettus ;
while Thucydides, in the next chapter but one, says that it
was under the Acropolis (ro re TeXaoycxov xadovpevov To v0
Ti» axpéwolsy, 11. 17): and Pausanias says the same thing

86 THE JOURNAL OF PHILOLOGY.

(77 5é€ aGxporrodcs meptBareiy 1d DRovrcy AéyeTas TOU TeEtyous
IleXaaryous olxnoavras tore uiro THY axpoToy, I. 28, 3). Now
this fact, if Thucydides was right—and he was likely to know
more than Herodotus about Athens—suffices to destroy Leake's
inference, that Enneakrunos “was on the side of Athens
towards Hymettus.” Further, even had the Pelasgi been
seated under Hymettus, that circumstance alone suffices not
to make the inference conclusive. For if they could have gone
to the Tlissus, where Enneakrunos is supposed to have been,
they might easily have proceeded to the Acropolis, where, as
I hope to shew, the fountain really was.

There is another contradiction between the two authori-
ties, which, however, is perhaps more apparent than real
Thucydides says that the fountain was not called Enneakrunos
till the time of the Pisistratide ; yet in the passage cited, we
find Herodotus applying that name to it at a much earlier
period. The only way in which we can explain this contra
diction is by supposing that Herodotus used that name, instead
of Kallirrhoé, by a prolepsis, in order to prevent confusion with
some other fountain, or stream. Nor was this precaution ur-
necessary, as there evidently were, or had been, two Kal-
lirrhoés at Athens, and of these one, no doubt, was at the
Ilissus. For in the Platonic dialogue entitled Aztochus, So-
crates is represented as having issued from a gate of the city,
and as having got to the Ilissus (which, therefore, was out-
side the walls), when he heard somebody calling him, and on
turning round, beheld Kleinias running towards Kallirrhoé
(E&toyvre prot és Kuvécapyes xal yevopévw pot xata tov ‘I\e-
cov, Sinke dwvy Boavrds tov, Lwxpates, Laxpates. as Se ext
otpadgels mepieaxcrouy omdbev cin, KXewiay opa tov ‘A foxov
Géovra émt Kaddiponv, Axioch. init.). Now as Thucydides tells
us that, after the time of the Pisistratids, the spring Kallirrhoé
came to be called Enneakrunos, and as we find that this
latter name was in use several centuries after, since Pausanias
employs it, we are compelled to the conclusion that the author
of the Aziochus could not have been alluding to the foun-
tain, but to some part of the Ilissus called Kallirrhoé, For
it would have been indeed absurd to have continued calling

ON THE ENNEAKRUNOS AT ATHENS. 87

Foneakrunos, Kallirrhoé, after its conversion into an artificial
fountain had concealed those natural springs, which, when
they lay open to view, had occasioned its original appellation.
And here probably may be detected the source of the error
which led writers of a very late period to place Enneakrunos
on the Tlissus.

That Enneakrunos was distinct from, and coexisted with a
Kallirrhos, may also be shewn from other authorities. Pliny,
enumerating the Attic fountains, says: “Cephisia, Larine, Cal-
roe, Enneacrunos” (NV. H. lib. tv. 7, 11): where modern editors,
inluding Sillig, have printed: Calliroe Enneacrunos: without
4comma between the words, assuming that Pliny wrote them
down as identical, and, as Meursius says, by apposition. But
as Pliny was reckoning up the actual number of the fountains,
he would surely have given his readers notice that these were
Only two different names for the same object, and have inserted
#eu, or some such word, between them.

Solinus, in a manner still more marked, mentions Kal-
Yirrhoé and Enneakrunos as two distinct fountains; “Callirhoen
stupent fontem: nec ideo Enneacrunon, fontem alterum, nulli
Tei numerant” (Polyhist. cap. X11): for which he has been
blamed by several eminent critics, and defended by one or two
others. Duker (ad Thucyd. 11. 15, note 8) takes part against
him, but does not seem to have been aware of the passage
which we have quoted from the Aziochus. He appeals to the
testimony of Harpocration, Hesychius, and other grammarians.
But when these writers say under 'Evveaxpouvos: xpyvn tis
éy "A@nvass: aporepov 8 éexadeito Kadndzpon, they only repeat
what we have already learnt from Thucydides, and do not
exclude the possibility of another Kallirrhoé.

When Statius writes :

Et quos Callirhoé novies errantibus undis

Implicat, et raptae qui conscius Orithyiae

Celavit Geticos ripis [lissus amores—(Theb. x11. 629),
he confounds, as a Latin poet easily might, the original spring
with the subsequent fountain ; but his mentioning it separately
from the Ilissus, shews that he considered it to have been un-
connected with that river.

ws

3 THE JOURNAL OF PHILTOLIOGY.

If we examine the passage of Thacrdides adduced br Leake
a Little closer, we shall find, I think, nothing contrary to thy
vicw, of in favour of placing Enneakrunos at the [lieu
Leake translates the words: dxeay re eyyts otoy, «1.2: “this
spring. being near the sanctuaries, was resorted to for all the
In’at important offices of religion.” This, however, is not what
Lis author says, bnt, that the ancient Athenians, who dwelt in
the acropolis, usel the spring because it was near them’. This
inakes a very essential difference ; since, as the temple of Zeus
Olympius iz one of the sanctuaries named, if the historian
really said what he is made to sav, it would be a strong argu-
ment in favour of Enneakrunos being on the Ilissus, and nest
the temple in question. But, properly interpreted, his words
afford as strong an argument the other way. For the assumed
site of the fountain is at least three quarters of a mile from the
western, and only, entrance to the acropolis (it took me a full
quarter of an hour to walk thither at a fair pace) ; and it could
not therefore, with propriety, be called near those who dwelt in
it. The passage, in fact, is in favour of the fountain being at
the seropolis.

Let us further remark that the Olympium lies S. E. of the
acropolis, and therefore when Thucydides mentions that and
other sanctuaries as lying about S. of it, it is probable that be
named the Olympium first (together with the Pythium, a kind
of adjunct to it,—Strabo, rx. 404), because it was the eastern-
most of the group, and that the other temples mentioned lay to
the west of it. And it is a confirmation of this inference that
the temple of Dionysus in the Limne, which is one of them, 18
known to have been under the acropolis, Thucydides then al-
ludes generally to other temples in this vicinity without naming
them, and mentions the fountain last ; whence a fair inference
may be drawn that this was the westernmost of the objects

' It may be obverved that Leake the temples, but to the houses of the
makes another little wlipin translating bride and groom. Thus Photius: &os
de ANA rev lepov, an well as for other qv rots yapodouw ’A@y»mor Aovrpa pera-
nncrad purpowen in the temples.” The wéuweoGae davrois xara rhe rod yduov
Inat words are not in hia author; and = tudépay. Voo. Aovrpodépes, 231, 17.
in {not the Aourpd were not brought to

ON THE ENNEAKRUNOS AT ATHENS. 89

named, and consequently near the entrance to the acropolis, as
we have already inferred from Pausanias, as well as from the
words of Thucydides himself. On the other hand, had it been
the easternmost object, and near the temple of Zeus, he would
turely have mentioned it in connexion with that sanctuary.
We may also observe that Thucydides names among the group
atemple of Gé, or the Earth, and we know from Pausanias (I.
22, 3) that there was really a temple of that divinity near the |
south-western extremity of the acropolis. There was also in-
deed in the Olympian enclosure a temenos, or piece of ground,
consecrated to Gé; but Thucydides is speaking more particu-
larly of temples (fepa), and uses the word %Spuraz in connexion
with them, are built, or founded. But—not to press the mean-
ing of iepdv too closely—were he even alluding to the sanctuary
of the Olympian Gé, still that also was to the W. of the temple
of Zeus, as it lay towards the Itonian Gate and monument of
the Amazon (Plut. hes. 27).
_ The next piece of evidence adduced by Leake is the follow-
Ing passage from Hierocles, in the preface to his Hippiatrics:
| Tapayrivos S8 ictoped tov tod Avés ved xatacxevatovras ’AGn-
mous "Evveaxpovvov mdnciov eicedabnvar Wodicacba ta &x
Ts ’Arrins eis TO Gotu beiyn Gmravta: which Leake takes to
Mean, “that when the Athenians were building the temple of
Jupiter near Enneakrunos, they ordered all the beasts of burden
i Attica to be brought to the city.” And he proceeds to re-
Mark: “There was no temple of Jupiter at Athens, of any cele-
brity, except that of Jupiter Olympius, and its remains are
found near the source of water at the south-eastern extremity of
the site of Athens” (p. 174).

But, as there was no temple of Jupiter, or rather Zeus,
at Athens, of any celebrity, except that of the Olympian—
and that indeed was of world-wide notoriety—where was the
necessity for identifying it as being near a fountain? It would
seem very absurd to define the site of St Paul’s cathedral
as near Peel’s statue, or of the Tower of London as near the
postern well; though on the other hand we might naturally
indicate to a stranger the statue or the well as being near
structures of such universal notoriety. Wherefore I take it

90 THE JOURNAL OF PHILOLOGY.

that mAnolov is not to be construed with caracnevalovras baat
with eicerabyjvat, and that the sense is: “they ordered th—#é¢
beasts of burden to be driven into the city near Enneakrunos— "
And so I find the passage translated by Meursius, who puf®—3
a comma between exstruentes and Enneacrunum: “Tarentinm 4
vero refert, Athenienses, templum Jovis exstruentes, prop=©
fontem Enneacrunum decreto mandasse, ut omnia tota Attic#=™
jumenta in Urbem abducerentur” (Ceram. Gem. c. 14, Operm—»
t. Lp. 493).

The next proof is from the Etymologicum Magnum : "Eyvea——
xpouvos, Kpnvn ‘AOnvnot wapa tov "Iktcody mporepoy KadX——
Mpon éaxev. Now the author of this lexicon, who flourished
about the tenth or eleventh century, is the first authority~—
who says disertis verbis that Enneakrunos was at the Ilissus
and if this was a fact, it is a singular circumstance that it
should not have been mentioned by Harpocration, Hesychius,
and the older lexicographers, Even in the lexicon of Photius,
who lived a century or two earlier than the author of the
Etymologicum, we read: ra 5¢ Aourpa exdutlov cx ris viv per
"Evveaxpovvov xadoupévns xpnvns, mpdtepov 5¢ Kaddspéns (voc.
Aovtpogopos, 231. 23): where the more recent hand of a person
who thought that he knew better than Photius has added (“ad-
didit m. recens”) adda xal viv abrn Kaddupon xadetras. And
in like manner Suidas, also a late lexicographer, reversing the
words of Thucydides and the earlier grammarians: KadAspon
xpnvn » év ‘AOnvats iris mporepoyv "Evveaxpouvos éxaetro.

What conclusion can be drawn from these variations, which
occur only in works belonging to the tenth or later centuries,
than that between their time and that of Photius, who flou-
rished in the ninth century, the true site of Enneakrunos
had fallen into oblivion, probably from the spring having been
. diverted, or become extinct ; and that the grammarians of that
later period, because they found that a spring called Kallirrhod
and a fountain called Enneakrunos had once been identified,
now began to imagine that the fountain was that other Kallir-
rhoé on the Ilissus? For that the real Enneakrunos, an artificial
fountain, should ever have regained the name which it bore
when its sources lay open, surpasses all belief.

ON THE ENNEAKRUNOS AT ATHENS. QI

Leake’s last proof, from ancient authorities, is the following
fragment of Cratinus (Schol. in Arist. Eg. 523) :
"Ava “ArrodXov, Tay exay Tay pevpatov’
Kavayovos mryai, Swdexaxpovvoy (1d) oropa,
"Tasoods ey (19) hapvyy. Té ay elrrouml co;
Eé x) yap émBvce tis avtov 1d ordpa,
“Awayta Taira xataxdvoes Trotnpacty.

We need not dwell, I suppose, on this passage. Twelve are
not nine ; nor does it follow because two objects are mentioned
in the same lines, that they were therefore together in place.
On the contrary, it rather affords a presumption that they were
separate and distinct objects.

It remains to examine the proofs which Leake adduces
(p. 175) from modern appearances and names.

There is, it is said, near the Olympium, a streamlet of water
issuing from the foot of a ledge of rock, which here crosses the
bed of the Ilissus, so that in times of rain the spring is en-
veloped in a small cascade of the river falling over the rock;
but which, when the bed is in its ordinary state, that is to say,
dry, or nearly so, forms a pool, which is permanent in the midst
of summer. The spring is still called, as well as the river itself,
Kallirrhoé [KadX:6p0n], so that there cannot be any question
of the identity.

To the same purpose Dr Wordsworth, speaking of Callirrhoé,
says, “The current of the river, or torrent rather, is here di-
vided into two streams; the one nearer the left bank comes
down over a stone bed cut and worn into a large and deep
trough ; the other division of the stream finds its way through
the rock by subterranean artificial xpovvos, or pipes, bored
through it, which suggested the description of Cratinus: seven
of them are yet visible’.”

On these passages it may be remarked: first, that there is
nothing surprising that the pool, or rather the river itself at this
point, should still be called Kallirrhoé, seeing that it bore that
name ages before, and at least as early as the time of Socrates.

1 Athens and Attica, ch. 31. Col. Mure also observed seven orifices.— Tour in
Greece, Vol. 11. p. 86.

g2 THE JOURNAL OF PHILOLOGY.

But this circumstance does not identify it with Enneakrunos, as
Leake supposes, on the assumption that there could have been
only one Kallirrhoé.

Secondly, with regard to the pipes mentioned by Dr Words-
worth, which I was not fortunate enough to see, it may be
observed that seven are not nine.

Again: Enneakrunos was originally a natural fountain, and
its sources when it retained its first name of Kallirrhoé, were
open to the view; while the object described by Dr Wordsworth
could never have been a natural spring at all, but merely, as he
himself says, an apparatus for conducting the water through the
rock from the upper stream into the pool. And this apparatus
was evidently nothing more than part of a Turkish fountain.
For Wheler, who visited Athens towards the end of the seven-
teenth century, saw at this place two Turkish fountains (* De-
scending yet a little further by the river, some rocks seem to
stop its course; whereby the water begins to appear again, and
settle in a kind of pool upon the rocks....The Turks after their
mode have accommodated two fountains to this spring,” Journey,
éc., p. 379). And these were still to be seen when Chandler
went there about a century afterwards. For he says: “The
current is now conveyed into the town, and only the holes, at
which it issued into the cistern, remain. These are in the rocky
bank next to the temple of Jupiter Olympius, which is in the
way to the gate dividing the cities of Theseus and Hadrian, and
not remote. At a little distance is a modern ruinous fountain.”
(Travels in Greece, Vol. 11. ch. 16, p. 95. ed. 1817.) Whence
it is clear that the holes which Chandler saw were the re-
mains of the other Turkish fountain seen by Wheler.

In fact, a fountain in the middle of a river, which in ancient
times, too, was much more abundantly supplied with water than
it is now (though even at present it would be subject at times
to be enveloped in a cascade), seems a palpable absurdity. On
this subject Col. Mure very justly observes, although he accepts
the Ilissus for the site of the fountain: “From this arrange-
ment of the pipes it may be inferred that any little moisture
the bed of the Ilissus occasionally afforded, was also made
available for the supply of the fountain. Its purity, however,

ON THE ENNEAKRUNOS AT ATHENS. 93

could hardly fail to be disturbed by the waters of the stream
when in a swollen state’.”. The Turkish fountains, constructed
“after their mode,” when the stream was become drier, were
evidently only a sort of water-works. Leake, however, at the
conclusion of his argument combats such a notion as follows:
“That Enneakrunos, or the ancient Callirrhoé, was a separate
vein of water, and not an artificial derivation from the Ilissus,
was proved by an excavation which the primates of Athens
made about the year 1804, at the pool above mentioned, when
abrisk stream of water made its appearance, evidently distinct
from the Ilissus, and having a course from the northward into
the above-mentioned pool of water. In fact the Ilissus receives
several subterraneous veins of water from Hymettus and An-
chesmus: these form pools in the dry bed of the torrent, which
ni resorted to by the Athenian women for the washing of
en.”

This paragragh proves too much, the second sentence nulli-
fying the first: for, as there are several such veins and pools, it
is evident that Enneakruni might be produced ad libitum.
Nobody denies that water might be found near the Ilissus by
digging for it; but such a proceeding would not make a natural
spring, as the Kallirrhoé was which existed before the time of
the Pisistratidee.

To recapitulate. It has been shewn that Pausanias places

eakrunos at the western extremity of the Acropolis; that
the evidence of Thucydides corroborates this position; and that
Herodotus says nothing which may not be reconciled with it.
It has been further shewn, from the dialogue named Asiochus,
that a point on the Ilissus was called Callirrhoé, and that it
Could not have been identical with Enneakrunos because, as we
ae told by Thucydides, the Callirrhoé which was converted into
that fountain afterwards lost its name. Yet that an Enneakru-
Bos and a Callirrhoé subsequently co-existed is evident also

fom Pliny and Solinus enumerating them as distinct fountains.
Solinus represents Callirrhoé as much the more magnificent one ;
and so of course it would be, from the natural cascade, and the

1 Tour in Greece, ubi supra.

04 THE JOURNAL OF PHILOLOGY.

largeness of the pool at this point of the Ilissus. It has also
been shewn, that, about the tenth or eleventh century of our
era, writers began to confound Enneakrunos with this Callir-
rhoe. Lastly, the appearances which still exist at this spot,
do not coincide with a fountain of nine pipes, nor indeed with
any fountain at all, properly so called; and are, in all proba-
bility, the remains of Turkish water-works.

If these things have been proved, then the following ad-
vantages are gained: Pausanias is found to be consistent in his
topography, which increases the value of his evidence regard-
ing other matters; while the group of temples, &c., which he
places in the vicinity of Enneakrunos, are discovered to have
been within, instead of without, the city walls; a situation much
more probable, and more suitable for them, especially in the
case of the Odeum.

THOS. H. DYER

April 25th, 1870.

ON THE LENGTHENING OF SHORT FINAL
SYLLABLES IN VERGIL.

Tar fact that Vergil allowed himself certain licences in the way
of lengthening short final syllables, licences which were wholly
rin great part avoided by his immediate predecessors in
poetry, has, as was natural, often been noticed. The most
detailed discussion of the matter is that of Philip Wagner in
no. XU of bis Quaestiones Vergilianae. Gossrau has a para-
graph upon it in the “ Excursus de Hexametro Vergilii” affixed
to his edition of the Aeneid of 1846: but this paragraph is, as
the writer himself professes, little more than a simpler repro-
duction of what Wagner had said. The subject is treated
briely by Lachmann (on Lucr. 2. 27) and comprehensively by
lucian Miller (De Re Metrica, p. 324333): but A. Weidner
(Commentar zu Virgil’s Aeneis I und 11) takes no notice of the
instances occurring in those books. While Ph. Wagner and
Lacian Miiller would account for these licences almost entirely
oa the ground of the position of the word in the verse, the
Plutine critics ('Ritschl, Fleckeisen, and W. Wagner) have
thought that in some cases at least Vergil was not unconscious
of the same uncertainty of quantity which prevailed in the
earlier period of Latin poetry. The object of this paper is to
show that neither explanation is wholly true: that Vergil, while
probably unconscious of any grammatical or etymological pro-
prety in the employment of these scansions, still did not em-
ploy them without due selection and a regard to the usage of

' Ritschl, Prolegomena to Trinum- p. 17, foll. W. Wagner, Introduction
mus. Fleckeisen, Neue Jahrbticher, ux1. to Aulularia.

96 THE JOURNAL OF PHILOLOGY.

the earlier writers, however imperfectly this usage was under-
stood in his own day.

The most decided innovation’ introduced into the hexameter
by Vergil, the lengthening of the first que in verse-beginnings
like “ Liminaque laurusque Dei” or verse-endings like “Noe-
monaque Prytanimque”, need not detain us, as it is an obvious
imitation of Homer's Aaprovy te KAutioy te, Ilpofonvep te
KAcuos te x.7.4. In Homer ce is mostly lengthened before
double consonants, liquids, and sibilants; and Vergil has scru-
pulously followed his master. Of the sixteen instances collected.
by Wagner fourteen present gue lengthened before a double
consonant: the other two are “Liminaque laurusque” (A. 3-
91) and “ Eurique Zephyrique” (G. 1. 371). Neither is it neces—
sary to dwell upon endings like “molli fultus hyacintho”, “lim—
quens profugus hymenaeos”, which, like Catullus’ “non de—
spexit hymenacos”, “novo auctus hymenaeo”, are mere imita-—
tions of the Greek.

The rest of these licences are distinctly traceable to Romax—*
sources, and require a longer consideration.

The early poetry of Greece and Rome is marked by con——
siderable uncertainty of quantity: thus in the Homeric poems===
we have both ’avjp and ‘avyp, didos and didos, "arovéco Oat ance
"atrovéecOar and so on. This uncertainty is observable in Latin
chiefly in the final syllables of nouns and verbs: a fact probably
due in great measure to the rule of Latin accentuation, which
forbade the accent to fall on the last syllable. Final syllables
which were long by nature were obscured by the backward
position of the accent, and gradually became short. This pro-
cess did not stop at the Augustan age, but continued till even
the final o of the present indicative was shortened by hexameter
poets. Verse-writing at Rome began at a time when the ten-
dency to shorten final vowels originally long had commenced,
but had not nearly prevailed over the natural quantity. This

1 Lachmann, 1 c. ‘‘quo primo Maro _instances in the remaining fragments
usus est”. Lucian Miiller, p. 822, of Ennius or Lucilius, nor in Lucre-
quotes a verse of Attius (ap. Festum, ius, Catullus, or the remaining verses
p. 146): ‘“Calones famulique metal- of Cicero.
lique caculaeque”: but there are no

OF SHORT FINAL SYLLABLES IN VERGIL. 97

state of things 1s most clearly discernible in Plautus: but it is
sufficiently obvious even in the stricter measure of Ennius. Lu-
cilius, as was natural, allowed himself, to a certain extent, a
similar freedom ; but the poets of the later republic, Catullus
and Lucretius, became much stricter. Except in Greek endings
like “despexit hymenaeos” &c. Catullus never lengthens a
short final vowel, unless we are to count the much-emended
line 100. 6 “Perfecta exigitur unica amicitia”, to which Mr Ellis
apparently does not object. Two instances have been restored
to Lucretius by Mr Munro: 2. 27 “Nec domus argento fulge¢
auroque renidet” and 5. 1049 “Quid vellet facere ut sciret
animoque videret”: but even these were altered by Lachmann
or with his approval, for they are solitary in his author. There
is nothing of the kind in the fragments of Cicero’s verses.
Vergil deserted the strictness of his immediate predecessors, and
recurred, to a certain extent, to the practice of Ennius’. It
will be worth while to compare the usages of the two poets in
detail.

(1) Lengthening of final syllables in r. (a) Nouns. Mas-
culines in or. As far as I can ascertain there is no instance in
the fragments of Ennius where this ending is short® either in
arsis or thesis. Ennius writes not only :

‘‘ Postilla, germana soror, errare videbar” (Ann. 42),

“O pater, O genitor, O sanguen Dis oriundum” (Ann. 117),

“Qui clamor oppugnantis vagore volanti” (Ann. 408),

“Tollitur in caelum clamor exortus utrimque” (Ann. 422),

“Imbricitor aquiloque suo cum flamine contra” (Ann. 424),
but also

“Clamor in caelum volvendus per aethera vagit” (Ann. 520),

! Horace is much freer than Catul-
lus, as Vergil is than Lucretius. Ex-
cept ‘‘Teucer et Sthenelus sciens"
(1 C. 15. 24), which he altogether ro-
jects, and ‘Si non periret immisera-
bilis” (3 C. 5.17), and “ Ignis Dliacas
domos” (1 C. 15. 86), about which he
has doubte, Mr Munro admits the rest
of these scansions in Horace without
hesitation. These amount to about

Journal of Philology. vou. 111.

ten: but it should be remarked that
none of them occur in the fourth book
of the Odes, the Epistles, or the Ars
Poetica, in which Horace was writing
at his best.

* So in Plautus, according to Fleck-
eisen (ap. C. F. W. Miiller, Plautinische
Prosodie, p. 42 foll.), it is exclusively
long.

98 THE JOURNAL OF PHILOLOGY.

unless with Lachmann we follow the indication given by Quin-
tilian' and read clamos. Compare with the lines of Ennius
above quoted the following from Vergil :
“Omnia vincit Amor, et nos cedamus Amori” (E. 10. 69).
“ Aequus uterque labor: aeque iuvenemque magistri” (G. 3.118).
“Nam duo sunt genera, hic melior, insignis et ore” (G, 4. 92).
“Luctus, ubique pavor, et plurima mortis imago” (A. 2. 369).
“Et Capys, et Numitor, et qui te nomine reddet” (A. 6. 768).
“ Considant, si tantus amor, et moenia condant” (A. 11. 32).
“ Quippe dolor, omnis stetit imo volnere sanguis” (A. 12. 422).
“Et Messapus equum domitor, et fortis Asilas” (3b. 550).

Lucian Miiller thinks the caesura sufficient to account forall
these cases both in Ennius and Vergil, denies the possibility of
clamor in thesis, and asserts that in the second part of the sixth
century A.U.C. this syllable was mostly shortened. No case of
such shortening, however, as has been seen, can be quoted from
Ennius. Vergil, who was probably ignorant of the reason which
made Ennius write as he did, viz. the original length of this
syllable, which corresponds to the Greek -wp or -wy, and who
only wished to give an antique flavour to his verse by suggest-
ing such echoes of the Ennian hexameter, would never have
dreamed of using the final or long except in arsis: but Miiller
can hardly be right in applying the same measure to both poets.

How purely a matter of form this licence was with Vergil
will become apparent when we consider how far, and (from an
etymological point of view) how unjustifiably, he pushes his
employment of it. Ennius, using tubar masculine, could write

“Interea fugit albus iubar Hyperionis cursum” (A. 547),
but no grammatical propriety can be alleged for such scansions
as
“Desine plura, puer, et quod nunc instat agamus” (Verg. E. 9.66),
“Si quis ebur, aut mixta rubeat ubi lilia multa” (A. 12. 68) ;
still less for

“Pingue super oleum infundens ardentibus extis” (A. 6. 254).
The lines |
“ Ostentans artemque pater arcumque sonantem” (A. 5. 521)

11, 4.18. “ Arbos, lahos, vapos ctiam et clamos aetatis faerant.”

10Oo THE JOURNAL OF PHILOLOGY.

Acipo paynoopevos, eet odrt prot alriol eiocy.
Xwopevos, br’ dpirrov "Ayacay ovdey ericas.
“Fatalisque manus, infensa Etruria Turno” (A. 12. 232=_)
and
“ Sicubi magna Iovis antiquo robore quercus” (G. 3. 38=)
may perhaps be considered an extension of this licence. 50
A. 3. 112 “ Idaeumque nemus: hic fida silentia sacris”.

Whether Ennius lengthened the dative plural in -bus cannot
be ascertained, and such a scansion is infrequent in Plautus
But Vergil does not hesitate to write (A. 4. 64)

“ Pectoribus inhians spirantia consulit exta”.

(b) Verbs. The only case is “Terga fatigamus hasta” (A
9.610), a quantity for which no analogy can be proved in Ennius
or Plautus.

(3) Endings in é Third person singular of verbs. Tb®
-at of the indicative present lst conjugation, though long DI
nature and frequently scanned accordingly in Plautus, is <
variable quantity in Ennius, but mostly long.

Compare
“Solus avem servat: at Romulus pulcher in alto” (Ann. 83) s=
“Inde sibi memorat unum superesse laborem” (Ann. 159),
“‘Quae nunc te coquit et versat in pectore fixa” (Ann. 340),
“Tum timido manat ex omni pectore sudor” (Ann. 399),
with
“ Missaque per pectus dum transit striderd¢ hasta” (Ann. 365)—
Vergil has no imitation of this.
-At of the imperfect is long in Plautus, and so in Ennius
even in thesis, Ann. 314,
“ Noenum rumores ponebat ante salutem”:
but short Ann. 141,
‘Volturus in spinis miserum mandebdt homonem”.
So Vergil, (but only in ‘arsis) E. 1. 39, A. 5. 853, 7. 174, 10. $83,
12. 772.
“ Tityrus hinc aberat: ipsae te, Tityre, pinus” :
1 It would be very rash with Fleck- Gyas revocabat: ecce Cloanthum” (5.
eisen and Ladewig to attribute to Ver- 187), or ‘“‘Arduus, effractoque inlisit

gil, on the sole authority of the Codex ossa cerebro'’ (5. 480).
Romanus, such lines as ‘‘Cum clamore

OF SHORT FINAL SYLLABLES IN VERGIL. 101

“ Nusquam amittebat, oculosque sub astra tenebat” :

“ Regibus omen erat: hoc illis curia, templum ”:

“ Per medium qua spina dabaé: hastamque receptat”:

“ Hic hasta Aeneae stabat: huc impetus illam ”.

-Et in the present and future indicative and imperfect sub-
junctive is long in Plautus and so in Ennius even in thesis,
Ann. 86:

“ Omnibus cura viris uter esset induperator ” :
in arsis, Ann. 100, 171, 349, 409:

“Nec pol homo quisquam faciet inpune animatus” :

“‘Inicit imitatus: tenet occasus, iuvat res”:

“ Pugnandi fieret aut duri finis laboris” :

“ prandere iube¢ horiturque ” :
but decét Ann. 229:

“ Nec me rem decet hanc carinantibus edere chartis”.
Compare the cases from Lucretius quoted above and Vergil A.
1. 308, 651:

“ Qui teneat, nam inculta videt, hominesne feraene ” ;

“ Pergama cum peteret inconcessosque hymenaeos ”.

-It of the present (3rd conjugation) is constantly short in
Ennius, but long Ann. 123,

“ Mensas constituzé idemque ancilia”

(if this be the present), 346, 484,

“Sensit, voce sua nictit ululatque ibi acute”:

“Multa foro ponit et agea longa repletur”.

So occasionally in the comedians (C. F. W. Miiller, p. 79).
_Vergil, E. 7. 23, A. 9. 9, 10. 433, has

“ Versibus ille facté; aut si non possumus omnes”:

“Sceptra Palatini sedemque petzt Evandri” :

“Tela manusque sintt. Hic Pallas instat et urget”.

-It of the fourth conjugation is long in Ennius Ann. 258,

“ Alter nare cupzt, alter pugnare paratust”

(if cugt be from cupire).
Comp. Ann. 419
“ Tt eques et plausu cava concutit ungula terram” :
432 “ Configunt parmam, tinnit hastilibus umbo”:
386 (thesis) “Infié, O cives, quae me fortuna ferocem”.
Vergil has no instances.

102 THE JOURNAL OF PHILOLOGY.

-It of the first future is short Enn. Ann. 153
“ Hac noctu filo pendebtt Etruria tota”,
and there is no instance in his fragments of its being length-
ened. Vergil however has erst twice; E. 3. 97, A. 12. 883:
“ Tpse ubi tempus erié, omnes in fonte lavabo” :
“Te sine, frater, ertt? O quae satis ima dehiscat ”.

-It of the present subjunctive and second future is long in

Plautus: so also Ennius has fueri¢ and dederit Ann. 128, 165,
“Si quid me fuertt humanitus ut teneatis” :
“ At sese, sum quae dederié in luminis oras”.

Compare velit Ann. 200,

‘“Vosne velit an me regnare era quidve ferat Fors”.

No instances in Vergil.

-It of the perfect indicative is often long in Plautus (refer-
ences in Miiller, Pl. Pr. p. 71), but Ennius, though he writes
(Ann. 599)

“Qua murum fier voluzé, urgentur in unum”,
makes it mostly short: a strange fact, as the original length of
the vowel is unquestionable. The long scansion was afterwards
taken up by Ovid in the case of words compounded with eo
(subiié &c.), and Vergil writes (G. 2. 211, A. 8. 363)

“ At rudis enituzt impulso vomere campus” :

“ Alcides subitt, haec illum regia cepit”.

In A. 10. 394 Vergil extends this licence to lengthening the
last syllable of caput. Prociil (“arcemque procul ac rara domo-
rum” A. 8. 98) stands by itself.

It will be seen from the instances quoted that Vergil, though
on the whole following the lines marked out by the early
Roman poetry, never allows himself these licences except in
arsis, and but seldom where there is not a slight break in the
sentence’. By Ennius these limitations were far less rigorously
observed. Vergil considered such scansions as antiquarian orna-
ments, and as such they were to a certain extent taken up from
him by Ovid, Propertius, Tibullus and the later poets.

1 Comp. Haupt on Ov. Met. 8. 184. ses voretund aut.” This remark would
‘Ovid setzt kurze Silben statt langer cover a great many, though byno means
in der Hebung vor griechischen Wor- all, of the cases quoted from Vergil.

tern oder in der Cisur des dritten Fus-
H. NETTLESHIP.

ZNEAS’ VOYAGE ROUND SICILY.
Vera. “n. u1. 687—706.

Tue death of the lamented Professor Conington must awaken
afresh interest in the study of those authors which he called
peculiarly his own. The third volume of his edition of Vergil
seven now in the press, and the eagerness with which every
Vergilian scholar will open the new work, will be sadly dashed
by the thought that the pure and noble soul that dictated it is
hot here to listen to the praises which it will no doubt receive.
In venturing to criticise some of Professor Conington’s past
Work on Vergil, it is assuredly from no wish to set up my own
knowledge of the language or the author against his. The
passage, however, named above as it appears in his translation
was selected by more than one reviewer as particularly suc-
cessful both in sound and sense. Now Conington himself has
Well spoken of Vergil’s language as*extremely sensitive; and
this is precisely what his ballad metre is very often not. In
this passage he, after Dryden, Pitt and Symonds, leaves out a
Variety of little words, which not only give the lines much of
their life and point, but the omission of which conveys a posi-
ively false idea. His translation is thus:
When lo! from out Pelorus’ strait
The northern breezes blow!
We pass Pantagia’s rocky gate,
And Megara, where vessels wait,
And Thapsus, pillowed low.
Before Sicania’s harbour deep
Against Plemyrium’s billowy steep
Ortygia’s island lies.. ...and then,
Pass rich Helorus’ stagnant fen.
Pachynus’ lofty cliffs we graze,
Projecting o'er the main,

104 . THE JOURNAL OF PHILOLOGY.

And Camarina meets our gaze,
Which fate forbad to drain,
And Gela’s fields, and Gela’s wall,
And Gela’s stream, that names them all.
High towering Acragas succeeds,
The sire one day of generous steeds :
Selinus’ palms I leave behind
And Lilybeum’s shallows blind.
Any one would suppose from these verses, that the same north
wind took®Aineas from Aitna to Pachynus, and from Pachynus
to Lilybeeum, and round to Drepanum; and that he saw all the
towns and harbours named under the same circumstances. In-
possible fas this would be in an ancient ship, it would not
trouble most Vergilian critics, who are fond of asserting that
Vergil is entirely careless of such delicate points as the direction
of the wind. Now I believe that Vergil has throughout this
passage very accurately defined the wind throughout ness
course, and just how it would carry him along or off shore, by
those very little words which Professor Conington leaves out—
probably as injuring the run of his metre.
Let us consider exactly where Aineas is and how he must
get round Sicily.

panne

Pachynus

ENEAS’ VOYAGE ROUND SICILY. —‘105

He is off the land of the Cyclopes, he has cut his cable
-and rowed till Polyphemus is out of his depth. Here he pre-
pares to take the favorable wind (ventis intendere vela secundis,
L 683). This however being contrary to his orders, he is
debating a return when the wind changes to the north by
special divine aid. The course now is given thus:

Ecce autem Boreas angusta ab sede Pelori

Missus adest, vivo pretervehor ostia saxo
Pantagis Megarosque sinus Thapsumque jacentem.
Sicanio pretenta sinu jacet insula contra
Plemmyrium undosum, nomen dixere priores
Ortygiam * * * et inde

Exsupero prepingue solum stagnantis Helori.
Hinc altas cautes projectaque saxa Pachyni
Radimus, et fatis numquam concessa moveri
Apparet Camarina procul campique Geloi
Immanisque Gela fluvii cognomine dicta.

Arduus inde Acragas ostentat maxima longe
Mcenia, magnanimum quondam generator equorum.
Teque datis linquo ventis, palmosa Selinus,

Et vada dura lego saxis Lilybeia cecis.

It is perhaps needless to say that Boreas with Vergil as
with Herodotus and Strabo is the cold northern wind, so well
known in Italy. Not always however due north, as those com-
mentators seemed to think, who are puzzled on Ain Iv. 442, to
know how “Alpine Boresw” could blow at once hine and ¢lline.
Vergil is there describing what he must have seen yearly at
Andes, the winds coming down from the Alps and attacking
the trees, now from one peak or gorge, now from another, now
N.W., now N.E., but always Borex, in the northern quarter.
Here Boreas is North—further defined as angusta ab sede
Pelori. It carries Aineas rapidly along the shore—prastervehit
—near enough to see Thapsus, although low—jacentem.

In fear of the well-known passage round Plemmyriuwm undo-
sum (undosum = IIAnpputpiov) he pays his vows to the Syra-
cusan deities, and now runs nearer the coast, s0 as to force his
way over the oozy estuary of Helorus, for this I believe is the

106 THE JOURNAL OF PHILOLOGY.

only meaning of prepingue solum consistent with Vergil’s wi-
form use of exsupero, and then still closer, grazes the cliffs of
Pachynus. He has now turned the corner, and let him sail
near to the wind as an ancient ship could, at the utmost seven
points, the right-hand coast will always be farther and farther
off. As he beats up, Camarina is procul—far off. The plains of
Gela may be distinguished at the bottom of their bay, and %
their city on account of its huge size—immanis,—but he does
not say, as Professor Conington makes him, that he sees the
river. Next is seen Agrigentum, from a distance—longe—be-
yond ordinary vision. But in a line filled with words implying
vastness, he tells us that only its immense size and towering
position made it visible. This is very nearly the last point on
the coast which Aineas could reach by beating against a north-
ern wind. His course lies to Lilybeum, the wind must change
to serve him. Accordingly we find that he is enabled dats
ventis to pass Selinus, run close to the Lilybsean shoals and
enter Drepanum. Mr Conington, it will be observed, entirely
leaves out in his translation any hint that Aineas ran nearer t
one coast than the other, or that the wind even changed, If it
be said that dats does not necessarily imply change, neither
does it imply continuance; it does imply such a wind as he
wanted, and that could not be the existing one.

Can any one say why Syracuse is not named in the above,
when so many other towns are, founded long after the supposed
date of the voyage ?

W. EVERETT.

Oct. 1869.

Di a

ON THE CHRONOLOGY OF ST JOHN V. AND VI.

Tae cardinal question which Gospel Chronologers have to
€nswer is this:—does St Jobn’s narrative of the ministry imply
‘Aree Passovers or four? Among modern commentators Gres-
Well may be taken as the leading advocate of the “four Pass-
Overs” theory, and Wieseler of the “three Passovers” theory.
As both sides admit the genuineness of John vi. 4 (qv 5é éyyus
To wacya 7 éoptn Tay ‘lovdaiwy), the question, as between
these two, is commonly narrowed to this:—was the nameless
feast of John v. 1 a Passover or not?

Greswell ‘in his 23rd Dissertation) proves to his own satis-
faction that this nameless feast was a Passover (assuming that
St John passes over a whole year in silence between his fifth
and sixth chapters), and so makes out his case for four Pass-
overs within the limits of the ministry.

To this hypothesis, which is the commonly received one,—
besides the objection so lightly met by Greswell, that it sup-
poses a whole year to be passed over in silence in St John’s
narrative,—there is surely a fatal objection in the fact that
it interposes eighteen months between the Bethesda miracle,
and our Lord’s allusion to it, as to something quite fresh in the
minds of his audience, in John vii. 21—23.

To avoid this, as well as for other reasons, Wieseler, fol-
lowed by Tischendorf and Bishop Ellicott, adopts Kepler's
suggestion that the nameless feast of John v. 1 was Purim;
and thus shortens the period between the Bethesda cure, and
the allusion to it in vii. 21, to seven months. The Passover
of vi. 4 thus becomes the second instead of the third, and the
ministry embraces in all three Passovers instead of four.

108 THE JOURNAL OF PHILOLOGY.

But there are several difficulties in the way of this theory:

1. Not only the Paris Codex, but the Sinaitic also, insemmrts
the article before éopr7 in John v. 1.

2. No one ever suggested Purim before Kepler. All tie
fathers assumed naturally that a feast which thus drew om!
Lord to Jerusalem must have been one of the three grezmt
feasts.

3. It is improbable that our Lord should have absente<i
himself from Jerusalem at a Passover. This objection applies,
of course, equally to the other theory.

4. Our Lord at this feast seems to allude (v. 35) to the?
Baptist as to one recently dead, and we have reason to believa™
that the Baptist was murdered just before the Passover’ oS
this year.

5. The persecution of our Lord for allowing his disciples
to rub the ears of corn seems to connect itself with this charges™
of Sabbath-breaking at the unnamed feast; and St Luke's
careful date, év caBBatw Sevreporpote, difficult at best, is
most plausibly explained by Scaliger to mean the first Sabbath
after the great Morrow of the Passover, before which it was
unlawful to gather ears of corn.

6. And there is a sixth and far greater difficulty, which must
strike every one who glances over Wieseler’s scheme (p. 280),
or the harmony of Tischendorf founded upon it, viz. that
all those events which in Greswell’s scheme occupy a full year,
including three tours through the towns of Galilee, and indeed
nearly the whole of the great Galilean ministry, have to be
compressed into the brief space of three weeks between the
feast of Purim March 26th and the Passover April 16.

This last difficulty seems to me, I confess, insuperable.
No one can read Bishop Ellicott’s 4th Lecture, occupying
50 octavo pages, without being almost painfully aware of the
ingenuity required in order to bring such an extended narra-
tive within the three weeks prescribed by his theory.

Besides all these difficulties, there is yet one other which,

1 To Wieseler’s quotations in favour but the Accession-day of Antipas, Plate
of yerto.a meaning not the Birth-day 1, Alcib. o. 17 may be added.

ON THE CHRONOLOGY OF ST JOHN V. AND VI. 109

so far as I am aware, has never received the attention to which
it seems to me entitled, and to the consideration of which
without further preface I will now proceed. To my own mind
it has suggested a@ third solution of the question, which, though
doubtless open to objection, may perhaps be thought worthy of
further discussion.

The difficulty to which I allude lies in the tncoherency of
context in this portion of St John’s Gospel.

In the fifth chapter we find Jesus attending a feast at
Jerusalem, and he ts left there at the end of the chapter in
bitter controversy with the Jews. The sixth chapter begins,
** After these things Jesus went over the sea of Galilee.”
Surely the inconsecutiveness must strike every one.

Again, the sixth chapter leaves Jesus in Capernaum; and
yet the beginning of the seventh seems to imply a migration
from Judea to Galilee, assigning as a reason for it the Jerusa-
lem persecution narrated in the fifth chapter. Here is a
®S€cond inconsecutiveness’.

Then, proceeding with the seventh chapter, we find our
Kon!’s discourse taking up the broken threads of the con-
troversy of the fifth chapter, in a way that must make the
Most cursory reader surprised that the Evangelist should have
@llowed the whole of that long sixth chapter to intervene.

“Why go ye about to kill me?”

“T have done one work, and ye all marvel.”

‘Are ye angry with me, because I have made a man every
Whit whole on the Sabbath-day ?”

As one dwells on these inconsecutivenesses of the narrative
as it now stands, one is almost forced into a wish to believe
that the fifth and sixth chapters have got transposed.

Now let us transpose v. and vi. Read in this new order
(iv, vi, v, vii) the coherency of context is at once seen to be

perfect.

1 A friend has pointed ont to me capite relata, Hierosolymis dicta ap-
that Cardinal Cajetan in his Comment. pareant,* * * gesta haec quae subjun-
in Joan. seems to have felt this diffi- guntur non immediaté juncta fuerint
ealty :—‘‘ quam verba in praecedente cum gestis in praecedente capite.”

IIo THE JOURNAL OF PHILOLOGY.

At the end of the fourth chapter our Lord is at Capernaum;
the next chapter (vi.) begins most naturally by telling us how
he crossed the Lake. The sixth chapter leaves him at Caper-
naum, “the Passover the feast of the Jews” being “at hand.”
How naturally then does the fifth chapter follow:—jera tavra
jv [n] €éoprn trav ‘lovdaiwy, “after this was the feast” (to which
he had just alluded as near at hand, not needing therefore to
be again named), «at avéBn o ‘Incots cis ‘lepoodAupa.

Again, the fifth chapter closes leaving our Lord at Jeru-
salem, but under sentence of death for Sabbath-breaking.
Most naturally therefore does the seventh chapter open by
telling us that Jesus resolved to walk no more in Jewry “ be-
cause the Jews sought to kill him.”

Every one, { think, who has accustomed himself for a while
to read the chapters in this order will be unwilling lightly to
relinquish it, if it be only for the sake of what I have called
the naturalness of St John’s narrative.

But when we find further that the whole scheme of Gospel
Chronology is suddenly cleared up by it, that the difficulties
which beset Greswell’s scheme, and the difficulties which beset
Wieseler’s scheme, at once drop out, as it were, by the simple
adoption of this inversion, its inherent probability is to say the
least greatly increased.

Jerome’s words (commenting on the Greek version of
Isaiah xxix. 1) come to have a clear and distinct meaning:
“Scriptum est in Evangelio secundum Joannem, per tria
paschata Dominum venisse in Jerusalem, que duos annos effi-
ciunt.”—(Op. 111. 245.) And a scheme of Johannine Chronology
results with which the main point of the Synoptic Gospels can
(with one notable exception’) be more easily harmonized than
with any other.

But however much the wish may be “ father to the thought,”

1 The exceptional difficulty is the an insuperable difficulty to one who
anecdote of the rubbing of the ears of adopts their order as his clue to the
corn, which according to this scheme Gospel Chronology. But the difficulty
must be placed after the feeding of the of compressing all the intervening
5000. Its very early position in St events into three weeks seems to me
Mark and St Luke must of course be almost ar great.

112 THE JOURNAL OF PHILOLOGY.

8. Both are carefully dated (vi. 4, and xxi. 14), as would
naturally be the case in supplemental anecdotes intended to be
added to the original narrative. And thus a better reason is
suggested for the insertion of vi. 4 than that commonly assigned
—that it was merely to account for the multitudes.

On no one of these points taken singly can any stress be
laid; but taken cumulatively they may perhaps be allowed
weight as confirmatory of a hypothesis antecedently probable.

Its antecedent probability rests on the fact which all must,
TU think, allow, that the narrative of St John’s Gospel gains
most strikingly in coherency and natural consecutiveness, if we
suppose that in the first draft the seventh chapter was written
as the immediate sequel of the fifth; and that when the two
supplemental chapters (vi. and xxi.) came to be added, one of
them got inserted in the wrong place.

If this hypothesis be allowed, the Chronology of St John’s
Gospel is at once cleared of difficulty: the nameless feast of v. 1
is % passover, and is to be identified with the passover of vi. 4;
and our Lord’s ministry limited to two years and a quarter.

Since writing these notes, I have been interested to find
that Ludolphus de Saxonia, whose Vita Christt was the great
text-book of the fourteenth and fifteenth centuries, seems to
take it for granted (without assigning any reason) that the
sixth chapter of St John ought to precede the fifth. To those
disciples of Ignatius Loyola, therefore, who are said to have
drawn so largely in their preaching from the eloquent old Car-
thusian, the hypothesis here maintained would have seemed
in no way strange.

Greswell mentions somewhere, I think, that Mr Mann,
whose book, “de annis Christi natali et emortualt,” is unknown
to me, also inverts these chapters, in order to work out his
theory of a one-year ministry.

I may be wrong, misled by allowing the notion to become a
dominant idea in my mind for some years; but, so far as I am
at present informed, the point seems to deserve further dis-
cussion.

J. P. NORRIS.

NOTE ON THE ‘ARZARETH’ of 4 Espr xu. 45.

‘Nam regio illa vocatur Arzareth.’ This passage has ap-
parently hitherto defied the ingenuity of commentators. From
the MSS. and versions no help is to be obtained. The Latin
MSS. of any value only vary between the reading Arzareth and
Arsareth; the Syriac gives Arzaph; the Aithiopic has Azaph:
Ockley from the Arabic translates Acsardri Karardwin, which
is evidently an attempt at interpretation, and is first trans-
formed by Gutschmid (in Hilgenfeld’s Zeitschrift fiir wissen-
schaftliche Theologie for 1860, p. 75) into Aqsarat kozarawin,
and then rendered as Turkish. But the Arabic text printed
by Ewald has Ascaréri Farardwin. The Armenian 'Apodpata
of Ptol. v. 12, is appealed to as possibly the Arzareth of Esdras
(Gutschmid, p. 76). Volkmar (Handb. d. Einl. in die Apokry-
phen, 2 Abth. p. 193) conjectures that Arzareth =’Aplapad
= (?7D58) AW y™® erets drdt, ‘the land of Arat or Ararat,’
ie. Northern Armenia, Le Hir (Etudes Libliques, 1. 214, note)
suggests that the first part of the word is the Hebrew “W,
har, ‘mountain, and with the remainder ‘Sareth’ he compares
‘Seres,’ the name by which the Chinese were known to the
ancients. Bretschneider conjectures YP ys ‘Land des
Schreckens.’

On this point Dr Schiller-Szinessy writes to me as follows:

“Now let us simply give the words of the Mishnah Synhed-
rin, x. 3 (Talm. Bab. Synk. fol. 110 6), JOR DAWN Mwy

pym> mime yo bs osbe spay nin pony
Journal of Philology. VOL. 111. 8

114 THE JOURNAL OF PHILOLOGY.

pon pn AN Nn ow PAN AN DY Ab oan
seo) SBD DY TD DW ROT Dn oN
jm> oeend ctny Jo om> Sexy pawn mey ax
However R. Eliezer may differ from R. ‘Aquiba with respect
to the fate of the ten tribes, both agree as to the application of
Deut. xx1x. 28 to them. Is not the Arzareth of our Apocrypha
simply the FMS YW (A. V. ‘another land’) of that pas-
sage, corrupted by an ignorant translator into a proper name?”
The conjecture has the double merit of ingenuity and sim-
plicity, and will appear even more probable than it does at first
sight if we refer to ver. 40, where the same phrase occurs, ‘et |
translati sunt in terram aliam.’
[W. A. W.]

LUCRETIUS, BOOK VI.

Is the last number of the Journal, pp. 219—228, Professor
R Ellis has given criticisms and explanations of various pas-
&ges in the 6th book of Lucretius. As he has been kind
enough in his discussion of many, indeed most of these, to
refer to my edition, I offer the following remarks in the hope
that they will assist in the illustration of our author. None of
the passages discussed by Mr Ellis is here passed over: in
order to be as brief as possible, I will sometimes assume a know-
ledge on the reader's part of the text of Lucretius and my edition.
48 49 are very corrupt in our Mss.: Lachmann’s reading,
&% well as those of older editors, will I think be disallowed by
al, Bernays makes the very improbable assumption of a
lacuna both before and after 48; and then he leaves 48 still
uncorrected. My own arrangement and emendation of these
ves, of which I have given an elaborate explanation, I did
think and still think very plausible. But Mr Ellis says: ‘for
Ventorum exirtant, placentur omnia rursum, I would read Ven-
torum existant (so Bernays) placentur momina rursum, which is
sufficiently justified by 474, Posse quoque e salso consurgere
momine ponts’. It is possible that Lucr. might have said momina
ventorum, though elsewhere he only uses momen in the singular.
But this reading renders Bernays’ improbable assumption of a
double lacuna necessary ; and then the verse is quite disjointed ;
and in my opinion not so near the Ms. reading, as my correc-
tion is.
52. ‘Munro makes quae... the subject of faciunt. May it
not be homines ‘and when they humble their spirits through
8—-2

116 THE JOURNAL OF PHILOLOGY.

fear of the gods’?’. But ‘Depressosque premunt ad terram
follows; and surely something external must be the subject to
this, not the men themselves.
68. Quae nisi respuis ex animo longeque remittis
dis indigna putare alienaque pacis eorum ;

‘now unless you drive from your mind with loathing all these
things, and banish from you all belief in things degrading to
the gods and inconsistent with their peace’, So I translated;
and wrote this note: ‘putare is for an accus. subst.: see n. to
1 418 repetam pertexere and 331: tndigna putare is not a
common construction, but Lach. illustrates it from Cic. de sen.
4 ‘quis coegit eos falsum putare’ and Aen. vill 522; Fore.
also exemplifies it from Virgil and Terence’. As in the notes
referred to I have given at least six instances of the infinitive
used by Lucr. for an accus. subst. and more than twenty for his
use of it as a subst. generally, I certainly never anticipated my
explanation being called in question. Older editors had got to
the same meaning by reading putando for putare. But Mr-
Ellis says: ‘it seems to me that this is not the first impression
the words convey; dis indigna putare with quae preceding must
surely be ‘think them unworthy of the gods’; to separate the
two clauses looks like an after thought, occasioned by the
difficulty of longeque remittis’. But quae nisi is simply quod
nist haec, the relative serving at once for a connecting particle
and a demonstrative pronoun, as in almost every page of any
good author; in Lucr. as often as in any. Next remitto with
an infin. is very good Latin, though not Ciceronian. But then
for longe you want prorsum or the like. Then prorsum remitts
putare may mean ‘quite refuse to believe that they are un-
worthy of the gods and inconsistent with their peace’. But
this is Just the contrary of what the poet means. Well then
andigna must be for digna; aliena for non alvena; on which
principle you might construe Cicero's falsum putare not ‘to
believe what is false, but ‘to believe that this is true’. But
Mr Ellis goes on: ‘I think that the negative idea in these
two words led Lucr. into a construction more Greek than Latin.
As in 399 parcit in hostis is, not ‘refrains against his ene-
mies, but ‘spares it to attack his enemies’-—’. But I must

118 THE JOURNAL OF PHILOLOGY.

wants, the alteration too. being exceedingly slight, as in our
archetype T and P must have readily interchanged: its first
blunder is tergis for pergis. I would compare too in 237 tellens,
for which I read pellens. And with reference to this I will for
convenience take here out of its order Mr Ellis’ note on

563. ‘ Inclinata minent is retained by Lambinus and I think
rightly ; it recurs perhaps in 1195 frons tenta mebat, Le. mine-
bat’. But, as I have there shewn, Heinsius and Lach. have
independently made the certain correction ‘frons tenta tume-
bat’, ‘the brow tense and swollen’, which admirably suits the
sense: the tu being omitted by perhaps the commonest of all
errors in good Mss. My own correction of 563 ‘Inclinata tu-
ment I look upon as almost equally convincing. For ‘minent’
is no Latin word, because ‘inminent, eminent, prominent’ are
Latin, any more than ‘cumbunt’ is Latin, because ‘incumbunt,
procumbunt’ are. The same may be said of many other com-
pounds, where the simple verb is simply non-existent. But this
which I thought was now universally admitted is not Mr Ellis’
theory; for he says: ‘for the same reason I would change tellens
in 237 to cellens, as Wakefield, rather than pellens Munro, or
pollens Lachmann’. Why ‘for the same reason’? because the
word is non-existent? surely not a satisfactory reason for ac-
cepting a mere conjecture of Wakefield ; though the only reason
I fear to be given for many of his conjectures. Then if cellens
existed, not in Lucretius’ rerum natura, but in rerum natura
at all, why should it have the sense whjch is required here,
and is exactly given by my pellens? For in my edition
I unfortunately omitted to quote Pliny, xiv § 136, ‘Campaniae
nobilissima [vina] exposita sub diu in.cadis verberari sole, imbre,
ventis, aptissimum videtur’: to a custom of this kind Lucretius
I doubt not here refers.

154. If Mr Ellis will examine the context, he will see that
what the poet emphasises is not the burning, but the noise
made in burning; that therefore this emphasis is better given
by my ‘burns with a more startling sound’, than by his ‘ burns
more decidedly’: comp. especially 149 and 151. I quite agree
with him that magis belongs to the whole sentence.

258. For ‘et fertus’ of Mss. Lach. gives the simple and

120 THE JOURNAL OF PHILOLOGY.

for opprimere. Instead of changing the genuine-looking oppri- -
mere, I made what I still think an almost certain correction,
videatur for videantur, a very slight change, the scribe having as==
so often adapted the verb to the adjacent plural: in 467 the =
Mss. give videatur for videantur: ‘close upon it follows so
heavy a clap that it seems to crush down from above the quar-
ters of heaven which have all at once sprung asunder’: in my
note I refer back to a former note in which I give several
parallel instances from Lucr. and one from Terence: ‘ At quem
_deum! qui templa caeli summa sonitu concutit’, which Lucr.
imitates in 11 1100. The whole point of our passage is that
the clap is so loud, it seems to bring down the sky on our heads.
But Mr Ellis says on all this ‘videantur which is retained by
Lach. need not be changed to videatur as Munro’. Yes, but
Lach. kept it in order to make a much less probable correction,
as he saw plainly the passage required essentially the sense

I have given. ‘Here the subject to videantur seems to be
lumina’. Any plural in the preceding paragraph might just as
well be chosen for the subject as lumina. With respect to
11108 to which I refer in my note, Mr Ellis goes back to the
old explanation without a word as to what Lach. says against it.
To me omnes ‘all men’, in that passage seems as absurd as ‘all
flies’, ‘all sheep’, ‘all mice’, or any other living thing you like.

296. I feel no doubt of Bernays’ correction “calidam
maturo fulmine’, which I hit upon quite independently of him,
being right: the Ms. reading, even after Mr Ellis’ explanation,
seems to me to be neither Latin nor sense.

370. At all events the insertion of res, so often omitted in
our Mss,, is critically as easy as that of sese for se; and to say
the least makes the construction more easy and natural.

428. ‘roused’ or ‘stirred’ were chosen by me as more
poetical than, and yet almost the same in meaning as “set in
motion”,

475. It is just possible that omnibus might have the sense
which Mr Ellis gives to omnis, but which omnis cannot have.

I feel little or no doubt that ollis is necessary.

483. slli could not possibly refer to halitus which occurs

five lines before and is given as a mere illustration.

122 THE JOURNAL OF PHILOLOGY.

573. inclinatur enim retroque recellit
et recipit prolapsa suas in pondere sedes.

As Mr Ellis quotes me here against myself, I must in justice
quote myself in defence of myself. In this passage, if any-
where, I thought I had done something. I shewed that pondere,
not pondera, was the reading of our archetype: I observed
‘ prolapsa answers to inclinatur, rectpit sedes tn pondere to retro
recellit: falling forward out of its place is the natural force of
prolapsa: see Fore. and comp. 1006: reciptt sedes in pondere
then is a proper expression, not prolapsa in pondera: a thing
prolabitur trans pondera, tumbles beyond its balance or centre
of gravity: thus Livy etc.’ Then, in answer to Turnebus and
Lach. who asserted that only the plur. pondera had this mean-
ing, I ransacked the language to prove the contrary,-and to
every instance they gave of pondera, I adduced a closely
parallel example of pondus in the same sense. Not to be need-
lessly prolix, I kept back many other instances, some of which
I have given in my note to Aitna 324. The passage I thus
translated: ‘it leans over and then sways back again, and after
tumbling forward recovers in equal poise its fixed position’,
Mr Ellis after quoting my note asks ‘granting that prolapea
has this meaning, why should this necessitate pondere? recipit
sedes in pondere seems to me slightly unnatural, tn pondera not
so’. But I must repeat once more that pondere is the real
reading of Mss.: pondera is the conjecture. Then pondus is
almost or quite synonymous with suas sedes, as in Lucan’s
‘mutataque sidera pondus Quaesivere suum’; so that with pon-
dera the meaning would be ‘brings back its state of stable
equilibrium into its state of stable equilibrium’; whereas my
reading gives this sense: ‘recipit pondus suum et manet in
pondere suo’, the sense which Lucr. intended beyond question.

G00. ‘Jdque is perhaps right.. “the void it has made”’: so
Wakefield ‘nempe id quod hiatu fecerit’» But Mr Ellis will
find Wakefield and his followers alone to support him: ‘tdque
solus Wakcfieldus se intellegere professus est’ says Lach.

624. Mr Ellis here defends the second vent#t, which Wake-
field calls a ‘iucundissima repetitio’; Lach. ‘inanis et sine
pondere repetitio’. Here too I side with the latter.

124 THE JOURNAL OF PHILOLOGY.

‘ branches of boiling service-berries’. ‘Service-berries of which
Pliny mentions four kinds, three of them vinous in flavour..
were actually made into a kind of wine (Plin. xtv 16).’ Yes,
Pliny there, § 103, mentions that wine was made out of the
sorba or fruit of the service tree, as out of mala of all kinds,
corna, mora, nuclet pinet, etc. etc. But sorbus is a service-tree,
and anything more strange than the notion of boiling a service-
tree, a tree of large size, or even ‘branches’ of one, I cannot
conceive. Fancy a fervida malus or a fervida pinus, or even
what would be less odd, fervida witis: a boiling apple-tree,
pear-tree, pine-tree, even vine.

But though this will never do, it gives me an opportunity
of recurring to this corrupt v. ‘At cum membra domnus per-
cepit fervida servis’; as it strikes me the domnus would be a
confirmation of what I proposed ‘At cum membra domus per-
cepit fervidus, nervis Tum fit odor cet.’: the n of nervis was
omitted and then written over the line and got afterwards
attached to domus instead of nervis. Comp. too Horace Sat.
11 4 51 ‘Massica si caclo supponas vina sereno, Nocturna, si
quid crassi est, tenuabitur aura Et decedet odor nervis inimicus.’
This perhaps would tend to defend vint, which I said in my
edition might be right; though I still think vert true.

851. Mr Ellis defends partim, apparently much as Wake-
field does; but to me the word conveys no meaning: I cannot
but echo Lachmann’s ‘ Lambinus rectissime raptim’: the change
is next to nothing.

951—958. Mr Ellis quotes the whole of this difficult pas-
sage: on 954 he says Gallt lorica ‘the Gaulish cuirass’ is a
rather forced expression. But he has failed to observe that this
is the technical prose term for a steel cuirass, used by Varro in
his description of it quoted by Lachmann and by me: the
proper meaning of Wwrica being a leathern cuirass. Galli lorica
I look on as a quite certain correction of Lachmann: the qua
gives sufficiently the object of coercet. But says Mr Ellis ‘it
scems to me that caeli lorica may possibly mean “the enclosing
sky”, something like Spenser’s baldrick of the heavens bright’.
But surely Spenser means simply the belt of the zodiac; no
enclosing sky, but Varro’s ‘limbus Bis sex signis stellimicanti-

126 THE JOURNAL OF PHILOLOGY.

pointedly drawn to a matter like this, it ought not to be passed
over without notice.

958. ‘raro corpore necwm’ says Mr Ellis ‘is in effect a
more poetical form of raro corport nexu etc.’: Lach. had said
‘haec absurda sunt: nullum est enim rarum corpus quo res
inter se coniunguntur. immo res iIpsae raro corpore sunt, sive,
quod idem est, raro corport’ nexu’: and I beg to side with
Lach.

971. First as to ambrosias: which Mr Ellis says is the
Greek genitive, ‘though Lach. denies this because Lucr. else-
where writes harmontiam harmontat; an inadequate reason,
and one which would banish every individual peculiarity of
grammar or construction’. Had Mr Ellis known that ‘vir pau-
corum verborum’ better, he would have seen that he meant
much more than this; that he meant something like the
following: ‘ambrosias might be taken for the Greek genitive;
but I think this can hardly be; for long before Lucretius’ time,
when they adopted such Greek words, classical writers regu-
larly gave them their own genitive: musae, musat, not muses;
though later writers say nymphe, nymphes and the like. Lucr.
himself has harmontas.’ Take this v. made almost wholly up
of Greek words:

Et cycnea mele Phoebeaque daedala chordis :
- we are sure that Lucr. used the Greek form mele; but feel no
less sure that the Latin form chordis is his: chordats certainly
we should reject. But be this as it may, leaving the form un-
certain, I come to Mr Ellis’ reading of 971:
Effluat ambrosias quasi vero e nectare tinctus :

“as if it were an offset tinctured with the true nectar of am-
brosia whence it is drawn’: ‘because tho tree is supposed to be
dipt in ambrosia and then drawn out’. But how can effluat
signify ‘to be an offset’? and how can tinctus e nectare mean
‘tinctured with nectar whence it is drawn’? and how is this
less prosaic than ‘the prosaic’ linctus, which assuredly will not
‘recede’ before this? But this gives me an occasion of mention-
ing my own latest surmise: can efiuat be said of the oleaster
itself instead of the ambrosia and nectar coming from it? comp.
Persius 11 20 ‘effluis amens’; and see Jahn there and his

A THEORY OF JOB XIX. 25—27.

THE following extract from the Authorized Version contains
the passage to be discussed, together with a portion of the con-
text which will be seen to have an important bearing on the
argument :—
xix. 20 My bone cleaveth to my skin and to my flesh, and I
21 am escaped with the skin of my teeth. Have pity upon me,
have pity upon me, O ye my friends; for the hand of God
22 hath touched me. Why do ye persecute me as God’, and are
23 not satisfied with my flesh? Oh that my words* were now
24 written! Oh that they were printed in a book! That they
were graven with an iron pen and lead’ in the rock for ever!
25 For I know that my redeemer liveth, and that he shall
26 stand at the latter day upon the earth: and though after
my skin worms destroy this body, yet in my flesh shall
27 I see God: whom I shall see for myself, and mine eyes
shall behold, and not another; though my reins be con-
28 sumed within me. But ye should say, Why persecute we
29 him‘? seeing the root of the matter is found in me. Be ye
afraid of the sword: for wrath bringeth the punishments of
the sword, that ye may know there ts a judgment.
In ver. 26 it appears from the italics that the words
though, worms, body are interpolations and do not correspond

1 Or simply, “thus"—5N"%1D3 for “ fortwihrenden Betheuerungen seiner
nox-iws, which occurs in Chap. xii.8. Unschuld” (Dilimann).
But the rendering in the text is usu- * The letters being cut in the rock,
ally preferred. and the lead then infused (Rashi).

* Some hereby understand the de- 4 “For ye say, How will we persecute
claration in ver. 25 sq.: others, his him! whereas...” (Ewald, #c.).

4 THEORY OF JOB XIX. 25—27. 129

directly to anything in the Hebrew. The translations which
wil next be given are professedly literal, and they express the
three views most generally received, viz. that Job expected, (a)
a Resurrection of the Flesh; (b) Immortality, or an Incorporeal
Future life ; or (c) Restitution or Vindication in the present life.

A. Resurrection of the Flesh.

This view is maintained by Dr Pusey in his Dantel the
Prophet, p. 508. ed. 2:

“The great passage in the book of Job is a confession in-
fended for all times :

Q that my words were written, O that they were graven in

4 book, were cut with an iron pen and lead in the rock for
ver!

Their most literal translation is;

And I, I know that my Redeemer liveth;

And that, the last, He shall arise upon the dust;

And, after my skiv, they have destroyed this body,

And from my flesh I shall behold God,

‘Whom I, I shall behold for myself,

~And mine eyes shall behold, and not another [lzt. a stranger, ]
‘My reins are consumed within me.

No doubtful meaning of any words can efface from the
Passage the doctrine of the resurrection of the flesh.”

B. Incorporeal Future Life.

In a note upon the words “with the eyes of his flesh,” Dr
Pusey then proceeds to controvert the view adopted by Ewald
and expressed with some variations in the words of Conant :

But I, I know my Redeemer lives,

And in after time will stand upon the earth;
And after this my skin is destroyed,

And without my flesh, SHALL I SEE Gop;
Whom I, for myself shall see,

Journal of Philology. VoL. M11. 9

130 THE JOURNAL OF PHILOLOGY,

And my eyes behold’, and not another,
When my reins are consumed within me.

“That the language (continues Conant) here refers to an
existence beyond the grave is asserted by the latest and best
interpreters. Ewald, in many respects the ablest of the recent
translators and interpreters of the book, regards this as unde-
niable ; and the view which restricts the language to an earthly
hope, is opposed to the proper force of the words, to the connec-
tion of thought, and to the spirit and tenor of the whole book.”
Cocceius, Vaihinger and Schlottmann are referred to in corro-
boration of this second view.

C. Restitution [or Vindication] in the Present Life.

Mr J.J.S. Perowne, in the Appendix to his Hulsean Lectures,
states the third view, which limits Job’s anticipation to the life
present, and in support thereof refers to Bernard (ed. Chance)

and Havernick.

“T would render the passage as follows :

I, even I, know that my Redeemer liveth,

And that at the last He shall stand upon the earth (it. dust);
And after my skin has been thus pierced through,

Yet from my flesh’, I shall behold God,

Whom I shall behold for myself,

And mine eyes shall behold and not a stranger's.

My reins are consumed within me.

1 This strong expression recurs in
Job xlii. 5, and there it is not literal,
nor does it refer to a future life:—“I
have heard of Thee by the hearing of
the ear: but now mine eye seeth Thee.”
Cp, xxxili, 24—26. In Exod. xxiv. 10,
11, seeing God does not refer to life
beyond the grave:—‘‘ also they saw
God, and did eat and drink.”

* Of those who limit Job’s anticipa-
tions to this world some, as Hiivernick,
here adopt the form B, and make

Job assert his confidence that though
reduced to a mere skeleton, without
skin or flesh (xxxiii. 21), he should
yet behold God. Umbreit formerly
held this view, but has since adopted
B as a whole. MRosenmiiller in his
second edition maintains A. Bishop
Warburton is conspicuous as a de-
fender of C. Delitzsch adapts the
form B to the theory A, laying streas
on the expression 18 '3°?}. But com-
pare note 1.

zz Pak esl EN4Ll OF PHILOLOGCY.

SS te weet begs Te ragscaly imserted, it still does not
Suc as Teter: str te restroed to the first hemistich;
tl ie SET 6 oeeird reiatle. both from the parl-
2ST Lvl > DAS SHewIere. tha: the said “body” would
Supie ite ‘sao ” cé tte Sos: Lemcistich and the “flesh” of
ke Dis SIC Eeee i Parcs
t= ES tte Tumcecsr cresertai. the Verse in question being
Pos Sek: (fa Tasca cf Gai to be enjoyed spintually
co ies obs Wil As SOOT SCL: have been destroyed; but it
I STU tbs Lie temcerims - eittest my flesh ” ‘is in accor-
ipo wit Estrnw aa The renderings of A, B and C, 3%
THELTIs Li TuaooTuar. mar té inclcded in the one expressioD,

» -
wee «= = a.

pa a. tie Do

i

teinz taken to imply in the one cas@

Fomor iesc*. ins
taht lL. teing in mr ssh s+) look ont therefrom and se©
Gniv” ani in tie seemed ame that, +I being out of my fle,
soa ke Gal” The ‘CmeE is no doubt a possible meanings;

be second is admissible.
cf these meanings is adopted, and
2 time preserved, both ~ flesh” and
~ being reed o the same ‘mortal’ body; but if Job is
expressing Lis awcrance of restiration im this life, it is not quite
cbhvicus kaw to aqoourt for his intense anxiety that his words
piizts te written”, and graven in the rock for ever, (ver. 23, 24).
Morecver the angument fmm the ending of the book is not
improtabiy falacious, for it does not appear that Job was really
expecting such a Vision of Gud as is there described. Does it
not rather tude Aim by surprise’? Ifso we must reconsider our
conclusion.

1 Given im tke A. V. marcn.

2 But this objection disappears when
by those ** words” we understand kis
former assertions of his innocence, and
take ‘3) “SN! antithetically: ‘ Yet I
know, &c.” The meaning would then
be: “Yet I shall be satisfied without
their being placed on record, for I
know that I shall be vindicated open-
ly.” It still remains, however, that
the use of D1’ JINN is unfavourable

C. Op’ is used naturally of a suc-

cessor, as in Deut. xxix. 21; Dan. vii.
24. For 375 in connexion with YVR
see Isai. xxx. 8; Ps. cil. 19, MX? INN
;YVIN TT. On INI see Dillmann.

3 In the sequel, “ the Lorp answered
Job out of the whirlwind” (xxxviii. 1;
xl. 6). From the marked correspond-
ence of ANT PY ANY. (xii. 5)
with IN °3'S) (xix. 27) we may infer
that the aspirations in xix. 25—27 were
then satisfied; but against the notion
that Job was expecting that satisfac-

134 THE JOURNAL OF PHILOLOGY.

when the new rendering of the clause is adopted. With thas
remark I pass on to discuss the passage in detail.

VERSE 25.

sop sey by poms on Ss ny oon

ben The full phrase BIA by) occurs in Numb, xxxv—
19 sq. and elsewhere: “The revenger of blood himself shall slay~
the murderer, &c.” For alone in the same sense Gesenius
quotes’: “And they shall] be unto yon cities of refuge from the
avenger.’ It has indeed been denied that the meaning “avenger
of blood”’ is here appropriate; but (1) the balance of testimony
seems to be in favour of the view that Job is contemplating a
vision of God to be enjoyed after death, and (2) there are reasons
for concluding from the context’ that the hope of an avengsng,
in some sort, is appropriate; for in ver, 22 Job’s “friends” are
described as his persecutors’ and as devourers of his flesh, "YA
yagn x, and in the immediate sequel (ver. 29) he warns
them to beware of the punishments of the sword.

PANN] “Et postremum super pulvere staturum.” He will
“stand at the latter day,” or “remain‘ last,” upon the earth.

The meaning of }}"IN is made to vary with that of bys, where-
by some understand God, and others a human avenger®. The
former view is adopted by Bernard (ed. Chance): “I know Him
to be a Being whose existence will have no end or limit. So in
Isa. xlvili. 12, God says, PANN “INI PWN “IW, I am the
first, yea, I am the last.” But, “Selon plusieurs commentateurs
ON signifie celui qui dans la postérité me lira, me vengera et
me déclarera innocent, et par ,INN) on entend la génération,
qui, dans un temps éloigné, me rendra la justice que me refu-
sent mes contemporains” (Cahen). Elsewhere }TNN is an
auljective, but some make it here a noun: “ein Nachmann auf
dem Staube wird erstehen.” It is however too much to say,
with Hahn, that if the parallelism be observed, the word “kann
nur substantivisch genommen werden.”

1 Numb, xxxv. 27. * But can 01), used of persons, have

3 See also xvi. 18. this meaning?
3 Comp. ver. 28. ® See note 2, p. 144.

136 THE JOURNAL OF PHILOLOGY.

fact that 15/3 does not follow immediately upon “MS, is uni
portant: suffice it to remark that we are dealing with a poeti
passage, and that a changed order of words is sometimes re=—
quired by the law of emphasis,

1Bp3] The word is generally allowed to mean, pierce, smite?»
destroy, and Is, x. 34 is referred to—51933 Wi %32D ApsIs—
But what is the construction? On this too there is something
like a general agreement, but it may be doubted whether the
commonly received opinion, mz. that the word is to be takem
impersonally’, is quite satisfactory. If Job is represented ass
expecting a 2NJ to arise, whose function it would be to take
vengeance on his “friends,” why should not THEY be the sub-
ject of 1593? In ver. 22 they are described as devourers of
his flesh ; why not therefore keep up the figure and render:

And after that, as to my skin, THEY have destroyed this,
&c. 7.e. when they have quite made an end of me, my avenger
will yet arise? The clause quoted from Is. x. 34 favours the view
that 53) may imply destruction by external violence at least as
naturally as the wasting of disease, for the action is there per-
formed with an iron instrument, 5t923 Fj23. The language
of Job is figurative throughout the passage, which follows (be it
observed) immediately upon the words YAWN ZAIDI
“and are not satisfied with my flesh?” For this remarkable
figure compare: “ When the wicked, even mine enemies and
my foes, came upon me fo eat up my flesh, they stumbled and
fell” (Ps. xxvii. 2); “ Have all the workers of iniquity no know-
ledge? who eat up my people as they eat bread, and call not
upon the LorD” (Ps. xiv. 4); “Therefore all they that devour*
thee shall be devoured” (Jer. xxx. 16); “ who pluck off their skin
from off them, and their flesh from off their bones: who also
eat the flesh of my people, &c.” (Mic. 11. 2,3). We see then

1 We may notice in passing that the
expression “impersonal” is sometimes
used inaccurately. In such a phrase,
e.g. a8 ‘‘man sagt,” there is an inde-
finite but not an impersonal usage, for
personal speakers are referred to, al-
though it is left undetermined who

they are. Le Clerc takes an indefinite
‘‘ homines, nempe mali” for the nomi-
native to IBD). For the “impersonal”
use Job vi, 2; xv. 28; xvi. 10; xxiv.
2—4, &c., are referred to,

® Compare Gal. v. 15.

138 THE JOURNAL OF PHILOLOGY.

flesh is clothed with worms and clods of dust; my shai
broken, and become loathsome”’...... “Thou hast clothed m
with skin and flesh” (Job vii. 5; x. 11); “My flesh and my
skin hath he made old” (Lam. i. 4). See Lev. viii. 17; Numb
xix. 5. The expression “skin of flesh" occurs several times in
Lev. xiii, See tuo Ezek. xxxvii. 6, 8, 107; Mic. iii, 1—4*, &.

2. The authority of versions and commentaries likewise
favours the view that WP and WS are not to be contrasted,
but to be taken as parts of one and the same body.

In the obscure LAX. rendering the contrast is avoided:

25. oida yap Sts aévvacs éotiy 6 éxAXvey pe peddAwy er
yns 26. avagcrinoas ro Séppa (al. caja) pov to avaythos
Taita’ Tapa yap Kuptov Taita pot ourereXéaOn, 27. a ee
€uauTa ouveviatapat, G 0 opOaduos ou Ewpaxe Kai ove adda,
marta O€ por ouvreTeXeoTas ey KOATT.

In the Syriac there is a direct parallelism between skin and
flesh:
ims “so dor yal] on ureato yo

and so tov in the Arabic:

The Vulgate does violence to the first hemistich in order
to preserve the parallelism, assuming that the second points to
@ resurrection :

Et rursum circumdabor pelle mea, et in carne mea ridebo
Deum meum.

With this compare Schlottmann’s argument for the nega-
tive rendering® of the preposition in "Y3D:

“Nun bemerkten wir aber bereits ver. 26 dass die negative
Auffissung des {2 in 2730 fiir den Zusammenhang natir-
lichere ist. Anders wir’ es freilich, wenn man mit Hierony-
mus in dem ersten Gliede das FX} 1/3 “1Y von dem
Umgeben mit einer neuen Haut verstehen kénnte, was aber
jetzt noch schwerlich irgend jemand als sprachlich méglich
vertheidigen wird.”

Aben Ezra takes {73 as well as “YY to be included in the

1 Quoted on p. 140. ® Quoted on p. 149. 5 See p. 141.

140 THE JGUBSAL OF PHILOLOGY.
flesh f# <be:r danzhters, and thev shall eat every one the flesh

of Lis frend in the siege amJ straitness, wherewith their ene-
rri=3, arr] thev that seek their lives shal] straiten them” Ger.
xix 9: ~ True deal belies of thy servants have they given to
be meat unt the fowls of heaven. the desh of thy saints unto
the beasts of the earth” Ps lxxix 2). We may add that "73
Is oppmcd to PE5—“ And shall consume the glory of his
forest, and of bis fruitful field, both soul and body “Ty! W5ID
wa" Is x 15; and to PM; “Now the Egyptians are men,
and not Gel; and their horses flesh, and not smrit” (Is. xxxi. 3).

A very striking illustration still remains to be quoted’.

“And I will lay sinews upon you, and will bring up flesh
upon vou, and cover vou with skin, and put breath in you, and
ye shall live... And when I beheld, lo, the sinews and the flesh
came up upon them, and the skin covered them above: but
there was no breath in them...So I prophesied as he com-
manded me, and the breath came into them, and they lived,
and stood up upon their feet, an exceeding great army” (Ezek.
xxxvil. 6, 8, 10).

Here we have first the “flesh,” &. making up the mass of
the bedy: next the “skin” giving completeness to its form:
but as yet no breath nor life. In Job xix. 26 we may suppose
this sequence reversed: the “WY being destroyed, a dead shape-
less mass of WW remains:

And after that, as to my skin, they have destroyed this,
Even from my flesh
And from my dead-flesh T shall see God.
Thus Job is made to express the assurance that when his
“skin” had been destroyed, and when he should have been
reduced to a shapeless mass of “ flesh,” still from that flesh” he

1 See also Numb. xii. 12: ‘‘as one
dead, of whom the flesh is half-con-
sumed.”

2 If ‘‘ to seo Giod” (=to see His day
xxiv. 1) means to see traces of divine
retribution in the world [p. 144], the
proposed rendering would amount to
the following: “I shall see that day
whon I am in my gravo;”’ or, ‘I shall

have the satisfaction of being vindi-
cated when I am dead.” Compare the
‘“‘ qudacissima prosopopaia” of Is. xiv.
4 sq. In Ezek. xxxii. 831 we have a
very striking parallel:— ‘“Pharaoh shall
sce them, and shall be comforted over
all his multitude, even Pharaoh and
all his army slain by the sword.” Here
emotions are poetically ascribed to the

142 THE JOURNAL OF PHILOLOGY.

On this Dr Pusey remarks' that, “The rendering of "W3D>
without my flesh, adopted by Davidson, ii. 227, from Ewald, &€-
is unidiomatic and unnatural. { can no more, of itself, mea”
without, than our from. Where we might render without, the
meaning is gained from the context.”

Various passages, which are quite inappropriate, have beem
loosely quoted in support of the simply negative rendering
“without,” which we must be careful to distinguish from the
common privative rendering of ‘the 1. An example of the
latter occurs in Ps, ]xxxiii. 5: “Come and let us cut them off
from [being] a nation;” where the ceasing to be a nation is to
result from the cutting off—a construction clearly unsuited to
Job xix. 26, where the being without flesh does not result from
seeing God. In Mic. iii. 6, (wrongly quoted for the negative
rendering) we have another good example of this privative sense :
“There shall be a night to fIMD, from vision t.e. hiding al
vision from you,” as Dr Pusey rightly remarks, This appears
plainly enough from the parallelism, as shewn by the English
version: “Therefore night shall be unto you, that ye shall not
have a vision; and it shall be dark unto you, BDI, that
ye shall not divine; and the sun shall go down over the pro-
phets, and the day shall be dark over them. Then shall the
seers be ashamed, and the diviners confounded: yea, they shall
all cover their lips; for there is no answer of God” (Mic. iii. 6, 7).
Here the lack of vision results from the “night” which falls
upon the prophets, and there is a necessary connexion between
the two.

But the passages most frequently quoted in support of the
negative rendering are :

Job xi. 15 DIDS J35 NON IN

Job xxi.9 “INBD DY omns

The former of these seems at first sight to the point, but
perhaps the } here should rather be taken in connexion with
the verb NYJ, in the sense of taking away from :—“Then
shalt thou lift thy face aloof (or away) from blemish.” In Job
xxi. 9 we find a still more precarious illustration of the required

1 In his Daniel the Prophet, ed. 2, p. 509.

A THEORY OF JOB XIX. 25—27. 143

hegative rendering, for there is a natural contrast between
peace and fear, the former producing an absence of the latter.
“Their houses are peace, without fear”"—or as we might say,
“at peace from fear,” +.¢. without fear by reason of their being
at peace.

Others have quoted Is. xxix. 18,

Ayn ony sy sw Sex

“A caligine et tenebris, 1.¢. remotis tenebris, ocult cecorum
nidebunt.””

But it may be doubted whether the foregoing explanation
(Rosenmiiller’s) of the verse is the true one. There is indeed
an obvious contrast here described between the former gloom
and the succeeding light, but is not this expressed solely by
the natural antagonism between the ideas of seeing (AN) and
darkness (“2DN), without the help of the supposed negative use
of j5? In Is, ix. 2, such a way of expressing the same con-
trast is chosen : “the walkers (1°3711) in darkness have seen a
great light,"—-not, the people that walked éc. So in Is. xxix.
18, we may take the rendering of the Authorized Version:
“And in that day shall the deaf hear the words of the book,
and the eyes of the blind shall see out of obscurity, and out of
darkness "——where it is predicted that persons described as
“deaf” shall hear, and persons described as “blind” and in
darkness, looking out of that darkness shall see: “the blind
thall see FROM AMID their darkness,” not, “remotis tenebris.”
The removal of the darkness is implied by M}°NN, just as in
the first hemistich the removal of deafness is implied by
Ow), whereas explicitly the persons there in question are
described as “deaf.” Neither this, then, nor any one of the
foregoing illustrations can be said certainly to favour the re-
quired negative use of {D.

Partitive renderings of "WD.

(i) Taking the jD partitively we may read: “After they
have destroyed my skin and my flesh (lit. of-my-flesh) ;” the
partitive accusative being used because the flesh is not so
completely destroyed but that something of it—a mangled

144 THE JOURNAL OF PHILOLOGY.

corpse—remains. For partitive uses of } see Gesen. Thesaur.
800. b.

(ii) Or we may suppose a zeugma and aposiopesis. The
same word W/2\ occurs in ver. 22, followed by PAWN ND.
Would not ver. 26 have seemed to end very naturally: “when
they have penetrated my skin, and of my flesh HAVE HAD
THEIR FILL, IY "W351"? Perhaps then we may suppose
13 to carry with it another verb, which the recurring "W725)
marks out as the PSY of ver. 22, from which same verse this
whole passage springs. Thus the construction would be ex-
plicitly incomplete :—

After that, as to my skin, they have destroyed that,

And of my flesh **** I shall see God.

mon rUNMN] It seems best to understand this of the traces
of Divine retribution, as shewn in Job’s vindication ; although
Rabbinic commentators and others have referred it to afflictions’:
“the hand of God hath touched me”’ (ver. 21). In chap. xxiii. 9,
the same verb is used: “On the left hand, where be doth work,
ats neds, but I cannot behold him;’’ and there Job’s anxiety
is that his innocence might be established. It recurs in xxiv. 1:
“Why...do they that know Him not see His days (1 Tih nd)?”
t.e. His days of vengeance on their oppressors. Notice especially
ver. 12, cp. xxxv. 14. I shall assume then that in chap. xix. 26,

Job looks to have his innocence asserted: a bee would arise
after his death to vindicate him, and tn the fact of that vindica-
tion® he would “ see God.”

VERSE V.

sr xdy wn op > mime oN WR
spna tna ys
nN b5}. Consumuntur renes met in sinu meo.

This clause is probably indicative of strong desire, sc. for the
realization of what had been before described. The word

1 Rendering INN asa present. So dictive or optative. But see P.8. v.
Rashi, oppon ofr 2A 53". 7 God would be seen indirectly,
But xlii. 5 shews that Job had not scen _— through the action of the Goél. Com-
God before. Hence MIM must be pre- _ pare ii. 5—7 with xix. 21.

A THEORY OF JOB XIX. %5—97. 145

sb) is used elsewhere of ardent longing: “My soul longeth,
yea, even fainteth for the courts of the LorD” (Ps. Ixxxiv. 2) ;
“My soul fatnteth for thy salvation : [but]’ I hope in thy word.
Mine eyes fail for thy word, saying, When wilt thou comfort
me?” (Ps. cxix. 81, 82). Cp. Ps. cxliii. 7. “ Bene igitur Hierony-
mus verba nostra vertit: reposita est hec spes mea in sinu meo”
(Rosenmiiller). Compare further, ‘YD "3", Ps. xvi. 7.

‘) TUM SN WRN]. This clause, as commonly render-
ed, seems fatal to the view that Job simply looks for vindica-
tion after death ; but another rendering may be proposed, which
3 consistent with that view, and which has the advantage
of joining the two clauses of the verse harmoniously together.
Instead of reading them disjointedly,

Whom? I shall see for myself...
My reins are consumed—
We hay thus connect them :—
Whom that I foe see
may

have nt
are

My reins consumed.

1. whom to sec has been, or is, my consuming desire.

There are two slightly different ways of arriving at this
rendering.

1. It is remarked in the grammars that the infinitive may
stand for the future &c., and vice versa. Subjoined are some
examples °.

In Prov. i. 2—6 YOY? breaks in upon a series of infinitives,
and is perhaps best rendered as below; “To know wisdom and
instruction ; To perceive the words of understanding ; To receive
the instructions of wisdom, justice, aud judgment, and equity ;
To give subtilty to the simple, t» the young man knowledge
and discretion. Tat the wise MAY HEAR and increase learn-
ing, and the man of understanding attain unto wise counsels.

1 There is no contrast in the ori- Shall see Him..."?
ginal. The parallelism shews that 3 The small capitals will shew where
ands expresses desire. I have given the quasi-infinitival rend-
3 If this verse is a mere reiteration ering by way of suggestion and without
the relative seems otiose. Why not “I authority.

Journal of Philology. VoL. 111. 10

146 THE JOURNAL OF PHILOLOGY.

To understand a proverb, and the interpretation ; the words of
the wise, and their dark sayings.” Similar parallelisms occur
in Prov. ii. 8, and v. 2: “That thou mayest regard ene)
discretion, and that tby lips may keep (\"¥3*) knowledge”
Compare Prov. xx. 25: “It is a snare to a man that he should

devour (Yo) that which is holy, and after vows to make or

quiry ”’ cpr). Prov, xxviii. 21: “To have respect of per
sons (0°32) “rl) is not good: and THAT for a piece of bread a
man SHOULD TRANSGRESS” (Y2/5*). Is, lviii. 5, 7: “Is it to

bow down (bm) his head like a bulrush, and to spread
(Y°¥') sackcloth and ashes under him?...Is not this the fast
that I have chosen? to loose (MMB) the bands of wickedness, to
undo (“W") the heavy burdens, and to let the oppressed go

(ndw) free, and that ye break (\MIN) every yoke? Is it not
to deal (O"5) thy bread to the hungry, and that thou bring
‘ (N'3NM) the poor that are cast out to thy house? when thou
seest the naked, that thou cover him ((f\"D3)), and that thou

hide not thyself (DOYMN) from thine own flesh?” Compare
1 Sam. ii. 3; Prov. xxiii. 35; Hos. i. 6; Ezek. viii. 6; Esth.
vill. 6; Lam. iv. 4; Job xxxii. 22; and conversely Ezek. xxi. 20.
In Lev. ix. 6 we read: “This is the thing which the LogD
commanded that ye should do, WYN MN MY,” and in Deut.
xxxlii, 11: Smite through the loins of them that rise against
him, and of them that hate him, that they rise not again”
(lit. from that-they-rise, PSy2""1). Let one more example
(from Ps. xvi. 7) suffice: “THAT I SHOULD BLESS [= to bless]
the LorpD who hath given me counsel, even in the night-seasons
have my reins admonished me.”

2. The construction above illustrated may be regarded
from a slightly different point of view. Of MINN W*S, tf
isolated from the context, a not unnatural rendering would be,
“that I may, or might, see.” Compare Gen. xi. 7; Deut. 1”.
£0; Dan. i. 8. Lee indeed asserts boldly that the relative ren-

1 For the full form of this construc- should be] called” (Hos. ii. 1); #4

tion compare ond TDN7WR OIPHI. see the next paragraph.
‘‘ Instead of their being [lit. that they

A THEORY OF JOB XIX. 25—27. 147

dering of the words as they occur in Job xix. 27 is ungramma-
tical, and adopts the construction MINN WR--Myt (ver. 25,
27), which is simple enough, only that two verses intervene.
But to return, MINN “WN may certainly be rendered con-
Jctively, “that I might see.” Now let it be required to
express, “whom that I might see.” To do this we should
prefix another "YN taken relatively; but the cumbrous phrase
mN WN AWN would at once reduce itself by ellipsis to
IAN “WN, the TWN serving at once for relative and con-
Junction.

T dy] LXX. xai ove ddXos. Compare: “ Let another man
(4) praise thee, and not thine own mouth; a stranger, and
not thine own lps” (Prov. xxvii. 2). Job trusts even after
death to be vindicated, and thus ipso facto to see God. But
this is only a partial satisfaction, for to see God’ with the eyes
of his own living self had been his consuming desire:

Whom that I might see* for myself,
And mine own eyes had beheld,
And not another's,

My reins have consumed within me.

This gives the full force of VI ND’ “et non pas un étranger,
tandis que s’i] arrive un vengeur aprés sa mort, ce n'est pas lui
qui le yoit” (Cuhen); and we may account perhaps slightly
better than on any other theory for the use of the past tense
W' if we suppose him to be expressing the disappointed hope
that by that time he might already huve seen.

Recuprtulation.
The vindication theory which has been advocated must of
course be rejected if the usual rendering of the clause "IN "WN

' Thus I take the expression in two tion Bernard’s rendcring of Job xiv.
tnses_a higher and a lower. Huf- 14, 15: “...can I hope...{that] Thou
Ragel, not disaimilarly: “Die Redensart mayest [still] call,..." (S1PM...0M).
Gott sehen ist ciner doppelten Erkli- This is specially to the point because
Ting fahig. Nach diesem Leben zu an expression of “hope” precedes. So
Gott kommen, oder, cinen gniidigen in xxxiv. 36: ‘My desire (is that) Job
Gott haben.” In connexion with the may be tried". Cp. moreover Ps. cxix.
litter he refers to Job xiii. 25; Pa. 17, NK ay dy Spr In xvi. 21,
M.17; xix. 18; civ. 29. MD), that one might plead, follows an

* Compare further for this constrac- expression of longing.

10—2

148 THE JOURNAL OF PHILOLOGY.

‘S) FUMN be correct; but (this objection to it being supposed
surmounted) it agrees well with the general tenour of the con-
text, and may be said to account more naturally than any other
theory for some of the more remarkable expressions employed
in the passage.

1. “Oh that my words were now written, &c.” (ver. 23, 24).
“Désespérant de se faire écouter par ses adversaires, il émet le
voeu qu’au moins la postérité lui rende justice” (Cahen). He is
confident of ultimate vindication, and is anxious to have ¥
known that he all along expected it. When it comes he wil
not be alive to speak for himself, and for this cause he is ans
ous that his confidence might be placed imperishably on recor

2. His adversaries are devourers of his flesh (ver. 2),
and he threatens them with “punishments of the swort
(ver. 29). It is then natural to understand by Goél (ver. 22):
an avenger of blood', who should maintain his cause againssss
those adversaries after his death.

3. THEY are the destroyers® of his “skin,” &c. (ver. 27 —)-
This is more forcible than to take \5j3 “ impersonally.” TEaiihe

verb is used of external violence in Is. x. 34: “And he shaxmall
cut duwn the thickets of the forest WITH IRON.”

4. Shin and flesh are to be taken as parts of the san——2¢
body, whatever be the precise significance of the FWY.

5. “W3, deud-lesh, that which remains when the "Wy, which
completes the form of the body, is subtracted. Conversely mm 2
Ezek. xxxvii. 6—10, first flesh is added to the dry bones, they 2
skin covers it, and lastly life is breathed into the bodies thie
completed.

6. “Even® from my flesh,” though only a mangled corpse?
I shall see God, sc. in my vindication. This expresses a hop
against hope suited to Job’s tone and condition.

7. Thus to see God, viz. by being vindicated after death, i=
i. partial satisfaction: “It has been my earnest desire to be vin—
dicated while yet alive, and thus to see God with my owns

1 O earth, cover not thou my 3 This does not necessarily exclude
blood” (xvi. 18). a lingering hope that it may yet be
2 So Rashi. Compare xvi. 11. well with him in life.

A THEORY OF JOB XIX. 25—27. 149

bodily eyes: others will be the actual witnesses of that justifica-
tion, which I have longed to see for myself.”

8. The conjunctive quasi-infinitive rendering: “ whom to
we (lit that I may or might see),”” may indeed be adapted to
the theories A, B, C, but the vindication theory has the advan-
tage of giving a very pointed contrast, and enables us to render
the put tense WD literally: “I shall see God (in my vindica-
tion’; Him whom I had hoped, alas, to see for myself—yea,
that mine own eyes might ere this have gazed upon.”

9. The conjunctive rendering in ver. 27' joins the classes
harmoniously together.

Isubjoin a translation of the whole passage, which is intended
to bring out the vindication theory.

I know that my avenger liveth,

And hereafter shall arise on earth,

And after that, as to my skin, they have destroyed that,

And from my dead-flesh—I shall see Eloah,

Whom that I might see for myself

(Yea, that mine eyes had beheld),

And not another,

My reins have failed within my frame.

PS, (i) Allusion has already been made to the following

“And I said, Hear, I pray you, O heads of Jacob, and ye
Princes of the house of Israel; Is it not for you to know judge-
ment? Who hate the good, and love the evil; who pluck off
their skin (ONY) from off them, and their flesh (ANNW) from
Of their bones; who also eat the flesh of my people, and flay
their skin from off them; and they break their bones, and chop
them in pieces, as for the pot, and as flesh within the caldron.
Thea shall they cry unto the Lorpb, but he will not hear them:

' Cahen in some sort anticipatestho this further, nor had I consulted Cahen
thie treatment of the sccund MAN, when I formed the theory advocated in
‘en he writes on the jirat: “Selon the text. In x. 18 YIN stands for:
Teeljues commentateurs c'est unvau: “O that I had [or, I ought to have]
P.urgaoi_ ne verrais-je Dieu qu'aprés given up the ghost." So in xix. 27
Mimort? Je vondrais avvir cette joie wo might render: ‘Whom I would
Pendant ma vie.” I have not traced fain have seen, &c.”

es
- - - -— << ~ —— - - -— -——- . mwocm #%
- - — -_—_ — - — -——e on a? cate
«
- - _ - - . °
— . — —_ - — —— [io = — - -_ a,
— - — — - = ~- ® bad =A
- ” ee
- - - -~ —_ = _ — nom « -
— - = — aoe sm oe - ene oe <>
— — *. as ee ee Oe —_ ae .
|
- - ~ ——w. ~- = «awe
o- - - -~ ° - — = - “ —_ 7", re ame,
- 7
°
- -. ~ _ me _—— >
—e “we =- - _— - - ~. i ve
-=
—_— - 7 me vr
— ow - a - - — —_ «ae mene —_ ae
-
——p my i?
- - e. - ~~
- - —=_ -_ . > om
ae me ew oe OT ee - . -s, _. 5 a
~ - - - a- LS
=-— - - . - ~ 6 ee el
_ - - ~ - oon ~.3
- — -_ ae te - —_-- = ~~ E
-_
—_ - -
. — - ° .
_ enme -= - e
-= - - = - o-*
* . «
. .- . - o "ete
. - ~- — Pad
—_ - - —- « ee - _—-
a6
- ee Li
se - - = ee - ary) of
_ one — ste
° a
- - - - - me - ° - ° wv
- —., —_ os -_ =" .
7 enwtece—m eer
- . e . . wwe. + _- Nea
-~ ~- > aap «--- -—— o-
~ — . °~)
~ = . -e. ;
~ - - - fd -_— —— . - 7 -_—" a —-" .
= = —_ . Ne
- . _. a a ee Sa oe
- °
- La . - = sx @ « ~s- oe! ay '
- - — —_ = *. wee ee tl * e
-. . Ele Gp Secekh - < - ~ +e Wye
eed — . wa’
- -— =. a ~- —s °
- -— - . - ~ _- ee ee ee) oo
- . - - _ 7 - ed — . i Sd *
. q
- . ° er)
. .™ on = . » = 8 - ~™~ rast
° . =e. - - ® we — —_ oe -.—" oh lee .
ot
~ — ° ~- ~ . ~ - = - ——- @ . “=e « mee eh
- - - - =- =m: . eo Tee we ohee®
-
- —_ » - . ° -3 7
7 .? ——~—= — = -~— oow te Or * Bre hd
— . = - - ro Le ww ee et ee aNhy
— vo . : ]
~ se - . ~~ 3 «+ -_ ~ - — a == =pe.r* y
- - - - - - ——= ~ mee > Ct cant)
= *
. - - - » sm
. - - . - — = .° enm -me we = ae ree yye ge i
—~ . -- - 7: —s --_ as sme en eee eo! en hawt ast Te

yan Ber PS te ls tit mis whith intraduee a pout

Pyrite. uci tush.” This argument

‘oes 7
a ec er on

eee ote “SKIDS

A THEORY OF JOB XIX. 25—27. 151

the first hemistich of ver. 20 a process of destruction is de-
xribed, while the part which escapes destruction is mentioned
by way of contrast in the second: “My bone cleaveth to my
skin and to my flesh, and I am escaped with the skin of
my teeth; so in ver. 26 we may contrast the remaining WS
mith the “iy already destroyed. But lastly, it may be urged
(s above on “IN) that the use of a preposition is subject to
sme modification from its noun; granted then that a certain
meaning of }} is in the abstract admissible, we should have still
toak whether that meaning is admissible in connexion with a
Particular noun’. And would not the rendering, “without*® my
flah,” sound strange to the Hebraist, even if there were one or
two different connexions in which | might mean “without”?

(iv) The rendering, “from (=in) my flesh,” is psycho-
logically not without harshness. Simpler in itself, if hard to
acommodate to the context, is the meaning: “of my flesh, or
hidred,” which would naturally describe the Goél. See Gen.
1.23; xxxvil. 27; xxix. 14; Jud. ix. 2; 1 Chron. xi. 1; Neh. v.
9. Omitting Eloah, we may read: “And hereafter he shall
and upon this dust (yea, after my skin, &c.), and of, or from,
ny flesh? I shall see ** *”; sc. the kindred Goél, who would
sing from his ashes. As regards Eloah, we may now suppose
(1; that Job suddenly rises above his original conception of a
human avenger into the unexpected climax: “ werd’ ich schauen
—Gott” (Ewald. See Dillmann) ; or (2) that by seeing God he
Means seeing His just judgement executed by a human instru-
ment‘; or (3) that Eloah may stand directly for a human
utiter. So Wolfssohn, quoted by Bernard, on xvi. 20. For
the non-literal seeing, cp. again Ezek. xxxii. 31. If Job’s hope
in posterity, xx. 10 is a natural retort. Cp. Ps. cix. 13.

(v) Perhaps greater prominence should have been given to

"Could we (as Chance puts it) write preceding clause. Compare Job xxxiii.
OND eee for “A man without eyes"? 21: “His flesh is consumed away, that
"Simpler perhaps than this would it eannot be scen.”
be the privative rendering, “ So that I 3 Elsewhere, I think, }® follows MIN
™ longer hare any flesh,” where we in Ex. xviii. 21 only.
the affix (Ps. cxv. 17), and carry 4 “Thou shalt be to him instead of
@ the idea of destruction from the God” (Ex. iv. 16).

152 THE JOURNAL OF PHILOLOGY.

the rendering: * From ‘the state of? my flesh I can see God.”
It is grammatically simple and has been adopted by many.
Against it compare xxi. 9, quoted on p. 144. But in Numb.
xxiv. 7, according to a common rendering. “I can see,” means
“Tcan see in prospect, or foresee” “I can see him (or it’, but
not now; I can bebuld bim .or it), but vet nigh” ie. I can see
tn the distance. So Job might say: “I can see {in prospect) a
manifestativo of Gud from or with respect te. my flesh = I ean
foresee my Vindication after death.”

(vi) Some details in Job xix. 25—27 being exceedingly
obscure, I have thought it well to propose for consideration
various expedients which have suggested themselves, although
in some cases I do not myself think them very plausible.
Details apart, the theory D has the twofold advantage of being
sugcested by the context and not contradicting anything which
occurs elsewhere in the book”, Against A is the fact that the
argument proceeds precisely as if Job had no idea of a resurrec-
tion: also A contains a solution of Job's difficulties, and thus
makes the actual ending of the book, with its appeal to his
inorance, an anticlimax. The argument against B is similar,
unless “Future Life” means an imperfect and shadowy exist-
ence, in which case B would approximate to D. Against C,
unless limited to mere Vindication, is the absence of any sub-
Kequent trace of the hope involved: moreover it is not required
by (if consistent with) the plan of the book that Job in the
midst of his perplexities should know what was to be their end:
while there are independent arguments for the view that the
(ot) ix conceived of as one who should appear after Job's death.
Thus much as regards the form of Job's utterance ; but what
allowance is to be made for the poetical nature of its expres-
nion, and for the style and purport of the book as a whole, and
whether the theory A, if wrong as an interpretation, be not a
right (or the only possible) application for a believer in the
feativreetion, are questions of importance which still remain to
be clincusxed-questions however which are more or less un-
nuitedt for discussion in the Journal of Philology.

C. TAYLOR.

1 Main con Gclairs sont toujours suivis de plus profondes ténébres.”

THE HISTORY OF THE RAVENNA MANUSCRIPT OF
ARISTOPHANES.

THE now celebrated Ravenna MS., the only one which contains
all the extant plays, was first made known to modern scholars
by Invernizi, who professed to have collated it for his edition
of Aristophanes, published in 1794. Of his collation Bekker
speak» in the followiig terms: ‘Ravennatem qui ante me ver-
savit incredibili socordia cum pari inscitia conjuncta &.’ The
collation which Bekker himself made in 1818, with all his dili-
gence and knowledge, is far from accurate, probably because
the time at his disposal was too short for the due performance
of his task. His collation of the Venetian MS., which is second
only in antiquity and importance to the Ravenna, is even more
imperfect, though, as he tells us, he examined it twice, at Paris

in 1812 and at Venice in 1819. Dindorf relied entirely upon
Bekker; and no subsequent edition of the whole of Aristo-
planes’ plays has been based upon a new collation of these MSS.

In 1852 I spent three weeks at Ravenna, noting all that seemed

'v Me important, and in 1867 I again made a minute collation
fthe MS. in the Acharnenses, Equites, and detached passages
fctherplays. In 1866-1867 my friend Dr Adolf von Velsen, of
Starbriick, made a thorough and complete collation of both the
Ravenna and Venice MSS., with a view to an edition of the
“t. The Equttes, which we noticed in the last number of

*~ the Journal, is the only play he has yet published. The
tholia of the Ravenna MS. were transcribed in 1837 for
Uindori’s edition by M. Miller’. So far as I know no serious

' Now librarian to tho Corps Légis- ing. His Meélanges de Littérature
latit at Paris, a gentleman as distin. Girecque is the product of long and
Rixhed for his courtesy as his learn- diligent research.

154 THE JOURNAL OP PHILOLOGY.

attempt has been made to trace the history of the MWS |
have endeav-ured to do this. and propose here to give brely
An @oreint wfmy researches. which have perhaps been more
Interesting te: myself than I can make them to my readen,
espeiany as they have led to so little in the way of definite
Tem it.

The MS. 15 a larze folio of parchment consisting of 191
loaves. OM cuding the fv-leaves at either end. It contains the
Glee Jays in the pdlowing order: 1. Plutus, 2. Clouds
3. Fe ce. 4. Borda, 5. Aviaits. 0. Peace, 7. Lysistrate. 8. Achar
nea Wings, 10. Thesmophoriasuse. and 11. Ecclesiazue.

Vie text is iu a cursive hand. the scholia in the mangin for
th. miest part in sma unelal characters, which were probably
adopted in order to diserimiuate tie commentary from the text.
The wh La were net written at the same time as the text, be
case vetenmiy the ink is of a different colour, but they were
probally added by the same hand. because we frequently find
lines of the text, which had been aceidentally omitted, given in
the maryin in the same ink as the scholia and the same cursive
writing as the orginal text.

Tie MS. has been corrected in parts by at least three dif-
ferent hands, one a tremulous hand of nearly the same date as
the MS. itself, another in blacker ink of the 14th or early in
the loth century, whose alterations are particularly frequent in
the Clowds, and one if net two still later in the Lysistrute
aul Thesmephortazuse, of which I shall have to speak more
particularly by and by.

On the fly-leaf at the beginning we read: ‘ Aristophanis
Codex Optimus cum argumentis et scholiis anonymi. Seriptus
seculo xX. Ita censebat Cyrillus Martinius Florentinus’’ By
the kindness of my learned friend Signor Francesco Palermo,
late librarian to the Grand Duke, I am informed that this
(‘yrillo Martini was a Priest, coadjutor of Biscioni who was
made librarian of the Laurentian in 1741. He was intrusted

1 This has given rise to a curious most precious is the celebrated Aristo-
error in Murray's Hand-Bovk, where —phanes, copied in the 10th century by
Apropos Of the Ravenna Library we Cyrillus Machirius, a Florentine.’
read: ‘Among its MS, collections the

RIVENNA MANUSCRIPT OF ARISTOPHANES. 155

by Biscioni with the task of describing and cataloguing the
Greek MSS. in the Library, and was doubtless a competent

Judge.

Bekker indeed says that he does not see why the Ravenna
YS. and those of AXschylus, Sophocles, and Demosthenes in the
laurentian, which have a strong resemblance to it, should be
thought older than the llth century. But Herr Miiller, of
Florence, who has spent his life in the thankless labour of copy-
ing and collating MSS. for other editors, and Dr von Velsen
bith assign them to the 10th century. We shall probably be
net if we suppose that it was written some time during the
lst century of the Basilian dynasty, which came to an end in
157, and in one of the monasteries, so richly dotated by the
hter princes of that family, ‘in which’ (to use Mr Finlay’'s
words} ‘the monks were living together rather like clubs of
wealthy bachelors than as holy societies of virtuous cenobites.’
(History of the Byzantine Empire, B. 111. c. 1.) Such persons
Were more likely to select the works of Aristophanes for their
library and to pay for the production of a costly and sumptuous
bok, for such it must have been, than their successors, when

Comnenus had confiscated the endowments of the monas-

teries and when consequently they were tenanted no longer by

e younger sons of noble houses, but by the sons of peasants

Completely ignorant of pagan literature and fanatically pre-
JUdiced against it.

Nuw for the history of the MS. It is at present in the
Biblioteca Communale of Ravenna, also called Biblioteca Clas-
S€nse, because it belonged to the monks of Classe of the Camal-
Jolite order. At the dissolution of the monastcries under the
French both convent and library were made over to the town,
4nd thus the books were saved from dispersion.

The convent, which owes its foundation to 8S. Romualdo, was
Orginally adjacent to the church of S. Apollinare in Classe, two
Miles outside the walls of Ravenna. In 1512 it was attacked
by the French troops, and its Abbot, Andrea Secchim, slain in a
Vain attempt to defend it. The monks for safety removed to a

place within the walls, and built the stately convent which still
bears their name. According to the Annales Camaldolenses it

156 THE JOURNAL OF PHILOLOGY.

was begun in 1512, and over the principal entrance is the date
1523, indicating, I suppose, its completion in that year.

There appears to be no record of the time when, or of the
person by whom, the library was founded. Perhaps it was the
Cardinal Giulio della Rovere, Archbishop of Ravenna from 1566
to 1578, who is mentioned in the Annales as having been
‘Insignis Benefactor ordinis Camaldolensis. He may have in-
herited something of the bibliomania which distinguished the
. Dukes of Urbino, to whose principality his family had succeeded
by favour of Sixtus IV. In the library itself is a portrait of a
former monk, the Padre Canneti, under which is an inscription
recording that he enriched the collection ‘selectis et copiosissi-
mis codicibus.’ His ‘floruit, as I was told, was in the begin-
ning of the last ceutury. There is nothing to shew how or when
the Manuscript of Aristophanes was added to the library. The
present librarian told me that he had heard from his predecessor
a tradition that it had been bought for a very small sum at a
book-stall in Rome. Terhaps it was among the acquisitions of
the Padre Canncti. But though the clue to its recent history
thus fails us, let us sce whether we cannot recover it at an
earlier period.

The Aldine Edition, the Editio Princeps, of Aristophanes
was published at Venice in 1498. It contains nine of the
Comedies, z.e. all except the Lysistrate and Thesmophoriazuse.
In the Latin Preface Aldus says ‘Decimam Lysistratam ideo
pretermisimus quod vix dimidiata a nobis haberi potest.’ It
does not appear that he or his editor Musurus had even heard
of the Thesmophoriuzuse. In this edition there was an im-
portant omission in the Paz (lines 947—1011, ed. Dindorf, from
To xarotv...to rov 8 orotvtew) which was indicated by the word
Nevret, and instead of the three concluding lines of the same
play,

@® YalpeTe, YalpeT’ av-
Spes, xav Evvérrnobé por
mWraKotvtas edeabe,
we read Actes.
mraxoitvras eae.
These lacuna were not supplicd in the second, or Juntine,

RAVENNA MANUSCRIPT OF ARISTOPHANES. 157

edition, published at Florence in 1515. In the preface Bernard
Junta, dedicating the book ‘nobili patritio domino Francisco
Accolto electo episcopo anconitano,’ says: ‘Putabam, vir duc-
tissime, duas quoque notioribus his addere posse nondum ab
aliis impressas, que cito forsan abs te nostra ope his novem
comitate legi poterunt, in forsan Euphrosyni bonini praceptoris
tui et aliorum tuorum pariterque nostrorum amicorum promissa
irrita quod credere nequeo in leves abibunt auras.’

The printing of this edition was completed, as the co!ophon
informs us, in the month of September, 1515.

Early in the following year: the same printer put forth for
the first time the two plays alluded to in his preface to the
former volume, the Thesmophoriazuse and Lysistrate. ‘His
summa manus imposita est quinto kl’ Februari M.p.xv. Leonis
Pape nostri anno tertio,’ t.e. according to our modern reck-
oning, January 28, 1516. In the preface, also addressed to
Francesco Accolti, Bernard Junta says: ‘ Venit, mi Francisce,
expectata dies illa in qua ex urbinate bibliotheca antiquissi-
mum Aristophanis exemplar nacti sumus ibique inter alias
Avowotpatny cai Gecpodpopiafoveas id est Lysistratem et Cereri
sacrificantes feminas non alias visas comedias invenimus hasque
et tuo nomine cudere tibique dicare, amicorum optime, visum
est.’ He then complains of the corruption of the text, and
in a note at the end adds: ‘Habes candide lector nusquam
hactenus impressas binas Aristophanis coinedias...quas ex codice
adeo vetusto excerpsimus ut altera interdum dictionis pars ibi
desideretur. This is by no means a correct description of the
MS. for it implies that its leaves had been worn or its writing
defaced by age, which is not the case. Is this mere carelessness
or deliberate mystification? Euphrosyno Bonini, above mention-
ed, was a native of Florence, and, as Poccianti in his Cutulogo
deglh autort Fiorentini tells us, at one time Professur of Greek
Literature in the University of Pisa. He also translated Galen,
and Bandini in his Catalogue of the Laurentian Library says of
him: ‘Fu dei piu distinti discepoli d’ Angelo Poliziano, tanto
che nel 1497, assai giovane, scrisse e recité nel duomo di Firenze
lorazione inaugurale per la solenne riapertura del pubblico

studio e piu che fece stampare a’ Giunta non pochi autori Greci
e Latini.’

158 THE JOURNAL OF PHILOLOGY.

Franceso Accolti was no doubt the same person whom Bembo
in a letter to Billiena, dated April 19, 1516, speaks of as dancing
attendance at Urbino upon the Duchess Dowager and the Lady
Einilia and professing to the former lady that he had been in
love with her for five lustres and a half. according to the
morals of that time a Bishop elect might thus conduct himsdf
without blame. At all events the dedication seems to impy
that Francesco Accolti had some interest at the court of Urbino.
Perhaps also Giuliano dei Medici, who had been sheltered &
Urbino in troubled times by the Duke Francesco Maria and who
was himself distinguished for his love of letters, was induced to
exert his powerful influence with the Duke, who, as mattes
then stood, could not well refuse anything to one of the Medic.
Giuliano died on the 17th of March following ; the troops d
Leo invaded the duchy, and entered Urbino on the 30th of May.
On the 18th of August Lorenzo, the Pope's nephew, was made
Duke of Urbino in place of the deposed Francesco Maria. Thus
the precious manuscript was borrowed, and in consequence of
the troubles which followed, neither restored nor reclaimed.

This is more probable than that it was restored and sub-
sequently stolen from the watchful guardianship which in
peaceful times protected the library of Urbino. At all events
this was not one of the hundred and sixty-five Greek MSS.
which were in the library when it was transferred to the
Vatican by Alexander VII., in the year 1658.

How and when the MS. came into the Library of Urbino
are questions as obscure as how and when it was carried away.

Duke Federigo, the founder of the library, commenced
making his collection of books about the middle of the 15th
century. He spared, we are told, no pains or cost in securing
MSS. cither in Italy or abroad, and he had 30 or 40 persons
employed as copyists. Vespasiano, of Florence, who was one of
his agents, writing about the year 1463, gives a list of the
authors whose works were then in Federigo's collection. Of
the Greek classics he mentions Aristotle, Plato, Homer, So-
phocles, Pindar, Menander, Plutarch, Ptolemy, Herodotus, Pau-
sanias, Thucydides, Polybius, Demosthenes, Atschines, Plotinus,
Theophrastus, Hippocrates, Galen, and Xenophon; but Aristo-
phancs is not named. (Dennistoun’s Dukes of Urbino, Vol. 1.

RAVENNA MANUSCRIPT OF ARISTOPHANES. 159

2158; Zanelli, Za Biblioteca Vaticana, ch. v.) We may there-
fore conclude that the MS. had not then been acquired.

It may have been added to the library subsequently either
by Federigu or by Guidobaldo L, who succeeded in 1482 and
ded in 1508. The latter was also an accomplished scholar and
Cnversed with fluency in Greek, but Aristophanes is not men-

Uoned in the list of his favourite authors given by Castiglione
(Dennistoun, Dukes of Urbino, Vol. u. p. 81). His youthful
4nd warlike successor Francesco Maria della Rovere had pro-
“ably no more inclination than money to spare for making
‘iditions to the library, at least during the early years of his

“€2m. The MS. was therefore in all likelihood brought to
C gbino not later than 1508. On the other hand if it had ex-
isted in a library so well-managed and so liberally thrown open
€ <> students as was that of Urbino before the year 1498, the date
© € the first Aldine edition, Aldus himself could scarcely have
€£zxiled to be aware of its contents’. On the whole therefore I
<-onclude that it was added by Guidobaldo to his collection
%e<treen 1498 and 1508, probably during the comparatively
© wanquil years which followed his restoration to his Dukedom
im 1503. It had a brief sojourn there, having been borrowed,
45 we have seen, in 1515 and in all probability never returned.
“What the Thesmophoriazusw and Lysistrate published by Ber-
Mard Junta early in 1516 were printed directly from the Ra-
Yenna MS., and not from any transcript of it, is to my mind
Clear from internal evidence by a comparison of the texts. In
the MS. itself I noticed a curious confirmation of the fact.
Faint pencil marks have been drawn across the text, corre-
‘ponding with the pagination in the Juntine edition, with (so
far as I observed) only one exception, and that was when
the unusual length of two lines had deranged the calculation.

Noticed also on one page of the MS. a printer's black thumb-
Mark,

I mentioned before that in these two plays the MS. had

€n corrected by at least one later hand not found in the rest

OF the volume, and chiefly employed in inserting the names of
© interlocutors omitted by the original writer. These inser-

. , The great Dukeandthegreatprinter cated to Guidolaldo his editions of
©Te on friendly terms. Aldus dedi- Thucydides and Xenophon.

160 THE JOURNAL OF PHILOLOGY.

tions are made in a reddish ink and are due as I believe to two
different hands, the later corrector having used an ink like but
not identical with the ink of the former, and also having imitated
his hand.

In the Royal Library at Munich, among the books which
formerly belonged to the Fugger family, the great merchant
princes of Augsburg, is a MS., numbered 492 in the catalogue,
eight inches long by six wide, on good paper, containing, tnter
alia, the Thesmophoriazuse and Lysistrate. The writing seemed
to me to belong to the early part of the 15th century, and M.
Halm, the eminent librarian, whom I consulted, agreed with me
as to the date. On the binding inside are the words iwavvou
Tov potSavou eye (sic), indicating doubtless the name of it===
possessor previously to its acquisition by the Fuggers’.

I have no doubt that this MS. is a transcr_pt of the Codex——
Ravennas, made by some one who had pretensions to scholar——
ship and therefore ventured on emendations, while he was not=—
so accurate in mere transcription as an ordinary copyist would
have been. If 1am not mistaken the transcript was made from—
the Ravenna MS. after it had been corrected by the earlier of
the two hands I have spoken of, and before it had been correct—
ed by the later. The earlier corrector may have been the
writer of the Munich MS,, the later, the editor employed by~
Bernard Junta. The selection of these two plays by the copyist-
shews that he was aware that they were not found in the
ordinary MSS. of Aristophanes. The writer was probably a
Greek, one of those who were induced to turn their attention t»
the copying or commenting of the ancient authors, because the
newly awakened enthusiasm of the west had made it a profitable
trade. The paper itself seems to be of Italian manufacture, but
this does not militate against my hypothesis, because from the
middle of the fourteenth century paper imported from Italy
seems throughout the Greek empire to have superseded the
inferior paper manufactured at home.

W. G. CLARK.

1 The wrong accent on elu seems to About this ‘John of Bevagna’ I have
show that the owner of the name was _ not been able to find anything in our
not a Greek. The name is probably to University Library, even with the as-
be translated ‘Giovanni di Bevagna.’. sistance of Mr Bradshaw.

\OTES ON THUCYDIDES AND THE ACHARNIANS
OF ARISTOPHANES.

THUCYDIDES, L. 68.

ww Se ti Set paxpryopeiv, av tors pev Sedovdopévous Spare
70 5 ériBouXetovtas avrovs nal ovy HxtoTa ToIs HmeTéepots
Funwayors.
_ Amold gives tyuerépors in his critical note as the reading of

one MS, This, though unnoticed by other editors, seems far
Preferable in sense, and the confusion between the two pro-
nouns in MSS. is frequent. It is a more telling argument for
the Corinthians to urge before the Laccdaemonians that “the
Athenians are plotting especially against your allies” than
“they are plotting against our allies.” And the whole ques-
tiva is of wrongs done to Hellas and to the confederacy (Eup-
pay.c) of which the Spartans, not the Corinthians, were the
heal. To urge wrongs done specially to Corinthian allies seems
out of place. And in the next chapter those on whom the
Athenians are encroaching are again mentioned, and the Lace-
daemonians charged with being virtually their oppressors, thus :
Tap 8€ Upets altiol...€s TOde aEl aTooTEpObVTES OV povoy Tols UT’
Exeiney SeSovrmpévous GAAG Kai Tos UmeTéepous Hy Evppcdyous.

THUCYDIDES, I. 8+.
ae 5¢ ws mpos ev Bovdevopevous Tors évavrious Epyw Tapa-
SkevalopeOa xai oun e& Exeivwy ws apapTnoopevwy Exe Set
Tas €\ridas, GAN’ ws NuaY aLTaY dadadds Tpovooupévwn.
It is strange that nearly all editors have preferred vapa-
cxevatapeOa, interpreting it as mapacKevalec@a: det. The ear-
Journal of Philology, vou. 11. 11

162 THE JOURNAL OF PIILOLOGY.

lier part of the chapter has described by a series of indicative
(ove &éEuSpifouer—otx émratpopeOa—et'Bovroe syyvopeba) the
actual conduct and character of the Peloponnesians. Archids
mus then proceeds “And in action we always prepare agans
our opponents on the supposition that they are taking we
counsel: and we de not need to vround our hopes on their pe
sumed blunders, but on our own secure foresight.” Goler
keeps wapacxevaloueba, but devs not shew how the followng
ov...de¢ is tou be understovd.

THUCYDIDES, I. 141.

oi TondTot ot're vats TANpOdHTEs ovTE Teas oTpaTLaS TAM
xis exrréprew Stvavtat, amd tov iStwv te Gua arévres Kai ct
Tov avta@y Saravevtes.

Nearly all editions have amo rev avtéy without comment
Poppo has atrév, but in his note says “ vulgo atréy, quo ambt
guitas evitatur.” Other passages support atray as Thucydides
Greck (though Bigys says that Thucydides uses ta atte
here only), but the distinction or variation “their private po
sessions,” © their own possessions,” is unmeaning. Ter avTov On
the contrary yields the very sense wanted. “The Peloponne-
sians are workers of their own land (avroupyot) and have no
forcien possessions” says Pericles; “their resources are all at
home: such a nation cannot send out fleets or armies often,
since they have at ene and the same time to be away from thet
own property and yet to draw their expenses from the same”
The Jand and property at home must deteriorate by the absence
of the cultivators, and yet this same must supply the sinews of
war, The advantage which the Athenians on the contrary had
in their foreign possessions is set forth later im this speech by
Pericles, and by Archidamus in Thue. 1. 81.

ARISTOPIIANES, churn. 988.

Tai tT émt to Setrvoy Gua Kai peyadra 87 dpovel,
tod Biov & é&éBare Setypa tude ta wrepa mpotav Oupar.
The amount of words lost here must be the equivalent of
elSes & eldes @ in v. 971, and the general sense appears to be
rightly given by the scheliast: Atxatomodes eretyet, orrerdet,

NOTES ON THUCYDIDES, éc. 163

omvdates rept to Seirvov. Hence Meineke (in his Vinditciae)

proposes ottoat S émrontal +’ émi «.7.r., which seems better
thn Bergk’s eides @ rovd; éreiyet rep) «7.4. But Meineke
goes on to say “recentissima aetate Henricus van Herwerden
ticavit Aristophanis locum in Rav. sic scriptum legi ézré-
perai t éxri TO Setrvov, quod sane mirum est Bekkeri oculos
fugsse.” If the Ravenna manuscript does contain this, the
luna is half supplied: for the rest we might take Meineke’s
avoci 8, or read eldes of err. But any way érrépwra: even as
aconjectural emendation suits the passage well: “see you how
he is all in a flutter (eager, excited) for the feast?” Similar
wes of srrepodobat, avamrepovcOas are referred to by the lexi-
cons; and especially to the point is the play on the various
senses of rrepovy, avamrrepovy in Aristophanes’ Birds, vv. 1436
lis:

2X @& Saude pr) vovOérer ps’ adda wrépov.

I. viv tos Aéywv wrepo ce. YX. cal mas av Noyots
avdpa wrepwceas av; II. mavres tot Noyots
avarrrepovytTat, %. mavres; II. ovx dxneoas
Grav Aeywouy of Trarépes ExdoroTe
Trois Gudétais ey tose xovpelos tradi;

Servers yé prov TO petpaxioy Autpédns

Aeyor averrrépwxey GoD imwndarety.

0 5é Teg Tov avtov dyoly eri Tpayodia

aventepwcbas cat trerotjabas tas ppévas.
This use of avattepovoOa: in Aristophanes (and the passage
thers it to have been a common one at Athens) recommends
frépetas as a good verb to fill the gap in the line of the
Atharnians. Nor need the occurrence of wrepd in the next
line offend. For even if it be thought that it suggests a weak,
iliterative sort of pun, Aristophanes is so often guilty in
this kind, that it is scarcely an objection.

W. C. GREEN.

11—2

NOTES ON THE SUPPLICES OF ASCHYLOUS.

“Esch. Suppl. 336.

tis 8 dv hirous wvotto Tos KEXTHPEVOUS ;

I cannot sce that Boissonade’s conjecture évorro is prefer-
able to the reading of the MSS., which Dindorf retains, provided
that a proper explanation be given to the word ¢iAous. If
gidous be understood as in Hom. Jl. u1. 163,

odppa idn mpéotepov Te Tocty mous Te Pidous TE,

where anot are Helen’s relatives by marriage and fAoe her
blood-relations, the meaning will be: ‘But who would purchase
relatives as possessors?’ t.e. ‘Who would give anything for a
relative as a husband? To this the king replies:

aOévos pev ovtw peifov avkerat Bporois,
which Mr Paley rightly explains: ‘hoc modo, nempe consoci-
andis familiis non modo propinquitatis, sed ctiam nuptiarum
vinculo, magis valent homines. The chorus answers :

kat Svotvyourtwy y evpaprs atradXayn,
where again I am quite ready to accept Mr Paley’s explana-
tion: ‘Mihi ita videtur intelligenda. Imo et si tis (sc. maritis)
res male evadant, haud multum morantur divortium; h.e. fa-
cilius a cognatis sanguine uxoribus quam ab aliis discessuri sunt
si velint mariti: propinquos enim non punient propinqui.’

Thus, with the reading of the MSS. qvotro, all is connected,
whereas a disturbing clement is introduced by the emendation
ovotro, which, though equally good as regards the preceding
question of the king:

motepa Kat éyOpav 7 Td pu Oeues Aéyers ;
is utterly foreign to the argument that follows.

NOTES ON THE SUPPLICES OF ZSCHYLUS. 105

For éAoz in the sense of blood-relations, compare also Asch.
Ag. 1219. maides Oavdvres wotrepet Tpos THY hidrwv.

fEsch. Suppl. 455.

ToANay dxovoov téppat aidoiwy Aoywv.

I cannot agree with Mr Paley and Mr Linwood that aidoiwv
here means ‘respectful.’ The ‘finale of many respectful words’
Was a threat on the part of the chorus to hang themselves,
which appears to me anything but respectful. For my own
pzart I understand aido‘wy as appealing to the aidds due to
stu ppliants, and as implying a claim on protection:

‘Hear the finale of many protection-claiming arguments.’

Zevs aidotos in 192 is surely Zeus, who presides over aiéws,
amd thus is practically equivalent to Zevs adixrwp in line 1.

f¥sch. Suppl. 461.
Ein TL TioTOY TOS VITOTTHTEL OTOAY.
_ I regret that Mr Paley has introduced vioornjcess, suggeres,
Im the place of varoorncet, promittes. What can be more natural
4nd suitable than that the chorus should say: ‘Unless you
make some reliable promise to this band, we shall do so and so’?

Esch. Suppl. 1018—1049.

This chorus is composed of pure Ionic @ minore lines varied
by or interspersed with lines or phrases, in which an avaxdaots

es place. It is my purpose to endeavour to show that two
“Mendations, made and generally accepted for the purpose of
restoring the metre, have really been prejudicial to it.

I think it will not be disputed, that a metrical phrase, in
Which an anaclasis takes place, must necessarily be equivalent
‘2 temporal value and ultimately reducible to the corresponding
20rmal phrase, in which the metre appears without the ana-
Clasis, I think too that I may assume that, when the phrase
1M which the anaclasis is found occurs at regular intervals, it
©Oncludes the stanza, to which it gives an agreeable variety,
Much as the versus paremiacus in the otherwise monotonous
AXapestic system. Thus the conclusion of the synaphca of the
SY stem will always coincide with that of the anaclastic phrase,

166 THE JOURNAL OF PHILOLOGY.

and the last syllable, whether long or short, will always have to
be considered as possessing the proper value of the last syllable
in the pure or normal phrase.

Let us first consider the lines or phrases, in which the ana-
clasis occurs, in which there is no doubt as to the reading, and
then apply the principles thence obtained to the settlement of
the metre and reading of the disputed passages.

In line 1025 we have:

éxéra p15 | Ere Neddou | mpoxoas céBwpev vpvoss.

In order to reduce the latter half of this line to the Ionic a
minore (or Anacreontic) metre, we have merely to reckon half
the long syllable Bw in céSwpev to the first and half to the
second Ionic foot in the phrase, thus:

mpoxoas o&B8a | Bad pév vuvois.
2 2

Similarly in line 1033 ;
yapos ENOor | KuOepetas | orvyepov wédXet 10d dAdo»,
we obtain two Ionic a minore feet by dividing ef in aéXe be-

tween them, thus:
ariryépov aédet | Net 785’ aODov.
3,

In line 1083 we have:
pedipa tpiBot | Bot +’ éparwy.
3 2

And in line 1051:
mpotepay qeéXol | Aol yivaixay.
a 2

Let us now consider the disputed and emended lines,

Line 1021 stands in the MSS. zepiwaiere maXarov, which does
not make up two Ionic a minore feet. This is given by Mr Paley
in his first edition qepivaieras madaov, an emendation, which
satistics the requirements indicated above, but necessitates the
alteration of of in the preceding line into ofs. In his second
edition Mr Paley (after Hermann) reads aépivacov| rai raraioy |,
which gives the value of a short syllable too much, unless the
oy of mraXacov be reckoned short, which I do not think it can be
at the conclusion of a system.

For my own part, taking into consideration the phenomena
exhibited by the uétis of Catullus, where an effect almost

NOTES ON THE SUPPLICES OF ESCHYLUS. 167

identical with that of an anaclasis is frequently produced by
the simple resolution of the last long syllable of an Ionic foot,
e.g. celéri rité | maria|, and considering how easily the article
to may have been absorbed by a preceding te, I venture to
propyse tu read :
wépivaseteé | 75 madacov.

ludeed I think that the insertion of the article will be
fuund an improvement to the sense and spirit of the passage :

ive pay aa|tuavaxras | paxapas Oeous | yavaovres |

Toloryous | Te Kai of yedu | "Epacivor |

wepivaiere | TO TaNracoy.

The other disputed passage is the corresponding line of the

antistrophe (1030), where the MSS. give,

TOOE peNlooovTes ovdas,
which scans at once in accordance with the above principles,
is ieiluws :

TO0E péALooorT | dvTés ovdas,
2 Z

whereas the generally accepted emendation,

TUOE peiAtaa|OvTes ovdas,
elves the value of a short syllable too much, unless the as of
orcas lb: reckoned short at the end of the system.

T must nut however conceal that, whether accidentally or
uot, the uther anaclastic phrases do not afford any clue to the
~hitien of the question as to the quantity of the last syllable of
au auaclastic phrase, as they all end with syllables either long in
themselves or made long by a consonant commencing the next
lin, Acainst this doubtful point I have to set (1) the reduc-
tim of the number of emendations from two to one, (2) the
steater inherent probability of my emendation, which inserts
a syllable easily absorbed or lost, but alters nothing, (3) the
Choe analury of the manner in which I explain the phrases in
qkstion with the phenomena of the undoubtedly authentic
waclistic phrases, (+) the improbability of the introduction of
“ very different a movement, as a ditrochaic, in an Tonic
‘Vstem, aud (3) the analogy of the sides of Catullus,

Every line of the Attis of Catullus is similarly reducible
ty an Iomie a ménore tetrameter cataleetic upon the principles,

168 THE JOURNAL OF PHILOLOGY.

(1) That each line is divided into two halves, which never |
run into each other.

(2) That each half-line is made up of syllables equivalent
in temporal value and reducible to two Ionic feet, the last half-
verse in each line being catalectic.

(3) That no liberty is ever taken with any syllable but the
first in the 2nd and 4th feet of any line.

(4) That the last syllable in every line is always consi-
dered long. For example:

Siipér alt& véct | véctiis Attis || c&léri rat% | maria.
3s: 6
Jam Jim ddlét | let quéd gi || jam jamqué poe | poenitét.
3 3 2 2

N.B. In the last line above quoted the Ionic @ minore is
replaced in the Ist and 3rd feet. by its inversion, the Ionic a
majore.

Indeed the Attis of Catullus, the metre of which I have
reduced to a very simple tabular form—six lines representing
every variation—in my and the late Mr F. N. Sutton’s Selec-
tions from Cutullus, Tibullus and Propertius, presents pheno-
mena sv similar to those of the Chorus in the Supplices of
“Eschylus which I have been cxamining, that I think they may
be fairly considered as not only illustrating, but actually ex-
plaining cach other.

A. H. WRATISLAW.

ON THE ATHENIAN PROEDRI.

FEw points connected with the political arrangements of the

Athenians have given occasion to so much discussion as the
regulations respecting the Proedri and Prytanes. A natural
cunivsity is felt to have the fullest possible information as to —
the mode of conducting business in the Athenian Boule and
Ecclesia, while an exceptional interest attaches to this par-
ucular point from one of the most striking episodes in the life
of Socrates—his conduct upon the occasion of the trial of the
eight generals. The question is one which we may venture,
even at the risk of repeating much that others have said before,
to review, with the hope of adding something towards its
elucidation.

The earlier enquirers into Athenian political antiquitics
drew most of their information from the Orators and from
Grammarians, especially such of the latter as had written com-
Mentanes upon the Orators. It is obvious, however, that
political phrases in Demosthenes or Aschines must often be
understood as applying only to the state of things in their
times; and as for later writers such as Harpocration or Li-
banius, valuable as they are when confirmed by other testimony,
Yet in many cases they have nothing to tell us beyond what
they have thought to be employed by the authors they illus-
trate, or they quote second-hand from authorities they only
half understand. It is not to be wondered at therefore if the

statements of the Grammarians respecting the office of the
Proedni are at variance with each other. The author of the
second argument to Deinosthenes’ oration against Androtion
has the following statemont: ‘pyov ovy of mevtaKxocwe tas

170 THE JOURNAL OF PHILOLOGY.

tptaxoclas mevrnxovra nuépas. aA érretd) woAAOt Hoay, Kat
Suvoyepas yvvov ta mpaypata, SteiNov éautous eis Séxa pepidas,
Kata Tas gvAas, ava TevTnKovTa, TovTovs yap éxaoTn ud}
mpoeBadreTO. woTe ouvéBawe Tovs TevTiKOVTAa apyew TAY
G\Awy ava TplaxovTa nuépas. avTat yap ai Tpiaxovra mTévte
nucpar eiot TO SwoéxaTov pépos Tod emavtTov ... GAr ézedy
Tad of TevTnKovTa Todt joay eis TO dpyew dua, oi Séxa,
KaTa KANpoy pds nuépas Tay era, opoiws dé Exactos Tav
GdXwv amd KANpoU NXE THY EavTOU nuLépay, axpts ob WAnpwOwoW
ai émta 1épat. Kat ovvéBawe toils Gpyovet Tpets jer) apyeuv.
Exactos S€ dpywy ev mid Hycpa exadeito émiatatns...... totéov 5
dre of ey evtnxovTa éxadovvro mputaves’ ot 5é déxa mpdedpor'
*6 5é ely emictatns. It has been generally taken for granted
that this gives an accurate account of the arrangements of the
Prytanes and Proedri, at least as they stood in the times of
Thucydides and Socrates ; and as such it has passed into all the
handbovuks of Greck political antiquities. But other explana-
tions by the Grammarians are to a different effect. Besides the
fifty Prytanes of the wputavevovoa van, we hear of nine daily-
appointed Proedri, one from each of the remaining tribes. Thus
Pollux (VIU. 96. fin.) says :—6ray ot mwputaves tov Sipov 9 TH
Bourn ouvaywow, ovtos (i.e. 6 émecratns Tov mpuTavewy) &£
éxaoTns puAns mpocdpov Eva KANpOL, MovnY THY TpUTavEevovcay
dies Suidas also (s.v. éruetarns) describes the same trans-
action in almost the same words. These statements, coming
from writers who probably followed directly or indirectly the
authority of Aristotle, are deserving of great weight. Nor is
Harpocration (s.v. wpéedpot) less explicit. Tpoedpoe éxAnpodvto
Tav tputavewv Kal’ éxaotny mputavetay els &€ Exaaotns purrs,
WAY THS TpvTavevorvons, olTivEes Ta Tepl THY ExxANTIav Sewxour.
€xarovvto Sé mpoedpot, erretditrep apondpevoy Tov GAXwY array
Tw. Toddakis & ott ToLvopa Tapa Tots pHTOpaLY, WS Kat Tapa
Anpoobéver év TO nat ’Avdpotiwvos, kal Atoylivyn év te xaTa
Krnoipdvtos. o7t 5€ 6 KaXovpevos emtotaTns KANpOt avTors
elpneev “AptatotéAns év “A@nvaiwy qodtteta. Here mpvraveiay
in the first clause is by most critics altered to 7épav, which
is certainly the word which we should rather have expected
to find, and ure ought probably to be inserted before tay

ay

ON THE ATHENIAN PROEDRI. 171

Tputaveoy'’. If we leave the text as it stands we perhaps might
construe as follows: “Proedri of the Prytanes (not ‘from among
the Prytanes,’ which would obviously contradict the words that
directly follow), were appointed by lot during euch prytany,
ue from every tribe except the prytanising tribe.” For wo
know from other sources (as we shall presently sec), that the
Dine non-tribal* Proedri were chosen by lot daily, and held
their office only during that portion of the day in which busi-
hess was being transacted by the Senate or the Assembly.

But now the question arises, what was the precise relation
in which these nine non-tribal Proedri and their Epistates
sod towards the fifty Prytanes of the prytanising tribe?
When and why came these nine non-tribal Proedri to be ap-
Pinted at all? Again, what is to be said of the ten daily
tnbal Proedri of whom we learn from the author of the argu-
ment to the speech against Androtion ?

The acevunt which Mr Grote (Vol. m1. p. 118 fol. 2nd ed_),
flowing mainly Schémann (De Comitiis, Bk. 1. ch. 7), has
sven, may be said to represent the popular view of the iatter.
lt is an account in itself indeed sufficiently intelligible and
probable. It is doubtful, however, whether it can be said to
rest on sufficient documentary authority. On the other hand,
Schimann himself, who is chiefly responsible for it, has seen
reauns for changing his opinion’, now that a more extended
anl careful examination of documentary evidence has thrown
4 diferent light on the matter. That evidence is twofold:
I, that of authors contemporary with the institutions we are
tamining, such as Thucydides and the Orators: and 2nd, the
“idence of contemporary Inscriptions. It can never be tvo

tten repeated in all questions of this kind, that the statements
Grammarians, valuable as they may be in connexion with
Other testimony, are as nothing compared with the authority

I : - . a .

; See Seliimann, De Comitiiz, p. 85; — the expressions tridulex and non-tribu-
: Mice, De Epistatia, p. iv. les, which I have translated ‘tribal '
he earlier writers on this ques- 9 and ‘non-tribal.’

no Uecd the terius contribules, and 3 See his Gricchische Altcrthimer,

Coan )

eantrihule a. (So Sehimann, De Vol. 1 p, 3.
But Meier with reason prefers

172 THE JOURNAL OF PHILOLOGY.

of contemporaneous documents, whether of Authors or of In-
scriptions.

First of all then we find in earlier writers no mention of
Proedri whatever. Thus when Thucydides (v1. 14) introduces
Nicias dissuading the people in the Ecclesia from the Sicilian
expedition, he makes him thus address the President :—«al ov,
@ mpvtavt, ravta (elmep nyel oot mpoonxew KndecOal Te TIS Wo-
ews Kal Bourer yevéerOar trorltns ayabos) émupndile cat yvropas
mpotibe, avis "A@nvaiows. The obvious inference from this
passage is that at this time the President of the Ecclesia (as
also of course of the Boule) was one of the fifty senators of the
dvd) wpuvtavevovoa; and further, that in his capacity as Epis-
tates he had the sole power of putting questions to the vote
(erinpitev) and of submitting subjects for discussion. To
the same effect is the narrative of Socrates’ conduct in the
assembly when the subject of debate was the conduct of the
generals after the battle of Arginuse’. Here, to add greater
force to the argument, we are furnished with the twofold ac-
counts of Xenophon and of Plato. "Eruyev nuay (says the
Platonic Socrates in the Apology, page 32 B) n duvAn 'Arreoyis
MpuTavevovaa, OTE Vets TOUS Séxa OTPATHYOVS TOUS OUK avEdope-
vous Tovs ex THS vaupaylias éBovrAecGe aOpdous Kpivetv......TOT
Ey@ povos THY TpuvTavewy nvavTiwWOny viv K.T.A4. To the same
effect is the account given by Xenophon (Memorabilia, 1. 1. 18).
BovAevoas yap mote, nal roy Bovdeutixcy OpKoy opocas, ev @ HV

1 Mr Grote (IIist. of Greece, Vol. v. state or a magistrate. His position

p. 527, note) thinks it not absolutely
certain that Sucrates was Epistatcs, as
this fact is asserted only in one pas-
sage of Xenophon. But even apart
from this explicit statement, it is im-
plied in the other passages which men-
tion the occurrence, It is noticcable
that in Thucydides (vr. 14) Nicias ad-
dresses the Epistates as 3 ITpdran.
The E)istates possessed his whole au-
thority as Prytanis: i.¢. as chairman
he simply summed up in himsclf the
the collective authority of the board of
Prytuncs. Ho was not an officer of

might be compared with that of the
Speaker in the House of Commons, or
of the foreman of ajury. Hence it is
that Socrates is spoken of as refusing
to put the question rather in his cha-
racter of Prytanis than of Epistates.
But that he really was Epistates seems
certain, both from the passages already
quoted, and the following one from tho
Gorgias, 473 &, wépuot Boudederww Naxwy,
éwecdh 4 Gu\h ewpuraveve xai Ua pe
émipnpives, yé\wra wapeixory Kal ovx
yriorduny cCwipyel icy.

ON THE ATHENIAN PROEDRI. 173

cava Tos wonous Bovrevoesy, eriotarns ev TH Enum yevouevos,
érBuuncavtos tov Snpov Tapa Tovs vomous évvéa (sic) oTparn-
yas pid Yndw Tors audi OpacvAdov nal Epacwvidny atroxreivar
wurtas, ove nOéeAnoev eminpndicat «7.4. Is not the obvious
inference from these passages this: that, at the time when
Secrates was a member of the BovAy, the President in meetings
ci the Senate and the Assembly was one of the fifty prytancs of
the dud wpytavevovea, chosen daily by lot, and having the
power not only of putting but of refusing to put questions to
the tute (éxepnpifev) 2 But neither in Thucydides nor Plato
tor Xenophon is any mention made of mpoeSpot. And the same
thins may be said of the following passage of Antiphon (De
Clar. p. 146, 37) quoted by Schémann (De Comitzis, p. 93 note),
apitarevoas THY TPOTHY WpuTavelay Arracav Av Svoiv nuépary,
Kai icpoworay Kai Ovwv varép Tis ToAews Kal erubnditev Kal
Lejov wopas wept Tay peyloTtwv Kal TAciotou akiwy TH TOXeEL
Garepcs fv. As far then as the writings of that period have come
demu to us, it may be said with apparent certainty that no
tention 1s made of Proedri in any writer before the Archonship
‘f Encleides. The statement accordingly of the scholiast we first
(Wuted receives from this a partial confirmation and a partial
ucpreciation. He is certainly right in saying that the presid-
tury of the Senate and the Assembly was held by an Epi-
Mates appointed daily by lot from among the fifty members of
the spyravevouca gud. But of any subdivi ision of those fifty
Inty five batches of Procdri we find not a word in confirma-
tn Certain it is that all the notices of the Prytany arrange-
Ments before the Anarchy are perfectly intelligible without
‘Ppsme any such subdivision. Nay more, had so elabor-
ale an arrangement existed, we should have expected to have it
Hcntiuned by name, or at least implied, in connexion with so
luteresting a story as that of Sucrates’ presidency.

The opinion we have been maintaining does not however
depend merely upon this argutum silentium of conte mporary
"ters, but receives also remarkable confirmation from inscrip-
Un Before the Archonship of Eucleides the regular intro-
thetory furmula in Athenian yodiopara is as follows. First is
“iven the date, by mentioning the name of the Archon and

9 Cena, TpwTos Cypaypateve. This heading
iS Lea. Ver Inspucnt.y Wanting from mscriptions that have come
ewn tus Neat comes the decree itself, invariably mtroduced
by the frisua:s—"Ecogev 79 Bouin wat to Sriuw, “Epeybris
erpuTusere. Leovas €ypauparere, Tima@mdns exeoratet, Arozei-
Ors eves For examples of these formule one may refer to
Barcah. (acy. Inger. 74.51; Rangabé, Antiguités Helléniques,
Volo on Nos 240, 257. 2459, 2§3—4, 294; Franz, Hlementa
Eperph. Gears po 3k Te will be seen then that in insenp-
tens of this po riel the President of the Senate and Assembly
is always indicated by the phrase o éetva émreotate, 1.e. he is
That he was a member of the dud» awputave-
ovca We alrealy koow frum the passages cited above from
Thuevlides, Plate and Menophon. In the inscriptions this
latter point is net specified; the demotic name of the Epistates
not being subjoined. as we shall tind it to have been in the in-
scriptions of a later date. But the important thing to notice
is, that no mention is anywhere made of Proedri, nor anything
said to suggest the idea that the daily President was chosen out
of anv intermediate subdivision of the Senate other or less than
the fitty members of the tube then holding the prytany’.

If now we proceed to examine the Athenian wniters and
inscriptions of the period succeeding the Archonship of Eu-
eleides, an alteration will be observed to have taken place in
these arrangeinents, and a corresponding change in the formulas
and phrases relating thereto. It would be wasting time to
quote passages to prove how continually the word mpoedpos 1s
used in the Orators, Passages like the following occur almost
on every page :—Excdnaola ylyverat, ev f Anpoobévns Aayyaves
Tpucpevev (.Vschin, Fuls. Leg. 259). ’Avaoras €« tov tmrpoédpev

Caled éemiorarys.

I Sinee writing the above, I find that
ax early os 1843 C) FF. Hermann had
contended aeaingt the existence of Pro-
cdr before the Archonship of Evueleides.
(eprenteisde procdria apud Athenic nsea,
Gottingen, 1843.) The opinion is how-
ever worth reasserting now, from the

additional force lent to the argument
by the very numerous inscriptions dis-
covered and published within the last
twenty-five years, Schiimann ((riech-
iache Alterth. Vol. 1. p. 391) and Meier
(De Epistatis, p. iv.) also adopt the
fame conclusion,

—_ — oe ee ge ee ee --

ON THE ATHENIAN PROEDRI. 175

Anpocbevns ove Edm TO Wodiopa errupndueiv (tb. 260). Anpoo-
bens ev TH Siw mrpondpeve TovTou Tov pijvos EBSoun POivovros,
(1. 268.) Od apoedpevovres tis Bovdis Kal 6 tabt’ epindivev
émotamns (Dem. adv. Androt. 596). In these and similar
cases it is impossible not to identify these apcedpo with the
nine non-tribal proedri described by Pollux and the other
Grammarians, and the Epistates with the president daily chosen
by lot from among these nine. Still as Schimann says (De
Conitiis, p. 87 note) although these passages tend to such a
conclusion yet they cannot be said to prove the point distinctly.
What further proof however is wanting is abundantly supplied
by inscriptions. From them it may be demonstrated first of
all that the Epistates of the Proedri in the time of the Orators
¥as invariably of a different tribe from that which held the
prtany. The number of Athenian decrees which have come
down to us from the period between the Archonship of Eu-
cleides and the times of the Diadochi is so larye that ample
illustration is at hand. One need but refer to Beeckh’s Corpus
Iuscriptionum, Vol. 1. or to Rangabd’s Antiquités Helléniques,
Nos. 876—674. Thus for example the decree published in
Beckh, Corpus Inscr. 105, is headed as follows:—Emt Nuxo-
eipoy “Apxovtos, emi THs Kexporridos Exrns mputaveias, Tapnru-
Gos evSexaty, Extn Kai eixooTH Tis mpuTaveias, éxeAnola’ TAY
Tpocdpwr éemeyendilev’ "Apiotoxparns “Apiatodyuou Oivaios xat
Sunt poedpot, Opacvarns Navoietpatov Opiaccos elev, x.T.r.
(See the commentary of Backh ad loc). Here then the Epi-
States of the Proedri is of one of the Demos Odvon, and there-
fore of the tribe Hippothoontis or sEantis, whereas the @vA7
*puravevovca is Cecropis. The same thing is observable in the
other decrees of this date: in all, where the stone is suffi-
Clently entire for us to recover the heading, the Epistates is
found tu be of some tribe other than that which is holding the
Prytany. Further, after the mention of the Epistates there
fullows invariably the phrase xal cupmpoedpot, which M. Ran-

" The right reading of the stone here ed, and the testimony of Rangabé (.{n-
“3 in ¢. 1.97. 1.5) is the imperfect tiguités Helléniques) is to the same
mn Dut the aorist. The imperfect oc- effect.
275 in all inscriptions I have examin-

176 THE JOURNAL OF PHILOLOGY.

gabé rightly translates ‘un tel et ses collégues mettaient aux
voix. The Epistates and his brother Proedri are classed
together as on the same footing. It seem implied that as he
was one of the non-prytanising tribes, so also was each one
of them. In other words we have here an illustration of the
statements of Pollux and others, quoted above, respecting
the nine non-tribal Proedri. But we are not left to be con-
tent with this presumption alone. There happen to have come
down to us several inscriptions in which, by a slight ampli-
fication of the customary formula, after the words rav mpo-
épov ereynpilev 6 Setva xai cuptrpoedpor, there follows a list
of the other eight oupmpcedpor with their respective demotic
names appended. ‘This was first noticed by Bosckh im the
Corpus Inscr. No. 111, where he shews how the Epistates and
aupmpoedpot are each members of a different tribe, i.e. of each
of the tribes save uA) mputavevovoa. As that inscription 18
posterior to the institution of twelve tribes (B.C. 306), accord-
ingly the cusmpoedpor are ten in number, the Epistates making
an eleventh. It is sufficient to refer the reader to Boeckh’s
lucid commentary upon this document. We may with more
advantage turn to one or two other inscriptions bearing on the
point, which do not seem to have been sufficiently noticed.
Beeckh on Corpus Inscr. 111, says of the enumeration of the
cuptrpoedpot, ‘additi hoc loco erant reliqui proedri, quod nus-
quam alibi repperi.’ A parallel example has since been publish-
ed by Rangabéd, Antiquités Helléniques, No. 427. Unfortunately
the stone is much mutilated; but what remains is a valuable
illustration of our subject. Lines 10 to 14 contain only propet
names; and as M. Rangabé says, judging by the original length
of the lines, they cannot have contained more than ten name*
at the most. We shall see that as the inscription was anteriO!
to the establishment of the twelve tribes, no more than te?
names are wanted. M. Rangabé arranges them as follows:

1. 6 eriabndivon, (1. 9.)

2. Wholly lost, (1. 9—10.)

3. Qvpuoxapns Ne... (1. 10.)

, ere Kuda6nvacevs, (1. 11.)

Se 0), re (I. 11.)

178 THE JOURNAL OF PHILOLOGY.

identified ; viz.’Ayapvevs, PAvevs, "AvagdrAvorios, Demes belong-
ing to the tribes Aineis, Cecropis, Antiochis respectively. One
would feel tempted to consider this a list of ovpmapoedpos but
for the fact that the proper tribal order is not observed. If
however this be not an insuperable objection, and this be really
a list of apoedpor (the lacunas would just suit that number of
names), then they were certainly non-tribal.

It is therefore certain in every way that in the days of
Thucydides and Socrates the presidency of the Senate and As-
sembly was held by the fifty senators of the @uvA1 apuvTavevowwa
with their értorarns. It is equally certain that by the time of
Demosthenes and earlier this arrangement was altered, and that
the chairman of the Boule and Ecclesia was an Epistate chosen
by lot from among nine mpoedpot, themselves daily chosen by
lot from each of the non-prytanising tribes.

When did this change take place? This has been found @
hard question to answer. Schémann (De Comat. ch. vii.) assigns
the new arrangement to no particular date, but seems to think
the non-tribal Proedri whenever instituted did not till after B C-
307 acquire the privilege of putting questions to the vote-
Boeckh however (in Corp. Inscr. 90) points out that in Bc. 31
the érupndivwv is not a Prytanis (comp. Corp. Inscr. 105); nay
that in B.c. 332 the change must have already taken place»
from Aéschines’ words (In Ctes. 385), xat tavra Erepot tives T
Wndicpata émupndifovary, ovx ex tov Sixatordtov tpomov A—
xovres Tpocdpevery, GAN éx TrapacKeuns KabeCouevon. av Se TeS
TOV dAdwv BovrcvTay (not mpuTavewv) drvTws AaYN KANpolLEOS
mpoedperew, x.7.4. This description says Beeckh, suits not th&
Prytanes, but the nine Proedri “ qui—in unum comitiorum vel
senatus diem tumultuaria haud dubie sortitione constituuntur »
ut facile aliquis in hanc unius diei proedriam irrepere potuent—
Further Atschines, just after this (p. 387), distinguishes th ©

Proedri from the Prytanes; ov& of mpvutavets, ov of mpocdpo#—
Beeckh in accordance with Schémann (Ibid p. 92) thinks th<&
érupnditov in B.C. 347 is still a tribal Prytanis; he woul <4
place the change in question somewhere about Ol. 109—11 &
(B.C. 344—333). Schémann since then, in his Antiguttate?%
Juris Publici Grecorwm (p. 222, note 4) expresses himself as £3

180 THE JOURNAL OF PHILOLOGY.

there remain the letters ...... txdeidov. Now there are two
Archons only in the Fasti who fulfil this condition, viz. Phrasi-
cleides, B.C. 371, and Charicleides, B.c. 363. ML. Rangabé for
various good reasons prefers the later Archon. In either case
the date is not very far removed from the Archonship of Eu-
cleides, and it is noticeable that the formula bears a strong
resemblance to that which was said above to be usual in In-
scriptions anterior to Eucleides. But is the Epistates here
mentioned the president of the tribal Prytanes, though himself
of a different tribe, or is he the president of the nine non-tribal
Proedri? The former is the view of Meier. With deference to
his authority I think it untenable.

I have collected some examples from Rangabé, Antig. Hellé
of the use of the old and the new formulas in decrees as close as
possible to the Archonsbip of Eucleides, The later formals
occurs in the following:

No. 393......... date B.C. 357
00 OTT cccecveceeeceereees 392
wo BIG ...ceccceececsreees 399 (probable date).
We find the older formula retained in the following :
No. 381 (page 3735)......... B.C. 378
re 2] eee re 377
BSS .....cecececesceceeecceeeees 372
BBG ....ccccceceescccecceareecens 371 or 363
186 ....ceccseccccecccccecescsecs 362
eS eee 349 (probable date).
AOL oo. ce ecececceeneeeeees 347.

It appears from this list that the old formula was not, as
Meier’s view would imply, at a precise date superseded by the
newer one: but on the contrary that for some time the two
were used indiscriminately. So that the date of the change
from Prytanes to Proedri must be fixed from other indications
than the employment of these formulas.

C. F. Hermann, in his Political Antiquities of the Greeks,
(127. 9) makes the Archonship of Eucleides the date of this
transition, This opinion receives strong confirmation from

Inscriptions. No. 376 of Rangabé’s Antig. Hellén, where the

182 THE JOURNAL OF PHILOLOGY.

this time those two united into one board of ten men, one from
each tribe, (see Bockh on Corp. Inscr. 150, and Kirchhoff a
the tayias rev adrwy cov in the Abhandlungen of the Berlin
Academy, 1861). In a similar way, we may well conceite, it
was thought that the fifty tribal Prytanes were at once tw
unwieldy and tvo exclusive a board to suit the needs of th
occasion. A new board was required which should poses
greater pliancy and activity by virtue of being smailer in tk
number of its members, and which by having a more represet
tative character should command a more complete deference.
Thus a board of nine from different tribes took the place of th
fifty tribal Prytanes.

It appears that the old board of Prytanes with their Bp
states continued still to exist, although its more important
function of controlling the debates of Senate and Assembly ws
made over to its successor. It still retained its more forml
functions, such as calling an assembly of the Ecclesia, and cor
ducting the ballot for the Proedri. Aristotle is quoted by Har
pocration (sv. 'Emvrrarns) as distinctly asserting the existence
of two officers of state bearing at the same moment the title of
Epistates. Avo eiciv of xaStorayevot emiotatas 6 Mev Ex Tp
Tavéwy KAnpoupevos. o O€ ex THY Tpoédpwy. wy ExaTEpos TW
Stoixnow Scouxet, SednrAwxev o "AptororérAns ev tH "AOnvaiwr
montTea. We may well regret, with Meier, that Harpocration
did not deem it necessary to add what Aristotle said of the
respective functions of the two. Still more must we lament
the loss of Aristotle’s Tlodsreias, a treatise whose preservation
would doubtless have rendered the discussion we have been
engaged in wholly unnecessary.

E. L. HICKS.

184 THE JOURNAL OF PHILOLOGY.

leane and Mr Pretor. Hunc ipsum librum means the present
volume, as it does in another passage, editum librum continuo
mirari homines et diripere ceperunt, and also in the well-known
quotations, Multum et vere gloriz, quamvis uno libro, Persius
meruit, Quint. x. 1, 94, and

Seepius in libro memoratur Persius uno

Quam levis in tota Marsus Amazonide,

Mart. tv. 29, 7. It would then be forced, to make ultimo
libro in the very next sentence signify the sixth Satire, as
opposed to the completed whole. As to the words of the bio-
grapher, it is by no means certain that corrextt is the true
reading. MSS. P.1, and 2, and M. 2, read recitart ; L. con-
tractarit ; and W. contrazit. If we suppose, with Jahn and
Hermann, that retractavit was the true reading, re...ctavit for
retractavit would give recitavit, and the variant contractavit
contraxit, and finally correxit. The words leviter retractavit
would then mean that Cornutus gave a few finishing touches
to the completed whole. And Jahn, pref. p. 45, shows that
retractare was the term for final revision, even after publication.
The Latin of the biographer, versus dempti sunt ultimo libro,
taken by itself, is more in favour of Jahn and Mr Macleane than
of Mr Pretor, for aliqui versus would more naturally mean ®
substantive fragment than detached lines, while it must be
allowed that quasi finiturus esset at first sight favours Mr Mac—
leane. But if we take finiturus absolutely, as in Ovid, Aré-
1. 755, finiturus eram, they will mean, “as if Persius intendect
to conclude with the sixth,” quasi, as usual, denoting that, ==
reality, Persius did not conclude as we have him.

II. Asto the Satire itself: the connexion is as follows:——
You are enjoying your retreat, and so am I,1—11. People hr©&
different ways of treating their income: there are misers in th©
world, but I am not one, 11—20. I, for my part, mean to enjoy
my wealth, without being a prodigal, or a gourmand, 20—2@-
Take my advice Live up, you, to your income, and don’t sparev€:
another crop is on the way. Besides the consoling reflexion of
my favourite,

Cuncta manus avidas fugiunt heredis, amico
Que dederis animo—

186 THE JOURNAL OF PHILOLOGY.

So in Juv. x. 78 sq.
Ex quo suffragia nulli
Vendimus, effudit curas :
and
Exinde per amplum
Mittumur Elysium...... ut convexa revisant
Rursus et incipiant in corpora velle revert1
Zin, V1. 743 8q.
and Eur. Hipp. 1328 sq.
Beoias 5 aS Exes vopos’
ovdeis atravray Bovreras mpobuula
TH Tov BéXovtes GAN aduotapeO ael.
As to Vende animam lucro, the imperative’ denotes hypo-
thesis, as we may see by comparing Propertius, lI. 25, 36,
Jam bibe: formosa es: nil tibi vina nocent,
with simul obligasti
Perfidum votis caput enitescis
Pulchrior multo.

In the former, we might, metre apart, say si bibis, and in
the latter obliga perfidum caput, so that the passage in
Persius is really equivalent to etiamsi vendas animam lucro,
depungam ubi sistas, quum finitor infiniti imventus fuero.
As to the re in repone, it seems to have its proper force: it is
frequently used to signify, not repetition, but mere relation.
Here a person volunteers advice, which seems to imply that a
want exists on the part of the involuntary client. Fe, thus
would signify, don't apply to my case the precept stored within
your breast; Repono is used in Prop. Iv. 4. 37, with a similar
sense of re,

Tile equus, ille meos in castra reponet amores
Cui Tatius dextras collocat ipse jubas—
t.¢. The horse of my lover will transfer from hence my love
into his camp. So in Horace, classe cita reparavit oras, t.¢. in

1 When writing the above, I had not Mayor quotes several instances of the
~ seen Mr Mayor's Note, Juv. 1, 155, hypothetio imperative.
pone Tigellinum, lucebis. p. 161. Mr

THE SIXTH SATIRE OF PERSIUS. 187

place of what she had. This force of re is very little attended to
although it has been pointed out by Wagner: nempe re—in
qubusdam verbis compositis significat rei alicujus in contrarium
mutationem : tn contrarium, however, seems too strong; change
@f sequence, as opposed to simple repetition, is what is here
contended for: and in the passage in Persius, the contrast would
be between the state of the precept in the breast of the adviser,
and its successive state when applied to the case of the client.
So in Virgil's sua nunc promissa reposci, the re denotes the
elation between the promise given and the promise performed.
The Bestius of the piece is the representative of good old
Roman notions: keep up the House, and don’t mind what these
Stoic fellows tell us, that all men are brethren under the Law of
Nature. Parsimony was a Roman Virtue: Roman boys learnt
atthe same time their money-tables and the value of money:
Hor. A.'P. 325—330: the Roman Paterfamilias was expected to
keep accounts with his own hand: and the regard of the Roman
for the pecuniary honour of his House is shewn by the common
practice of instituting a slave, as his heir, when his circumstances
were embarrassed. Bestius, then, represents the Roman view, as
opposed to the calls of charity, officitum, the 16 xaOjxoy of the
great ethical school to which Persius belonged.

THOMAS MAGUIRE.

ADDENDUM.

Mr Pretor, in a private letter, objects to laying stress on the
juxtaposition of ibrum and libro:

“T cannot help thinking that (contrary to your argument)
the juxtaposition of librum...libro rather makes it likely that
they are used loosely—the first of the entire Satires, the latter
of the last alone. I cannot think he would say ‘this last book
he left unfinished (sc. the entire Satires) from the last book
(meaning the same) certain verses were withdrawn.’”

But, surely, the Biographer uses Hunc tpsum librum not, in

188 THE JOURNAL OF PHILOLOGY.

opposition to ultimo hbro, but in illustration of bis remark, e
raro et tarde scripsitt. I translate: this very volume he left
unfinished. Some verses were removed from the end of the
volume. Jtbro and labrum thus denote the same thing—the
present edition, as edited by Cornutus. Ultimus is thus used
in Cic. Att. v. 16, 4 and in Ter. Heaut. v. 1, 29.

THEBAN INSCRIPTION AT THE FOUNTAIN OF
DIRCE.

Towarps the end of May 1864, I made a transcript of an
inscription upon a stone which is built into the wall above
the fountain of Dirce at Thebes. It was not till last summer
that I looked into Bockh’s Corp. Inscr. Gr. in order to see
how it was given. I had taken it for granted that an inscription
in such a prominent position and so legible must certainly
have been copied, and correctly, and that my own copying had
been mere waste of time, except so far as it had given me
some amusement.

The letters, though a good deal rubbed, were quite legible
all through, with the exception of a very few: nowhere were
there gape of more than one or two letters together except in
the first line, where the last two feet of the verse were missing,
the stone being quite chipped away in that place. I identified
the lines I had written out with No 1654 in Beeckh, but was
astonished to find that the inscription as there given was
hopelessly corrupt. I then went to look for my own copy, but
could not find it, and have been unable to do so since. As
I fear the sketch-book in which I had written down the lines
is lost beyond the probability of recovery, I think I may fairly
ask forgiveness if I give my version of the inscription from
memory. My recollection of it is I believe quite clear except
with respect to lines 5 and 6, in which I can only recall a
couple of stray words.

The extreme inaccuracy of Pococke’s transcript in the ex-
ample before us may well make us suspicious of him in other
cases. We have here a good instance of the way in which

190

THE JOURNAL OF PHILOLOGY.

patience and ingenuity may be thrown away by scholars in an
attempt to emend passages which have been converted into
absolute nonsense by the carelessness or ignorance of tran-

scribers.

I will give the inscription and commentary as they appear
in Boeckh, and then my own recollection of the thing. The

omega in the original has the form of our W.

Boeckh, Corp. Insc. Gr.
1654.

Thebis. ed. Pocockius Insc. ant. P. I. c. 5. 8. 3. p. 50.

OTOZQKAPTH2ZFOPTIMOZIONO....

TAI... TOTHZAPETH2ZEZ0OKOAHNIOPOE

OMNH2HIZENETIKKENH
HYKOMOZOENIHPEPIAIQ

6 EYNOH2ZINQHTPATIKAZTEIMHIOIZTE ...

KAIITTAEMITANYTIAATEIN
ANAAENO...EMMIAKEK...OIPAAOBEIA

Holz

HPQAHBAIAENTTIAIAEZAEZ=OMENOY

TOYTOYTOYPIAIHAAOIO.. AHAEIAY!
10

ESTHZEYIMPAEIZ . KYAOZEHMATPIAt

OYTATPIZKATAFAIANOZAIZAIOEIEQF ENO
MEIZONATHAETTOAEIKOZMONEOHKE!I

8 peri

[M]oipa S [w]xeia?
ripe. .ev mard[o]s? [aleEoudvou ?

Vs. 3. Sanderus
conjecit wavyrolys dpe-
rijs €foxo...lacunam a
Pocock. nimiam sig-
nari censens.

Vs. 9. AHA mihi
est ANA (xedva p.
xééy', ut solet in in-
scriptionibus) San-

erus conjecit [x}jSer
[A] of ye¢.]

horam distichorum
on attingo.

tovtou [x ]oupidin[a]Ao[x]o[s]...[«é]5[v"] ebui[a. Eotpate est nomen

éorne Bumpalt list xvdos é9 matpib..

uliebre (n. 709,

| 1161).

[olu yalp tis xara yaiav Scars Acc[O]e[v] yévos [eoriy,

petLova rade Trove Koopoy EOnKe [yuvn.

THEBAN INSCRIPTION. IQI

OYTOZZWKAPTH2IOPTYNIOZ.........
TIANTOIH2APETH2ZE=OXOZHNIOXOS

ONMHTHPMENETIKTENENIKPHTHEYPEIH
HYKOMOZZOENIHDEPTATONAYZONIWN

bd bd s bd # s
bd bd s # s bd

AAMWAENOY2OZEMAPYEKAKHKAIMOIPABAPEIA
HPWATIPINIAEINTTAIAAZAEZOMENOYS
TOYTOYKOYPIAIH AAOXOZKEANEPFEIAYIA
EZ THEEYTTIPASIZKYAOZEHTIATPIAI
OYFAPTIZKATALAIANOZAIZAIOOENFENOZESTI
MEIZONATHAETTOAEIKOZMONEOHKELYNH.
Otros Lwxaprns Toprvmos......
wavtoins aperis EEoxos Nvioyos.
Sy pntip pev Erixrevy evi Kprjrn evpein
nixopos, oOevin héptarov Avooviwv.
Kacvyyntous te [xpata]ious
xai [waytwy] vratwy [xpéccovas 7ryepnovas. |
GrAd € voicos Evapye xaxn nai Moipa Bapeia
jpwa, wpiy dew traidas acEopévovs.
Totrou Kxouptdin adoyos Kxéby’ Epy’ eidvia
é€atno Evarpakis xidos é9 marpid:.
ov yap Tis Kara yaiav Coats AcdOev yévos earl
petlova THd€ trodes Koopov EOnKxe yurn.

In line 5, KAZTEIMHIOIS TE is xacvyyjtous te, and
I think the following word is xpataious.

In line 6 I can remember the word vratwy which IJ take
to be represented in Beckh by TITAATEIN, the first word
is of course «ai. I believe the line is xat mavtwy imratwv
apéacovas ryyepovas, but I cannot quite trust my memory in
this particular case.

I cannot restore line 5, as Pococke’s reading only helps to
perplex me, and I have forgotten the clue which would enable
me to correct it.

W. E. CURREY.

THE JOURNAL

OF

PHILOLOGY.

THE EPISTLE TO THE ROMANS.

Ix the last number of this Journal (im. p. 51 sq.) Mr Hort
criticised and condemned a theory which I had suggested in the
preceding number (It. p. 264 8sq.) to account for certain facts
connected with the text of the Epistle to the Romans. The
facts, it will be remembered, were mainly these; (1) One or
more ancient writers used a copy of the Epistle containing only
the first fourteen chapters, with or without the doxology which
in the common text stands at the close of the whole (xvi. 25
—27;. (2) In the existing copies this doxology appears some-
times at the end of the xivth chapter, sometimes at the end of
the xvith, sometimes in both places, while in some few in-
stances it is omitted altogether. (3) At least one text omits
ev ‘Popy ini. 7,15. The theory, by which I sought to com-
bine and explain these facts, was this; that St Paul at a later
period of his life reissued the Epistle in a shorter form with a
view to general circulation, omitting the last two chapters,
obliterating the mention of Rome in the first chapter, and
adding the doxology, which was no part of the original Epistle.
Mr Hurt impugns some of these assumed facts and explains
away others. Having done this, he attacks the theory itself,
and endeavours to show that it is untenable.

No one, who is really anxious to ascertain the truth, would
object to such a criticism as Mr Hort’s, even though it should

Journal of Philology. vou, 111. 13

194 THE JOURNAL OF PHILOLOGY.

lead to the rejection of a darling theory. I am_ especiall
obliged to him for the thoroughness with which he has applie
the test of textual criticism to my hypothesis. And, if I veo
ture, notwithstanding his arguments, to maintain that the facta
themselves are stubborn and in some respects even stronger
than I had supposed, and to uphold my theory as the mos
probable explanation of the facts, until a better is suggested, I
trust that I am not blinded by partiality. At all events I will
give my reasons as briefly as possible, taking the facts firt
and then proceeding to the theory.

I. The first and most important of the facts is the ex
istence, in early times, of copies containing only fourteen
chapters. Of this the indications are various, and (as it seems
to me) conclusive.

(i) The statement of Origen respecting Marcion has bees
‘universally understood,’ as Mr Hort himself allows (p. 60), #
mean that this heretic struck out not only the paragraph co-
taining the doxology, but the two last chapters also; ‘Caput
hoc [t.e. the paragraph containing the doxology] Marcion, §
quo Scripture evangelice atque apostolic interpolate sunt
de hac epistola penitus abstulit; et non solum hoc, sed
ab eo loco ubi scriptum est Omne autem quod non ex fae
peccatum est (xiv. 23) ad finem cuncta dissecuit. In ali
vero exemplaribus, id est, in his que non sunt a Marcio?
temerata, hoc ipsum caput diverse positum invenimus. A
universal understanding may be wrong, but most frequently
is correct; and I cannot doubt that this is the case here. B
Hort however adopts a reading of a Paris MS (Reg. 163
which has ‘zn eo loco’ for ‘ab eo loco,’ and himself alters ‘hé
into ‘hic.’ Thus he makes Origen say that Marcion cut o
the doxology, not only at the end of the xivth chapter, b
also at the end of the Epistle. Now my reply to this is thre
fold; (1) Though we may allow the general value of the rea
ings in this MS, whose date however is not earlier than abo
the 12th century, yet its text is far from faultless, so that on
a slight presumption is raised in favour of a reading from -tl
fact of its being found there. In the present instance howev

THE EPISTLE TO THE ROMANS. 195

the reading ‘in eo loco’ has no meaning, unless with Mr Hort
we likewise change hoc into Aic—an alteration for which there
48 to MS authority. (2) Mr Hort’s reading and interpretation
destroy the force of individual expressions in the context.
“ Unue ad finem cuncta dissecuit’ is natural enough when
applied to two whole chapters, but not to the doxology
alone; and again in ‘hoc ipsum caput’ the ipsum becomes
Meaningless, unless it is contrasted with some other portion.
If the words be taken as they stand and interpreted in the
Ordinary way, the sequence commends itself; ‘Caput hoc...non
Solum hoc sed...usque ad finem cuncta.. hoc ipsum caput’; but
it is entirely broken up if they are read and explained as
Mr Hort wishes. (3) One who reads continuously not only
the passage quoted above, but the whole paragraph of Origen as
Ziven by Mr Hort (1 p. 59) or by myself (11. p. 265), will
hardly fail, I think, to see how Mr Hort’s interpretation
Zuvolves and confuses the natural order of the topics.
When again Mr Hort supposes the statement of Jerome
(on Ephes. iii. 5), that the doxology was found in plerisque codi-
eibus, to have been derived from Origen’s commentary on the
same Epistle, I allow that this supposition is probable. But
I do not see that Mr Hort’s view gains strength thereby. Com-
menting on Ephes. iii. 5, Origen would be concerned only with
the doxology in which ‘the mystery’ is mentioned, and he would
be going out of his way, if he said anything about the omission
of the xvth and xvith chapters, with which he was not in any
way concerned. Moreover it must be observed that, when
there is a question of a various reading, Jerome sometimes
manipulates Origen’s statements in such a manncr as entirely
to disfigure their meaning. Such is the case for instance with
the opening verse of this very Epistle to the Ephesians, where
Origen, having before him a text which omitted év 'Ed¢éca, in-
terprets tois ovoty in an entirely lucid though highly artificial
way, but Jerome, repeating his great predecessor's comment,
holds language which can hardly be called intelligible.

As regards the statement of Tertullian, when arguing
against Marcion (Vv. 14), that the threat of the tribunal Christi
(Rom. xiv. 10) occurs in clausula of the Epistle, I agree with

13—2

196 THE JOURNAL OF PHILOLOGY.

Mr Hort that the inference which supposes Tertullian to refer
to a copy of the Epistle wanting the xvth and xvith chapters,
though ‘natural, is not ‘conclusive. Let the fact that the
inference is natural have no more than its proper weight.
I should not have laid much stress on the expression, if it had

stood alone; but in connexion with Origen’s account of Mar.

cion it cannot be overlooked.

(ii) For the negative argument that the last two chapters
are nowhere quoted by certain early writers I claim a suppk-
mental value. More than this it does not deserve. The fact
however remains that neither Irenzeus nor Tertullian nor
Cyprian (except in a very doubtful allusion) refers to them I
will only add that this omission occurs in Western writer’,
whereas they are more than once quoted by Clement and
Origen. The importance of this fact will appear hereafter.

(iii) I owe it to Mr Hort’s candour that my attention was
directed to the capitulations of the Latin Bibles, and the eu-
dence derived thence seems to me to strengthen my case enor-
mously. In my former article I had referred to Wetsten’s
note: ‘Codex Latinus habet capitula Epistole ad Romanos 51,
desinit autem in caput xiv; ex quo conficitur ista capituls ad
editionem Marcionis fuisse accommodata’; and, misled with others
by his careless expression desinit (where desinunt would have beex2
clearer), I had naturally supposed that the MS itself, to whicks
he refers, ended with the xivth chapter, and accordingly re—-
marked that ‘later critics had not been able to identify the MS
and thus verify the statement.’ I have no doubt however that
Mr Hort is right, and that Wetstein refers to such a phenome-
non as the Codex Amiatinus exhibits, where (though the
Epistle itself is complete) the capitulations end with the end
of the xivth chapter, there or thereabouts. I have since bee?
investigating the subject’; and the results of this investigation

1 The first distinct quotation by any
Western writer, so far as I can discover,
occurs in Victorinus c. Arium iii. p.
280 o, a treatise written about a. p.
865—where xvi. 20 is quoted. Even
Hilary of Poitiers (if the index may be
trusted) cites nothing from these two

chapters but the doxology. The ‘vey
doubtful reference’ in Cyprian is gives
by Mr Hort, p. 65, note 2.

* After I saw Mr Hort’s article i
type, I began to look into the matte;
and, before it was finally struck off, |
mentioned the remarkable phenomemn

THE EPISTLE TO THE ROMANS. 197

seem to be sufficiently important to justify my taking up a few
pages in recording them.

In fact, there is evidence of two distinct capitilations—both
ending with the xivth chapter—the first very widely spread,
tthe second only preserved in a single though very early MS.

Of the first of these, the Codex Amiatinus affords the oldest

and best example. In this MS the table of contents prefixed
to the Epistle gives 51 sections, the 50th section being described
“De periculo contristante fratrem suum esca sua, et quod non
sit regnum Dei esca et potus sed justitia et pax et gaudium in
Spiritu Sancto,’ and the 5lst and last ‘De mysterio domini
ante passionem in silentio habito, post passionem vero ipsius
revelato.’ Corresponding to these, the sections are marked in
the text, and agree with the descriptions in the table of con-
tents as far as the 50th. The 50th is marked as beginning at
xiv. 15, and here again the description is accurate; but the
5lst commences with xv. 4, and has no connexion with the
description. The description of the 51st in fact corresponds
to the doxology (xvi. 25—-27), and to nothing else in the re-
mainder of the Epistle. The natural inference therefore is,
that the capitulation was made for a copy of the Epistle,
containing only fourteen chapters and the doxology; and that
the scribe who first adapted it to a full copy with the sixteen
chapters, not finding anything corresponding to the 51st section
in the immediate context, extended the 50th section as far as
the subject allowed him and made the 51st section include
all the remainder of the Epistle. This solution, which Mr Hort
allows to be certainly possible, seems to me to commend itself
as in the highest degree probable.

This capitulation appears to have prevailed very widely.
It is found in not less than seven MSS enumerated by Card.
Tommasi (Thomasii Op. 1 p. 388 sq. ed. Vezzosi), and dating
from the age of Charles the Great downwards. It occurs again
in the British Museum MS Add. 10,546, an Alcuinian copy,
generally called ‘Charlemagne’s Bible,’ but really written in
one of the succeeding reigns; in the important MS Harl. 1772

of the capitulations in the Codex Ful- in a note appended to his article
densis. To this conversation he refers (p. 80).°

198 THE JOURNAL OF PHILOLOGY.

belonging to the 8th century; in the Oxford Bodleian MS
Laud. Lat. 108 (E. 67) of the 9th century (in which however
the number is expanded from 51 to 67 by a subdivision of one
or more of the earlier sections); in the MS B. 5. 2 of Trin
Coll. Cambridge, belonging to the 11th or 12th century’; and in
the Cambridge University MS Ee. 1. 9 written apparently late
in the 13th century*. In Add. 10,546 the sections correspond
in number and position with those of the Amiatinus, but the
words are occasionally varied, e.g. de non contristando fratre for
de periculo contristante fratrem suum. In Harl. 1772 the
number of sections in the table of contents is reduced to 49
by combining §§ 43, 44, 45 in one section, while (except unin-
portant various readings) the words of the Amiatinus ar
strictly followed. In the text however the whole 51 sections
are marked ; of these the first 49 correspond to those of the
Amiatinus, but the 50th commences not with the beginning
of xiv. 15 Si enim propter, but with the middle Noli cabo
(while on the margin in a later hand stands xlviiij opposite
Si enim propter), and the 5lst not with xv. 4 Quacumque
enim, but with the middle of xiv. 22 Beatus qui (the Q of
Quoscumque being however illuminated). And again in Cambr.
Univ. Ee. 1. 9, where the number of sections is simularly re-
duced to 50, the beginning of the 50th and last section ‘de
mysterio etc.’ stands at xv. 1 Debemus autem nos, i.e. at the
precise point where it would have stood, if the MS had cor-
tained only the doxology after the xivth chapter. These
variations show the difficulty which was felt in adapting the
end of the imperfect capitulation to the complete Epistle: and
they answer any objection founded on the fact that in the
Amiatinus itself the last section does not commence at the
exact place in the text which the hypothesis seems to require.
In more than one MS however, which I have examined,

1 In the older Trin. Coll. MS of 2 In the Cambr. Univ. MS Fr. 4. 40,
St Paul’s Epistles B. 10.5, of the 9th which came from the Library of Christ
century, the Epistle to the Romans Church Canterbury and was wnitta
and part of the First tothe Corinthians probably early in the 18th century,
are wanting. The Amiatinian capitu- though the Amiatinian capitulations
lations are given for the other Epi- are not given, I find this note ‘Hee
atles. epistola capitula li dicitur habuisee.’

200 THE JOURNAL OF PHILOLOGY.

MS B. 5.1 of Trin. Coll, Cambridge, belonging to the 12th
century. Here the scribe has retained all the Amiatinian sec-
tions, including the doxology; but by combining two in the
earlier part, he reduces them to 50 in number. Thus the 49th
is ‘de non contristando fratrem, etc.’, and the 50th ‘de mysterio
domini, etc. To these he adds two new sections, which are
the same as those described in the last MS:

li obsecratio pauli ad dominum, ete.

lii salutatio pauli ad fratres.
In the text the 49th section begins at xiv. 50, the 50th at xv. 4,
the 51st at xv. 30, and the 52nd at xvi. 1. The inequality of
scale in these superadded sections shows that they did not pro-
ceed from the same hand as the rest’.

These facts have been elicited by an examination of such
MSS as came conveniently within my reach*. Doubtless a
wider investigation would produce more striking results. But
I have seen enough to convince me that the Amiatinian capitu-
lation, though originally framed, as will be seen hereafter, for a
short copy of the Old Latin, yet maintained its ground as a
common mode of dividing the Epistle, until it was at length
superseded by the present division into 16 chapters in the latter
half of the 13th century.

The second capitulation, of which I spoke, is found in the
Codex Fuldensis which, like the Amiatinus, was written about
the middle of the 6th century. The sections in the text cor-
respond exactly with the Amiatinian. Not so in the table of
contents. Of the latter Ranke remarks (Codex Fuldensis, p.
xxill, 1868): ‘Que epistole ad Romanos preemissa sunt capitula
duabus in partibus constant, quarum altera (i—xxiii), tottus

1 The relation between the two MSS
last described is curious. For, while
other indications would suggest that
the capitulations of Brit. Mus. Reg. 1.
E. viii. were derived from those of
Trin. B. 5. 1, the former presents the
older form of the Amiatinian 50th seo-
tion ‘de periculo contristante fratrem,’
while the latter substitutes the amend-
ed form ‘de non contristando fratrem,’

which perhaps appears first in the Al-
cuinian copies.

4 My examination has not extended
beyond the British Museum MSS to
the 11th century (inclusive), and the
MSS in the Cambridge University and
Trinity College Libraries. The infor-
mation respecting Bodl. Laud. Lat.
108 I owe to Mr Coxe, the Librarian.

—_— am ia be

202 THE JOURNAL OF PHILOLOGY.

wish to deny that there is force in this argument; which never-
theless does not seem to me conclusive. The strongest expres-
sions in this direction are ‘pro fide romanorum...deo apostolus
gratias agit ut probetur fidem in deum muneris est divint,’ and
‘in Christo Jesu qui solus sic humana [humanam] naturam re-
cepit ut eum contagia veterts origints non tenerent.’ The
African fathers were more or less Augustinian before Augus-
tine’s time, and (so far as I can see) might have held such
language’.

On any showing however the Latin Bibles bear strong testi-
mony to the existence of the shorter forms of this Epistle at an
early date. The alternative hypothesis, that these sections were
determined by the lessons read in Churches, is devoid alike of
evidence and of probability. With this single exception, the
Amiatinian capitulation in the New Testament includes, I be-
lieve, the entire book in every case. It does not bear the
slightest trace of being intended for lectionary purposes. Nor
indeed is there any reason why the 15th chapter should be
excluded from the lessons; for it is much more fit for public
reading than many sections elsewhere, which are retained.
Even the 16th chapter would be treated with exceptional rigour
on this showing, for in other epistles the paragraphs containing
the salutations are religiously recorded in the capitulation.
Moreover, the oldest evidence which we possess on the subject
exhibits lessons for Sundays and Festivals taken from the 15th
chapter; and if so, a fortiors it would not be neglected in the
daily lessons, supposing (which seems improbable) that daily
lessons had been instituted at the time when this capitulation
was made.

When my attention was first directed to the Amiatinian
capitulation, I naturally inferred that it had belonged originally
to the Old Latin and was later adapted to the Vulgate. <A fur-
ther examination has shown this inference to be correct. The

1 e.g. Cyprian Ep. 64, says ‘Secun- Christian Dogmas, 1. p. 185 sq. (Eng.
dum Adam carnaliter natus,contagium Trans). Augustine's own dogmatic
mortis antique prima nativitate con- views on these points were enunciated
traxit.’ Compare also Tertull. de Anim, before Pelagius took up the subject: ib.
40, 41; and see Neander Hist. of p. 847 sq.

204 THE JOURNAL OF PHILOLOGY.

Now, waiving for the present the consideration of its origi-
nal position, I wish to point out two great improbabilities
involved in the other assumptions in this sequence. First.
There is no such obvious connexion between the paragraph at
the end of chapter xiv and the doxology, as should lead to
their being connected together’, if separated in their original
position by two whole chapters, while on the other hand these
intervening chapters present material for more than one excel-
lent lesson. Bengel indeed suggests, as Mr Hort points out,
that the severa sententia apaptia éorly, with which chapter
xiv closes, would be deemed unfit for the end of a lesson and
that this inauspicious termination was got rid of by tacking on
the doxology. But how much more easily would the difficulty
have been overcome by continuing the lesson a little further and
closing with the 2nd or 4th or 6th verse of the next chapter. The
instance which Mr Hort quotes (p. 72, note 1), Acts vi. 8—vii.
2 combined with vii. 51—viii. 4, as a lesson for St Stephen’s
day, will hardly bear out his hypothesis, for there the combina-
tion is naturally suggested by the subject. Secondly. This solu-
tion requires us to believe that all the three steps numbered
(2), (3), (4), had taken place before Origen’s time, so that he can
speak of some MSS as having the doxology in the one place
and some in the other, without suspecting how the variation
had come to pass. This supposes such an early development
of the lectionary as (I believe) there is no ground for assuming.

III. Lastly there are the phenomena in the first chapter to
be considered. Here the important fact is, that in one extant
MS (G) certainly, and in another (F) probably, the mention
of Rome has been obliterated in two distinct passages. In i. 7
Mr Hort explains the omission by the fact that ‘a Western cor-
rection substitutes éy ayamn @eod for ayamnrois @eod,’ so that
the words would run enpwmHenarann, where the repetition of éy

1 In a note (p. 71) Mr Hort remarks Constantinople and from which the

that ‘the Synaxaria, valeant quantum,
give Rom. xiv. 19—23, plus the doxology
as the lesson’ for the Saturday before
Quinquagesima. But since the doxo-
logy occurs here in the vulgar Greek
text which prevailed at Antioch and

Synaxaria are taken, they would na-
turally read it here. I would add that
the Synaxaria (see Scrivener’s Intro-
duction, p. 68 sq.) present no parallel
to the omission of two whole chap-
ters.

206 THE JOURNAL OF PHILOLOGY.

It will thus be seen that Mr Hort denies some of my facts,
and impugns the significance of others. As the facts give him
no trouble, it follows that the hypothesis, which has no other
ratson d'étre but to explain them, should not find favour with
him. But,if (as I think I have shown) the facts are even more
cogent than they appeared at first, being reinforced by the
Latin capitulations, an explanation is still demanded. I cannot
indeed say that my hypothesis is free from objections. But
a priors improbabilities could be detected by the keen eye
of criticism in the most certain events of history; and a theory,
which is based on circumstantial evidence, cannot hope to
escape objection on this ground. But, if no other hypothesis
has been offered which does not involve more or greater im-
probabilities, and if some hypothesis is needed to account for
the facts, I must still venture to claim a hearing for my own.

In Mr Hort’s criticism of the theory itself, as distinct from
the facts which evoked it, there are three points especially
which call for a reply.

(1) I had assigned the doxology (xvi. 25—27) to the
shorter recension of the Epistle, which I supposed to have been
issued by St Paul himself at a later date, and had produced
parallels to show that its style very closely resembles that of the
Apostle’s later Epistles. Mr Hort himself considers it to have
been the termination of the original Epistle. His argument is
threefold: (a) that it is appropriate ; (5) that St Paul at the time
entertained the idcas contained in it; (c) that it presents num-
berless close parallels of expression to the earlier Epistles.

(a) As regards its appropriateness, I entirely agree with him.
I cannot indeed assent to Baur’s opinion which he adopts, that
the main drift of the Epistle is revealed in chapters ix—xi.
The central idea, as I conceive it, is the comprehensive offer of
righteousness to Jews and Gentiles impartially, following on the
comprehensive failure of both alike before Christ’s coming.
After this idea has been developed, the objection arises that,
however comprehensive may be the offer, the acceptance at all
events is partial and one-sided ; that while the Gentiles seem
gladly to accept it, the Jews stand aloof; and that thus the
promises of the Old Testament appear to be nullified, and indeed

208 THE JOURNAL OF PHILOLOGY.

in the same paragraph (i. 2). On my hypothesis the opening
portion was read over and altered, when some years later the
Epistle was issued by the Apostle in this second and shorter
form ; and it was therefore natural that the thanksgiving which
was then appended, should embody not only thoughts but also
expressions taken from the commencement, thus binding toge-
ther the beginning and the end of the Epistle.

(ii) The character and condition of the text of the twin
MSS, F and G, is one of the points on which Mr Hort lays
most stress; and certainly, if his account of my theory were
correct, I should find it difficult to answer him. Expressing
my hypothesis in his own words, he represents me us holding
(1) that ‘the scribe of G copied i—xiv from one MS and xv,
xvi, from another,’ and (2) that ‘the scribe of F copied in like
manner from the same two MSS, though he left no mark of the
transition from the one to the other’ (p. 68). He then remarks
that ‘If the first of these hypotheses were true we ought surely
to find some evidence of it in the respective texts; whereas
the closest study fails to detect a shadow of difference in the
character of the readings before and after the blank space’;
and that ‘when F is taken into account, fresh embarrassments
arise. But I did not for a moment contemplate the scribes of
F and G each of them copying directly from these two MSS,
containing respectively the shorter and the longer recension of
the Epistle. I was well aware that the phenomena of these
MSS would not admit of such a supposition. And I venture
also to think that my language, which Mr Hort himself quotes
just before (p. 67), cannot be taken in this sense: ‘The copy-
ist of an earlier MS, from which it [G] has descended, tran-
scribed a MS of the abridged recension till the end of chapter
xiv, and then took up a MS of the original Epistle to the
Romans’; ‘ Either their common prototype {i. e. of F and G] or
a still earlier MS from which it was copied, must have pre-
served the abridged recension.’ This language was expressly
intended by me to leave open the question, as to the length of
the pedivree which connected F and G with the scribe who first
combined the two recensions ; and the idea of direct parentage,
which Mr Hort has imposed upon me, never once entered my

210 THE JOURNAL OF PHILOLOGY.
inadequately to express the degree of coincidence between D on
the one hand, and FG on the other. Certainly in the two last
chapters of this Epistle, with which we are mainly concerned,
by far the greater number of the important deviations from the
standard text are shared by D in common with FG. (2) These
three are the only’ three Greek uncial MSS which, whether on
external or internal grounds, can be assigned to the Western
family. Whatever distinctive features therefore they possess in
common, it is reasonable to set down to the Western type of
MSS generally. The Old Latin Version (with the exception of
a few fragments) is only known to us through these same MSS,
which are bilingual; for other independent copies, which contain
a more or less pure Old Latin text, have not been collated:
and its phenomena entirely accord with this supposition. The
remaining source of evidence—the early patristic quotations—
does not offer any obstacle to this conclusion; and indeed in the
last two chapters of the Epistle, this evidence, as has been
mentioned, is entirely wanting. On the whole then, I think
it may be said that the coincidence of D with F and G repre-
sents very fairly the Western text.

The second class of readings, those peculiar to F and G, are
in the xvth and xvith chapters comparatively unimportant. The
divergences of these twin MSS from D may be taken as ap-
proamately representing their peculiarities, though in the
course of the analysis it will be seen that in many cases these
divergences are supported by other, and especially by Western,
authorities’.

These are as follows:

XV. 1 apecnov [apecxesy]; 3 ove [ovx]; 7 dpas[(D* qyuas, but D** dues
with most authorities, including Western]; 11 exawecare [D exauwece-

1 I pass over E, which is now ac-
knowledged (at least so far as regards the
Greek) to be a direct copy of D, and
therefore to have no independent value.

? I have not recorded either the ac-
cidental errors of G when these have
been corrected at the time when the
MS was written, or the divergences of
F from GQ. Mr Hort’s view, that F was

copied directly from G, deserves con-
sideration, and may prove true, though
his arguments do not seem quite con-
clusive. So far as it has any bearing
on my hypothesis, it is rather favour-
able than otherwise. The converse
proposition, that G is copied from F,
could not be maintained for a mo-
ment.

212 THE JOURNAL OF PHILOLOGY.

chapters, would find a text substantially such as we actually
have here; and secondly, that no long pedigree need have been
interposed between this archetype and FG, in order to develope
the phenomena which they exhibit in these chapters; but that
the intervention of a single scribe, or two at most, would ex-
plain everything. If so, the argument from the character of the
text cannot be considered a substantial objection to my view.

(iii) Mr Hort advances another argument against my hy-
pothesis based on the assumption that the textual phenomena
on which my theory is built are gathered together from tncon-
gruous sources; and he even goes so far as to ask, ‘ How is it
that every authority, which supports, or may be thought to
support, some part of this combination [i.e. the Short Recen-
sion, involving (a) the omission of the word Rome in the first
chapter, (b) the omission of the xvth and xvith chapters, (c) the
presence of the doxology] contradicts some other part?’ (p. 76)

To this statement I demur. I allow indeed that all these
phenomena do not coexist in any extant authority. If this had
been the case, I should not have had to frame a hypothesia,
for the existence of this Shorter Recension would have been an
absolute fact. But that there is any contradiction in my au-
thorities, which prejudices the hypothesis, I cannot allow.

This attack has led me to marshal my troops to better
effect. I wish especially to call attention to the fact, that
the authorities, on which I chiefly rely, have for the most
part a close affinity to one another and that they belong to
the Western type. The Latin capitulations derived, as I have
shown, from the Old Version are essentially such. The copy
or copies, to which they refer, presented two (6, c) out of the
three phenomena, and (for anything we know) may have pre-
sented the third (a) also. The remarkable absence of quota-
tions from the last two chapters in the earlier Latin Fathers
points in the same direction. The MSS FG, which are the
only indisputable vouchers for (a), are essentially Western.
Their relation to (6), (c), is a matter of dispute between Mr
Hort and myself; but the fact that there is a great break in G
at the end of the xivth chapter (however explained) cannot
but be held to favour my hypothesis to a greater or less

214 THE JOURNAL OF PHILOLOGY.

to transpose the benediction (9 yapes «.7.X.), which properly
stands at xvi. 20, to xvi. 24, might even more easily induce
him to treat the doxology in a similar way, inasmuch as he
would still leave it the end of the Epistle as he found it,
though the Epistle had been lengthened out by the two ad-
ditional chapters. Thus the fact that the Western authorities
place the doxology after ch. xvi, seems to me to prove nothing
as to the want of ufinity between the several authorities for
my hypothesis.

But this investigation leads me to observe (and I thiok
the observation is pertinent) how entirely this Western cha
racter of the authorities coincides with my hypothesis. I sug-
gested that ‘at some later period of his life, not improbably
during one of his sojourns in Rome, it occurred to the Apostle
to give this letter a wider circulation’; and that for this pur-
pose he made the alterations which resulted in the shorter
edition, so that it was rendered ‘available for general circulation
and perhaps was circulated to prepare the way for a personal
visit in countries into which he had not yet penetrated’ (p. 294).
This hypothetical change is made in the West and for the
West; and it cannot be considered a matter of indifference
that to this same region we owe the authorities which sug-
gested the hypothesis, though at the time when I propounded
it I did not see the full significance of this fact.

With these remarks I will leave the theory. For a replr
so thorough and so suggestive as Mr Hort’s I can only feel
grateful. It has led me to consolidate the different elements
of my hypothesis, and, unless I am mistaken, to present s
stronger front to attack. From criticisms of inferior merit |
might have found less to dissent, but I certainly should have
found less to learn.

J. B. LIGHTFOOT.

THOUGHT, WORD, AND DEED.

Prorgsson WEBER in his very interesting article on the
Jaina treatise, the Bhagavat{ (published in the Abhandlungen
der Konig]. Akademie der Wissenschaften zu Berlin, 1866), has
the following note (p. 173), on the so-called Yoguvaoga, or
‘addiction or attachment of the mind, speech, and body to
any act :’

“Diese alt-arische, auch im Veda bereits mehrfach sich
findende Dreitheilung hat, wohl von Persien aus (durch den
Avesta, resp. speciell etwa durch die Manichier), auch in die
Christliche Liturgie Eingang gefunden, findet sich resp. in der
aageblich auf Papst Damasus (Mitte des 4 Jahr.) zuriickgehen-
den Confessions-Formel der gregorianischen Messe, ‘quia pec-
cavi nimis cogitatione verbo et opere’ (s. Daniel, Codex Litur-
gicus, pp. 50—51,115. Nitzsch, Praktische Theologie, 2, 281),
und von da aus dann schliesslich in dem ‘mit Herzen, Mund,
und Handen’ unsres neuern Kirchenliedes wieder.”

Prof. Koeppen, in his Religion des Buddha, I. p. 445, had
previously claimed this ethical division as a Buddhist discovery ;
at least he adds, “ findet sich tibrigens auch bei den Parsen und
Manichaern, wie bie den Brahmanen, bei den letzteren jedoch

8 vereinzelt, dass man wohl voraussetzen darf, dieselbe sey
nicht urspringlich brahmanisch.”

But there can be no doubt that it is older in Indian litera-
ture than the rise of Buddhism, as we find it in the Sanhité of
the Black Yajur Veda, vi. 1. 7, where we have the following
mystical explanation of certain words addressed to the cow
which is given as the price of the soma plant, and which is
supposed to represent the goddess Speech: “He addresses her,
“thou art thought, for what he thinks by his mind (manasd),

that he says by speech (vdchd); he addresses her, ‘thou art

218 THE JOURNAL OF PHILOLOGY.

manuum et sinus. Quid est hoc? ut ore et manibus et sinu
castus et innocens sit homo. Cum os, inquit (ML), nomino,
omnes sensus qui sunt in capite intelligi volo; cum autem
manum, omnem operationem; cum simum, omnem libidinem
seminalem.” The Greek church appears to have adopted this
practice when a Manichean was received as a catechumen, a
the priest oppayifes To pérwmov avrovU nai td oTopa xa TO
ornOos. (See the formula Receptionis Manicheorum in Tollius)
Augustine finds fault with the division as being inaccurate and
confused. But amongst the Manicheans two of these signacula
seem to have assumed a more narrow and technical meaning,
and Baur would explain the signaculum oris as the abstaining
from eating flesh and the signaculum manuum as abstaining from
injuring living creatures; but this seems a needless refinement,
when we have Augustine's express testimony that the Mazi-
cheans of his time gave a higher and wider meaning to the
phrases. No doubt a thorough Manichean would include the
liniited meanings, as Manicheism borrowed much of the Bud-
dhist ahinad, but, hke Buddhism, it would not confine itself to
them, but would aim at addressing and satisfying the common
conscience of mankind.

We might perhaps be justified in supposing that the Maui-
cheans derived it from the Persians or the Buddhists; but it
is not so easy to determine whether these derived it from the
Brahmans; and the question arises, are we obliged to suppose
that it was borrowed by any? Could it not have been it
vented by any people which had eyes to examine their own
consciousness ?

The passage I have already quoted from Plato, though it 8
used there as a psychological rather than an ethical division, is
very closely connected with the ethical application. The two
modes of looking at human actions inevitably run into each
other, and we can easily conceive that Plato's words might sug-
gest the ethical use to any one who was versed in his expret
sions. Now Plato is clearly free from any Buddhist influence
Buddha is now supposed to have died B.c. 477, and therefore
we are quite safe in maintaining an independent division here
I do not think that Plato ever again recurs to the division; he

220 THE JOURNAL OF PHILOLOGY.

apafeotv, ered) ta Kade nal dpovety nat rAéyerv wat oseiy
dvaryxaiov, supmAnpovpeva Ex te evBovdas Kai evrrpakias «al
evroyias. (Cf. also in v. 245). In 11. 217, de Mutatione nomi-
num, we have it applied to a division of sins, with a similar
reference to Deuteronomy: oyeddov roivuy nai td duapripata
Kat Ta KatopOapata cupBéPyxev ev rpiciv eEeralerOas, Stavola,
Aoyors, mpakectv.

In 111. 74, de Congressu quer. erud. gratia, § 1, he uses the
division without any mention of the Old Testament: apery Se
ov Sadettrovea avedturras Sé nai adiactatws Kata Tovs apepeis
xpovous det yea, Bpépy pev ovdapmss, Noyous 5¢ acrelous wai
Bovdads avemtXrarous Kai érraiveras mpakecs.

In v. 135, de Judice, § 3, he seems to have the same division
in his mind when he explains the Urim and Thummim, or, as
in the Sept., tiv Snrwow nal rHv adrnOeay,—ayugotépwy tay
éy nuiv Noyov -eixdvas, évdsaBérou te xal mpodhopixov. Aciras
yap 6 ev mpodopixds Snrocews, 7 Ta apavy trav «a0 Exacroyv
nav evOvpia yvwpiterat TO médas’ Oo Se évdiaBeros arnOeias, eis
tedeioTnta Biov xai mpakewy, 5: dv én’ evdatpoviay ddd; avev-
ploxerat.

The same division had indeed been all but directly ex-
pressed in the 139th Psalm, “Thou knowest my downsitting
and mine uprising; thou understandest my thoughts afar off.
Thou compassest my path and my lying down, and art ac-
quainted with all my ways. For there is not a word in my
tongue, but, lo, O Lord, thou knowest it altogether.” Here we
have the thoughts, the actions, and the words described, but we
miss the sharply-defined division into three; nor is it more clearly
brought out in Hosea vii. 1, 2.

Common as the phrase thus is in Philo, it does not seem tc
have obtained general currency in the early Christian Church
though the Sept. rendering of Deuteronomy xxx. 14 must have
often brought it before the readers of that version. It is founc
however in the opening prayer of the liturgy of St Mark (Re
naudot, Lit. Orient. Coll 1. p. 132), etre coe nuaprouev ev oye
9 épyp 7 Kata Scavorayv. This liturgy belongs to the early par
of the third century, and was used by the Church of Alexan.
dria. Origen also several times employs the phrase in hi

222 THE JOURNAL OF PHILOLOGY.

‘peccata manus. He also concludes by objecting to the division
as new: “videtisne quomodo novitatis appetitio, comite errore,
in magnas deducatur angustias? Tribus namque istis signacu-
lis, que nova guadam dtvisione preedicatis, quomodo includatis
omnium peccatorum purgationem non invenitis.”

I would thus briefly sum up the facts as far as I have been
able to collect them. The ethical division in question is found
in the later Vedic period of Sanskrit literature, and is there-
fore in India of Brahmanical origin; but it never attained
any great currency in India until it was adopted by Buddhism.
After this period we find it frequently used in the classical
Sanskrit authors. It also appears in the Zendavesta, and from
Persia it passed to the Manicheans.

In Europe it appears first in Plato’s Protagoras, but here
again it never attained any currency, until it was brought for-
ward by those Jews who endeavoured to unite Jewish faith and
Hellenic culture. But it was a division which was not wholly
strange to the Jewish mind, as several places in the Old Testa-
ment could readily suggest it to a thinking reader. It does
not seem to have been adopted by any early Christian writers
except those of the Alexandrian school, and Augustine evidently
thought it an unphilosophical division. Its popularity with the
Manicheans would no doubt tend to throw it into discredit, but
its adoption in the confession of pope Damasus, and subee-
quently in the Gregorian mass, can surely be sufficiently ac-
counted for by its presence in the Sept. and Philo, without our —
having to assume that the Church borrowed it from the Mani—_
cheans. We are all familiar with the phrase from its use imae—~
the Confession in the Communion Service, and in Bishop Ken’ ===.
morning hymn,

Direct, control, suggest this day
All I design, or do, or say.

Its adoption in the liturgy of pope Damasus may be paral——
lelled by the adoption in our own liturgy of the old Peripatetic
division of ra ayaa into ra rept wuynv, ta wept copa, ance
Ta éxtos, Which reappears in the well-known phrase of “ afflicte—l
in mind, body, or estate.”

E. B. COWELL.

224 THE JOURNAL OF PHILOLOGY.

fulfilment of the Law; the other desired to oppose Christian-
ity as true to Judaism as false; but it was equally necessary
for the purpose of both to prove the antiquity of the Messianic
faith. To these causes we must chiefly look for the origin of
the Pseudepigraphic writings, but there is yet another consider-
ation which may assist us in determining their nature and
mutual relationship. We know what a tangled web of tradition
was woven around the patriarchal history, until the simplicity
and truth of the original was almost obscured by the Talmudic
accounts. In fact, the tendency in the East has always been
thus to elaborate the facts recorded in Holy Writ, and form a
second or traditional history, which by the vulgar is accepted
even more readily than the first. We can easily imagine
therefore that a history of such vital importance as that of our
Lord would hardly escape a similar corruption. The Eastern
Christians would love to dwell upon the minutest details of His
life, they would, in their zeal to assert His divine nature, be
unwilling to admit that any incident of that life was without its
distinctive and miraculous character, and above all, they would
be, perhaps insensibly, loth to abandon their national traditions
and prejudices, and would leave no means untried for reconcil-
ing the latter with the faith which their hearts and convictions
compelled them to confess. In this way a secondary and tradi-
tional Gospel History did spring up, and the identity of many
incidents in the various Apocryphal Gospels which have reached
us, as well as the general similarity of tone observable in them
all, would seem to indicate that they are not so much the pro-
ductions of individual] persons and times as remnants of a once
popular, and if I may use the expression, parasitic account.
This will account for the fact that such of these writings as have
been preserved in their Eastern versions, Carshunic, Arabic
or Syriac, are always more full and accurate than the Greek, al-
though probably posterior to these in date. The Greek writers,
who took their accounts from the current traditions of the
Eastern Church, would naturally reject much that did not appeal
so strongly to a Western mind, which however the Eastern
translator would as naturally again supply from his own national
lore. Accordingly we find that the Oriental versions do often

ss. MIP SD CORSE OF Ferbocovy.

@s Qs .
Moe as hs eek ee me GL Ie which exactir cor-

moaote

ee tet we Wt. lhe cee Sto WooTuN €@s ors ascees. Whether the

ese eet an Crem. tue tee ce Escenas muty
bee Te eV feeriigasy eas Lath -nal Umeitiens
Do ee te EO WetT™ ode va ters aud wond sem

- Soke 7 a VG Totes sutetest°i -g that ther werr

Le Take Lr at ae) List cw Whos: existence I hare

ent. ats Chet Shatamemsts of tae Apoerrphal wus
> te . =" 2. VL LD: ra Soht or u their oThgn, punt
See ee a ofthe cite Eastern Church. Thee

wre let Ta Moar cacewe te totes ov tem sin and that tof
Ro TT ee tet Wien has ostaamistead the Eastern and
la “Mab baron wi b- the latest time. 2nd. The
fo DAL. lett Loot the Apemties as civen in the book
OR et ett Smetis UNL tue received account in the
Neeoran (nur. Ao m lestioa of Syrian christian poems
wren an Aref ocnotne death oor 17th century, by a monk
th Lptw nonates Gra’ ral an aatiber similar book of poems

Py A Ret No cas pree ct toe same features az those which
Oot tate the Apeompnal writings: the ade of Gabriel upia
Sie tate ot the Twe ive Ate ste. micht} have been suggested im-

Ie ate oN Ghe be kw : buat. Sri. The frazmeutary Latio
we (ate Rove Tsous of the Pscud-piztaphie writings present the
rater ber us the & heiie, For example, a Latin version of
tie Peetenents of the Twelve Patriarehs is preceeded in the MS
bythe ¢ Death of Adana” and appears to have been translated
Ghatrdyed tet s dae comypocte cilection of Apocryphal Books
$00.0 Theo aitisccus aud traditions contained in these works are
suehoas would mere Laturaiv linger in the Eastern than in the
Western © houreh.

Fier these considerations, I should infer that in the earliest
nates the Eastern Chureh, a legendary and popular history of
Qur Lord was already springing up beside, and parallel with
the Gospel aceount, which with a tendency to Christianize more
decidedly the Old Testament History and Prophecies had 4s-
sumed a definite and generally accepted form; and that the
authors of the Apucryphal Books of various ages which have

eee yg

CHRISTIAN PSEUDEPIGRAPHIC WRITINGS. 227

come down to us, drew their materials from this source, while
they elaborated them according to their individual idiosyncra-
sies or the tenets of the sect whose doctrines they desired to
support.

The Eastern origin of the Apocryphal Books being once
admitted will assist us materially in determining more accu-
rately their several dates and authorship. To illustrate this,
I will examine briefly a work which may be regarded as a
fair type of the class, the Testaments of the Twelve Patriarchs,
&work too the origin of which has been the subject of much
diguisition, and many conflicting conjectures.

The Testament of the Twelve Patriarchs belongs to the
second of three classes, which I have enumerated at the com-
mencement of this paper, namely, to the Apocryphal produc-
tions of the second century of the Christian Era. It consists of
4 series of discourses, put into the mouths of the Twelve Patri-
archs the sons of Jacob and purporting to be their dying in-
junctions to their children. The Patriarchs are made to con-
fess each the gravest errors of his life, and the discourses
invariably conclude with a prophecy of our Lord’s coming and
mission. Now the fashion of inculcating moral precepts or
Promulgating philosophical speculations under the form of
lestaments delivered by the illustrious personages of antiquity

long been prevalent in the East; it is with the Oriental
Philosopher as much a stereotyped rhetorical artifice as the

Dialogue was with the Greeks. The origin of this custom as
Grabius has remarked in his Praefatio may no doubt be sought
bor in the idea to which Cicero gives expression in the words
dvinare mortentes, and Holy Scripture furnishes many such
stances in recording the dying words of Abraham, Isaac,
Jacob and other holy men. So also we have other Apocryphal
Testaments ascribed to Adam, Moses, Joseph, &c. as well as
Many books in Arabic, and other Oriental languages composed
°M an exactly similar plan. Of these, I may instance the
J&vidgn Khirad of Abu’alf Maskawf, in which not only are
“€rtain moral precepts put into the mouth of Hosheng, one
of the earliest kings of Persia, but Testaments attributed to
Rearly all the wise men of antiquity are contained in the same
15—2

228 THE JOURNAL OF PITILOLOGY.

volume. It is worthy, of note that the phrase yd bani, ‘Oh my
son,’ is of constant occurrence, this according exactly with the
use of the words tréxvia pov in the Testaments of the Twelre
Patriarchs, The word 6:a6«7n, is far less precise than wariysh,
which the Arabs use to designate this species of literature, and
which is restricted to this one idea. The appropriate nature af
such a formula would at once commend itself to the Pseudepi-
graphic writer as it seems to have done in the case of the book
under consideration. There exists a similar Testamentary al-
dress of our Lord to His disciples on the Mount of Olives before
His ascension, in an ancient Carshunic MS. in the library of
Trinity College, Cambridge. In this, after a series of precepts
repeated from the Gospels, our Saviour is made to predict the
future of His Church, and the fate of His disciples
But besides the evidence of an Eastern origin, which is
afforded by the form of such compositions, these peculiar
dogmas and = speculations enunciated in them conclusively
establish their oriental character. It will at once appear ta
the student of Oriental Philosophy that a spirit of Gnosticism
closely allied to the Persian system of mysticism pervades them
all. The Testament of Adam, to which I have before referred,
is called by Pope Gelasius in his decree, A.D. 494, “ The Peni-
tence of Adam,” and by Epiphanius (adv. Heres.) “The Apoc-
lypse of Adam,” and lastly it is spoken of by Cedrenus in the
following words :—'Adau t@ é€axocwooT@ Eres petavonoas ey
5c’ amroxadtrews Ta wept Taév ‘Eypryopwv Kai Tov Kataxdvo pe,
Kal Ta Tepl weTavoias Kai THS Belas capx@cews Kal Tepi TaY Kal
EXATTHY @pay nueptvry Kal vuKTEpiY avaTrepTroLereYy EvyiY Te
Gem ato Tavtwy Tay Kticpatov 8 Oupepr tod él rhs petawia—
apxayyedov. “Adam in the six hundredth year of his ag =
repented and knew by revelation the secrets of the Vigilané<—
(the angels who dwelt on the earth before the Deluge, t(—-™=
‘Sons of God’ spoken of in the Pentateuch) and of the Flo=—=
and of Repentance, and the Divine Incarnation, and of t 8
prayers offered up by all creatures each hour of the day axv¢
night, being taught these things by Uriel the Archangel of
Repentance.”

Now these expressions furnish a clue by which we may

CHRISTIAN PSEUDEPIGRAPHIC WRITINGS. 229

arrive at an exact appreciation of the nature and origin of the
book. The phrase’petavonoas éyvw 8: avroxaduews indicates
a Gnostic source, for with these mystics the word peravoia came
to be regarded as almost synonymous with arocaduyis, and
it is in this sense that we find the Repentance of Origen, of
St Cyprian, Mambré, &., and in this sense it is certainly em-
poved in the Testament of the Twelve Patriarchs. In the
latter work each is made to express his repentance for the
besetting sin of his life immediately before enunciating the
prophecy which is put into his mouth. Cedrenus, in the passage
quoted above, proceeds to describe the Universal Liturgy for
erery hour of the day and night in words which precisely cor-
respond with the extant fragments of the Testament of Adam.
This mystical division of the twenty-four hours is merely a
reproduction of the ideas embodied in the Jeschis Sadés and
Sirouzd of the Zend Avesta', from which most of the mystic
doctrines of the Gnostics were borrowed, especially those of
the Sabaran sect who are also called Mendaites, Nazarenes or
Christians of St John, as they existed at the close of the fourth
century. Now there are evident traces of this same Persic
fom of Gnosticism in the Testament of the Twelve Patriarchs;
for example, the expression ¢as yvdoews in Levi 18, the an-
glography, the allusion to the spirit of ecstasy (Reuben 1),
crags xal eixov Oavarov, which is the same as the Hdl
of the modern Persian mystics. Reference is also made
© the Tabula Celi, an idea which has pervaded the whole
fystem of Oriental Mysticism whether in its Sabwan, Zoroas-
tnan, or Muhammedan form. The Si, co the Tablets of
Etemity, on which the Pen, 1.e. the Spirit of God, wrote the
Order of the Universe, is still a common-place with the Sufis,
a sect of Mystic Philosophers who, though existing amongst
the Mahommedans, borrow most of their doctrines from the
game source as the Christian Gnostics themselves. This Persian
Mysticism, becoming subsequently strongly impregnated with
Greek Ethics and Egyptian superstition, did not for some time

? Ernest Renan, Fragmens du livre Gnostique intitulé Testament d'Adam,
p- 11.

230 TUE JOURNAL OF PHILOLOGY.

assume the formidable aspect which appeared in the later
Essene and Elchasaitic sect, though it seems to have found
favour in the very earliest ages of the church amongst both
classes of Judaizing Christians, the Nazarenes, and the Ebion-
ites. In the first and second centuries both these sects main-
tained much more of their Jewish character though in an
entirely different degree and manner. The former, to quote
Dr Lightfoot’s words, “held themselves bound to the Mosaic
Ordinances, rejecting however all Pharisaical interpretations
and additions. Nevertheless they did not consider the Gentile
Christians under the same obligations, or refuse to hold com-
munion with them; and in the like spirit is this distinguished
from all other Judaizing sectarians. They fully recognized the
work and mission of St Paul........ They were distinguished
from other Judaizing sects by a loftier conception of the Person
of Christ, and by a frank recognition of the liberty of the
Gentile churches, and the commission of the Gentile Apostles.
These distinguishing features may be traced to the lingering
influences of the teaching of the Apostles of the Circumcision.
To the example of these same Apostles also they might have
appealed in defending their rigid observances of the Mosuic
law. But herein while copying the letter, they did not copy
the spirit of their model; for they took no account of altered
circumstances,”

Now these are just the points which distinguish the Testa-
ments of the Twelve Patriarchs from other writings which
bear the impress of Gnosticism. The Levitical sympathies of
the author, his respect for the Jewish Hierarchy generally, his
constant thanksgiving for the admission of Gentiles to the
Covenant, and above all his prophecy of the birth and mission
of St Paul (Benjamin 11), are ideas so eminently characteristic
of the early Nazarene sect, as strongly to support the theory
advanced by Ritschl, and supported by Dr Lightfoot, that the
author was a Judaizing Christian of the Nazarene persuasion.
But it would be tedious were I to point out the innumerable
instances of Orientalism which appear in the several Pscudepi-
graphic writings. The theory which I am desirous of establish-
ing, and which I believe will be borne out by an impartial

CHRISTIAN PSEUDEPIGRAPHIC WRITINGS. 231

consideration of almost any one of these works, is that they
are not merely isolated tracts, and the invention of individuals,
but portions of a Cyclic narrative of the events of Sacred His-
tory, which if it did not actually exist in the shape of a com-
plete written work, at least formed a harmonious traditional
whole originating in the East, and accepted as an authentic
scripture by the early Eastern Church.

EK. H. PALMER.

PROF. MUNRO'S NOTES ON JUVENAL I. 13, AND ON
AETNA 590.

THESE two notes deal with the same question, whether a causal
ablative of the person can be used without the preposition.
The note on Juv. 1. 18, may be found in Mr Mayor's Juvenal,
p. 93, 2nd ed., and is in explanation of

adsiduo ruptae lectore columnae.

The latter occurs in Prof. Munro's Aetna, p. 77, in explana-
tion of
extinctosque suo Phrygas Hectore.

The sum of the former is that the ablative may stand without
& preposition, if accompanied by an adjective; that we may say
adsiduo lectore ruptae, because it is equivalent to lectoris adsi-
duttate, but not lectore ruptae. Of the latter, that the bare
ablative may be used if it is equivalent to per ; hence extinctus
suo Phrygas Hectore = per extinctum H. = extinctos extincto
Hectore, or to use a form suggested in the former note = re
Hectore extincto. It is obvious that ab Hectore would not do
here.

This explanation would be unexceptionable, if it covered
every case of the disputed ablative. It is virtually that of
Scaliger, who explained scriberis Vario Maeonii carminis alite
as=cum V. Maeonu carminis ales vit. But, I venture to
think, that Prof. Munro is nearer the mark when he makes
Juv. 11. 240, tngents Liburno = vi. 351, ingents vehitur cervice
Syrorum. For his analysis of the adjective with a noun will not
suit a passage like Hor. Epist. 1. 19, 12 sqq.:

Si quis, voltu torvo ferus et pede nudo
Exiguacque togae simulet textore Cutoncm,

NOTES ON JUVENAL I. 13 AND ON AETNA 590. 233

where there is no adjective to analyse. Textore seems rightly
rendered by Prof. Lincoln of Boston, by the help of his tailor, or,
thanks to his tailor. Nor will the analysis apply where the
aljective is not a predicate, but merely ornamental, as in
Statius, Ach. L 219:
per undas
an magno Tritone ferat.

Here Tritone really = AMonstro, and denotes the mechanical means
of transit. But this latter passage leads to a solution which
will suit every case, viz., that ab must be used with either per-
son or thing if we wish to call attention to the original source
of the action, but that the bare ablative, with or without an
adjective or participle, may be used to denote instrumentality
either of person or thing, animate or inanimate, and that
whether the animate being be purely passive, as in Juv.
XIV. (+:
serpente ciconia pullos
nutrit et inventa per devia rura lacerta;

or an éuuyov dpyavoy, like the tall chairman in Juvenal 11.
Both usages in the case of inanimate things occur in Ovid,
Art. 1. 723:

Candidus in nauta turpis color, aequoris unda
Debet et a radiis sideris esse niger,

as we would say by the action of the sun’s rays on the water;
the sun, to speak popularly, being the more efficient agent of
the two. In the same way, 7b. 510:

a nulla tempora comptus acu,

@ expresses strongly the agency of the curling pin, it was to no
Pin, to no dressing that he owed his success. There is therefore
hothing mysterious in the combination of ab with the person or
thing in certain cases: it still preserves its strict force of on the
Part of, either in time, space, or action; a force which will, I
think, take off, elegance apart, every usage of ab. Hence the
Propriety of omitting that preposition in the instances collected
by Mr Munro and Mr Mayor. Take, for example, jacent suis

234 THE JOURNAL OF PHILOLOGY.

testibus, Cic. pro Mu. 47, by the involuntary admissions of their
own witnesses, whereas a suis testibus = by their direct evidence,
and so suis testibus = their reluctant evidence on cross-examina-
tion, If this distinction be sound, in Hor. C. 11 12, 27:

quae poscente magis gaudeat eripi,

poscente must be governed by magis, as in such a case the
postulant could be neither reluctant nor unconscious.

Some of what Mr Paley calls Propertian ablatives may be
explained in this way, viz., 111. 26. 91:

et modo formosa quam multa Lycoride Gallus
mortuus,

done to death, not by the malice of Lycoris, but by her fatal
beauty. So iv. 14. 30:

Nec digitum angusta est inseruisse via,
because the way is crowded. So Iv. 6. 24:
Si placet insultet Lygdame morte mea,

‘Let my death be food for his mockery.

As to the so-called dative of the agent, I do not believe in
its existence. Every instance alleged will turn out to be a
Dativus commodi. The dative of the consecrator after verbs of
consecration is a good illustration. In the examples collected
by Orelli on Tacitus Germ. 3, aram Ulixi consecratam, viz., Ann.
Xv. 41, aedes Statoris Jovis Romulo vota; Suet. Oct. 1, ara
Octavio consecrata; Jul. 88, ludos consecratos ei, sc. ab e
Venert Genetrici, the dative seems to denote that the foundation
will enure to the memory or benefit of the mortal founder, the
real agent or final cause of the consecration being the God or
Hero’. Virgil has:

Causam lacrumis sacraverat aras

1 Conington on Verg. G. 1.14, says: sages the notion is strictly that the
‘cui seems to imply that the process divine agency is manifested in the
goes on for him, because he is its works which suggest to us the notion
patron and author, thus denoting cau. of the divine agent, who thus receives
sation indirectly, Comp. 1.5. So due honour at our hands. 8So in G.
Luer, 1. 7, 8.” But in all these pas- uz, 16, in medio mihi Caesar erit tem-

NOTES ON JUVENAL 1.13 AND ON AETNA 590, 235

=arae consecratae lacrumanti=to Hector as a relief for the
tears of Andromache. This in Greek would be ém) Saxpvar, as
in Eurnp. Hipp. 32:
‘Imrmodute 8 éme
70 Aowrrov avopatey iSpicbar Oeav

=in the name of Hippolytus, and the dedication would run
INMOATTOS KTIPIAI. Similarly, in the legend from
Philustratus V. Ap. 1. 16, p. 19, quoted by Mr Munro in his
Aetna, p. 41, xumapirtou te épvos n yn avadédwxev emi Kurra-
pirre daaciy épn8e Acoupiy = which bears the name Cyparissus.
And in Tac. Agr. 2, Cum Aruleno Rustico Paetus Thrasea,
Herennio Senecioni Priscus Helvidius laudati essent, the dative
means, when Rusticus and Senecio had had the ill-luck to
praise, or something of the sort.

Passages however occur in which the ablative of the noun
without either adjective or participle is a genuine modal or
cunditional ablative. So, lumina morte resignat, rightly ex-
plained by Turnebus, La Cerda and Henry as in death, rns
murte tarapyotons. So, Caesar, B. G. 1. 18, imperio R. Populi,
which is evidently opposed to si quid accidat Romanis, 2b. supr.,
and therefore = 7ov imperio P. R. vaapyovros; and perhaps,
Lucr. 111. 928 :

Maior enim turbae disiectus materiai
Consequitur leto,

leto = rod leto vrrapyovros.

The distinction between the accusative with per, and the
bare ablative of the person amounts to this, that per keeps up
the notion of a delegated task, while the bare ablative of the
person points out that the person is either by his presence or
absence a necessary condition of the result. Both forms agree
in this, that they exclude altogether from the person specified
the notion of any initiative as principal.

THOMAS MAGUIRE.

Quezn’s Cottecr, GaLwar.

plamque tenebit, mihi =I shall be I shall be celebrated as the means of
immortalised as the dedicator of Cac- gathering all Greece together in Cae-
ear's temple; and in ib. v. 19, mihi= sar’s name.

THE ROMAN CAPITOL, AS LAID DOWN IN MR BURN’S
“ROME AND THE CAMPAGNA.”

THE recent discoveries on the Palatine Hill, due to the lbe-
rality and literary taste of Napoleon III, have given a new
impulse to Roman topography; hence the more recent works
on the subject excel their predecessors in interest. This beau~
tiful book does credit to the spirit of the publishers; the sub—
ject has at last found a form worthy of it. The wood engrav—
ings are exquisite, and represent with vivid reality sume of the
most interesting scenes in the eternal city. It is only to be
regretted that the author does not seem to have been aware of
the private house discovered on the Palatine in May 1869.
As belonging to one of the few remains of the republican period
it is highly interesting, and still more so because in it, tO
gether with the house of Asinius Pollio, also recently dise<-
vered near the Baths of Caracalla, we have the only vestiges ==
Rome of the private life of the Romans. In its style, the hou =
resembles, but excels, those of Pompeii; and some views of # t
from the photographs published by Cav. Rosa, would have bee=!
a valuable addition to Mr Burn’s book.

I do not propose to enter here into any general criticism <?
Mr Burn’s work. The following remarks are confined to tk
question of the Capitoline temple, which the author appears ©
think he has satisfactorily settled.

At p. 185 Mr Burn observes that there are some few pa»
sages of ancient writers relating to this question “which haw
never been fairly discussed, and these appear to point so plair= 1:
to the conclusion that the Capitoline temple must have been OF
the south-western height, that it seems surprising to find the
contrary any longer maintained.” |

THE ROMAN CAPITOL. 237

But, if these passages have never been “fairly discussed,’
that might be an apology for those who, until now, have held
the contrary; though they can plead no such excuse after
reading Mr Burn’s remarks on them, in which, for the first
time, they are submitted to a fair discussion. We will there-
fore endeavour to follow, with the greatest attention and impar-

taality, the three “decisive arguments” by which Mr Burn has
convinced himself that the Capitoline temple was on the south-
western height.

i. “In the first place,’ Mr Burn proceeds, “the evidence

derived from the bridge of Caligula, mentioned by Suetonius
(Cal, 22), seems decisive as to the situation of the Temple of
Jupiter. Suetonius says that Caligula in his madness imagined
that he held conversations with the Capitoline Jupiter, and
wed to whisper in his ear, and apply his own ear to the lips of
the statue for an answer. He is said to have threatened to
expel Jupiter from the Capitol unless he listened to his
advances, and the monarch of gods was at last obliged to
appease the Emperor's anger by inviting him to share his
femple. Caligula then, in order to connect his palace with the
femple, built a bridge across the intervening valley over the
‘emple of Augustus. Now it is allowed on all hands that this
bridge could not have been thrown across to the height of Ara
Celi, as it would then have passed over a part of the Forum,
‘ad no alternative is therefore left us but to conclude that it was
‘armed from the northern corner of the Palatine to the Caffa-
relli height, and that the Temple of Jupiter stood upon that
height,” ;

To this passage is appended the following note: “It was
Plainly the temple of the Capitoline Jupiter to which Caligula
Made his bridge, and Dr Dyer is mistaken in contradicting

Cher, Dict. Geogr. 11. p. 766.”
This is only the lady’s argument, “it must be so, because I
Know it is.” To retort it is to refute it, and needs only the
“unter-assertion—if it would not be rude—that Mr Burn
'S Mistaken in supporting Becker. For, putting aside the asser-
ion, he has not thrown a single new ray of light on the ques-
‘Om. He does nothing but advance the old arguments.

238 THE JOURNAL OF PHILOLOGY.

Oue of these is. that the bridze could not have been camed
t> Ara Ceh berause m that case it must have crossed the
Fuaum Ami what then? The only proof that it did not is
tha: this is “albowed on all hands” Because, I suppose, that
gach a bevdze woobi have been more difficult or unsightly than
coe t) the SW. herzht Granted. But who shall say whats
maiman w:th all the world at his feet might have done? Are
we t) make him “cum ratione insanire.”

This sbjection alone suffices to prevent Mr Burn’s argument
from being “ decisive.’ At the same time it may be admitted
to be more probable—that is all—that the bridge was thrown
from the Palatine to the S.W. height. But, when this is
granted, by what art of divination does Mr Burn jump to his
conclusion that, in that case, no other alternative is left bat
that the Temple of Jupiter must have been upon that height ?
He might at least have communicated the process, and so have
taken us with him.

For my part I can imagine more than one alternative. I
still hold that the true interpretation of the passage in Sueto-
nius is, that the bridge was thrown from one Aull to the other.
Had Suetonius meant what Mr Burm asserts, he would have
said “ Palatium templumque Capitolinum conjunxit.” That he
uses Palatium and Capitolium for the hills, I have shown
from his Life of Augustus (c. 29): “Templum Apollinis in
Palatio (extruxit) aedem Tonantis Jovis in Capitolio ;” where
it is impossible but that the hills must be meant. Indeed
Mr Burn allows this sense of the words, and even himself trans-
lates, in the passage first quoted, that the bridge “ was carried
from the northern corner of the Palatine to the Caffarelli height"

How, then, does this prove that the temple was on that
height? To do this, Mr Burn must assume that he knows pre-
cisely the direction and length of the bridge; that it began at
the palace, and ended at the entrance of the Capitoline temple,
on the Caffarclli height. Now what is this but a begging of
the whole question ?

Assuming for the moment that the temple was on the Ar
Celi height, there are two ways in which Caligula might have

got to it.

THE ROMAN CAPITOL. 239

First: the bridge might have proceeded across the back of

the Capitoline Hill to the north-east summit. Those who have

walked through the covered bridge at Florence leading from
the Pitti Palace to the Uffizi, and even crossing the Arno—a
toute quite as long, if not longer and describing an angle—will
admit that there is nothing improbable in this.

Secondly: after landing on the south-west height, Caligula
night have walked to the north-east height. And that this
® ot improbable appears from what Suetonius proceeds to
wy, but which Mr Burn does not quote: “Mox quo propior
met, in area Capitolina nove domus fundamenta jecit.” Cali-
gula therefore was by no means content with his bridge, and the
further the temple was off the more reason would he have had
for his discontentment.

This argument from the bridge, then, so far from being
decisive, affords at best a mere presumption, which must utterly
Vanish before any stronger presumptions that can be produced
in favour of the opposite height.

i. Mr Burn’s second decisive argument runs as follows: “A
second argument, which appears strongly to support the same
Cucusion, may be drawn from Cicero's account of the statue of
Jupiter Capitolinus. The Capitol had been struck by lightning,
aud the statues and other works of art, especially that of the
Capitcline Jupiter, placed on a column, had been much injured.

Haruspices, when consulted as to the means to be taken in
ler to avert the calamities thus portended, advised that a

statue of Jupiter should be made and placed on a higher
Pedestal, and that the face should be turned towards the East,
“in the hope that if the statue which you see before you,” says
Cicero, addressing the people in the Forum from the Rostra,
“should overlook the Forum and Curia, the designs of traitors
Zainst the state would be brought to light and discovered.”
The alteration, he adds, had only just been completed during
bis own consulship, and on the same day the Catilinarian
Conspiracy had been detected.

“If we place the statue on the Ara Celi height, and draw a
line eastwards from it, the line will not pass through any part
of the Forum; whercas, if turned to the South, it would have

240 THE JOURNAL OF PHILOLOGY.

overlooked at least that angle of the Forum where the Temple
of Saturn stands. But by placing the statue on the Caffarelli
height, with its face eastwards, it is at once seen that the
Forum and Curia would he nearly in a direct line opposite to
it, and Cicero's words become at once intelligible. That the
alteration of position was scientifically and carefully made can-
not be doubted, as it was done under the inspection of the
Haruspices, and in consequence of a general consultation among
the most learned members of that body; and there is no reason
whatever for supposing, as Preller does, that the orientation
of the statue was not accurate. Dion Cassius, a careful and
critical writer, gives exactly the same account of the change of
position made in the statue. “It was made to face the East,”
he says, “and the Forum, in order that the conspiracies then
causing so much agitation in Rome might be detected.”

On this I would remark, is it so certain, to begin with,
that Cicero is alluding to a statue of Jupiter Camtolinus ?
Mr Burn says in a note that in the passage alluded to (Jn
Cattl. 111. 8. 8. 20) the whole context shews that the statue
of Jupiter Capitolinus is intended; and refers also to Cicero
De Div. 1. 12. 8. 20, 21. I have carefully considered these
passages, and do not see my way so clearly to the same con-
clusion. In the first passage the words employed are merely
“simulacrum Jovis;” and though the god alluded to had no
doubt the same attributes as the Capitoline Jove, he was not
the Jupiter Capitolinus properly so called; for the statue of
this god was in the interior of the temple. Nor is there
anything in the passage from the De Divinatione which throws
any further light upon the subject. The statue that was turned
stood originally on a low column, so that it could not be seen
from the Forum; the alteration consisted in putting it upon
a higher one, and turning its face from the West to the
East. Now from these indications it may be pretty safely
inferred that it did not stand before the Capitoline temple,
on whichever height we place that building; for as the temple
faced the South, had the statue stood in front of it, it would
surely have looked also to the South and not sideways, or
askance either to the West or to the East. Jove was the

THE ROMAN CAPITOL. 241

presiding deity of the whole hill, and in this capacity it seems
nit improbable that he may have been called the Tarpeian
Jove, to whom Solinus refers when he says that the horses
of the quadriga ran three times round him—not the temple,
as Pliny says in adverting to the same occurrence (“relicto
certamine ad Capitolium quadriga prosilivit, nec ante substitit
guam Tarpeium Jovem trina dextratione lustrasset,” p. 195,
ed. Mommsen). It seems probable that it may have stood in
the middle of the hill, in what has been called the inter-
montium; which however would be the lowest part of the
summit: and this would account for the necessity of raising
it on a higher column, to make it visible from the Forum.
In like manner the colossal figure of Athene in the Acropolis
of Athens stood quite detached from her temple, and fronted
tke west. If this was the position of Jupiter all difficulty
about his view over the Forum and Curia would vanish at
once ; but also all arguments drawn from it as to the site of the
temple.

We shall only add that the passage in Dio Cassius (XXXVII.
9; to which Mr Burn also refers, throws no further light upon
the sulject. Dio merely calls it a statue of Jove erected upon
a column, and repeats Cicero's account of its being raised
higher and turned from west to cast (dyaApa Atos és Kiovos
iéprpévov...cat Ta Act dyadpa peitov, mpos Te Tas dvaToNas Kal
apes Tv ayopay Bréerov). Nor can any argument be drawn
from the fulluwing sections (21, 22) of the Catilinarian oration,
where Cicero says that the work had been accomplished with
the assent of Jupiter Optimus Maximus, and where point-
ing with dramatic effect to the statue now for the first time
visible from the Forum—an arrangement no doubt purposely
contrived by the cloquent consul in order to give point to
lis Gration—he exclaims, “ille, ille Jupiter restitit.”. For Ju-
piter is there alluded to in his general character of best and
greatest and guardian of the city.

But even if the statue on the column should be thought
ty have stood before the temple—a most awkward position for
it—I still maintain my opinion that the north-east height, and
net the south-west onc, would best have afforded it a view of the

Journal of Philology. vou. 11. 16

242 THE JOURNAL OF PHILOLOGY.

Curia and Forum. Mr Burn, however, selects the latter, and |
contending that the statue must have been scientifically placed,
holds that a line drawn castwards from it with mathematical
accuracy would pass through the Forum and Curia; whereas
had the statue been on the Ara Celi height, the line would not
have touched the Forum at all.

Before we can consider this point about the orientatian
of the statue, it is necessary to determine the exact position of
the temple before which it is supposed to have stood. We
know in a general way that it faced the south; but as Mr
Burn observes, p. 189, rules about the orientation and arrange-
ment of buildings must always be considered as subordinate
to the exigencies of the site; and that “doubtless when these
rules proved inconvenient the ancient augurs had many ways
of evading them.” Whence it appears that though they were
so mathematically strict about the orientation of a statue
(which is necessary for Mr Burn’s argument), they were some-
what lax about what might be considered the more important
position of a temple, which is also necessary for Mr Burn’s
argument.

The augurs being thus somewhat lax about the site, I will .
suggest the probability that the front of the temple, instead
of looking due south, may have inclined rather to the south-
west. My reasons for thinking this possible, nay even pro-
bable, are, drawn first: from the configuration of the Capi-
toline Hill which lies in a direction from north-east to south-
west ; and therefore temples placed in the same direction would
harmonize better with the requirements of the site, than if
placed as it were transversely and askew. Secondly: this view
is corroborated by the remains of a very considerable temple,
whatever it may be, discovered by recent excavations on the
Caffarelli height, which according to Mr Burn’s own admission
(p. 188) looks to the south-west.

Now, if such was the situation of the great temple, and
if the statue stood before the main entrance—which, if it stood
there at all, may be fairly presumed to have been the case—
then I am afraid it would have had a very bad chance of being
seen from the Forum, and a still worse of seeing the Curia.

244 THE JOURNAL OF PHILOLOGY.

argument, brackets this with Fides, without referring to any
other authority than Appian.

Now in the passage cited Appian says nothing whatever
about the temple of Mens; and instead of showing that the
temple of Fides, where the senate met in the Gracchan sedi-
tion, was on the Capitol, it proves precisely the reverse, as I
shall presently show. But first of all I must quote some
further remarks of Mr Burn’s about this temple. At p. 192 he
says: “The Temple of Fides is one of the larger and mos
fréquently mentioned temples of the Capitol. It was first bult
by Numa, and then restored in the First Punic War by Atilius
Calatinus and Aimilius Scaurus (Plut. Num. 16; Liv. 1 21;
Cic. De Nat. D. 11. 23). Meetings of the senate could be held
in it, and it was here that during the Gracchan tumults the
sitting was held when, gradually excited by vehement denun-
ciatory speeches, the Senators at last rushed out, headed by
Scipio Nasica, and murdered Tiberius Gracchus, near the
statues of the seven kings, which stood at the door of the tem-
ple (App. B. C. 1.16; Val. Max. 11. 2.17).” And he subjoins:
“The passages of Cicero and Appian which vouch for the tem-
ple’s situation, are too distinct to be explained away.”

All these passages are also quoted by Becker (loc. cit.) with
the addition of Cicero, De Off. 111. 29, of Dio Cassius’ (xxv. 17),
and the Fasti Amiterni, Kal. Oct. Fiper 1n Caprrouio. And
to these I will add another from Pliny, which Becker does not
seem to have been aware of: Spectata est et in aede Fidei 10
Capitolio (tabula) senis cum lyra puerum docentis (N. H.
Xxxv. s. 36.100). Becker probably read the passages he quotes,
for he was not very scrupulous about perverting or mutilating
authorities in order to suit his theories. With regard to Mr
Burn, the most charitable supposition is that he did not read
them, but implicitly followed Becker, for whom he seems t®
have a great vencration.

It appears plainly enough from these passages that there
was a temple of Faith on the Capitoline Hill; but it appears
just as plainly, first, that 1t was not the temple built by Numa;

1 Kal wveipa peya dmvyevdpevoy ras = rov ris Ilerews vewp xpormernyti as
re orn\as tas wept rd Kpomov xal wept améppnte xal dieoxddace.

THE ROMAN CAPITOL. 245

and second, nor that in which the senate met in the Gracchan
tumult. Cicero, in the first passage, says that the temples of
Fides and Mens had been lately dedicated on the Capitoline
Hill by M. Aémil. Scaurus; and in the second, that the temple
was near that of Jupiter, Opt. Max., and was founded by our
ancestors’; a phrase which he would hardly have used of Numa.
Atius Calatinus had nothing whatever to do with the matter.
Mr Burn has confused the temple of Spes, previously founded
by Calatinus, with those founded by Scaurus. On the other
hand, Valerius Maximus, Plutarch, and Appian evidently speak
of the temple originally founded by Numa. The manner in
which Livy relates this is not very perspicuous. He mentions
that Numa introduced a cultus of Fides, and proceeds to say
that he ordered the famines to proceed to that temple—though
he had not previously mentioned any—in bigs, and to per-
form sacrifice with the hand wrapped up as far as the fingers
(Et soli Fidei solemne instituit: ad id sacrarium flamines bigis,
curru arcuato, vehi jussit, manuque ad digitos usque involuta,
rem divinam facere, I. 21). Sacrarium usually denotes a place
where sacred utensils are deposited ; in Lib. xxx1Ix. 9 and 10,
Livy uses it of the place where the unholy rites of the Baccha-
nals were perfurmed at Rome. But if the senate could be
assembled in the Sacrarium of Fides, it must have been a
temple ; that is, it must have been an inaugurated place. But
to return: —

Among the ancients these allegorical divinities, as well in-
deed as what we may call their more proper gods, had different
attnbutes ; thus as there was a Pudor patricia, and a plebeia,
8) also there seems to have been a Fides publica and a Fides
privata. Now that Public Faith was the deity established by
Numa we learn from Dionysius: fepov iSpvcato Tiatews dnpo-
Tas kai Oucias avtH Katextnoato Snportededs (11. 75). Where-
ure in the passage of Valerius Maximus, in which he relates

1“Ut Fides ut Mens, quasin Ca- jurandum violat, is Fidem violat, quam
“tolio dedicatas videmus proxime & in Capitolio vicinam Jovis Optimi
L Emilio Scauro, ante autem ab Maximi—ut in Catonis oratione est—
tllio Calatino erat Spes consecrata.” majores nostri esse vuluerunt.” De
: Nat. Deor. 1. 23: “ Qui jus igitur Off. 111. 29.

246 THE JOURNAL OF PHILOLOGY.

the Gracchan sedition, we should probably read publiew fa
publice: “in aedem Fidei public convocati Patres Conscripti,
&c. (111. 2. 17).

We learn from this passage only that the senate was assem
bled on that occasion in the temple of Public Faith, and thers
is nothing to show. the site of it; though it is a probable infer.
ence from the passage we have quoted from Dio (supra p. 244)
that it lay near the temple of Saturn. On the other hand the
passage referred to in Appian affords the most satisfactory
negative evidence that it could not possibly have been in the
Capitol, and for the purpose in hand this is enough. For that
historian relates that, on the critical day, Gracchus occupied
the Comitium and the Capitoline temple with his partizans
xatéXaBe Tod Karir@ Xiov Tov vedy évOa yetporovnces eyeddos
nal Ta péoa Tis éxxdrnolas (p. 612, ed. Tollii). In the midst of
the confusion: the senate assemble in the temple of Faith (vpe-
pévov Se TouTwy 1 Bovd cuvidOev eis Td THS Tliotees icp
p. 613); apparently because as the Gracchani had occupied the
Comitium, they were debarred access to the Curia; and still
less would they have been able to mount to the Capitol,
already seized by the rioters, without a fight. Having
some resolutions in the Temple of Faith, they mounted up
the Capitol under the conduct of the Pontifex Maximus, Cor?
Scipio Nasica (xplvavres 5 30a expivayv, és rd Karrerart?
avi¢goay, tb.). When Nasica had ascended to the temple (av€?
Oadv Sé eis 7d iepov, p. 614)—not therefore that which tP*
had left below, as Becker seems most wonderfully to have c©
strued, confounding this fepéy of the Capitol with the fors*
one of Faith—his followers began to attack the Gracchani
the summit, driving them onwards and hurling them over #
precipice ; and in the tumult Gracchus himself was slain né
the doors of the temple and the statues of the kings; wh#
therefore stood, not before the Temple of Faith, but, as w
much more natural, before the temple of Jove, the princir
one on the Capitol («ai [paxyos auros eidovpevos repli
tepov, avnpéOn mapa rds Ovpas, mapa rovs trav Baciéwy a
dplaytas, tb.). Indeed it is absurd to suppose that the Sena’
could have held a sitting in the temple of Fides, had it bee

THE ROMAN CAPITOL. 247

close to the Capitoline, while the Gracchani were in possession

of the latter. Plutarch’s account of the matter is another
proof that the temple of Fides was below. He tells us that
Flavius Flaccus, one of the senators, who seems to have been a
frend of Gracchus, hastened to communicate the decision which
the senate had come to to kill him, and called out from below,
but could not be heard; on which he made signs that he
wanted to speak privately with Tiberius. Gracchus ordered
his people to make way for him ; when Flaccus, having ascended
with some difficulty (avaBds pods), acquainted Tiberius with
the danger he was in (Tib. Gracch. c. 18).

Paterculus, in his brief account of the matter (I: 3), tells
the story rather differently, and makes Nasica exclaim “qui
airam vellent rempublicam se sequerentur,” after he had
mounted up to the Capitol, and as the signal for the actual
stuck; and not, as Appian with more probability relates in
the Temple of Faith, before ascending. But this is of no im-
portance.

The necessity therefore for encumbering the Capitol with a
temple large enough to hold the senate, arises only from
atotal misunderstanding, or wilful perversion of some not very
dificult texts. The temples in the Capitol erected by Scaurus
Fides—apparently private Faith—and Mens, were doubtless
Comparatively small. With regard to the “templum ingens”

of Jupiter Custos erected by Domitian (Tac. Hist. m1. 74;
Suet. Dom. 5), the evidence only goes to show that it stood
‘Somewhere on the Capitcline hill.

I have now examined Mr Burn’s three decisive arguments
Against the temple’s having been on the height of Ara Celt,
and am ata loss to discover on what this supposed decisive
Character is founded. On the other hand, ainong tho argu-
ments, which he regards as undecisive, there are two or three,
I think, which at all events are not so unimportant as he
imagines.

Most of the descriptions of the attacks upon the Capitol
I will at once abandon. I have indeed never used any of
them, except that of the Vitellians, as showing anything of a
Positive character; although Becker did, and with his usual

248 THE JOURNAL OF PHILOLOCY.

overweening opinionativeness, adduced them as completely de-_
cisive of the question. My remarks in the article on Rome _
were made mercly to show that this was not the case; and -
from the observations made by Mr Burn, p. 187 sq., I appear to
have succeeded. On this point I hope I may also be permitted .
to adduce the opinion of the late Lord Broughton, a very com-
petent judge of the matter. He says: “I confess that the
learned and candid writer of the article ‘Rome’ in Dr Smith's
Dictionary, seems to me to have demolished Becker’s arguments
in favour of the Caffarelli height (in which arguments are in-
cluded the three advanced by Mr Burn as decisive), although
perhaps he has not removed all the objections to the other
summit. Indeed Mr Dyer, the wniter of the article, with a
fairness that does him honour, and adds weight to his opinions
in general, confesses that the question will not. admit of com-
plete demonstration ;” but he adds, “we hope that the balance
of probability may be shown to predominate very considerably
in favour of the north-cast height’.” And in a note on this
passage Lord Broughton says: “The story of the famous
Vitellian attack on the Capitol, as told by Tacitus, is com-
pletely perverted in order to make it suit the German theory;
and Mr Dyer is fully entitled to exclaim, ‘Our chief objection
to this account is its impossibility.””” With regard to that
attack I still hold the opinion that I expressed in my article,
p. 765: “It is plain that the fire (which destroyed the temple)
broke out near the Lucus Asyh, and then spreading from house
to house, caught at last the front of the temple. This follows
from Tacitus’ account of the porticoes and the eagles which
supported the fustiqium, or pediment, first catching fire. The
back-front of the Capitoline temple was plain, apparently a
mere wall; since Dionysius (IV. 61) does not say a single word
about it, though he particularly describes the front as having
a triple row of columns, and the sides double rows, But as we
know that the temple faced the south, such an accident could
not have happened except it stood on the north-east height, or
that of Ara Celi.”

Mr Burn has not thought it worth while to notice this

1 See Lord Broughton's Jfaly, Vol. 11. p. 12.

THE ROMAN CAPITOL. 249

argument, and it still remains unrefuted. He has also passed
ut¥er my argument (p. 768 A), from Dionysius’ description of
the temple, that had it lain on the south-west height, it would
have presented its nude and unadorned back to those who
copruached it. Nor has he adverted to an argument which,
I believe, nobody but myself has advanced, drawn from
L:vy's narrative of the trial of Manlius (v1. 20). The Comitia
Centuriata were assembled to judge Manlius at the spot after-
wards occupied by the Circus Flaminius; whence, as a glance
at the map will show, the north-east height must have
Leen conspicuous, and the Arx also in sight. Manlius took
wivantage of the situation to appeal to these objects, and
especially to the Capitol, with its temple of Jove, “Capitolium
spectans Jovem deosque alios devocasse ad auxilium fortu-
narum suarum.” Where it is evident that, by ‘Capitolium,’
Livy means not the whole hill, but the Capitol in its narrower
scuse: first, because he alludes to the temple of Jove upon
it; secondly, because just afterwards he cnumcrates the two
summits distinctly (ut Capitolium utque arcem intuentes).
Tu deprive him of this appeal, the tribunes altcred the place
of assembly to the Lucus Poetelinus, a spot just outside the
Purta Flumentana, whence the Capitol with its temple could
but be seen (unde conspectum in Capitolium non essct). The
mup will show that this was the only spot in the Campus
Murtius where the temple, from its being hidden by the south-
Wet summit, which we assume to have been the Arx, was
enccaled from view. The tribunes would doubtless have been
glad to conceal the Arx also, had it been possible; but an
anpeal tu that alone would have wanted the effect of the
rligiy which so much swayed the superstitious Romans; for
the temples even on that height could hardly have been visi-
le, but only the towering edge of the precipice. They were
re longer in the presence of those rescued deities in whose
sicht Manlius had invoked their judgment (sce Dict. p. 751).
The anguraculum upon the Arx is another very awkward
cujeet for those who place the Arx at Ara Celi. Mr Burn
in. 195) allows that the argument drawn from it is not without
weight, yet contends that there is not much reality in it. His

250 THE JOURNAL OF PHILOLOGY.

attempts to evade it are, however, of a very slippery descrip-
tion. “The Ara Celi height,” he says, “is about fifteen feet
higher than the Caffarelli, and as the temple of Jupiter upon
the latter is known to have been a comparatively low struc-
ture, perhaps partly in order not to obstruct the view from the
opposite height, and the auguraculum was most likely raised
upon a tower, the augurs may eastly have been able to see over
the temple roof. Even if this be not admitted, is there any
impossibility in the supposttion that the Temple of Jupitr
enjoyed an exemption from the rules applied to ordinary ™“
temples and houses’?”
No impossibility, perhaps, but the very highest degree of—~_;
improbability. As the temple of Jupiter must have been in-_.»—
augurated, it 1s very unlikely that the augurs would hav ©
violated their own rules in its construction; especially as the—--y
might so easily have avoided the difficulty by making a new=mmmy
auguraculum in front of the temple on the south-west summi—wit.
This however they did not do; the auguraculum continued t—0
be on the Arx; and, therefore, the probable inference is, the==at
the Arx was the south-west summit. Mr Burn’s ‘ suppositiommam’
therefore I cannot admit, especially as we see that the augu. rs
were so particular about getting an unobstructed view, th —=at
they ordered a house on the Celian, a long way off, to “Wile
lowered. How much more must their prospect have be<«men
interfered with by a huge temple just before them! Wm th
regard to the augurs being able to see over it, it would haw.-ve
been desirable to have had a little more evidence about tkumat
high tower, which at present seems to have no more so» Jid
foundation than a castle in the air. A better way of evadm mg
the difficulty would have been to assert that the augur look ed
west, as Mr Burn does in his note (No. 5), after Becker (Ham et),
Iv. s. 357); though in the text he says, and I think more
correctly, that he looked generally towards the south.
But to quit these arguments from probability and adwert
to something more tangible. After the publication of my
urticle, some excavations were undertaken on the Caffarelli

1 The italics are of course my own, _reader's attention to the steps of the
and are merely designed to call the argumentative process.

THE ROMAN CAPITOL. 251

height, with a view to discover if any traces still remained

of the Capitoline temple supposed to have been seated there.

The labour was not altogether fruitless; for the foundations

of a very considerable temple, and from the nature of the
building a very ancient one, were brought to light. It mea-
sured 39°18 metres in length, and about 24 in breadth, or about
127 feet by 79: and was therefore no unworthy sister to the
Capitoline temple, whose length was only about 200 feet. But,
a8 these dimensions clearly showed that it could not have
been the Capitoline, the theory that the latter lay on the
Caf¥arelli height was abandoned by the more candid of the
German school, and M. von Reumont in his History of the
City of Rome, recently published, admits that the result of the
€Xcayvations is fatal to that assumption’. And accordingly in
his plan at the end of the volume, he places the Templum
Jovis on the north-east height and the Arx on the south-
West.

Not so Mr Burn; who, itpsis Germanits germanior, still
Sticks to his theory, though these extensive remains impinge
On the spot where the Capitoline should have lain, had it been
On that height at all. “Whether the foundations thus de-
Bcribed,” he says (p. 188), “be those of Domitian’s temple of
Jupiter Custos, or must be ascribed to the more ancient temple
of Fides, cannot at present be decided.” But, at present, I
think we are in a condition to say very decidedly that at least
they could not have belonged to the last; that is, to the large
temple of Fides founded by Numa, which Mr Burn supposes
to have been on the Capitol. But the exclusion of this temple
does not imply the acceptance of Mr Burn’s alternative, that
then it must have been the temple of Jupiter Custos. From
the size and site of it, it 1s much more likely to have been
that of Juno Moneta, which we know lay on the opposite
height to that of her brother and husband. And as these
remains extend from the Caffarelli palace to the Via di Monte

1 Die Ergebnisse der jiingsten Aus- Hihe gelegen habe, auszuschliessen.
grabungen im Garten des Palazzo Gesch. der Stadt Rom. B. i. Anmerk.
Caffarelli scheinen die Annahme dass__8. 800. Berlin, 1867.
der Haupttempel auf der siidwestlichen

B ]

252 THE JOURNAL OF PHILOLOGY.

Caprine. there could have been no room eastwards for a temple
of 200 fect. or about 60 metres; for the whole breadth of the
heicht trom that place is barely so much, and would have
left no rowm for roads and approaches. And for this we need
eniy refer the reader to Becker's plan of the Capitoline Hill
in his Huxabuch.

I wil row take my leave of the Capitoline question.
I am far frem presuming to say that I have decided it; but
I will reassert. with greatly increased confidence, the con-
eusicn at which I armved some fifteen years ago, that on
the whole the balance of probability inclines very consider-
abiv in favour of the north-east height. I am aware that
I mar be twitted with having departed in some of the remarks
wich I have made, and particularly with regard to the temple
cf Fides from what I had previously said in my article on
Rome in Dr Smith's Dictionary of Geography. But ai devrepas

dparzides codarepas. Iam not ashamed to acknowledge that
] have grown somewhat wiser, and perhaps the circumstances
under which that article Was written, may extenuate the com-
mission ef a few mistakes. Mr Bunbury having unexpectedly
decumed to write it, [ was requested to undertake it. I had
hot pr vicusiy given any special attention to the subject ; the
time ailowed for the completion of the task did not embrace
many months, and part of it was to be devoted to a visit to
Reme. I was strengly recommended to fellow Becker, then
in doigh vegzue: by doing so, I should certainly have saved
myself a yreat deal of trouble, and perhaps have carned an
equal modicum of reputation. But as I proceeded with my
task. [ found that I could not always implicitly trust Becker's
“admirable work.” I ventured to differ from him on three
capital peints of Roman tepegraphy; the sites of the Capi-
taline temple. of the Comitium, and of the Cuma Julia, which
last. indeed, is a natural sequence from the restoration of the
Comitium to its proper place. On the last two Mr Burn has
confirmed my judgment by adopting it; but, though I was
the first, at all events in England, to bring forward new views
on these points, Mr Burn has completely ignored me. Momm-
sen had adopted the same view as myself respecting the Comi-

THE ROMAN CAPITOL, 253

tum ; but it was not till my article was nearly ready for the
press that I lighted on his. Detleftsen’s paper on the same
subject was posterior to mine. With regard to the Curia
Julia, nobody, I believe, either at home or abroad had pre-
tivusly entertained my view of it. The establishment of these
points, and the investigation of the other multitudinous ques-
tions respecting the topography of Rome, demanded a great
share of my limited time; and I am afraid that in some
suboriinate arguments I may have placed too much con-
fidence in Becker. It may be said that I had an opportunity
to correct any oversights in a re-issue of my article in 1864,
for the use of travellers, a purpose for which it was never in-
tended But that reprint was made without my consent, or
even knowledge; and contains some original errors of the
press uncorrected. On my remonstrance, the Publishers placed
on the fly-leaf a notice that the book was a verbatim reprint
of the article published in 1856, and therefore only repre-
sents the views held by the author at that time. I regret
that the erroneous view respecting the temple of Fides has
also slipped into my History of the City of Rome, p. 37, and
I hereby recall it. It was not till 1 was writing my History
of the Kings of Rome, that I became fully aware of the extent
to which Becker could abuse and garble the passages of ancient
authors in order to suit his views.

THOS. H. DYER.

Jan, 21, 1871.

ACTS XXI. 37, 38.

°O 8¢ én “EAAnviotl ywaones’ ova dpa av el o Avyurtios,
6 Tpo TOUT@Y TAY NUEPAV K.T.Dr.

Dr ALEXANDER ROBERTS in his “ Discussions on the Gospels,”
deals with the question of the language employed by our
Lord and his disciples in a manner, which, to my mind, is
generally satisfactory. I fully go along with the general course
of his argument, when he contends, that the Jews in general
and the Galilzans in particular were to a great extent bilingual,
using Greek and Aramaic indifferently, just as the Welsh in
Britain and the Czechs in Bohemia, although -circumstances
would often arise, in which they would prefer to be addressed
in their properly national language, as in Acts xxii 2. But
when he deals with the passage, which I propose to discuss,
he has no better suggestions to make than that a “rude
Egyptian” might possibly have been unable to speak Greek.
Yet Egypt was undoubtedly the stronghold of Hellenism; the
Septuagint was to all intents and purposes the Bible of the
Egyptian Jews; the learned Philo himself appears to have been
ignorant of Hebrew; and if an Egyptian Jew was ignorant
of Greek, it is difficult to imagine what language he could
have spoken for the common purposes of life and ‘business.
I consider, that in this matter Dr Roberts, and also Dean
Alford, have been misled by the authority of Winer, who
objects to the rendering of ov« dpa by nonne tgitur? which
is adopted in English by the Authorised Version, and renders
the words by non igitur, “Thou art not then (as I thought, but
now see contradicted) that Egyptian.” It is true, that ove dpa
is most frequently used as Winer says, but I shall presently
prove by the only legitimate method, that of quotation from
Greek authors of undoubted weight, that it is not exclusively

256 THE JOURNAL OF PHILOLOGY.

than a Galilean, and the goodness of St Paul's language and
pronunciation would naturally suggest to Claudius Lysias the
hypothesis of his being an Egyptian Jew of influence.

Rom. ui. 30.

"Esreizrep els 6 Oeds, Os Sixatwoes mepiropny ex rictews al
axpoBvetiay Sia tHS WioTews.

Singular difficulties meet us in this passage as ordinamly
construed, which appear to become less explicable the more
they are examined. Why should the preposition é« be applied
to the justification of the circumcised Jew, while the prepos-
tion d:a is applied to that of the uncircumcised Gentile? Why
should the article be inserted between the preposition and its
noun in the case of the Gentile, while it is omitted in that
of the Jew? Answers exhibiting more or less acuteness and
power of hair-splitting have been given by various commenta-
tors to these questions, but the general result of their argu-
ments has been an increasing conviction in my own mind, that
the Apostle Paul had no such views, and entered into no
such subtleties, as they are severally compelled to ascribe
to him.

Can we not then begin de novo, and find an explanation of
the words, which shall simply put aside and ignore the antitheses
in question and the whole set of controversies founded upon
them? Can we not find an explanation, which shall entirely
get rid of the assumed antithesis between é« aiotews and a
THS TiaTEws 2

If we return to verse 26 in the same chapter, we find the
expression tov éx aiatrews "Incov, which appears to bear 8
singular relation to, and perhaps may have suggested epstopiy
éx MioTews, a8 a compound expression in the passage which we
are considering. It is pretty clear that in verse 25, itacriptoy
dia wiatews is a compound expression; why should not qepito
pny éx wiotews be one of a similar character ?

We thus obtain two classes, that God is willing to justify,
TEpLToNY Ex TioTews, “circumcision, that is of faith,” ie. Jews

258 THE JOURNAL OF PHILOLOGY.

contrary, we are establishing the law on a firm basis, as 4

law of faith and not of works, a law of Spirit and not of —
letter.”

Titus in. 8 and 14.

In these two passages we have the expression xadav épyw
xpoiotacOa:, which is translated in the text of the Authorized
Version “to maintain good works,” and in the margin “to pr
fess honest trades.” The commentators generally favour the
rendering of the text, and indeed I am not acquainted with oe
who takes his stand upon that of the margin.

Let us consider the context of the passages in which this
expression occurs. In Tit. ii. 8, translating literally, we have:
“Trusty is the saying, and about these things I wish thee
to be positive, that those who have believed God may be heed-
ful xadav Epywv mpolotacGa; these are xada and beneficial
to mankind.” What are «ada and beneficial to mankind!
Surely the épya in question. But all “good works” in the
ordinary sense of the word are such, and if xadd épya are to
be considered as merely equivalent to dya0a épya, a sense of
kados not unfrequent in late Greek, the clause tavra cot
KaNG «.T.d. is a mere piece of useless tautology. The contert
therefore appears to drive us to the rendering of the margin,
which makes the clause ratra dors: nada «.T.r., an excellent
explanation of the meaning of xaday in xadray épyov apo
coracba.

Let us now proceed to ver 14. Here we find pavOaveraca
5é xai ot nuérepor nadav epywv mpoloracbas eis Tas avaryxains
xpetas. Here it surely is much more natural to consider the
article ras as having a subjective meaning, and referring to the
subject of the sentence in the sense: “for their [own] nece®
sary requirements,” than to understand “the necessary require-
ments of the individuals in the community” to be implied. Or
we may understand the article as generalizing the words to
which it is prefixed, so that eis ras dvayxaias ypelas would
signify “for necessary requirements in general,” “for all neces-

— oe

TITUS HI. 8 AND 14. 259

sary requirements;” which would imply a direction to Chris-

tians living amongst heathens to confine themselves to reputable
and necessary employments, avoiding such as were unnecessary,
and the mere handmaids of luxury.

As to the word épyov, I need but quote 1 Tim. iii. 1, e? ris
étisxorns opéyetat, Kadod Epyou émiOupet: “If any one is
anxious for a bishop’s office, he desires an honourable occu-
pation ;” and Xenophon, de Vect. Iv. 6: apyupiris &é bo@
dy theiwy haivnrat, TocovT@ TAcloves emt TO Epyov ToUTO ép-
yovras: “The more silver-ore appears, the more persons come
to this occupation.”

With regard to wpolorac@au, it will be requisite to examine
at first hand, rather than accept at second hand the traditional
references of Lexicons and commentators. In the Electra of
Sophocles, 980, we have:

@ Torow €yOpois ev BeSnxcow trote
yuyns apednoavre mpovotytny povov.

Here arpovatnvas povov is clearly used in the sense of “in-
ficting death upon” enemies in prosperity. In Xen. Mem.
Il. 2.2, we find: €¢ tod éavrov Biov Karas mpoeotnxot, “if he
regulated his own life well.” These passages are as favourable
to the view against which, as to that for which, I am con-
tending. But I can scarcely believe that a passage in Athe-
neus, 612 a, has often been actually referred to by commenta-
tors, who favour the rendering of the text of the Authorized
Version, We find here that the art of a perfume manufacturer
vas not always considered reputable, 2¢Awvos rod vouobérou ovd’
exitpemovros avdpi Tovavtns mpolctacbat réxvns, “Solon, the
hwgiver, not even permitting a man to profess, or rather prac-
tue, such an art.”

Here we have at once an illustration of both the words and
the sense of the passages in question. Can we hesitate for an
instant between taking the injunctions of St Paul as trite
maxims of the driest kind, and understanding them as vivid
Practical precepts, bearing closely on the social relations of
his day ?

A. H. WRATISLAW.

17—2

ON LUCRETIUS, BOOK VL

As most of the suggestions on the 6th book of Lucretius
made by me in this Journal (1869, pp. 219—228) have been
recently impugned by Prof. Munro (Journ. 1870, pp. 115—217),
I wish to say something more on the points in discussion be-
tween us.

47—49. Most editors will probably accept, as Prof
Munro has done, Bernays’ view that there is a lacuna after v. 4/.
The two next vv. are in the MSS. as follows:

Ventorum exirtant placentur omnia rursum
Que fuerint sint placato conuersa fauore.
I proposed to read

Ventorum existant, placentur momina rursum,

Quae fuerint sint placato conuersa furore.
Exzistant is a conjecture of Bernays’, and seems to me neatly
certain; furore is as old as the second edition of Lambinus:
momina is mine. Prof. Munro objects that momen is else
where only used in the singular. I would not deny that Scalige®
has failed to prove the existence of momina in Manil. m1-
679, Iv. 207: but if Lucr. could use so unusual a plural a3
aeribus, 1V. 291, v. 645, he might, I think, with less licens©
use momina, a plural which has nothing objectionable in itS
form, and if it occurred in even one undoubted passage O
Lucretius would be accepted without hesitation. And this
reading does not necessitate another lacuna, though a double
lacuna is possible. Like existant and placentur, sint may be
dependent on some word in the lines which have dropped out ‘
‘whilst I explain how the agitations of the winds arise, and are
then again lulled; how those which have been, lull their age

2()2 THE JOURNAL OF PHILOLOGY.

where Munro translates guam ‘how’, I venture to think that
it is the relative dependent on suspicere, that in caels templa
is appended to this as epexegetic, and that there are thus, so to
speak, two clauses dependent on dignatur, (1) quam nemo dig-
natur suspicere, (2) nemo dignatur suspicere tn caels templa, ‘an
appearance which no one any longer cares to look up to—to
look up, I say, into the quarters of the sky.’ It is of couse
obvious that suspicere may be ‘look up to see,’ and there are

many to whom this explanation would seem sufficient ; but it

is worth while to put forward the bolder hypothesis, because it

would be difficult to name any one in whom a freer spint

breathed than Lucretius, and this is not unlikely to have found

something to correspond to it in the idiosyncrasies of his

expression.

More doubtful is 1v. 397—399, Ecstantisque procul medio de
gurgite montis Classibus inter quos liber patet exitus ingens
Insula coniunctis tamen ex his una uidetur. Munro makes
montis an accus. by attraction: I hold it to be a nominative;
the original outline of the sentence was montis...coniuncti tnsuls
widentur: and if it had consisted of two verses only, this out-
line would probably have been retained as it is, the intervention
of the defining relative clause, Classtbus inter quos, enables him
to give a freer, if I must say so, a more Greek form to the
sentence; the nomin. of the first verse is repeated in a different
shape—not hz contunctz, but ex his coniunctis.

To return to vi. 68: I did not deny any one of Prof. Munros
statements, but was not convinced by them, nor am I now.
The whole point of my remarks was to shew that they did not
settle the question. Resolve quae nisi into quod nisi haec, as Prof.
Munro does; that does not diminish the difficulty of separating
dis indigna putare from them ; and if they cannot be separated
without harshness, we are reduced to my explanation. Ther
are however other things of a similar kind. Take rv. 1088,
Quod fiert contra totum natura repugnat, ‘the direct contrary of
which nature protests to be the case’ (Munro), more literally,
‘which nature combats to be all done contrary.’ What does
nature combat? not that it is done contrary, but that it 8
done at all; it asserts that it is done contrary. But that asser-

ON LUCRETIUS, BOOK VI. 263

tion is a denial of the other theory; and repugnat conveys both
ideas ; it ‘fights away’ from the first theory, and asserts, equally
combatively, the contrary. Take again IV. 500, Eé sv non
polerit ratio dissoluere causam Cur ea quae fuerint tuxtim quad-
rata, procul sint Visa rotunda, tamen praestat rationis egentem
Reddere mendose causas utriusque figurae. You try to explain
the reason why square seems round, not to explain away (dis-
soluere). What then do you explain away? The fact that
square seems round. But as both notions, the negative of
explaining away a fact, the positive of explaining the reason of
that fact, are only the obverse and reverse of one medal, the
wey and Se of one notion, Lucr. combines them in dissoluere,
and makes a sentence which is prima facie illogical, how-
ever easy it may be to overlook the difficulty, by the use of some
tox media like ‘protest.’ This is true of parctt in hostes,
whether translated ‘spares it for his enemies’ (Munro), or, as I
should prefer, ‘reserves it to attack his enemies;’ either ver-
sion conceals the change which it has undergone, from its
natural meaning of ‘ withholds it against his enemies,’ 1.e. with-
draws it so as not to attack them, to an unnatural one of
“reserves it, so as to attack.’
116. To the passages quoted by me in defence of the MSS.
reading, may, I think, be added iv. 668, Fit prius ad sensum
quae corpora conueniebant Nunc non conueniant et cetera sint
magis apta. Lachm. alters Fit to ut; Prof Munro inserts ué
before quae, and says that for years he has considered the
Omission of u¢ in such cases impossible. Allowing all weight to
his authority, and it would be difficult to name any greater,
I cannot help rejoicing in the admission of his former doubts ;
doubts which are not quite extinct still, as his critical ndéte on
0. 1004 shews. There the MSS. read Inde aliis aliud coniungit
et efficit omnes Res ita conuertant formas, which Lachm. alters
to contungitur et fit ut omnes, Munro to contungit et effit ut
omnes, with these words: ‘no editor before Wak. would tolerate
the omission of ut: yet it is a strange thing that our MSS. so
often omit ut after eficere and jfieri, if the omission is not the
poet's own.’
129. Lachm. says that Lucr. is here speaking of the rend-

206 THE JOURNAL OF PHILOLOGY.

thing is of course true of proutncta, prouintia; the pronup-
ciation so far approximated, that both forms existed side by
side, though one, as having more authority, is preferable, and
I am quite ready to give way to Prof. Munro’s infinitely greater
experience in favour of proutncia.

285. Prof. Munro’s change of utdeantur to uideatur is a
very slight one, and is no doubt supported by many similar ex.
amples in Lucr. and elsewhere; still the change is not neces-
sary, and if not, lumzna is the nearest subject, though possibly
it is the combination of causes, wind and fire, which together
forms the nom. to urdeantur.

475. Whether omnis ratvo=‘the whole principle’ or ‘every
principle, as in omnis diuum natura, ‘every divine nature, I.
646, the meaning of the line seems to me substantially as I
translated, ‘in every case.’ Omnibus would probably have been
the prose construction; but omnis is, if I am not greatly mis-
taken, equally good Latin.

483. I should rather have said that «lls was aestui than
halitut; but I still think ¢lla weak, and tli intelligible.

490. Tam magni montis tempestas atque tenebrae Coperiant
maria ac terras, ‘if such huge mountains are the storm and
darkness which cover earth and sea,’ 2.¢. ‘if in such huge moun-
tain-masses gather the storm and darkness which cover the
sea;’ a construction no doubt more common in Greek than
Latin, but confirmed, I think, by another passage, rv. 140,
Interdum magni montes auulsaque saxa Montibus antire et solem
succedere praeter Inde alios trahere atque inducere belua nimbos,
where the mountain-like masses of cloud are described equally
personally. The constr. is quite the same of Aen. 1x. 182, Zo
milia gentes Arma ferunt Italae, ‘so many thousands are the
Italian tribes that advance in arms.’

548—551. It is surely premature to say that mihe cumgut
salue Rite uocanti is the only one instance where cumque stands
independently of a relative or relative adverb: there are maly
good writers whose text has not been critically edited; wb
can say to what extent this may operate in modifying ov!
present rules, or how many as yet undiscovered instances of
exceptions may increase the probability of exceptions know?

ON LUCRETIUS, BOOK FI. 267

already, but set aside as mistakes for want of further confirma-
tion? In the case of cumque, Prof. Munro himself admits that
the text of Lucr. presents many difficulties: for on 11. 113,
Contemplator ttem cum solis lumina cumque Inserti fundunt
radu per opaca domorum, he says, ‘I know no other example of
cumque following cum;’ in II. 21, pauca widemus Esse opus om-
nino quae demant cumque dolorem Delicias quoque uti multas
substernere possint, he accepts the MSS. reading doubtfully; in
I. [20—722, Nam ueluti tota natura dissimiles sunt Inter se
genitae res quaeque, ita cumque necessest Dissimili constare
fgura principtorum, he admits without comment the alteration
quamgue. More definite is Hand (Tursellinus s.u. cumque).
He assigns to it two uses; one with relatives; the other, a more
attique use, in which it stands alone, with the meaning of
(2) quandoque, quoquo tempore, (b) quoquo modo, nearly =in
“nuersum. This second use he finds in Lucr. 11. 20, 11. 113, as
Well as in vi. 85, 111. 548, 1V. 737, vi. 1017, and the certainly
corrupt passage Vv. 312. I agree with his general view, but not
in all his instances ; it cannot, I think, be denied that in 11. 113,
V1. 85, 738, IV. 737, cumque more naturally qualifies the verb
than any other word in the sentence. It may be so in II. 21;
it is not impossible in 11. 721, or, consequently, in v1. 550.
With regard to v. 312 where the MSS. give Quaerere proporro
sihi cumque senescere credas, I propose silicumque senescere pe-
tras, accepting Munro’s Aeraque for Quaerere. I did not men-
tion Lachmann’s arguments against plaustri, because I thought
them inconclusive; his words are ‘immo plaustri non magno pon-
dere concussa, id est leut plaustro. Editores quomodo tecta non
magno pondere patienter ferre potuerint non uideo. But plaus-
tri is separated by the length of a line from non magno pondere,
and these words are in obvious antithesis to tota; plaustri in
fact destroys the balance of the sentence, takes away from the
strength of its ending, and is, as I said, less Lucretian than the
abl. plaustris, followed by its second explaining abl. non magno
pondere,
568. The difficulty of respirare appears to lic in this; in its
more literal sense, it means to take breath after holding the
breath, as divers do when they come up to the top of the water.

268 THE JOURNAL OF PHILOLOGY.

Cic. de Fin. 1v. 23, 64: Quis enim ignorat si plures ex alto
emergere uelint propius fore eos quidem ad resmrandum, qui ad
summam aquam tam appropinquent, sed nihilo magis respirare
posse quam eos qui sint in profundo? And so Cicero talks of
the wind-pipe breathing back and returning the air in the
lungs, de Nat. Deor. 11 54, 136. In this sense ‘to breathe
back again or up,’ implying that the air has till then been kept
in the lungs, is intelligible enough; and as this respiration or
taking breath is a relief to the breather, respirare easily passes
into tho meaning of resting. But how can this be true of the
winds? They blow or breathe out, and then cease to blow;
but this ceasing is not strictly a respiration; they could hardly
be said sptrare, respirare, to blow and abate blowing; if they
respire or take breath, it cannot be by exhaling, which is their
normal condition, but by simple quiescence. But then resp-
rare would mean two different things, (1) to breathe back,
either in the ordinary process of respiration, or after a stoppage
of the breathing, (2) to rest after breathing. This is what I
denied; the passages quoted by Prof. Munro are metaphorical,
and respirare in them is in its second stage of meaning;
whereas in Lucr. it must, one would think, be in its first. And
if so, my view that it contains both ideas, ‘to blow and lull,’
may be right.

573. It cannot be considered certain that pondere is the
real reading of MSS.: pondera is found in B, and has therefore
almost equal authority. Prof. Munro’s examples no doubt show
that pondere is used much in the same way as pondera, but he
has not shown that recipit sedes in pondere is as natural as in
pondera ; and until some more clear instance can be brought,
prefer to follow my instinct in favour of the accus. ‘ Recipit
sedes in pondere is a proper expression, not prolapsa tn pon-
dera;’ but need it be prolapsa tn pondera at all? I took and
atill take an pondera with recipit.

624. The repetition of wents is made much more probable
by the recurrence of the line in v. 388, Nequiquam, quoniam
uerrentes aequora uenti, and again V. 266.

715, 716. The whole point of my remarks on these verses
was to raise the question of construction. In reading the two

270 THE JOURNAL OF PHILOLOGY.

side, the successive stages of a picture. Here the birds, on
reaching the pestilential lake, are seized with a dizziness,
which makes them forget the even motion of wing which
hitherto has borne them along as smoothly and quickly ass
sailing ship; their pinions move convulsively and with pain;
the motion is no longer placid, but broken, with a perceptible
alternation, like the toiling of a rower. This too does not last
long, the same dizziness which first changed their flight from
smoothness to unevenness, from sailing to rowing, gradually
paralyzes them altogether; their neck droops and they fall.
So in Ovid's description of the fall of Icarus, Met. v. 227:
Tabuerant cerae; nudos quatit tlle lacertos; Remigtoque carens
non ullas percipit auras. Oraque caerulea patrium clamantia
nomen Excipiuntur aqua; it is not till the wax wings have
melted, that the oar-like motion is mentioned or thought of;
it becomes conscious as soon as it begins to be difficult; and
the next step is the fall into the sea. Cf. A. A. 1. 89, 90: Ta-
buerant cerae, nudos quatit tlle lacertos, Et trepidat, nec quo
sustineatur, habet. Occidit. That the motion of rowing 1s ome
of struggling effort is shown by many passages where it is used
equally metaphorically; ¢.g. in wrepvyov eperpotow epercoperat
of the eagles circling wildly round their eyrie; Eur. L T. 289 4
& é« yitavwv trip trvéovea xai povov IItepois épéooe of a Fu wy
hovering fiercely in the air; Aesch. Supp. 541, "Iw olotpw ¢pe<z-
copeva. |

799. The passages where flustra is mentioned are (1) Pa wul.
Diac. Flustra dicuntur cum in mart fluctus non mouentur, quam
Graect paraxiay uocant. (2) Isidorus de Natura Rerum, xL-¥V.
(Sueton. Pratum de Naturis Rerum, 157 Reyfferscheid), Flees-
trum motus maris sine tempestate fluctuantis uelut Naeuius 1
bello Punico stc ait. QOnerariae onustae stabant in flustris: wé
8 diceret tn salo. (3) Tertull. de Pallio 11. Sic ef mari fides
unfamis, dum et flabris aeque mutantibus, de tranquillo probum,
de flustris temperatum, et extemplo de decumanis inquietat. ‘Si-
milarly the sea has a bad character for being trusted, for,
while the gusts upon it change as often as itself, it passes from
a calm sea to a sea good for sailing, from heaving water to sub-
sidence, and immediately after that from immense waves to

472 THE JOURNAL OF PHILOLOGY.

leaves or stem (it can hardly be a piece of myrtle wood, as
he talks of cutting up a piece of juniper wood and boiling it
in wine, c, CXxIu.), I think Lucretius might say feruida sorbus
without presenting to his reader’s imagination anything so
definite as a boiling service-tree. Sorbum, a service-berry is
as definite a word as can well be; yet Palladius, 1m. 15, 4,
talks of grafting sorba upon sorb- or other trees, where Pliny
more correctly speaks of sorbi, xvul. 75. So cerasus is strictly
a, cherry-tree, cerasum the fruit; yet Propertius says, Hic dul-
ces cerasos, hic auctumnalia pruna; pirus a pear-tree, pirum
a pear: yet Columella says, piros serito (de Arbor. XxXIV.), and
immediately after, Mala aestiua cydonea sorba pruna serio:
cf. amygdala s1 parum feracia erunt, perforata arbore lapuem
adigito (ib.), all tending to show that there was a freer use
existing side by side with the more strict.

954. Prof. Munro speaks as if the passage quoted by him
and Lachm. from Varro, L. L. v. 116, settled the question.
Miiller’s best MSS. give it as follows. Lorica quod e loris de
corio crudo pectoralia faciebant: postea subcidit Galliae ferro
sub id uocabulum ex anulis ferream tunicam. Lachm. changes
Galliae to Gulli e, ferream tunicam to fere tam tunica, and
translates ‘afterwards that of the Gaul (=the Gaulish breast-
plate) of iron came to be included under that word,’ and Munro
accepts this. I suggest, however, that the MS. reading ferream
tunicam may be right, that Galliae is not Galli e but Galha es
and that subcidit is from subcido, ‘afterwards Gaul cut away
from the meaning in making a lorica of iron, and included
under that name an iron shirt made of rings. Galli lorica
then has, gua expression, only a partial support from Varro:
but Lucr.’s text gives caeli lorica. It is possible that caelt
might be corrupted into Galli ; but in Iv. 936, Gallo the MSS.
reading is a mistake not for caelo but callo. But even if
Lachm.’s double emendation is right, would Lucr. have repre-
sented an iron cuirass as the last and greatest exhibition of —
the strength of iron? For this is the natural meaning of ~
the line as it would thus stand; though Prof. Munro, perhaps 4
aware of the difficulty, conceals it by referring Dentque to the=
whole sentence ‘fire which is wont to pierce even the strengths

274 THE JOURNAL OF PHILOLOGY.

urbe finibusque, daturos quod Lars Tolumnius dedisset respon-
deri tussit: but the reason of that may be that the word
is naturally a rough one and would be used therefore gene-
rally in commanding or threatening, directly or indirectly.
In 958, raro corpore nexum, seems to me as certainly right
as metu quae possint numine diuae (Lachm. numin?) 11. 623,
mente fruatur Iucundo sensu, u. 15 (Lachm. menty, ‘without
cause’ Munro). Even in v. 949, guibus e scibant umore fiu-
enta Lubrica profluve larga lauere umida saxa, v. 1410,
Maiorem interea capiunt dulcedine fructum, umorv and du-
cedini, though accepted by both Lachm. and Munro, can
hardly be considered certain, from the tendency to an abuse '
of ablatives which, as I said before, is so perceptible in Lucr.
As to the construction, Lachm.’s words seem to mean, that, a8
there is no case in which the condition of cohesion is the
rarity of the body, things being rare, and therefore raro cor
pore; nthil est nist raro corpore nexum is impossible, and must
be changed to nist raro corpory nexu; this would appear to
imply that he thought raro corpore nerum, must mean ‘united
by a rare body.’ But, this is to assume the very point: I
maintain that raro corpore nexum, as properly, if not more 9%,
means ‘possessed of a rare texture of body,’ lit. ‘woven with
a rare body,’ t.e. possessing a rare body, which forms its ter-
ture. Lachm. here, I think, puts a pressure upon language
which it will not bear; much as in the line, efluat ambrosas
quasi uero e nectare tinctus, he thinks to settle the question
by a triumphant ‘oleaster neque ambrosia et nectare effluit 1
que diffluit sed forte affluat.’ But then Lucr. knew the Cyclops
line, "AXAa TOS’ auBpoolns Kal véxtapds éotw drroppat, ad
with this in his memory, first thought of the wild olive 8
steeped in the gods’ drink: and then, having made his solid
sufficiently liquid, ventured to add a word which would be sure
to recal the Greek line, efluat; at the same time that it col-
veyed the notion of coming direct from the fountain-head of
sweetness. I translated this ‘offset’ as an ambiguous ter,
which would suit oleaster, and not be incongruous with the
general idea of a liquid. But having used efluat he might
add ¢ to nectare tinctus; and this is less prosaic than /ncis

ON LUCRETIUS, BOOK VI. 275

ecause it conveys more than one association, and could not be
understood without an admission that the language of poetry
is more complex than that of prose, even when the poet is as
matter-of-fact as Lucretius.

972. Escae is nearer exscet than esca; and surely would
not require guod to be changed to guo; nil est escae is practically
equivalent to nulla est esca. Ill. 498, Qua quasi consuerunt et
unt munita utat is somewhat similar, in the position of the
geuitive, though gua is of course adverbial.

1135. Prof. Munro says, ‘A strange atmosphere comes, say
from Egypt to Athens, which by being breathed engenders
disease. But the sun of Egypt does not travel with it; it
is bright or gloomy, as the climate into which it comes is
bright or gloomy. No; but the sky in any given place may
become unusually bright, and this may be connected with some-
thing unhealthy in the atmosphere, or at any rate may pro-
duce disease by merely being unusual, aliquid quo non consueut-
mus utd.

1199. I thought that ut est following an wt est in 1167 was
likely to be genuine, and proposed to translate it in each place
Similarly ‘as happens’: (so Munro, on 1167), 7.¢ in 1199,
Quorum si quis, ut est, uitarat funera leti, ‘if any of them, as
may well happen, had escaped death.’ I still think ut est
hight, and do not see any necessity for explaining it in 1167
on the forced view of Lachm.; it is certainly used as an inde-
Pendent phrase in Cicero, Fam. xvi. 18. 1, sed ut est, indulge
saletudint tuae, whatever it may there signify.

I take the opportunity to propose one or two emendations.

Lucr. v. 880, 881,

Ez alienigenis membris compacta potestus,

Hine illinc paruis ut non sat (sit A) pars esse potissit.

Read, par urs ut {rent par esse potissit.

It is difficult to say whether sat or non is more likely;
reads sat; on the other hand the opposition of par to non
ar, like that of tdem to non idem, is more forcible. Comp.
sels, 11. 8, ut quod idem est non idem esse urdeatur.
18—2

276 THE JOURNAL OF PHILOLOGY.

Aetna 120 (Munro), Nam mille (so Munro) ex tenut uocen-
que agitata necesse est Confluuta errantes arcessant undiu
uenas. For uocemque, read uiolensque: comp. uolentia=w0w-
lentia in 214; uolet, uocet in 246.

Aetna 432, Quamuts aeternum pinguescat et ubere sulphur,
read pingut scatet.

Aetna 294, Pellet opus collectus aquae usctusque mouere
Spiritus, read mouers.

ADDENDUM to 258.

Corssen speaks of ec in his first volume, p. 155, 2nd edition,
as usual in compounds beginning with / in old laws and frag-
ments of old poets down to Sallust and Cicero; and he quotes
twelve instances, Two of these, ecfatus in a line of Ennius ap.
Cic. de Div. 1. 20. 41, ecferunt Heaut. 745, seem not to be sup-
ported by the MSS., though they were very probably so written
by Ennius and Terence. Of the other passages those quoted by
him from Nonius have been collated by me with the excellent
Harleian MS. The first is written et fero; the second ec fen-
mur; the third ec ferant; a fourth not quoted by Corssen, also
from Cicero, ec ferunt. (The MS. gives also ecfere, ecferre,
ec ferre, in the three cases in the same article of Nonius where
ecferre is printed by Gerlach and Roth: and 80 ec ferte, 292. 19.)
It will be observed that the MS. in most of these instances
writes the ec apart like ab alienauerit (Corssen, p. 154); and
this is confirmed by the form which it seems sometimes to
assume, haec, e.g. in de Fato, xv. 35, Tusc. Disp. m. 16,
38, Sest. XLvIII. 102. I do not believe it accounts for the
frequent change of ec to ef.

R. ELLIS.

A PASSAGE IN CEDIPUS REX.

éx 5€ rrvOpévov
éxdive Kotha KAZOpa. 1260, sq.

WHAT xoiAa means is doubtful, but in two passages imme-
diately following, xAg@pa signifies the fastenings of the cham-
ber-door. In IL x1v. 167, we have,
muxwas Sé Ovpas otabmoiow Ernpoew
KAnids KpUTTTH,

he fitted the doors closely to the jambs with a secret fastening,

Le. with a means of shutting, and so of opening, and the

goddess having entered shuts the doors,

Gupas éréOnxe haewwas.

The fastening, therefore, was on the inside.

It would thus appear, that the folding doors, besides the
fastening in the middle, were further secured by fastenings
let into the jambs on the inside, as the Greek doors opened
outwards. In this way, the bolt and its box in the jamb are
correlatives, and we can see the meaning of xofAa in Theoc.
xv. 15 referred to by Wunder,

004 otabua Koirta Oupawr,
where were the jambs pierced for the bolts, i.e. Juno called up
the snakes inside the door, and so they make at once for the
two infants, v. 20, and no mention is made of any intervening
obstacle. But if the bolt and box are correlative’, they may
both be called «Aj6pa, i.e., means of security, and the meaning
of the passage in (Ed. Rex would be, that Cidipus gave the
doors such a drive with his foot, that he not only burst them
in, but dislocated the xotka «AgGPpa—the boxes or sockets in

1 Cf with massy staples

And corresponsive and fulfilling bolts.
Troilus and Cressida, Prologue.

278 THE JOURNAL OF PHILOLOGY.

the jambs—é« wvOévov, from their position. That is, the bolts

acted as levers, and bent the sockets, which were «otAa.
THOMAS MAGUIRE.

QurEn’s CoLLEGE, GaLwar.

TWO PASSAGES IN VERGIL.

Ecl. 1. 68—72.
En unquam patrios longo post tempore finis,
Pauperis et tugum congestum caespite culmen,
Post aliquot, mea regna, videns mirabor aristas?
Impius haec tam culta novalia miles habebit ?
Barbarus has segetes ?

THE general meaning is plain: My land will go to rack
when I am gone. Shall I ever see it again? If I do, the
change for the worse will astound me. That is, Shall I ever, 1
long time to come, be surprised by the state of my hereditary
farm, at the roof of my cottage (which will then be) deterw-
rated—rudely heaped with sod, (shall I wonder) when I se
the diminished crops, where I once ruled undisputed lord (and of
course brought cultivation to a high pitch)? Shall the lawless
pensioner hold as his own, my fields now so cultivated? Shall
the foreign mercenary own ground like this? To justify the
general interpretation we have only to explain v. 70:

Post aliquot, mea regna, videns mirabor aristas ?

by the lines which precede and follow. As to the sped
points:—I take pauperis as a predicate, and in its stnet
sense of diminished in value, damaged, a sense preserved iD
the action de pauperie under the Twelve Tables, mentioned
by Ulpian, D. 9,1. D. 19,5; by Paulus, S. R. 1.15; and by
Justinian, Inst. Iv. 9: viz. si quadrupes pauperiem fecisse dict-
tur, i.e. if a beast do damage, so that pauperis et tuguri = pat
peris tov tuguri. Pauperis appears to be used in its relative
sense in the business-like arrangement with Priapus:

280 THE JOURNAL OF PHILOLOGY.

esse sese memorat H., comparans Grecorum apmafey et mp-
apratew Ta Neyo ” Forb. Le.

So far all is clear; ®neas, when Iulus had made his joke,
plura alludens, v. 117, immediately continuo exclaims Salt:
that is the words mensas consumimus were, as a matter of fact,
immediately followed by Salve. Now, as A&neas did not inter-
rupt Iulus, who had finished, nec plura, the intermediate lines
describe the mental state of Aineas, ea vox—presstt. Nearly
all the difficulty seems to have been caused by stupefactus, the
modern notion of which suggests temporary coma. Hence, the
apparent contradiction between the presumed anesthetic state,
and the quick application of the casual joke—the apparent
contradiction being intensified by the ordinary sense of pressit
rocem. But stupefactus occurs in three other places in Vergil,
in none of which does it convey any notion of anzsthesia, but
quite the reverse: viz.

(1) Arrectae mentes, stupefactaque corda

Tliadum. En. V. 643 sq.
Here, the action of Juno, and the words arrectae mentes, both
exclude any degree of stupefaction. The Trojan women are in
a high state of excitement.
(2) Aristaeus, ingenti motu stupefactus aquarum,
Spectabat, diversa locis, Phasimque Lycumque,
Et caput, unde altus primum se erumpit Enipeus,
Unde pater Tyberinus, et unde Aniena fluenta.
Geor. IV. 365 sq.

The words italicised exclude comatoseness: Aristzeus was
looking with interest and wonder at the eastern and western
rivers in their sources and in opposite points. In fact, stupe-
factus spectubat might describe one

Like some watcher of the skies
When a new planet swims into his ken;
Or like stout Cortez when with eagle eyes
He stared at the Pacific

tingenti motu stupefactus aquarum.
(3) Quorum stupefactae carmine lynces. Ec. vit. 3.
The lynxes were charmed and not stupefied, that is, they

METHODISCHE GRAMMATIK DER GRIECHISCHEN
SPRACHE, Von Rupotr WestrHat Erster Theil. Erste
Abtheilung. Jena. Mauke’s Verlag. 1870.

THIS is the first section of the first part of a new Greek
Grammar. In a very interesting preface, M. Westphal tells
us that it was his original intention to publish a Greek Syntax
only. In this he intended to proceed upon more advanced
principles than those laid down in Hermann’s tract De emen-
danda Graecae Grammaticae ratione, and embody the results of
Comparative Grammar, so far as it can be said that there are
any results affecting syntax. But as the work progressed, he
found it impossible to separate the explanation of the use of a
word from the explanation of its form, especially in the cases.
The plan, at first adopted, of prefixing an account of the form
of a word to the account of the use of it, seemed unsatisfac-
tory as he proceeded with it. For such accounts must of
necessity be brief—too brief to satisfy those who are not ac-
quainted with the changes rendered necessary in accidence by
Comparative Grammar. To such the mere results of the
latest enquiries would appear as dogmatic and unfounded
innovations, unless some explanation were given of the rea-
sons why these changes are not only possible but requisite.
But to introduce long discussions on various points of acci-
dence would disturb the arrangement of the syntax. Hence
the author concluded to separate the two elements, and write
such an accidence as should be sufficient to explain his syntax.
The work has grown under his hands considerably beyond the
destined limits; so much so, that in this first section, comprising
merely the ‘ Lautlehre’ and declensions, we have a volume of
445 pages.

284 THE JOURNAL OF PHILOLOGY.

tic and factitious. In the place of the free growth and develop-
ment of language it presents us with a structure organized
after the will of the grammarian. Perfect arrangement is only
possible among lifeless structures; but in language we must
study not the forms merely but the life.

Moreover, he who would make his knowledge methodical in
the way that M. Westphal proposes would be in danger of spend-
ing some years in the study of Greek without being able to con-
strue a single Greek sentence of the language. For instance,
the account of the vowel declension is followed by a list tn
extenso of all the stems which belong to it, and not till these
are mastered do we arrive at the second or consonantal declen-
sion. This is as if anyone beginning the study of Greek
should learn Modga and Aodyos, and then look out in his lexicon
every word ending in the terminations -a and -os before he
proceeded further in the study of the declensions. Such a one
would be methodical, without doubt ; but his gain would not be
great. His memory would be taxed to the utmost, but his
analytical knowledge of the Greek language would not be in-
creased in the slightest degree. And so here. It cannot be
said that a classification of the stems belonging to the vowel
declension throws much light upon it, or explains any difficul-
ties attending it. There are certain types, xperys, ren, Movoa,
veavias, avOpwrros, véws, &c. When we have mastered these it is
indifferent whether we see them in one example or a thousand,
He would not be a good teacher of Euclid who insisted on
placing all the letters of the alphabet in turn upon the points
of his diagram in order to ensure completeness. No doubt,
we need a complete list of stems, but would it not be better to
gather them all together under a ‘Stammlehre,’ and arrange
them conveniently for reference according to their meaning and
form ?

It is not surprising that M. Westphal has fallen into diffi-
culties in his attempt to carry out such a severely methodic
arrangement. Thus in the first declension-class, he takes the o
nouns before the a nouns, in order to avoid the inconvenience
of teaching ten, Aoyos, &c. and ayaos, ayaOn. And yet it may
be said that o must on any theory be regarded as a modifica-

286 THE JOURNAL OF PHILOLOGY.

the parisyllabic declension ceased to be parisyllabic. Moreover,
the classification cannot be thoroughly carried out. In which
class are we to place words like aidws? They are parisyllabic
in form, and yet M. Westphal includes them in the second or
imparisyllabic class, because he regards them as formed from
stems in ¢, and therefore to be placed side by side with yévos.
The genitive therefore was aidoc-os, and has become aidois
merely by contraction. But what are we to say of av@parrov ?
Is it not a contraction from av@parroio?

ii. The various forms of the genitive singular are a source
of difficulty in Greek Philology. It is not easy to reduce them
all to one form, and yet why should we have two forms for one
and the same case? Oey it is true is sometimes used for the
genitive, so that we have two distinct forms of this case, but
it is also used for relations in which the common form of the
genitive is never used; and the meaning of the suffix is there
clearly different from that of the usual genitive, however nearly
the two may approach in other instances But the different
forms of the genitive, Movons, veaviov, avOpanrov (from avOpa-
moto) and tratpidos, do not display the slightest variation of
meaning. In his explanation of this case Curtius adopts two
original forms: one in -as, in which the vowel was long in
feminine nouns, and another in -sja; and he proceeds thus in
his analysis: Movaajas, Movoa-as, Movons, veamajas, vea-
yiaos, veaviao, veaviov, avOpwirocjo, avOpeoto, avOperroo, av-
Opwrov. warpidsos retains the original form. The 7 when it
occurs may be considered part of the stem, so that we have

Movea-j- as
veavia-)-as
Tatpio-|os

av0pwio-|ajo.

The evidence for these forms is taken chiefly from Sanskrit, in
which the feminine a-stems have a genitive in -djdas, and the
masculines a genitive in -sja. Bopp, with whom Schleicher is
inclined to agree, considers that the masculine nouns in a, like
those in o, had a genitive formed by sa. However this may
be, the number of hypothetical forms remains the same.

288 THE JOURNAL OF PHILOLOGY.

the agreement between the dual dative and genitive in -cv
with the Sanskrit in bhyam is too probable to be set aside until
we have something certain to put in the place of it. The ob-
jection to the comparison, of course, is that we have no relic
of a Greek dual in -fw; nor is the omission of ¢ a common
occurrence in phonology.

iv. In page ix. of the preface, where the author is explain-
ing why his book has taken the form in which we find it, he
gives an account of the difficulties which beset the attempt to
separate accidence and syntax. ‘In treating of the Sema-
siology of the Greek dative, I must give an account of the
peculiar Greek usage which combines the dative with locative
and instrumental prepositions, in opposition to the Latin use,
which never combines the dative with such prepositions. The
Greek said ody zatpi, év marpi, but in Latin cum patri, in
patri is impossible. It is impossible because in Latin the da-
tive is really a dative, whereas what is called a dative in
Greek is sometimes a dative, and sometimes a locative, accord-
ing to the difference of declensions: and therefore it can some-
times discharge the functions of a dative proper, and some-
times of a locative. It is as the latter that it is combined
with prepositions of locative signification. The criterion of
the distinction here drawn is the long vowel. The final « in
qatpt is short and marks the word as a locative: but in patrs
the final vowel is long, the remnant of an original dative ter-
mination ai, and marks the word as a dative proper, which
cannot therefore enter into combination with prepositions of
locative meaning.’

This is no doubt true: but is it the whole truth? If the
Greek dative, so called, takes the preposition as a locative, and
by virtue of the short «, what are we to say of év ofx@ when ev
oixot was possible (we actually find ev [Ipravorot)? In Latin too
locatives like ruri have the long vowel no less than datives
like patri; and they also, like the dative, are never used
with prepositions; on the contrary the locative signification
is just that which seems least to need the aid of prepositions,
within certain limits. The reason of this difference in the use
of cases in Latin and Greek seems to be that as cases become

GRAMMATIK DER GRIECHISCHEN SPRACHE. 289

more vague, the need of prepositions is felt, while they are not
required in those which preserve their original force. Now in
latin the accusative and ablative became vague, and there-
fore required to be further defined by prepositions. In Greek
the genitive and dative also in addition to the accusative lost
their original force. But it is noticeable that the genitive and
dative in Greek both perform functions of the Latin ablative,
which is used with prepositions. It would be interesting to
enquire how far the prepositional use in both languages is
parallel. .

v. M. Westphal would regard the § in pus, Eptd0s as eupho-
nic merely: and this is intelligible, inasmuch as it is sometimes
inserted and sometimes omitted; but it is not justifiable to
hold the same opinion of the 6 in Aapzras, Aaprrados. We
never find the form Aapzay, but the 5 is retained throughout
the declension except in nomin. sing. and dat. pl., where the
omission of it is required by the laws of euphony. Would it
not be more true to say that ayptas, aypiados for instance is a
separate feminine formation from a stem aypwo, which by the
addition of ja, as in the feminines of participles, would give us
ayvia-ja-s, and then by a not unintelligible process, ayprads,
than to regard the j as simply euphonic? Such an analysis
though attended with difficulties, e.g. the retaining of the final
¢ after the feminine ja, would clear up the origin of the 6, and
leave it part of the stem.

vi. Once more, on p. 323 will be found an explanation of
the difficult nouns in -w; on p. 363 an account of nouns end-
in -ws. Aida and jos are treated together with yéAas, iSpes,
warpws, pntpws, and the feminines in -w are regarded as for-
mations from stems in -os. This is contrary to the views of
Curtius, who deserves indeed more attention than he recvives
from M. Westphal in this particular point. We have here an
instance of a tendency which runs throughout the volume, to
group together words of identical terminativu without sufficient
regard to the stems. We may regard the r as euphonic in i¢pas,
just as it is in yapes; but that will not bring the stem into
harmony with the stem of watpws or npws. Moreover the pre-
sence of a euphonic letter constitutes a real peculiarity of

Journal of Philology. vou. m. 19

290 THE JOURNAL OF PHILOLOGY.

declension. Then the theory which regards feminines in -w as
formations with digamma does not rest ‘solely on the four
Ionic accusatives in -ovv. There is the analogous class of
words in -ws, which when compared with their Latin counter-
parts, tratpws with patruus, seem to require the digamma in
their formation, and to be quite distinct from the sigma stems
aides, jas, Apws with which M. Westphal classes them.

These are a few points among very many which the gram-
mar presents for discussion. They will show how independ-
ently M. Westphal has gone to work, how little he has allowed
himself to be carried away by prescription, or what may be
called orthodoxy in grammar, But they will give but a little
idea of the immense amount of materials collected even in the
first part, and the unflinching determination on the part of
the author to leave nothing without illustration or explanation.
The accentuation for instance is treated in a most interesting
manner, and the endeavour made to establish general princi-
ples. Whether we agree with the author or not, the book so
far as it has gone deserves respectful attention, and what is to
come promises to be even more interesting still, It is to be
regretted that the misprints are very numerous.

E. ABBOTT.

NOTES ON THE TRANSLATION OF GENESIS.

The warning of Cain. Gen. iv. 6, 7.

Very much labour has been expended, but without satisfactory
result, upon the concluding verse of the following passage :

“And in process of time it came to pass, that Cain brought
of the fruit of the ground an offering unto the Lorp. And
Abel, he also brought of the firstlings of his flock, and of the
fat thereof. And the Lorp had respect unto Abel, and to his
offering: but unto Cain and to his offering He had not
respect. And Cain was very wroth and his countenance fell,
And the LorD said unto Cain, Why art thou wroth? and why
is thy countenance fallen? Jf thow doest well, shalé thou not
be acccpted? and tf thou doest not well, sin lieth at the door.
And unto thee shall be his desire, and thou shalt rule over him.”
Gen. iv. 3—7.

This rendering is inconsistent with itself, as it may be well
to remark before discussing it grammatically. Sin expectant
does not truly correspond to evil accomplished. “If thou doest
not well,” sin no longer lurks at the door but has entered and
seized its prey’. Some have accordingly understood by “sin”
the punishment of sin. “If thou doest not well punishment
awaits thee.” Whilst others would render, “Si bene egeris,
acceptaberis: sin male, nihilominus sacrificium erpiatorium
pro peccato ad ostium cubat (solebant enim sacrificia poni ad
ostium Sanctuarii), ¢.e. Tibi poenitenti est spes venize.” These
two classes of commentators have recognized one condition of
the problem before us which is now commonly ignored ; but as

8 I suppose that in a paraphrase we sin lieth in wait for thee; yet thou
might read sinnest in place of doest mayest foil him, and avoid sinning.”
sot well, &c., thus: ‘If thou sinnest, See next note.

19—2

292 THE JOURNAL OF PHILOLOGY.

regards the word “sin” it is perhaps better to adhere to the
view which makes it symbolize under the form of a wild beast’
the principle of evil. If Cain successfully resists temptation
he “rules over” sin: if on the contrary he falls into sin, then
sin tpso facto has dominion over him’; and sin “lies at the
door” only so long as it is doubtful whether or not the man’s
passion will goad him on to evil.

With these remarks I pass on to consider the passage in
detail.

The LXX. reads:

oux éay opis mpoceveyens, dpOas Se yr) StérAns, Hpapres ;
novyacov' wpos aé 1 atroaTpody avrov, Kad od dpteus avroo.

This Greek version is allowed to be very inadequate in many
particulars, but its opening words suggest what is probably the
right’ construction of PRY 3O'R. If now the accents be re-
garded, there appears a symmetry in ver. 6, 7 which the Eng-
lish version obliterates; the original, after the introductory
clause, And the Lord said unto Cain, falling naturally into the

rhythmical form,
7? mn nab

pp woes nad)
new e'nbe xo
JON ND ON

y>7 mxon nnp>
mpwn pow
SISowbn AN

3 Compare 1 Pet. v. 8.

2 Compare Rom. vi. 12, 14: ‘‘ Let
not sin therefore reign in your mortal
body, that ye should obey it in the
lusts thereof...... For sin shall not hare
dominion over you.” Here we have an
exact counterpart of the 13 Seon nAx
of (sen. iv. 7. In neither passage are
we to suppose that the domination of
sin over a man is something subse-

quent to his doing evil. The two are
contemporaneous or coincident.

* And no doubt the simplest. Com-
pare j3) 13°D'7 (Ps. xxxiii. 8). See
for the same and some other construo-
tions, Deut. xiii. 15; 1 Sam. xvi. 17;
2 Kings xi. 18; Is. xxiii. 16; Jer. i. 12;
Ezek. xxxiii. 82; Jon. iv. 9; Prov. xv.
18, xxx. 29.

NOTES ON THE TRANSLATION OF GENESIS. 293

where we have a sequence of ternary lines concluded by two
binaries.

The meaning of the first two lines is clearly: “ Why art
thou inflamed-with-rage, and why hath thy countenance fallen ?”

The next line contains an infinitive of the common word
NO) to lift up ; and the context suggests that it refers to the lift-
ing up of Cain’s countenance which was said to have fallen. If
the falling of the countenance here denotes a yielding to anger
and vexation, the lifting up of the countenance must denote a
recovery therefrom.

It has been remarked above that the LXX rendering sug-
gests the propriety of connecting 3'O'N immediately with
may. An exact rendering of ANY 3'O'N ON is: “If thou
shalt do well to-lift-up'’” The word, “to do well,” when used
with an infinitive thus following, imports the vigorous, skilful,
or successful performance of the action expressed by the verb
which is in the infinitive. In such cases it is sometimes said to
be used adverbially, and may be replaced in English by an
adverb, as in 1 Sam. xvi. 17: “ And Saul said unto his servants,
Provide me now a man that can play®* well, and bring him
to me.” In Gen. iv. 7 the meaning seems to be: If thou shalt
well lift up (sc. thy countenance); i.e., If thou shalt thoroughly
recover (sc. from thy passion).

If the preceding clause has been interpreted rightly, it
scems evident that after the second 3°O’N we must supply
may. Now, making the } disjunctive, we have the alterna-
tives: If thou shalt succeed in lifting up (thy countenance), or
if thou shalt not succeed (in lifting up thy countenance), in
other words:

Whether thou shalt recover from thy passion,
Or whether thou shalt not recover—

the BN)---B8 corresponding as in Ezek. ii. 5: “And they,
whether they will hear, or whether they will forbear (for they

' The force of 3°H'*M might be pre- _ passion.
served by the colloquial form of ex- 3 The infinitive may or may not
pression, to get well over it, sc. thy have 5 prefixed. See note 8, p. 292.

294 THE JOURNAL OF PHILOLOGY.

are a rebellious house), yet shall they know that there hath
been a prophet among them.”

Compare further, Eccl. xii. 14: “for God will bring every
work to the judgment appointed over every secret thing,
whether it be good or evil” (Ginsburg), ine. to see whether tt be
good or whether it be evil’.

We now come to the line, “Sin lieth at the door.”

The alternatives of recovery and non-recovery from passion
having been stated, it follows naturally that siu, like a wild
beast seeking prey, awaits its opportunity: sin lieth at the duor
waiting to see whether thou wilt regain thy composure, or
whether thou wilt not regain it. Man’s passion is the Tempter's
opportunity, and it depends upon Cain’s giving way or not
giving way to his vexation whether or not he is to fall into the
hands of sin. “Cease from anger, and forsake wrath: fret not
thyself [for the result is] only to do evil” (Ps. xxxvii. 8).

The idea is precisely similar in Gen. iv. 6, 7*, not to men-
tion the verbal correspondence of 77M with Ww".

The two binary lines bear a striking resemblance to the
second hemistich of Gen. iii. 16,

srpwn sero
sa wa NT

Here two distinct things are said, (1) that the woman
should be actuated by ardent longing for ber husband, and (2)
that he should have the mastery over her. So in Gen. iv. 7 it
is said in line 6, that sin, under the figure of a wild beast, is
actuated by ardent longing for Cain, a desire to have him for
its prey; and, in line 7, that “thou Cain mayest have the
mastery over him, viz. sin.” There are two ways of connecting
these ideas :—

(1) “To thee is his desire ;” sin longs to have thee for a
prey, “ YET mayest thou prevail over him.”

1 5 DN) JID ON. tempter his opportunity. ‘Be ye angry,

* We have the same combination of and sin not: let not the sun go duwn
ideas in Eph. iv. 26, 27 as in Gen. upon your wrath: Neither give place
iv. 7. Wrath—even righteous indig- to the devil.” Cp. James i. 20.
nation unduly indulged— gives the * IND JR WINN A

NOTES ON THE TRANSLATION OF GENESIS. 295

(2) “To thee is his desire ;” this desire, or its gratifica-
tion, representing one alternative: ‘“‘oR thou mayest prevail
over him,” which represents the other alternative. Thus we
have an introverted parallelism, wherein lines 3, 4 correspond
respectively to lines 7, 6.

The passage as a whole is far from easy to translate lite-
rally, but the meaning which I have attempted to bring out may
be roughly represented as follows :

Why art thou wroth?

And why is thy look downcast ?

Doth not sin couch’ at the door,

Whether haply thou wilt look up,

Or whether thou wilt not look up?

And upto thee is his desire,

And thou mayest have the mastery over him.

PS. I find that Kalisch mentions as a rendering to be
rejected, “whether thou bearest it calmly or not (Solomon).”
This seems to give the construction above advocated. I cannot
say whether I had noticed it before forming my view of the
passage. I have not the opportunity of referring to this trans-
lation.

The sons of God and the daughters of men. Gen. vi. 1—4.

There are three points in this passage which I proceed to
notice :

e

I. Who were the sons of the Elohim ?

Kurtz thus states the leading views: “(1) They are repre-
sented as filit magnatum puellas pleberas ragentes. (2) They
are supposed to have been angels; or (3) pious persons, the
descendants of Seth, while the daughters of men are supposed
to have been descendants of Cain. The first mentioned is the
view of the Samaritan version, of Jonathan, Onkelos, Symma-

* This word is used for P21" in Gen. xlix. 9 and Dent. xxxiii. 13.

296 THE JOURNAL OF PHILOLOGY.

chus, Aben Ezra, Rashi, Varenius, &c., but is at present gene-
rally abandoned.”

It will be gathered from this that the first view has great
traditional authority: I may also remark in passing that the
root Elvhim seems to have been treated not quite exhaust-
ively ; something remains to be said about it which would tend
t» remove the first instinctive objection to the “abandoned”
view of the passage before us; but at any rate we ought tu
allow in translating for a view which is so strongly supported,
unless we are satisfied that the evidence against it is over-
whelming. This might be done by introducing a marginal
reading, “sons of the Elohim,” which exactly reproduces the
ambiguity of the original.

II. Who were the Nephilim? Were they the offspring of
these sons of Elohim and the daughters of men? or did their
existence merely synchronize with the unions spoken of? The
Authorized rendering of ver. 4, wherein the Nephilim are men-
tioned, is obscure and unsatisfactory :

“There were giants in the earth in those days; and also
after that, when the sons of God came in unto the daughters
of men, and they bare children to them, the same became
mighty men which were of old, men of renown.”

The Hebrew runs:
Me ON OA DD pwa ys DAI

pa oy pt ma Se DONT 123 WS WN
:pem wx phyp swe on AD

How is the first WS to be taken? Why not in the sense
whom, as an accusative after 1° thus:

“ [hom—the sons of God went in unto the daughters of
men and—they bare unto them” ?

It is scarcely necessary to remark that the particle "WN
may impress a relative sense on even much longer passages
than the foregoing; uor is it any objection to the proposed

NOTES ON THE TRANSLATION OF GENESIS. 297

rendering that a subordinate clause 3) \N3° intervenes. Com-
pere Gen. iii. 11; Exod. vi. 8, xviii. 3,4; Numb. xii. 12, xx. 13;
Deut. xxviii. 68; Nehem. ix. 29; Isai. xxviii. 4; Ezek. xx. 21.
The second passage of Numbers here cited is, as I arrange it,
stnkingly similar in construction to Gen. vi. 4.

savy my Seow 95 09 WN ADD °D ADA
;D3 wp"

“These are the waters of Meribah, which—the children of
Israel strove with Jehovah and—He was sanctified in them.”

There is no difficulty about supposing the effect of the
relative particle WYN to be transmitted, as here, through an
Ethnach. It may be carried on even from one verse to another,
as in Exod. xviii. 3, 4.

Now to return to the passage illustrated—if the proposed
arrangement be right, the Nephilim are identified with the
offspring of the “sons of God.”

“The Nephilim arose in the earth in those days and thence-
forward (ie. from the continued series of marriages); whom
(Le. the Nephilim)—the sons of the Elohim went in unto the
daughters of men and—they bare unto them. These be those
mighty men who from time immemorial were men of renown.”

It will be seen that the punctuation here adopted agrees
better with the accents than does that of the Authorized
Version, which would require the Ethnach to be on D7.

III. The third point to be considered is the rendering of
Wa XV nw/3 in Gen. vi. 3.

A prejudice, to which I must plead guilty, 1s felt against
the view that 03273 is a contraction for DJ WS; but after
considering the renderings by which it has been proposed to
supersede this traditional view, I am driven to the conclusion
that they are very feebly supported. Moreover the objections
to the traditional view are much exaggerated.

Rosenmiiller writes: “O33 plerisque est Particula Caus-

salis ex Preefixis 3, ¥, et 03 composita, sicuti plerique veterum

298 THE JOURNAL OF PHILOLOGCY.

statuerunt; atque verba “Y/3 NIN 03W3 sic vertunt: quia
etiam caro est. Verum ut taceamus [j, etiam, hic plane
otiosum esse, deberet vox ex tribus istis Particulis composita
aliis punctis vocalibus instructa esse, et 1933 efferri. Accedit,
quod hujusmodi Particularum compositiones seriori tantum
Hebraismo, seu potius stylo Rabbinico sint propriz.”

1. Now in the rendering guia etiam caro est, it may be
granted that D1 1s made “plane oticsum,” or at least that
no suitable emphasis is given to it. But why is the emphatic
NY not expressed? The sentence had gone forth against all
flesh: the Divine Spirit dwelt in man, and gave him a pre-
eminence: but he had degraded himself: “he also” like the
brutes “is flesh”: let him perish with them. It cannot be
said that this makes the Ea) otiose.

2. As for the contraction of WW into Y, and withal with
the required pointing, it is found explicitly in Judg. v. 7’,
NIT ‘HOPS “I, “until that I Deborah arose ;” and with
compensation for Dagesh in Judg. vi. 17: “shew me a sign
that thou talkest with me ,"3D ARN.” The only difficulty

about the pointing is in the last syllable, where we have O3
instead of 03. This however involves nothing more than the
ordinary lengthening of a short vowel into its corresponding
long vowel, and the difficulty is therefore not insuperable’.
It is proposed by Gesenius to assume an anomalous infini-
tive form 3%. For this, ‘4 (Is. xlv. 1) and 30 (Jer. v. 26)
are referred to. The latter illustration would be the more
effective, since letters of the same organ are liable to similar
phonetic vowel-changes; but (1) perhaps 2 itself is not an

infinitive’, and (2) the collocation of singular and plural—

And, of later books, in Cant. i.7. | own part, if the contraction be once
* Keil and Delitzsch seem to make granted, I cannot see any further diffi-
no difficulty about the form of D3W3, culty.
and object only to the supposed in- 3 Journal of Philology, No. 8, pp.
congruous emphasis in 0). For my 182, 4.

NOTES ON THE TRANSLATION OF GENESIS. 299

“in their erring he is flesh ”—is extremely harsh’. Fiirst makes
DY an adverb-form.

To conclude, whatever objection may be felt to the tradi-
tional view that D3Y9 stands for DI WRN3, there would seem
to be no other known explanation which could reasonably be
substituted for it. The Authorized rendering is well suited to
the context ; as would be the slightly modified rendering : “ with
(or tn) one who (W839) is, even he (NY O43), flesh.” It
should also be noted that the contraction which is objected
to may after all be comparatively modern. An _ original
Di WN may have come to be pronounced and afterwards
wntten briefly 03273.

The so-called Wixpow of the Ark, Gen. vi. 16.

sanb> mwyn wy

: mbynbp mbsn mos be

It is now commonly assumed that the amaf Neyouevor,
cohar means light; and hence is deduced the meaning aperture
for light: “A window shalt thou make to the ark, and in a
cubit shalt thou finish t¢ above.” In answer to the objection
that a single window would thus seem to be described, and
that too of such small dimensions as to be wholly inadequate,
it is argued that ¢ohar should be taken collectively : “usque ad
ulne longitudinem facias eam, i.e. lucem, /fenestras.” This
would however still leave the description of the “ windows” very
vague, (1) as regards position, and (2) as regards dimensions,
whereof only one would be given. Moreover the collective
rendering, though defensible in the abstract, seems here to
have been devised in answer to objections, and is not naturally
suggested by the context. Others take gohar to mean internal
light, rather than an aperture for light, and refer the words,
“shalt thou finish 1" to the ark, and not to the cohar. But it

1 As Keil and Delitzsch allow. First goes farther and rejects this construction.

300 THE JOURNAL OF PHILOLOGY.

seems unnatural to make the mention of the lighting of the
ark precede the directions given for closing it in at the top and
thus making some lighting apparatus necessary: the order
would be more natural if the clauses, “light shalt thou make
to the ark,” and, “to a cubit shalt thou finish 17,” were inter-
changed. Moreover the repetition of the word ark in the next
clause: “and the door of the ark shalt thou set in the side
thereof,’ 1s on the whole favourable to the view which refers
“IT” to the gohar’: at least, on this supposition it is necessary,
but on the other unnecessary, to repeat the word ark.

As regards tradition, the force of the argument in fa-
vour of the meaning aperture for light is overestimated.
Jewish authorities may favour the meaning light, but they dif-
fer in their application of it. The Greek versions shew signs
of perplexity rather than of agreement.

Theodotion is quoted for the unique rendering @'pay. The
LXX. does not favour the meaning laght, but reads, for: “A
¢ohar shalt thou make, &c.,”

Kal emicuvayov tomoes thy KiBwrov Kal eis whyuv cuvTe-
A€gess auTyy avwbev.

Here instead of the substantive ¢ohar’ we have the parti-
ciple éxricvvaywr, and the meaning seems to be that the ark was
to contract and grow narrower towards the top. This LXX
rendering may be indefensible as a whole, but it suggests a view ©
which has been proposed by Alb. Schultens, and dismissed per-
haps too summarily by Gesenius (Thesaur. 1152. b) :

“ Dorsum arce i.e. tectum (v. DID viil. 13) intelligebant
° Su~-

Alb. Schult. c. dial. p. 287 et J. D. Mich. in suppl. coll.
Sve
dorsum, %¢6 testudo: sed illud [i.e. the meaning lumen] cum
certo linguse usu magis convenit.”
But the sequence of meanings in Hebrew under the root in

1 Some say that, while the affix is feminina.” Compare noe nk (Prov.
feminine, ‘W1¥ from its form must be xy. 81). In Hos. i. 8, “ph is a woman's
masculine. But Rosenmiiller remarks, pamo.

“plura hujus forms nomina sunt 3 Or (as a paraphrase) for + any.

NOTES ON THE TRANSLATION OF GENESIS. 301.

question is far from being clearly ascertained; while the com-
parison of the Arabic root 4 is wholly favourable to the view
of Schultens. These points will be considered in the sequel,
meanwhile I proceed to shew that the meaning roof is very
suitable to the context of gohar.

I. The argument from the context.

The ark is thus described. “The length of the ark shall
be three hundred cubits, the breadth of it fifty cubits, and the
height of it thirty cubits. A gohar shalt thou make to the
ark and to a cubit shalt thou finish it from above; and the
door of the ark shalt thou set in the side thereof; with lower,
second, and third stories shalt thou make it” (Gen. vi. 15, 16).
Here (1) I have transliterated the disputed word "WY, (2) I
have departed from the Authorized Version by rendering
mor See literally, TO a cubit, and (3) I have rendered mbyndy
FROM above’, again departing from the Authorized Version,
which reads simply, above.

Now according to the usual interpretation we have, to begin
with, an exact statement of the dimensions of the ark, its
length, its breadth, and its height. Then comes a notice of a
window or windows, whereof only one dimension is given;
neither is it made clear where they are to be placed, as wit-
ness Kalisch: “It was to be provided with a door at the side,
and with windows in the upper part (ver. 15), or the roof
(viii. 13).” But if the ark was to be exposed to a heavy rain-
fall the roof would not have been a very natural place for
apertures*; and it would appear from the context that the ¢ohar

1 I do not lay stress on this, but
only on the rendering “roof.” See
also p. 327.

7 Some, as Lange, have conjectured
that the ark was glazed. ‘ We sup-
pone, therefore, with Baumgarten, that
it ["WT¥] must be regarded as a light-
Opening in the deck, which was con-

tinued through the different stories.
Against the rain and the water dashing
must this opening have been closed in
some way by means of some trans-
parent substance; for which purpose
a trellice o: lattice-work would not
have been sufficient. The expression
‘to a cubit’ denotes also precaution."

302 THE JOURNAL OF PHILOLOGY.

was not to be in “the side,’ whereof the mention comes in
quite supplementarily in the immediate sequel. Moreover, ac-
cording to the received view no plain mention is made of that
very important part, the roof. If however cohar itself means
roof, everything becomes clear: “Thou shalt make a sloping
roof to the ark, and this roof shalt thou finish off to the extent
of a cut measuring from above.” The roof was to project and
depend to the extent of a cubit from the top of the sides of
the ark: it would form eaves, under which doors or windows
would naturally be placed; and after the provision for a water-
shed by which the sides of the ark would be protected, we are
prepared for the direction which now follows: “the door of the
ark shalt thou set in the side thereof.”

IL On the Hebrew root WY.

We have next to inquire into the sequence of meanings
in the words:

noon pvr
otl ws

make ol (7) iT.

and to consider whether there is any thing in Biblical usage
which forbids us to render "W¥ in Gen. vi. 16 by roof

According to the usual view, “noon” is described as the
time of double or most intense light, so that the singular “fi¥

would mean light: “otl” is then thought to be named from
its brightness: and the verb is taken by some to be a deno-
minative from “V7¥*, and by others to be a denominative from

py . It is not however quite clear that the choice of

meanings for the verb lies between the two thus obtained,
viz. (1) make orl, (2) labour at noon. The verb itself occurs
once only, and all that seems certain about the Biblical ap-

304 THE JOURNAL OF PHILOLOGY.

sun is then farthest from setting. Compare Jer. vi. 4, where
noon is regarded as the turning point of the day. And indeed
the Hebrews must have had an idea of midday’ as they had of
midnight; although we do not actually find BY ‘¥M corre-
sponding to ab on ‘¥n. The Arabs use ia) for the middle
of both day and night. Lastly, if pews properly denotes

midday with reference to a dividing point? rather than to a
maximum intensity of light, it is easy to see how the same
root might come to be applied, as in Syriac’ (see Castell), to
denote midnight as well as midday.

“W1¥"] From this word for “oil” it is usually thought that

the verb “yJ¥ 1s derived, but the form of the substantive seems
rather to indicate that it comes itself from the verb: so First
takes it, making “V3¥ mean fo shine. But there is no evidence
. to prove that the verb means this; nor is the sequence of ideas,
(1) to shine, (2) that which shines, viz. oil, entirely satisfactory.
It would perhaps be more natural to suppose “W¥ to mean
press, or somehow to denote a process of making oil, and thence
to deduce “oil,” as being succus expressus.

In favour of the conjecture that “W1¥* may properly denote

a succus expressus, is its meaning “oil fresh from the press.”
So Gesenius: “ Oleum, idque recens et hornum (quo differt a
[Dy ut mustum a vino), a splendore pellucido dictum (cf. am
Zuch. iv. 12). Kimchi: SAX WYTM Aya jo jown NI
ney.” But if “VT¥" means properly oil fresh from the 73°
or treading, in contrast. with DY, its characteristic would
not, I suppose, be brightness: on the other hand, it is not
unlikely, @ priori, that its name would contain a reference to
the process of its manufacture, and this favours the view that
the true sequence is (1) press, (2) succus expressus. There is

1 Compare peonuBpla, and meridies ® Whother being that point, or in-
(as usually explained), mittag, midi, cluding it.

mezzogiorno. The Persian for noon is 3 po is used for midday, and
nim roz, half-day, as Mr Palmer in-

forms me. See too on 70, §1V. LAS SOLE for midnight.

306 THE JOURNAL OF PHILOLOGY.

2. The latter word is used however in ver. 14. How is
this to be accounted for? Simply by the symbolism of the
passage. The word “y7¥* is required because the oil is repre-
sented as new oil, coming directly from the tree. “Non dubium
est (writes Rosenmiiller on FWY) interpretes illos intellexisse
effusoria seu epistomia vasis alicujus, ex quo liquor effluit, hic
quidem epistomia lacus torcularis, oleum calcatum effundentia.
Aben Ezra AY WASN vasa esse ait in quibus calcantur olive...

In vv. 2, 3, describitur principium et causa olei, et in vv. 12,
13, 14 subjecta que ab eo succum ducunt (Gussetius).” Now if
the oil is represented as oleum calcatum, and if “V¥* means
oleum calcatum, this is the word which must be used even if the
“oil” is really jD%. This passage then is no proof that wy"

means shining oil, nor does it appear that there is any other
passage which can be adduced to prove it.

TE) This verb occurs once, viz. in the hiphil, in

Job xxiv. 11:
way omy a

SOY IST aap

The view commonly received is that 777¥* means to make
oil, from “VI¥" oil; but, as above remarked, the noun, to judge
from its form, seems rather to have been derived from the
verb. In order to satisfy the requirements of the verse itself,
it is sufficient to make the verb mean generally, to work the
press, and not specially, to make oil: the particular meaning
ow might none the less be appropriated to the derivative "yyy".
A comparison of vv. 6, 10, 11, in some degree confirms the
conjecture that in ver. 11 there is no reference to oil. ‘ In the
field they must reap his (the wicked man’s) gratn, and gather
the produce of the wicked man’s vineyard, (ver. 6).... They (that
is the poor) go about naked, without clothing, and hungry are
they, when they carry the sheaves. The wicked man’s sheaves
(ver. 10).... They press out otl(?) within their (the wicked men’s)
walls; they tread their wine presses, and yet suffer thirst (ver.
11).” See Bernard's Job. Here ver. 6 describes the out-door

308 THE JOURNAL OF PHILOLOGY.

ing. They leave the student with the impression that the com-
parison favours the assumed’ root meaning light, when in fact it
goes decidedly against it. Under 4 we find a great variety of
meanings which come at once from the idea dorsum, or upper-
most surface. For such meanings see Castell, Freytag, &c.; or
for a more complete list. see the Muhtt el Muhtt, written mn
Arabic, by Bust&nf, large edition, Beyrout. We find such
meanings as to spurn, cast behind one’s back, to mount or be on
the back or top of anything, (e. g. a house), @ stirrup, as used In
mounting, te recite memoriter, sc. on the back or “ tip” of one's
tongue, to ticket a thing writing its price on the back of it, &.
&c. We have also the meaning NOON, from the sun’s culmina-
tion’, and various other meanings of which the origin is suffi-
ciently obvious. It may be well to consider more at length the
particular class of meanings adduced in the Hebrew lexicons.
To explain the meanings prodiit, manifestarit, and the like,
it is altogether unnecessary to go out of the way and assume
& new root meaning “splendour.” We only want the idea of
externality, which the lexicons give plainly enough ; thus Freytag
(and see Castell),

S$ -
“ plb Apparens, conspicuus, externus.

ee
\alb Extrinsecus et manifeste.”

If . means back or surface, as it actually does, and if it is
constantly opposed, as it is, to a) (Heb. JOA, belly), we get at
once the contrast, externus, tnternus (used of sin, of meanings of
the Qoran, &c. &c.), a applying to that which is shut up,
inclosed, and therefore obscure, secret, profound, while xe ap-
plies to that which is on the surface, outside, and therefore mani-
fest, or to what is literally “ superficial.” The same contrast has
a variety of simply physical applications: 4 and ky are used
of the high and the low parts of a wady, i.e. of the raised sides

ve

and of the gl) or gravelly part where the water flows or per-
' Bee p. s. page 812.

NOTES ON THE TRANSLATION OF GENESIS. 309

colates: also of the outside of a garment in contrast with the
lining; of the two sides of a quill; of the hand, &c. &e.
In conjunction (like Alpha and Omega) the words imply
totality, being used of God. These applications are easily ex-
plained if x means simply outside, but if it meant bright},
splendid, it would not contrast so symmetrically with ye We
may conclude then that the meanings manifestus fuit, manifes-
tavit do not at all favour the assumed meaning splenduit. In
these as in its other applications the root is pervaded by the

meaning DORSUM.
I may add that Mr Palmer considers this to be a fair state-

ment of the case.
It must now be asked whether \W¥ and |. are really to be

identified. We have seen that both parties agree in the iden-
tification—Gesenius and Fiirst endeavouring to support by it

their meaning “splenduit,” and Schultens quoting ‘& dorsum.
Now there is an affinity between the letters ¥, , so that the
words are related in form’: they have also in common the strong
permanent meaning NOON. The Arabic form moreover appears
in the Chaldee, where it is allowed to be related to W¥. Thus
Buxtorf :—

“SVT, VIO Meridies, Medium diet. Derivatur ex He-
breo V7T¥ et BVI, commutatis § et ¥, ut fieri solet.”

There is also in Syriac 13014, and in Samaritan the con-
necting form O'V30. There seems then to be good reason for
identifying ee and “W¥; and we have seen that the all-per-
vading sense in the former is DURSUM.

IV. On the Chaldee "Wt, “ Medium?”

Buxtorf gives “ medium” as a meaning of “VO; and if this
were correct it would be natural to derive “V1'O in accordance

’ See p.s. page 318. Secu
3 The singular of D'INY¥ agrees even in vocalization with je: as ¢.g. does

pe win OF.

310 THE JOURNAL OF PHILOLOGY.

with its explanation sc. meridies, medium dies’. But see Bux-
torf s.v., and on the other side Levy. Not to enter upon a long
discussion of this point, I remark in passing that the saymg
quoted by R. bar Shila m the Gemara, 7275 “YH NVTD WO
NDT, when the “noon” shines it 1s the middle of the day, testi-
fies to the fact that “VW was actually conceived of as midday,
whether or not it meant etymologically midday.

V. On WO, to purify.

Under the Chaldee root “WWM we find the meanings, Purga-
tio, nitor, aqualiculus in balneo; Medium’, meridies. But
since & stands for the two Arabic letters b, &, it naturally
occurs to us that the meanings of two distinct roots may have
been here united under the one form W70. The Arabic 4,
means, purgatto, &c., and , means dorsum, meridies, &. When
we go to the Hebrew these roots diverge still further, for we
find “VW, purgatio, and BWI, meridies; and the forms "\",
‘VT¥ are not so strikingly similar as to make it seem necessary
that we should identify them. It appears rather that in their
later forms they have converged, and thus they may have come
to be regarded as more closely allied than they in fact are: nor
is it difficult to see how this convergence may have been
effected, for (1) as regards form, ¥ often passes into & in
Chaldee, and (2) from “mertdies” and also from “ purificatio”
may be derived the meaning brightness ; and thus, both in form
and meaning, “WW¥ and “WWM might approximate. The next
step is to reverse this process, and assume that they have

1 If W¥ meant medium, we might
suppose DY TINY (=O “YM, corre-
lative to AOSTA *¥N) to become con-
tracted into a spurious dual form
DvIN¥. So with DYINY. D'VW. In
Prov. vii. 9, OVW actually occurs.
But a more probable explanation of
the dual form is the following. Sup-

pose the two slopes of the sky, from
the place of sunrise to the zenith, and
from the senith to the place of sunset,
divided into a certain number of parts:
then the time during which the sun
traverses the two highest parts, one
of each slope, constitutes the OWS,
or fro noons.
* But this doubtful. Sce g IV.

THE JOURNAL OF PHILOLOGY.

312

Or again, for the sequence dorsum, oil'—in nature the facts
of flutdity and declinty go tegether: hence we might expect
that in language the corresponding ideas would go together. Ac-
cordingly under one root, viz. dee, we find (in Freytag): “ As-
cendit, descendit, liquefecit, sublimavit, suprema pars montia
difficilior ascensu, superficies terre.” Compare jo0, 6, ligua-
vit, fastiguum monits”,

Or, again, “\W¥* might mean, that which rises to or is
skimmed from the surface, like cream, as opposed to sediment.
Or from the idea of two slopes running up into a ridge (dorsum)
might come (1) contraction, (2) compression, &ec. &c.

p.s. For the ancient names of the divisions of day and
nicht? Mr Palmer refers to the following verses by Nasff el
Y&zijf, Maymd el Bahrein (Beyrout), p. 35:

2 eit SW eel ye LS
ae Sly ‘a ia ae ggeall Gat ol
JaF ry illy pat, Jalil oF pall ull
Now ¢€ yi is explained as “being high*, not yet begun to

decline”: Jly; means declension: and between these comes
4,46, which I understand of the moment at which the sun has
culminated, and is at the point of descending. The reference to
ascent may be dubious in the earlier y's but of the later

hours one is named from (hel, root, while yoo means, to descend.
Thus we see that there is a distinct reference to height or depth

+ Compare Bat ‘‘ oleum quo ungi-
tur, pluvia levior superficiem modo
torre humectans:” xpluw, ‘strictly to
touch the surfuce of a body slightly...
hence to rub, anoint with scented un-
guents or oil.”

* And see end -

3 As comparatively modern authori-

ties compare the list of treatises mark-
ed R. 13. 16, in the Trinity College
Catalogue of Arabic, Persian and Turk-
ish MSS. Here we find elaborate di-
rections for determining the hour of
the day by observations of the sun.

* Freytag begins with Altus fuit.

5 Freytag begins with, emicuit elata-
que fuit illustrior diei pars.

314 THE JOURNAL OF PHILOLOGY.

absolutely reject the singular rendering and all agree that it
would be at least exceptional. This being the case, the singular
rendering ought to be expelled from the tezt, unless the meaning
of the passage is on the whole clear enough to force upon us
a very exceptional usage of a common word.

The Hebrew of vv. 25—27, is as follows:
29 INN ODN"
sym ay may toy
my oe ot no oN
swoosy 29 mM
ow ‘Sera pen ne cnbs nee
10> Tay ID TM

First consider the last verse by itself. If this could be con-
ceived of as entirely isolated, we should not hesitate to take the
‘5? as a plural referring to the two individuals mentioned in
the preceding hemistich. Japheth and Shem having been men-
tioned, it is added that Canaan should be a servant to them, viz.
to Japheth and Shem. It would neither occur to us to make
{D5 a singular, nor to conceive of “Shem” and “Japheth” as
used collectively for the Shemites and the Japhethites'*.

Next take ver. 25 in connexion with ver. 27. In the one
Canaan is cursed and assigned as a slave to his brethren: in
the other, Japheth and Shem being mentioned, it is said that
Canaan shall be their slave. In these verses, taken apart from
ver. 26, it would seem obvious that the two second hemi-
stichs were substantially identical, the clause ‘Canaan shall be
their servant” being simply a modified expression of, “a servant
of servants shall he be unto his brethren.”

Thirdly let ver. 26 be introduced, the words OY sain
being for the present omitted. Then, giving to the} an asse-

' Even if we say (Keil &c.) that 105 that the singular is required in a trans-

though grammatically singular is exe- lation as opposed to a paraphrase.
getically collective or plural, we allow

316 THE JOURNAL OF PHILOLOGY.

may fairly predispose us to conjecture that they are to be con-
nected exegetically. It is strictly in accordance with Biblical
usage elsewhere to draw such an inference: in this very passage
paronomasia is employed with reference to Japheth: and there
is a peculiar fitness in saying that One who is emphatically
described as the God of Shem, should be thought of as dwelling
in the tents of Shem. May it not be said that, regard being
had to well authenticated usage, the marked literal agreement
which I have pointed out would in all probability have been
made use of unless (to introduce a not very probable supposi-
tion) + had altogether escaped notice? It seems far from
unnatural to suppose that, the name Shem being in itself less
suggestive, the expression God of Shem was introduced to Jead
up to the form assumed by the blessing of Shem. The whole
passage now becomes symmetrical. The curse of Canaan is the
burden of the song, coming in at the end of each verse like the
refrain of Ps. cxxxvi., “for his mercy endureth for ever.”
Shem and Japheth have each a direct blessing, whereas the more
usual renderings give to Shem only the indirect blessing im-
plied by the fact that Jehovah, to whom blessing is ascribed, is
called the God of Shem. And, moreover, as there is here a sym-
metry of substance so there is also a symmetry of form, since
in each case the medium of paronomasia is used. As regards
authorities: “the Targum of Onkelos interprets the Hebrew by
making ‘i the subject of j3&*, and renders it paraphras-
tically MYT IDV ANISY “!"). His Shekinah shall
dwell in the dwelling of Shem (or of the Name). Maimonides,
Rashi, and Aben Ezra, all follow this, though they also allude
to a secondary sense : that Japheth should learn in the schools of
Shem, which is also expressed in the Targum of Jonathan. So
the Judaico-Arabic interpretation of Arabs Erpenianus. The
interpretation, too, must have been very ancient, antecedent to
Targums and Talmuds, as it seems to have coloured everywhere
the poetry and language of the Old Testament. Hence that
frequent imagery of God’s dwelling with his people, or the con-
verse in expression, though essentially the same in thought,
his being his people’s “dwelling place to all generations.” See
1 Kings vi. 13, viii. 29; Exod. xxv.8; Ps. xc. 1; Ezek. xlin. 9;

NOTES ON THE TRANSLATION OF GENESIS. 317

Zech. viii. 3.” See for the above, and for further judicious
remarks in favour of this interpretation, a note by Professor
Tayler Lewis in the English Edition of Lange’s Genesis.

The Vision of Hagar. Gen. xvi. 13, 14.

“ And she called the name of the LorD that spake unto her,
Thou God seest me: for she said, Have I also here looked after
him that seeth me? Wherefore the well was called Beer-lahat-

rot.
We have to discuss,

we Oe An (i)
SOT TS ONT pos pon (ii)
MD ND (ii)

(i) means, Thou art a God of seeing; ie. either, a God
who sees, or a God who is seen. If the former, the Authorised
Version rendering may serve as a paraphrase: in illustration
compare Gen. xxi. 14, “And Abraham called the name of that
place MAY AVY.” The following ‘sm can only mean my seer,

or one seeing me, and this seems to turn the scale in favour
of the active meaning of *&",

(ii) This clause is literally rendered in the Authorized
Version. The accentuation shows that ‘N" is a participle

with an affix of the first person. She calls God a God of pro-
vidence as being one who saw and watched over her though
she had not looked to Him. The only difficulty is in the
combination “JIN MN, which I think does not recur. But,

1 ii Gesen. Thesaur. 843. a, on Gen. xix,
Compare a TW) (is. xvi 7) 26. “Dei providentiam (writes Vata-
VRE would be preferred because of the blue) nunc agnoscere incipit; quam
assonance "1 ‘fT, *RU NN. Or per- prius sibi visa esset fortuito raptari
haps the idea may be that of looking per desertum, nune sentit ac fatetur
after as in following a leader. See pivinituagabernari res humanas.” An-

318 THE JOURNAL OF PHILOLOGY.

“to look after,” is natural enough in itself, and we may illus-
trate it not unsatisfactorily from Ezek. xx. 24: “ And their eyes
were after, "MS YI, their fathers’ idols.”

(111) The meaning of this, as it stands, is plainly, The
well of (or fo) the living one who sees me’. The foregoing
interpretation is now commonly abandoned; it being assumed
that the feeling of Hagar must have been that expressed in
Judg. xiii 22": “We shall surely die because we have seen
God.” But against the inference see the remarks of Lange
in loc. At any rate it is not obvious from the context a priors;
and it is found, when the words are considered in detail, to
involve an alteration of the text, although its advocates are not
all of them aware of this’.

Subjoined is an extract from the commentary of Kalisch:
“Do I even still see after seeing? although I saw thee, I
still live and see the light of day. 1. Om is evidently the

same form which °N*) is tn pausa, just as my become in pause
my Ezek. xxvii. 17. 2. NT or °N" cannot be translated
he sees me, for this would require YW, 3. It signifies vision
or sight, (comp. 1 Sam. xvi. 12; Job xxi 21). 4 03 not

only adds emphasis to the word to which it refers, but intro-
duces a new and stronger notion, and often one of surpnise,
and is therefore here to be translated even (as in Prov. xiv. 20;

Eccl. x. 20; Ps. xiv. 3), 5. bon is in this concise passage
used instead of bon TY (2 Sam. vii. 18), with which it is in
fact always identical in sense; for Bon i is not here but htther,

and is only used after verbs of motion (Exod. i ui.5; Judg. xviii 3;
Ruth ii. 4, etc.); and it signifies, therefore, here, hitherto, to

other opinion is that she knew the * Compare on the other side Gen.
angel to be an angel because he va- xxi. 17 where the voice brings comfort
nished mysteriously so that she could and not fear.

not follow him with her eyes. See 3 First, Keil, &o. notice it and pro-
Gesen. on “Wit psn. For the Val. pose to move the accent. My atten-
gate compare Ex. xxxiii. 28. tion was called to the point by Mr

* See First a.v. ‘fh, Mason when I first read the passage.

320 YHE JsURS4L OF PHILOLOGY.

oem “ee” «Ps ixmcix 49. chaz it means f& die. Nor can we
Wit arzie &om a ws very cid ase of opaz to early Hebrew.
Azan. che pax tense “STH 3s unsuitable to such a phrase as,
=I, I sit Seve!” Merecver. it seems to me that making
SOO wear ~ ive.” reutraiues the evident play upon the idea
of smpe ~viscn” This dtEculty has been felt by some ad-
Veeares of the lateiy adopted meaning of "RO, who have felt
constrained to draz into the words “God of seeing,” the com-
piex meaning. ~A Geil who being seen, those who see him
remain aie. Lange mentions this as the view of Hengsten-
berg and Tuch.

Lasiy, those who render SX AP) "NS in the way ap-
proved by Dr Kahsch. should at least explain their construc-
tion of the words) They assume that ‘NM is a substantive,
(either changing or not noticing the position of the accent );
bat do they take it ta regimine with i, to which the article
is prefixed? Even if we suppose T? and ‘XN to be both
nouns, we should still expect the qualifying noun to be that
which stands in the second place, so that *X") ‘M would mean
a life attended with rion, rather than a rision attended with
life, Le. which one sees without dying. But if we translate the
received Masoretic text, ‘N"1 must mean one who sees me, and
it is then not easy to make ‘7? mean anything else than to
the liring one. The change of an accent is a small matter, but
here it seems to increase our difficulties. To conclude with a
two-edged remark of Kalisch: “If the Hebrew phrase should
be deemed obscure or elliptical it may be remembered that it is
intended as the etymological explanation of a name; and that
in such cases the choice of words depends on the latitude which
the name affords.”

P.S. Rosenmiillersums up: “Qui omnes tamen videntur in eo
errasse, quod "N" ceperunt vel pro 7IN7 ‘3X, vel pro ‘HN AN,
quum tamen vix dubium sit esse idem quod proxime precessit
numen ‘8, mutato tantum, propter pausam, Schva compo-

sito, in analogam vocalem longam.”’ This suggests an addi-

NOTES ON THE TRANSLATION OF GENESIS. 321

tional argument against the view which Rosenmiiller, Kalisch,
&c. defend. It happens that ‘ome is just as much in pause

as om) (ver. 13), and a (ver. 14). Now why does not "89

assume the pause form ¢ as we should expect it would do? It
stands anomalously in pause but not in its pause form: may
" not this be designed to distinguish i¢ the more plainly from
3 , With which, in an unaccented copy, it would otherwise

coincide in form ?

Sarah and Abimelech. Gen. xx. 16.

The words of Abimelech to Sarah have been variously
explained; but of the interpretations which have been pro-
posed none can be said to be thoroughly satisfactory, while
some may be fairly characterized as extravagant. The Autho-
rized Version is open to the @ priors objection that it dis-
regards the punctuation of the original. It also turns the
disputed words of Abimelech into a reproof of Sarah, when
the context would perhaps rather lead us to expect that his
tone to her would be apologetic, although in ver. 10 he natu-
rally expostulates with Abraham: “What hast thou done
unto us? and what have I offended thee, &c.”

The disputed passage is, in the original,

syme5 pps HON enn An (1)
ae ex Os) oy mos > Nn @)
srg 55 ne (3)

(1) Of these three clauses the first is clear; we have only
to notice the emphasis on S'Mxx. Abraham had been described
to Abimelech as Sarah’s brother, and Abimelech now alludes
to him qua brother, and as one who had represented himself
as such.

(2) Ny) not t, viz. the thousand pieces of silver, but, as
above, he, thy brother: Abraham, qua brother of Sarah:
Abraham, as having been so described to Abimelech.

Journal of Philology. vou. 1 91

322 THE JOURNAL OF PHILOLOGY.

"sy MDI] a covering of eyes. To cover the eyes is to
take away the power of seeing: and to “cover” them, meta-
phorically, with reference to any particular matter, is to take
away the power of seeing or understanding the truth of that
matter. Thus in Is, xxix. 10, 11, prophetic vision is taken
away: “The Lorp hath closed your eyes...the seers hath He
covered. And the vision of all is become unto you as the words
of a book that is sealed.” In Gen. xx. 16, I take the covering of
eyes with reference to Abraham's deception as regards Sarah,
z.e. to his attempt to keep others in ignorance of her actual
relationship to him. .

AN] with thee. Abraham by representing himself as her
brother becomes a covering of eyes (or, as we might say, A
BLIND) to all that are with her, +e. to all strangers who may
associate with her or be in her company.

by FN)] bué with any, 2.e. in the presence of any, or with
whomsoever thou mayest associate.

FNDI] thou wilt be recognized, sc. despite his attempts at
concealing thy true status. As in Gen. xxx. 15 we find nn
for FNM, so here we have nn for AND, the regular
form of the 2 pers. fem. past niphal from M5‘, with a \ pre-
fixed, which is here, as it is said, “mere conversiva.” The word
Mm’ signifies indeed in suitable contexts (though only second-
arily) to reprove, but here probably, as many take it, its mean-
ing is to potnt out clearly. Compare Gen. xxiv. 44: “Let the
same be the woman whom the LorD hath appointed out for
my master’s son,” where the Hebrew is 995 MYT PY WIN
JIN:

The meaning of the whole would thus be:—“ Behold I have
given a thousand pieces of silver to thy brother: behold, he
may be for thee a covering of eyes to any that are with thee;
but with any thou wouldest be recognized,"—¢.¢. although
Abraham, represented as thy brother, may serve thee for (or
attempt to make himself) a covering of eyes (=“a blind”) to
any persons who may chance to associate with thee, yet in the
company of any persons whatsoever thou canst not fail to be
recognized: thy true statue as the wife of so great a man

NOTES ON THE TRANSLATION OF GENESIS. '323

and a prophet (ver. 7) must be recognized, though God himself
should have to interpose in thy behalf.

This would make the address of Abimelech to Sarah com-
plimentary and consolatory, as we might expect it to be, for
his ground of complaint would be against Abraham alone, and
he would regard Sarah as one who had suffered wrong through
acting under her husband's directions.

The Blessing of Esau. Gen. xxvii. 39, 40.

“ Behold, the dwelling shall be the fatness of the earth, and
of the dew of heaven from above; and by thy sword thou shalt
live, and shalt serve thy brother; and i shall come to pass
when thou shalt have the dominion, that thou shalt break his
yoke from off thy neck.”

(1) In the blessing of Jacob (ver. 28) the same expres-

sions DYDWT ‘7H and poN" '2MWb had been used. It is
said that they must be taken differently in ver. 28 and ver. 30:
in the one partitively (of the dew, &c.), in the other priva-
tively (without the dew, &c.). But the latter rendering is at
least anomalous grammatically; nor is it a conclusive argu-
ment that “every blessing had already been given away to
Jacob; not dominion only, but also fertility and abundance
had been granted to him; and, therefore, nothing was reserved
or left for Esau” (Kalisch). Jacob was indeed to possess a rich
portion of the earth, but it does not follow that there was no
rich portion left which might fall to the lot of Esau. The
following contrast may be intended. Jacob was to settle in
and cultivate a rich country (cp. corn and wine): Esau was to
live a roving life in a rich uncultivated country, supporting
himself “by his sword.” This difference corresponds to the
difference of their dispositions.

(2) The words
APTA TAN WS TN
present considerable difficulty. The Authorised Version ren-
dering is unsatisfactory in itself: so too is the rendering which
21—2

324 THE JOURNAL OF PHILOLOGY.

merely substitutes wander freely for have dominion. It is
perhaps the construction which has need to be reconsidered.
On the word ‘)9 Kalisch writes that it “denotes the wild
attempts of an untameable animal to break through every
restraint, and revel in unchecked liberty: taken in a figurative
sense, it describes, therefore, well the incessant revolts and
attacks of a ferocious people, eager to shake off the yoke of
servitude or dependence (Hos. xii. 1; Jer. ii. 31).” In the
passage under discussion “t¥| comes aptly before P45, to break
off the yoke, if it denotes the antecedent impulse, the attempt
to break it off Perhaps then we may use the word revolt
as a rough approximation to 9. If now we take the con-
struction of the Vulgate: TZempusque ventet quum excutias et
soluas, we may render very suitably to the context:

“But there shall be [a time] when thou shalt revolt, and
break his yoke from off thy neck.”

The Vulgate here follows the LXX. which gives as a literal
rendering: éoras Sé¢ qvixa day xabédys nal éxrvons Tov Suydy
avToD amd Tov Tpayndov aod.

This construction is no doubt unusual, but compare Hab.
i. 3: “And there are [that] raise up strife and contention,
no” TD) 3 A” “Would not Job xxi. 4 run more smoothly

with a like construction ? THA “¥pn nd YTD BN, and is
there [any reason] why I should not be impatient? ¢.e. simply
“why am I not to, &c.” Compare Neh. v. 2—4, “there were

that said ([]""9N WN &).” Some make the *'}* WR By
of Numb. ix. 20, aliquando.

The last words of Jacob. Gen. xlix.

A. “Simeon and Levi are brethren; instruments of cruelty
are [in] their habitations. O my soul, come not thou into their
secret ; unto their assembly, mine honour, be not thou united :
for in their anger they slew a man, and in their self-will they
digged down a wall. Cursed be their anger, for it was fierce ;
and their wrath for it was cruel: I will divide them in Jacob,
and scatter them in Israel” (ver, 5—7).

326 THE JOURNAL OF PHILOLOGY.

nonnunquam cum tauris comparari, doceant Deut. xxxiii. 17;
Ps. xxii. 13; lxviii. 31. Cfr. Homer, Iliad, ii. 480. Itaque col-
ligimus, "W mpy dici de Schechemo, quia Schimeon et Levi
vires ejus circumcisione infregerunt infractumque occiderunt.
Cfr. xxxiv. 24.8.” The action of OY of course precedes that
of V5 which had been previously mentioned. Compare Job
xiv. 10.

This interpretation gives a very suitable sense to the pas-
sage. It is natural to expect that the reference would be to the
affair of Dinah; and the expressions used, if interpreted as
above, fit in very exactly with this view.

The nuptial contract is made an instrument of violence, and
advantage is taken of friendly relations (cp. }1¥") to disable the
Shechemites by inducing ‘them to be circumcised. If it were
required to express this last poetically perhaps it could not
have been done better than by the phrase "YY “PY, to ham-
string (=disable) an ox (or collectively oxen). There is no
trace in chap. xxxiv. of any literal hamstringing of oxen: it is
merely said (ver. 28) that the cattle were taken possession of.

Lastly, this devclopes a pointed meaning in ver. 6. “Come
not into their secret, &c.:” it is dangerous to enter into close
agreement with them, “for they savagely slew men (lit. a man),
when by the help of a friendly compact they had disabled
them.”

B. “But his bow abode in strength, and the arms of his
hands were made strong by the hands of the mighty God of
Jacob; (/rom thence is the shepherd, the stone of Israel)”
(ver. 24),

eT? YT WEY (i)
Spy’ “YIN *TD (i)
baer pox ys pep

(i) What is meant by the “arms” of his hands? The
word for arm comes from yf, to scatter, sow, spread: “ bra-
chium ab expandendo dictum” (Gesen. Thesaur.). The arm when
acting 18 /1'%)3, “stretched out” (Deut. iv. 34). Now as the
arms are to the body, with reference to extension, so are the

AN INTRODUCTION TO GREEK AND LATIN ETYMO-
LOGY. By Joun Prize, M.A.

ALL who are interested in philological studies must feel grate-
ful to Mr Peile for undertaking to put into a practical form for
English readers the results of the investigations which have
been carried on for many years with such success among Ger-
man scholars. Dublin has of late sent us a contribution to the
same subject in the Ist volume of Mr Ferrar’s Comparative
Grammar; but, previously to this, we had no books in English
which professed to give a systematic view of etymology ex-
cepting the translation of Bopp, and Dr Donaldson’s Cratylus
and Varronanus. Whatever may have been the merit of
these books, they were not very attractive to readers, owing to
their awkward arrangement and the want of good indices and
tables of contents, and they are also now to some extent anti-
quated. With regard to Dr Donaldson in particular, though
we should be loth to treat the dead lion as he has been treated
by a writer in the Academy for Dec. 15, 1870, who speaks of his
theories as being “as obsolete as those of judicial astrology,”
yet we entirely agree with Mr Peile, that “the mixture of the
proven and not-proven, makes his works unfit for students of
Comparative Philology’.”

Mr Peile’s book has no doubt been in part called out by the
introduction into the examination for the Classical Tripos of a

paper in general philology, the nature of which is determined

1 It is to be regretted that in ano-
ther passage (p. 40) we find even Mr
Peile making use of the depreciatory
phrase, ‘‘Dr Donaldson is enabled to
see, &c.” If odium philologicum has
too often characterized the behaviour
of scholars towards their living con-

temporaries, surely none are more
bound to show courtesy and respect to
the memory of those who have done
good work in their day and are no
longer able to defend themselves a-
gainst attack or misrepresentation.

330 THE JOURNAL OF PHILOLOGY.

(2) by explaining the various inflexions, particularly those of
- different dialects, (3) by showing the historical relations of Greek
and Latin to each other and to the Teutonic languages, (4) by
contributing to our knowledge of the prehistoric condition of
the two races. Supposing the interest of the student to have
been aroused by an introduction of the kind I have described,
the next thing, it appears to me, should have been to give a
slight sketch of the evidence on which the general science rests,
and then to state exactly what definitions and axioms are as-
sumed in the exposition which follows. Unfortunately Mr Peile
has chosen to confine himself to the department of ‘ phonetic
change,’ and has thus been able only to treat incidentally of
inflexions. This limitation of subject has, I think, given an air
of exaggeration and one-sidedness to a good deal which he has
written. Even within the sphere of phonetics his book would
have been more practically useful, if it had contained a larger
list of ascertained derivations, and if all of these had appeared
in the index. Again there seems to me a certain want of clear-
ness in the general arrangement: many technical terms receive
no explanation ; others are repeatedly used before receiving an
explanation: principles are assumed not only without proof, but
even without statement, until the reader who takes his first
ideas on the subject out of the book itself 1s utterly bewildered.

I am not here giving my imagination of what might be
the case, but my observation of what actually was the case
with a pupil of my own whom I had recommended to read
the book. Thus he was particularly puzzled by the employ-
ment of various metaphorical terms, such as hard, light, soft,
strong, heavy, weak in reference to sound, when there had
been no previous classification or arrangement of sounds, stat-
ing which should be considered to possess any of these quali-
ties, and no definition of the meaning of the terms themselves
What added to his embarrassment was to find these distinc-
tions insisted upon as the very key to the science of philology.
“Our one sure guide,” it is said in p 8, “in etymology is
never to derive a harder from an casier sound,” and so in
the conclusion of the book it is stated “my main object has
been to point out the common reason of all these changes of

GREEK AND LATIN ETYMOLOGY. 331

language ; to convince you that they all sprang from the same
desire for ease of articulation :” and, from beginning to end,
the one chief source of phonetic change is asserted to be the
laziness which prefers weak, light, soft sounds to the strong,
heavy, and hard.

I do not of course mean to throw upon Mr Peile the re-
sponsibility of this theory, but I think he has given greater
prominence to it than any one else. I do not know of any other
writer who has described it as his main object to show that
all phonetic change is the substitution of an easier for a harder
sound, and that the cause of such change is the natural laziness
of man. I confess that to me, whether this theory be true
or false, it seems to be a very unimportant appendage to the
ecience of language. For instance, I deny both parts of it: I
do not think that the various laws of phonetic change can all be
reduced to the one law that an easier sound is substituted for
the more difficult ; nor, if it were so, should I at all the more be
disposed to grant that the sole psychological cause for this was
laziness. Supposing my view to be wrong, if the theory is as
important as Mr Peile would make it, my mistake here ought to
vitiate all my conclusions as to particular etymologies. But
the fact is that belief in the particular etymologies is in no
degree dependent upon these hypotheses, which are presumed
to account for them, but upon a vast mass of generalized obser-
vations, which are absolutely certain whether we can account
for them or not. No doubt the aim of science and the tendency
of the human mind is always to reduce plurality to unity, to
substitute more general for less general laws: and if we can
be sure that our highest generalization is capable of being uni-
versally applied, it of course furnishes a vantage-ground from
which to carry on further investigations in fields as yet un-
explored. But in order that it may do this, we must be sure of
the universality of our principle: otherwise we fall into the
error of over-simplification, which has so often proved a pitfall
in the path of science.

It seems to me that if philologers, instead of imagining how
phonetic changes might be produced, had noticed how they
were being produced all around them, they could hardly have

332 THE JOURNAL OF PHILOLOGY.

persuaded themselves into the belief that ‘“‘man as a

_ animal is actuated only by laziness;” a belief which I venture
to call even more libellous than that which some would attri-
bute to the Political Economists, that “man as a social animal
is actuated only by self-interest.” If we ask ourselves what
are the causes of the differences in articulation which we hear
around us, we shall find that they may be roughly, classified as
mental, physical and circumstantial. Under the first head
would come excitability, vehemence, nervousness, preciseness,
artistic sensibility, the analogical disposition always seeking
after resemblances, and its opposite, which we may call the
analytical disposition, always seeking after differences; under
the second, dullness of hearing, defectiveness in the organs of
speech; under the third, external influences so far as they act
upon the other two. Thus cold diminishes our power over the
organs of speech, and makes sound less distinct; one who lives
much in the open air, as a country labourer, a hunter, a sailor,
who has to speak loud to make himself heard, will lose the
finer shades of tone which will be retained by those who live an
indoor life. Again, the art of writing and the existence of a
priestly caste are strongly conservative influences. Not to dwell
on this, let us consider what would be the effect of natural dis-
position on the manner of speech. I think it will be seen that
many changes which Prof. Max Miiller and Mr Peile would ex-
plain from laziness are really due to an entirely opposite cause.
If we compare, for instance, a vehement excitable child with one
who is rather slow and precise, we shall find the one in his
burst of eager volubility omitting half the unaccented vowels
or syllables which drop languidly from the lips of the other.
So a brisk man of business clips his words both in writing and
speaking, not in the least from laziness, but to save time and
spare his muscular energy, if he does spare it, for something
more important. If it is from laziness that we have shortened
senior to sir in speaking, it must be from laziness also that we
shorten Mister to Mr in writing, and an Oriental scribe might
trace the same degeneracy in our disgracefully easy characters,
and in the art of printing altogether. It is not really energy
of character which preserves the primitive or traditional sound,

334 THE JOURNAL OF PHILOLOGY.

nation destined regere tmperio populos, and to leave the arts of
peace to others; just as I hold the elaborate vowel system of
the Indians to be a mark of the indolent unpractical life of
the dreamy Oriental.

Still, is it not true that all known phonetic changes may
be described as substitution of an easier for a harder sound ?
I must own I should be surprised if it were true, assuming that
the causes of change were as various as I have endeavoured to
show; and in fact, what with sporadic change, exceptional cases,
and Prof. Max Miiller’s elastic admission, that “lazy people
take the most pains,” the operation of the law is so curtailed
even by its advocates, that its nominal admission becomes of
very little importance. But to one who accepts it blindly
it may be the cause of endless confusion. How are you to
know what is a harder or easier sound? In one page we find
it left to each man to decide from his own experience: p. 2,
‘It is clear to any one who attempts the sounds, that a is a
fuller and stronger sound than 1,” and therefore has a ten-
dency to pass into it. As to this I will venture to say there is
not one man in twenty whose consciousness will agree with the
scale of difficulty which is laid down by the philologers. To
take the letter a (by the way it ought to have been stated,
to begin with, how the letters should be pronounced), the be-
ginner wishing to test its strength from his own experience
will perhaps call to mind such words as art, wt, machine.
He quite agrees as to the a of art being fuller and stronger
than the 2 of zt; but what of the word machine? Is the ur-
vocal a to be considered stronger than the 1? Or his memory
may recall the sounds “ Unaty in Trinaty,” inflicted upon him
by some illiterate (or lazy?) reader of the Athanasian Creed.
Is he still to hold that the @ is stronger than the + which it has
displaced? But then what becomes of the principle that all
phonetic change is from stronger to weaker’? Compare also

1 Tt is really marvellous that one Ferrar puts it on the right ground
who has ever thought of the subject when he says that ‘‘all unaccented
should be so run away with byatheory vowels in our European languages have
as to speak of this a in the slang pro- a tendency to return to this sound.”

nunciation, fellah for fellow, as stronger Comp. Gr. p. 6.
than the original ow (see p. 282). Mr

336 THE JOURNAL OF PHILOLOGY.

ture of the tongue which alone is required for sounding e or 4.
Therefore, according to our definition, o is naturally a stronger
sound than e or 1.” P. 187, “Corssen concludes that 7, thin
though it be, requires for its pronunciation a considerable ten-
sion of the organs of speech differing herein from ¢. This seems
very unsatisfactory. It is this effort required in pronunciation
and nothing else, which is the mark of a strong vowel, and yet
nothing can be plainer than the fact that + is weaker than
@, 0, or %.”’

So far we have only incidental hints as to the meaning
of the terms strong and weak: it is not until the 217th page
that they are fully explained. There it is said that “though
every language has its own scale of strength, which is dis-
coverable only by investigating the facts of the particular lan-
guage, still we can lay down a few broad rules which seem to
be common to all languages, as they depend on physiological
facts. We may assert with confidence that a momentary sound
is stronger than a protracted one. It is, J think, quite clear
that the complete check given for a moment to the breath
must require a stronger effort on the part of the organs of
speech than is needed when there is no perfect stoppage.”
Hard sounds are stronger than soft. The rationale of this
“cannot be shown without entering more into physiological
questions than I propose to do.” “The aspirate is weaker than
the corresponding unaspirated letters...because the breath
heard follows a less permanent contact.” Hard letters differ
in strength according to the length of the air-tube. In the
case of vowels it is the whole exertion of both expelling and
partially checking the breath which measures the strength.
“We have thus got a tolerably definite idea of the changes we
may expect to find among momentary sounds. For protracted
sounds it is less easy to lay down rules. Curtius thinks that m
is stronger than n,” &c. As the aspirate has been so variously
spoken of, we will quote one more passage, p. 299, explaining
how unaspirated letters became aspirated in Greek: “The A is
produced by letting, as it were, a sigh of relief escape after the
pronunciation of a difficult sound.”

Here then we have at last a real definition of the terms

338 THE JOURNAL OF PHILOLOGY.

from the lungs, On the other hand, if I may judge from my
own consciousness, I should say that among English vowels, J,
which is the sound of least volume, and in that sense weakest,
is the one which requires most care and effort. The tongue
is raised so as to oppose a half check and shorten the air-tube
and the lips are narrowed (see Max Miiller, Lect. 2, p. 121).
Mr Peile seems to be aware that his definition will scarcely suit
his own scale of vowel strength, so he adds to the muscular
exertion of the check the initial exertion of expelling the
breath. But is there really any ground for supposing that
this differs in the case of the different vowels? A strong cur-
rent of air may be thrown into an J just as much as into an A,
as may be easily seen if after holding the breath for some time
we allow it to escape in either form. It might even be main-
tained that as the passage is narrower in the case of J the rush
of imprisoned air must be greater. However, we will not con-
tend any longer about the propriety of the term strength, but
accept thankfully the definition now that we have got it; the
strength of a sound varies according to the degree of muscular
effort required to pronounce it.

To make this definition of use we must have a scale of
muscular exertion, and this is supplied by the doctrine that
a complete check requires more effort than a partial check,
and that the exertion is greater, the sooner the check is ap-
plied, in other words the shorter the air-tube. This sounds
reasonable enough, but it is of less use in practice than might
have been supposed, from the uncertainty as to the exact point
at which the check is applied in pronouncing each letter, and
also from the fact that the same letter is pronounced with dif-
ferent degrees of force in different languages or under different
circumstances. Thus we read, p. 180, “/ was a strong sound in
Latin,” “especially powerful when followed by another conson-
ant; p. 237, “s was sounded strongly when initial, weakly
between two vowels.” Again, p. 226, “perhaps the Latin d
was not a true dental, the tongue may only have been press-
ed against the upper part of the mouth ;” and shortly after-
wards, “if r were sounded at the natural place, the top of the
palate, it would be less likely than J to be confused with a

340 THE JOURNAL OF PHILOLOGY.

and substituted, older and younger, whatever it might be, which
involves no hypothesis as to difficulty of pronunciation? It may
be well here to notice some of the exceptional cases alluded to.
Their importance may be estimated from the fact that, as
Mr Peile tells us in p. 11, Prof. Max Miiller proposes to account
for them by an entirely distinct principle, which he calls Dia-
lectic Growth. Though Mr Peile refuses to admit this, yet he
gives a lecture at the end of his book on Indistinct Articulation
of which he says, that “it is possible to alter a language in
another way than by merely substituting an easier for a more
difficult sound. It is possible to pronounce a word without
sufficient sharpness to give each letter its full and proper
sound.” From this indefinite sound a new sound may arise,
“and it is not at all necessary that the new form should be
really easier to pronounce than the old one.” This is very
well, but it is spoilt to my mind by the sentence which follows :
“the old saying is here justified that lazy people give them-
selves most trouble.” If laziness acts so blindly as to prefix,
and insert, the ‘auxiliary’ vowels in Greek, and change ya into
67 and yug into fvyoy (Peile, p. 294 seg.); and if, after doing
its worst for thousands of years, it leaves a language with such
forms as y@av in Greek and ‘twelfth’ in English, how can we
possibly draw any conclusions as to its action? But what a
singular view is given of the history of the world, by this
supposition of a continuous change for the worse, each nation
in its turn lazily dropping the strong sounds of the vigorous
primeval race, the less effeminate only proving their superior-
ity by retarding the rate of phonetic corruption for a while,
but all alike withheld apparently by some law of destiny from
retracing their course ; vestigia nulla retrorsum! One is curi-
ous to know how it was that our forefathers “with no more
flexible muscles than ours” (Peile, p. 18), should have selected
sounds which have proved such stones of stumbling to their
descendants. With regard however to Indistinct Articulation,
I venture to assert that it need have no more to do with lazi-
ness than the indistinct pronunciation of a nasal by one who
has a cold in the head, or the incapacity to perceive a discord
on the part of one who has no ear.

GREEK AND LATIN ETYMOLOGY. 341

Another class of apparent exceptions to the law of degener-
acy (whether to that of laziness I cannot say) is found in the
‘dynamic intensification’ of vowels, of which Mr Peile says
“the weakening of original a into a, e, o by the Greeks was
turned by them into clear gain,” p.6. Another famous example
is Grimm’s law, of which Grimm himself held that “it showed
@ certain amount of pride and pluck on the part of the Teutonic
nations to have raised the soft to a hard, and the hard to an
aspirated letter.” Mr Peile, following Prof. Max Miiller, en-
deavours, not very successfully, I think, to reduce it to a
case of laziness (p. 218). I need not however seek for further
instances. One sentence of Mr Peile’s allows all that I am
here arguing for; “in every speech amidst the greatest amount
of corruption, new forms are still constantly produced by the
inexhaustible vital force of language, nay often with vigour pro-
portioned to the amount of loss to be supplied,” p. 92. Only I
would beg that this principle may be remembered not merely
in dealing with the pigeon-hole ticketed dynamic change, but
also with that ticketed phonetic change. As long as a language
consists only of monosyllabic roots each syllable may have the
same stress laid on it: when the unifying process begins, and
the monosyllables crystallize round an accent, what one vowel
gains in emphasis another must lose, and the same rule holds
in later changes. It is the same human being, acting at the
same time, with the same object, whose actions we treat of
under these two heads. If we are to credit him with ‘inex-
haustible vital force’ in the one case, we must cease to describe
him as all ‘laziness’ in the other.

It is time now to return to our student whose puzzle as to
the use of the terms ‘strong’ and ‘ weak’ caused all this long
digression. In p. 4 he reads that the Greeks changed the final
m into n because they found it an easier sound: immediately
after it is called lighter; in p. 221 Mr Peile says that Curtius
holds it to be weaker, but he regards it himself to be merely
inconvenient at the end of a word. Would it not have been
better not to bring in the words weaker and easier at all, but
simply to say that m takes more time than nm to pronounce and
would delay the voice on an unimportant syllable?

342 THE JOURNAL OF PHILOLOGY.

P.'7. “The hot enervating climate of India weakened the
consonants in Sanscrit to a more than usual degree.” Yet San-
scrit is “eminently conservative” (p. 5); “the oldest known
language of the Indo-Germanic tongue” (p. 18); and yet again
“it is an erroneous belief that it is older than its sisters” (p. 23).
The same inconsistency of expression is found respecting Latin
and Greek. Thus, in p. 20, Greek is said to have “remained
closest to the original language;” in p. 26, it is called “undoubt-
edly the most rich and flexible,” while Latin is “ the most tena-
cious and unyielding of the entire family.” Yet of Greek we are
told that “it is the genius of this language to develope the
vowel-system and allow the consonants to decay” (p. 58); and
of the tenacious Latin (in that remarkable passage which I have
already quoted) that “its vowel-system passed down every step
of degradation—a degradation not too fancifully connected with
the weakening of the Roman character.” And yet again, in
p. 129 “the conservative Latin” is complimented on retaining
the true weight of vowels which had been lost in Greek; in
p. 166 the weakening of the vowels in Latin is said to have
“materially increased the force and precision of the written
Latin ;” while in p. 266, we read “even in consonantal combin-
ations, where the Greek seems to have changed so much more
than the Latin, it is in reality more truly conservative.” I
bring these passages together to show what caution is needed
in drawing inferences from particular facts to the general cha-
racter of a language, and still more in passing beyond the
sphere of language to the general character of the people by
whom it is spoken.

P.10. We have here the first use of what I think the very
objectionable names, soft and hard for the media and tenuts.
There is the less excuse for this, as there 1s a choice of really
expressive names such as sonant and surd, voiced and whispered,
or even the old flat and sharp. My objection to the words hard
and soft is that the metaphor intended is not appropriate: the
difference of sound observable between p and 6 is not like the
difference produced by striking a hard and a soft object, but
rather like that produced by striking wood and metal. In
the next place, hard is liable to be confounded with difficult,

GREEK AND LATIN ETYMOLOGY. 343

and soft with low-voiced, so that the latter would really suit the
whispered, better than the voiced letter.

P. 12. I think the experience of most people would go
with Prof. Max Miiller in denying that ‘glory’ and ‘cloth’ are
often pronounced ‘dlory’ and ‘tloth.’ No doubt a careless
hearer may easily mistake one sound for the other, but they
are perfectly distinct to the speaker; and since dl and él are so
far from being natural English sounds, that we have no words
beginning with either, while gi and cl are among our common-
est combinations, I see no reason why any one should prefer the
false pronunciation, and certainly I am not aware that I have
ever heard it. Though I cannot agree with Mr Craik in de-
riving ‘clever’ from ‘deliver,’ yet the change from dl to cl is more -
intelligible than the converse. The feeling of the Latin lan-
guage seems to have been the same as our own in this respect:
gt and cl are of frequent occurrence, while dl and ¢l are un-
known or extremely rare; in fact the forms latus and its show
how intolerable the sound él was felt to be. One more proof
which may be alleged in favour of gl is that it is a sound, as
I am informed by experienced persons, greatly affected by
infants even before they arrive at distinct articulation of single
consonants.

P. 19. What ground is there for the statement that Latin
is more like Keltic than like Greek ?

P. 23. Is the principle of exphony, which is mentioned here
and in several other passages, to be considered distinct from
the principle of laziness ?

P. 25. Is not the ‘rigorous observance of phonetic laws’
carried too far in the severance of @eds from Deus? These laws
seem to me to be merely the statement of a gencral tendency re-
sulting from many different causes, which, being to some extent
under the control of man’s free will, need not necessarily act with
the regularity of purely physical causes. Every language has its
anomalies, and with regard to the particular case of d =, the
difficulty seems a little exaggerated, as we regularly find this
correspondence in the middle of a word. Curtius gives aiéw
aedes, avOos, adoreus, kc. Nay, even Mr Peile, though here he
denies the possibility of an initial d in Latin corresponding to

344 THE JOURNAL OF PHILOLOGY.

-a Greek 6, yet in p. 42 tells us “no one doubts that ab-do,
con-do, &c. are formed from the root DHA, Gr. 6e.”

P. 27. Is the evidence for ‘vast prehistoric time’ derived
from language-reallyconclusive? Prof.Max Miller gives examples
proving the extraordinary rapidity of linguistic change among
uncivilized races; and in simple patriarchal times, it seems to
me, that any physical inability or trick on the part of the chief
was likely to be caught up by the clan and become hereditary.
One of the reasons given for believing in a vast prehistoric
period, is the ‘flexibility and lightness’ of the Indo-European
roots in the earliest stage known to us: but why need the
primeval language have been heavy and inflexible ?

P. 31. Among many things told of the orginal Indo-Euro-
pean race which are calculated to try the faith of the student,
perhaps their liking for the sounds BH, DH, GH, 1s the most
remarkable. It is some little consolation to find that the philo-
logers who discovered these uncouth combinations: are them-
selves puzzled how to pronounce them. On the whole, it
seems from pp. 55, 262 that we may hope that the-H was
not really anh after all, 2.e. not the spiritus asper, but only
a ‘breath.’

P. 33. It seems impossible to explain the scale of a here
given, unless we take the first a to be merely the ur-veeal, that
which is “least modified by the organs of speech,” which least
requires muscular effort, and is therefore (physiologically) the
weakest of the vowels.

P. 34. The word qualitative should have been explained;
and vocalism in the next page. Inthe latter page; Schleicher's
words are-quoted with approval : “the vowels can express rela-
tion as well as meaning, the consonants are nothing but ele-
ments in the expression of meaning.” I am not sure that I
understand this, but it would seem to deny that the s of domi-
nus expresses the relation of the subject, and the n of tango the
relation of the present.

P. 37. It scems to me that one of the weak points in the
book is the very inadequate list of examples of substitution.
Considering how slight is the attention given to etymology in
the ordinary Latin Dictionaries, full lists of classified deriva-

GREEK AND LATIN ETYMOLOGY. 345

tions would have been of great service to students. Why is
q omitted among the Latin equivalents of K? What is meant
by calling KI the antithet of AK? It should mean bluniness
in that case, not quietness. And would it not have been worth
while to compare the root of xivéw ?

P. 38. ‘Keita: not a perfect, in form, any more than in
sense. Rather say ‘in form analogous to adeiyas, in sense the
perfect of ri@nut, even though xoirn should induce us to agree
in calling it a Present.

P. 40. Why is not Latin tono given under root TA ?

Pp. 44. Should not b be given as a Latin equivalent to 7,
as we have ab, ob, sub’, bibo, &e. ?

“The root IIA, to protect, gave the Europeans the word
patar, a father.”

The converse is much more probable: nature’s word for
father suggested the more general word for protection, and such
seem to be Mr Peile’s second thoughts. See p. 66, 96.

P. 45. It would have been more convenient to have separ-
ated general discussions, such as this on the theory of secondary
roots, from the account of particular words. Is it worth while
recording Pott’s wonderful derivation for ios ?

P. 63. What does Mr Peile mean by speaking of the use
of ép7rw in the general sense of “go” as pecuhar to the
Dorians ? .

P. 65. The remarks made upon the diversity of names for
the sun &c. among the early races, are hardly such as we should
expect from a believer in- Comparative Mythology (p. 51). If
the whole mythology of the Aryan races may be traced back to
mistaken meteorological metaphors, there surely must have
been an original community of names, as well as a pitiable, in
fact a (to me) incredible, “dependence on atmospheric con-
ditions.”

P. 67. It is odd that nasals should have been unable to
stand alone in the original language, as the nasal ma comes 80
naturally to infants now.

1 I am not sure whether Mr Peile it as an example of incorrect aspira-
would connect this with rd, as on two tion. He should at any rate have ac-
occasions (pp. 281, 802) he mentions ocounted for the sin sub and super.

340 THE JOURNAL OF PHILOLOGY.

P. 71. The reign of fancy has certainly not yet come to an
end in etymology, if we are to accept the statement that maneo
is derived from a root meaning thought, because a man may be
“so filled with thought that he stands stock-still.”

P. 75. Mr Peile seems greatly to admire Corssen’s account
of exta and juxta, as he repeats it again in p.197. Without
further explanation it will be utterly unintelligible to the be-
ginner. He may possibly guess that the ingenious ec-ista
means ‘ most outwards, but what has that to do with the only
exta of which he has ever heard, the dictionary meaning of
which is ‘the inward parts?’ What is meant by denominative
in the same page? The only explanation yet given is in p. 41,
which will not apply either here or in p. 114.

P. 82. I do not understand the reason given for transposi-
tion of a vowel withr. It seems to me that if in a word
like @apoos the r is rolled, the preceding vowel is necessarily
slurred and a faint vowel sound audible after the r.

P. 180. Speaking of vowel change produced by assimila-
tion from neighbouring consonants, it is said that o is changed
into u in the word bubus through the influence of the following
6b. Is it not rather the vocalizing of v, bovibus passing into
bubus like denovo into denuo? I hardly understand how it can
be said that “there is a great gulf fixed between these vowels”
(p. 177), when, from the time of the Scipios to that of Quin-
tilian, there seem to have been many words which might be
spelt with o or w indifferently (p. 165), and when in fact we
often find an archaic u changed into o in the Augustan spelling
(p. 182).

P. 165. I am a little suspicious of Latin orthography in-
truding itself into English, and regret te see the form genetive
admitted in the later pages of the book. Surely genitive has
been naturalized long enough to be independent of whatever
spelling criticism may assign to genetivus.

P. 135. Is it necessary to suppose that the -es of cages is
weakened from -o¢? Why may it not have come directly from
as; as we read, p. 131, that the change from a to e was prior in
time, and spread more widely than that from a to 0? The
nominative of the neuter substantive (yévos) appears exceptional,

GREEK AND LATIN ETYMOLOGY. 347

not only when we compare it with its cognate adjective but
even with its own inflexions.

P. 202. ‘Agnitus’ is quoted as an example of the power of
accent to shorten a long unaccented syllable ; but if the original
form were ‘agnotus,’ as implied, then by the ordinary law of
accentuation the accent must have been on the penultimate
and preserved it from being shortened. Indeed ‘cognitus’ is
mentioned in the next page as a proof of Corssen’s law, con-
travening the old law.

P. 104. The term continuous is surely more appropriate
for “I am doing” than either protracted or permanent.

The discussion which follows is not by any means clear; but
it would take too much time to unravel it, and I have perhaps
given sufficient specimens of the kind of difficulties which are to
be met with in the book, and which are, I think, likely to inter-
fere with its usefulness as a handbook. Perhaps it would have
been better if the lecture-form had been given up. A lecturer
is apt to be loose and unsystematic, to run into digressions,
and to have recourse sometimes to devices for keeping alive the
attention of the class, which are hardly deserving of being im-
mortalized in print. For all these reasons I should hope that
in a new edition the volume may be recast, and appear in a
simpler and at the same time a more scientific form.

I may add that I have noticed the following misprints:
xix. 1. 8, for Latin read Later; 1.5 from bottom, for number
read member; p. 2, accents omitted; 52, marg. for specialization
read generalization; 81,1. 1, for s read v; 109, for strengthenod
read strengthened ; 131, 1. 4 from bottom, for are read 18; 148,
L 5 from bottom, for ¢aci read datot; 304, |. 14, for when read

what.
J. B. MAYOR.

‘DECADENCE.’

I aM afraid it is no longer possible to extinguish this barbar-
ous Gallicism which has been accepted now by so many of
our best writers; still I think it would not be unfitting that

348 THE JOURNAL OF PHILOLOGY.

Cambridge, which raised its protest against ‘telegram, should
at any rate express its disapproval of the far less excusable
‘decadence. ‘Telegram’ could plead in its favour the un-
doubted need for a new word, and its own superiority in point
of simplicity and convenience over any of its rivals, as well as
the fact that it seemed to be supported by English analogy,
and could only be proved incorrect by a reference to the laws
of Greek composition. ‘Decadence,’ on the contrary, obtrudes
itself into ground already occupied by ‘decay,’ ‘decline’ and
other words; it is entirely opposed to analogy, no less in Eng-
lish than in the original Latin—compare such forms as ‘ accid-
ence, ‘incidence, ‘coincidence, ‘occident,’ ‘deciduous’—and,
to heighten its barbarity, it makes its entrée into English with
an accented penultimate, that is, if we may assume that Drum-
mond’s line quoted in Latham’s Johnson,

‘doth in decddence fall and slack remain,

is the first example of its use. Dr Latham supposes it to be an
original English compound, but I think there can be little
doubt that it was borrowed from the French, which was itself
derived from the Low Latin ‘decadentia’ like ‘decadivus.’
The earliest examples of its use given in Littré are from
Calvin and Montaigne; but the title of Montesquieu’s famous
work was probably the means of making it generally known.
Thus we find Goldsmith recurring to ‘decadence, though Sir
Thomas Browne in the previous century had made use of the
correctly formed ‘decidence.’ Gibbon, notwithstanding his
French tastes, sticks to the English word ‘decline; and ‘ decad-
ence’ seems to have made little way in England until the last
quarter of a century, when, possibly owing to the influence of
Comte, it came into fashion, apparently to denote decline, and
connote a scientific and enlightened view of that decline on the
part of the user. One cannot ask enlightenment to forget
itself; but might it not learn to etymologize correctly, and,
retaining the same connotation, to use the form dectdence in-
stead of decadence ?
J. B. MAYOR,

HORATIANA.

HORACE, CARM. I 20.

Vile potabis modicis Sabinum
cantharis, Graeca quod ego ipse testa
conditum levi, datus in theatro
cum tibi plausus,

care Maecenas eques, ut paterni
fluminis ripae simul et iocosa
redderet laudes tibi Vaticani
montis imago.

Caecubum et prelo domitam Caleno
tu bibes uvam; mea nec Falernae
temperant vites neque Formiani
pocula colles.

Horace invites Maecenas to come and see him: ‘you shall
drink’ he tells him ‘cheap Sabine wine, bottled by me when
you received in the theatre such tumultuous applause’. Then
without the least connecting link he goes on to say: ‘Caecuban
and Calenian you shall drink: I possess neither Falernian nor
Formian’.

What is the meaning and connexion here? you shall drink
cheap Sabine: you shall quaff the most costly Latin and Cam-
panian wines; such wines do not fill my cups. Bzbes must be
synonymous with potabis, as the words themselves declare, as
well as the ‘ Vina bibes interum Tauro diffusa cet.’ of epist. I,
7, 5, the invitation to Torquatus in that epistle resembling

350 THE JOURNAL OF PHILOLOGY.

in many points that given in our ode. And what force is
there in the evidently designed antithesis of tu and mea?

The older editors, even Lambinus and Bentley, did not
feel, at all events they take no notice of the difficulty. Not
so recent editors. Peerlkamp in his usual fashion pronounces
the ode spurious. In the Rhenish Museum for 1837, p. 598,
Lud. Doederlein perceives the absurdity of Tu bibes, and pro-
poses Tum bibes. This has been adopted by those editors who
are most distinguished for their accurate knowledge of the
language and will not tolerate any solecism of expression, by
Meineke, Haupt, Luc. Mueller among others; even by Keller
in his elaborate critical edition. But if this change sets nght
the expression, it appears to me I confess to sacrifice the
thought and poetical truth. If the words have now any force,
they must imply that Horace will next give him a better wine
than Sabine, but has not the best of all in his cellar. Yet
surely no one who knows Horace will assert that the poet
held any wines to be more costly than Calenian and Caecuban :
‘Premant Calena falce quibus dedit Fortuna vitem’: ‘ Absumet
heres Caecuba dignior Servata centum clavibus, et mero Tinguet
pavimentum superbo Pontificum potiore cenis’. Caecuban in
fact, from its excellence and the smallness of the space on
which it was grown, was the most expensive of Italian wines:
Pliny (xiv § 61) tells us that in his time it had been destroyed
‘incuria coloni locique angustia’, but chiefly by Nero’s canal
from Ostia to Baiae; that however ‘antea Caecubo erat gene-
rositas celeberrima’, while ‘ secunda nobilitas Falerno agro erat’.

It seems to me clear that Horace, wishing to say Maece-
nas can afford costly wines, but he himself cannot, singles
eut in his usual manner Calenian, one of the finest of Cam-
panian, and Caecuban, the most precious of Latin wines, to
Maatch with Falernian, the most famous of Campanian, and
Formian, the next best of Latin wines; and that it would
have answered his purpose just as well, if he had said ‘you
can afford Falernian and Formian, I cannot afford Calenian
or Caecuban’. And this is felt by those whom we might call
the common sense editors, who wish to give a consistent mean-
ing to Horace’s words; but, with far less knowledge of the

HORATIAN A. 351

language than is possessed by the supporters of ‘tum’, fear
to change the words of the manuscripts, but often do not fear
to give these words a meaning which they cannot bear. Among
these editors are Mitscherlich, Orelli, Dillenburger, Ritter and
Macleane. The latter, following in the wake of the others,
thus construes: ‘you may drink, if you please, the richer
wines’: and for this impossible sense of dides refers with
the rest to other futures which have no analogy whatever
to this.

The passage can hardly be right as it stands; and as tu and
mea are in almost necessary contrast, the corruption would
seem to lie in bives. The mutual relation of Horace’s Mss.
cannot be determined, and with slight exceptions they appear
all to have bibes; but on referring to Keller I see that one
of the oldest Parisian Mss., which he designates by A, gives
tides. Assuming this to be an earlier form of the corruption,
it is natural that with such a context ignorant scribes should
change it to dibes. But bides would most nearly represent
not dibes, but vides, as for centuries before these Mss. were
written 6 and v in many words were used almost indifferently :
thus, to give one instance out of a thousand, in Lucr. 11 902
both the Vossian Mss. have bideant for videant.

Tu vides then I believe to be the true reading, with the
sense of ‘you provide’, ‘supply’, ‘can afford’. Dictionaries
shew that the word not unfrequently has this or similar mean-
ings of the compound promdeo: Ter. heautont. 457 ‘Nam ut
alia omittam, pytissando modo mihi Quid vini absumpsit ‘sic
hoc’ dicens, ‘asperum, Pater, hoc est: aliut lenius sodes vide’:

' Relevi dolia omnia, omnes serias’, Horace was familiar with
Terence, and I think it not improbable that this vide and
the very rare use of relevs suggested to him, perhaps uncon-
sciously, his own vides and lew. Cic. ad Att. v1, 3 ‘id autem
ex eo, ut opinor, quod antecesserat Statius, ut prandium nobis
videret’; Tusc. disp. 111 46 ‘eripiamus huic aegritudinem. quo
modo? conlocemus- in culcita plumea,...dulciculae potionis
aliquid videamus et vini’. The following passages too will
illustrate this usage: Ter. heautunt. 670 ‘ Nisi aliquid video,
ne...resciscat senex’; Cic. de orat. m1 2 ‘videndum sibi esse

352 THE JOURNAL OF PHILOLOGY.

aliud consilium: illo senatu se rempublicam gerere non posse’;
ad fam. vil 20, 2 ‘sed valebis meaque negotia videbis’; Livy
xx1I 4, 10 ‘nulla re, quae agenda videndaque magno futuro
duci esset, praetermissa’; Ovid ars 1 587 ‘Inde procurator
nimium quoque multa procurat Et sibi mandatis plura videnda
putat’. That Horace knew the Caecuban stowed away in
Maecenas’ palace, would appear from epod. 1x: Quando re-
postum Caecubum ad festas dapes, Victore laetus Caesare,
Tecum sub alta (sic Jovi gratum) domo, Beate Maecenas,
bibam.

Peerlkamp, contending that the ode is spurious, has a long
note upon the second stanza. He says, what is perfectly true,
that Virgil has vocis wmago, Silius clamorts tmago, and so on.
And then he adds ‘sed nemo unquam eccho appellavit tma-
ginem montis, silvae vel sazt; neque dicere potuit’. But has
not the writer of this ode been able to say ‘ montis imago’? or
is he a second Odris? or, if he is nonnemo and therefore could
not say ‘montis amago’, then this must be due to some Nobody
of a copyist. How does it help to prove then that Herace did
not write the rest of the ode? But in the first place, not only
Horace himself (1 12, 4), but Cicero, Varro and others use
tmago absolutely for an echo: and in the next place the follow-
ing parallel appears to me to justify and more than justify
montis tmago. Virgil, geor. IV 50, says ‘vocisque offensa re-
sultat imago’; but that does not prevent him from saying in
Aen. V 150 ‘ pulsati colles clamore resultant’; vi1I 305 ‘Conso-
nat omne nemus strepitu collesque resultant’. And he has
been followed by the writers of the silver age, prose and verse,
downwards to Ammian (xXxxI 13, 2) in this use of resulto,
Now surely ‘colles resultant’, ‘caelum resultat’ and the like
are at least as strange as ‘montis imago’.

H. A. J. MUNRO.

354 THE JOURNAL OF PHILOLOGY.

And there is nothing in the primary meaning of the word to
prevent its being so applied.
On these grounds I suggest as a corrected text,

Nanta Bosporum
Pronus perhorrescit.

The mariner in prostrate suppliant dread shudders at the Bos-
porus.

Sat. 1. 3. 25. Cum tua pervideas oculis mala lippus tnunctrs.

This line is unquestionably an imitation of the Greek frag-
ment cited by Bentley, ré radXorpiov...caxov ofvdopkeis 1d
S Wrov wrapaPrérrets.

But pervideas does not mean, as vrapafdéres, ‘you over-
look ;’ but ‘you look into thoroughly, keenly.’ (This difficulty
has commonly been got over by interpreting permdeas lippus
after the analogy of cesctor spectas in Sat. 1. 2. 91; but to this
there is the forcible objection, especially in a writer so exact as
Horace, that pervideas must in that case be simply = videas,
and the emphasis of per be lost.)

Now per, pre and pro are perpetually liable to be con-
founded and substituted for each other in MSS. and editions;
see for instance Cic. Off. 3. 75, pervidertt, where Manutius notes
that five MSS. have prev. and eight prov.; or compare the
conflicting prasectum and perfectum in Ars Poet. 294 It
would be easy to add to these instances, But it is needless.
No one would object to reading prevideas with Bentley on
the ground of its being a violent alteration, and there is some
MS. authority for it. Bentley proposed it on the supposition
ut prendeas idem sit ac pretervideas. Is there any reason
against this, except the somewhat curious fact that pretervidere
is nowhere found—a vox nihili? There seems to be some mis-
take as to the meaning of these prepositions, pre and prester.
Prajfiuo e.g. is said to be for preterfluo; precurro for pra-
tercurro. Why? Really they are two modes of expression
containing the same result. The river, gut regna Daunt pra-

356 THE JOURNAL OF PHILOLOGY.

evenio (©. 4! 4 65); examen: (C. b. 35:31); and with these may
be classed his numerous dzraE Neydpeva prodoceo, juvenor,
geterno, claro, inimico, intaminatus.

I have noted at various times usages in Tacitus which seem
to indicate in his style a remembrance and adoption of Hora-
tian language; and if it is allowed that he had any frequent
regard to it, weight should be allowed to this in the argument
for preevideas.

Compare dabat et fama; Tac. Ann:1.7, with Hor. 8. S. 2. 2. 94,
contusis, Ann. 4. 46, with C. 3.6.10; muto, Ann. 12.13. Sat.
1. 4. 29. 2.7.110.; wllarum molibus, Ann. 4. 67. C. 3. 29. 10.
additus, Ann. ib. C. 3. 4. 78.

additis veteranis, Anu. 13. 31, ts'relied on by Orelli as con-
firming the reading addzdit, C. 3. 4. 38.

falsum renidens,- Ann. 4. 60, seems like a prose adaptation
from perfidum ridens of C. 3. 27. 67; breve confintum, Ann. 4,
59 from exiguo fine, C. 2. 1, 19.

The Germania, 18. 19, contains general and verbal resem-
blances to Carm. 3. 24.

J. E. YONGE.

ON TWO TRIPLE READINGS IN THE NEW
TESTAMENT.

In 2 Pet. iii. 10 the Vatican and Sinaitic MSS, supported
by some ancient versions, preserve the reading xal ta éy avty
Epya etpeOncerar. For evpeOnceras we find in C apano6n-
covra:', in A the common reading xataxanceras, Can we
find any explanation which shall account for the strange fact
that the best attested reading is contrary to the sense, and for
the double variation from it?

A rough and ready remedy suggests itself at once. We
may suppose that ovy before evpeOnoeras was accidentally
omitted either in the original or in a copy from which all
our MSS have descended: and that the readings of A and C
are corrections to make sense. But surely copyists would
be more likely to restore the negative than to substitute quite
different words. For the reading of C especially we must not
admit such an explanation without searching for one less
improbable. ‘

Another suggestion is that evpe@jceras has somehow got
into the text from the Latin urentyr. But it is waste of
critical power to apply to a triple variation a hypothesis
which only explains two readings: and this suggestion leaves
agavicOncovrat unaccounted for. Besides, would xataxanoeras
have been translated by® urentur ?

1 Wetstein and Tischendorf quote
Syr. for dgancbycorvra: but I do not
know what can have been Wetstein’s
authority for this: my friends who
are learned in Syriac can find no such
reading. As usual I have failed to

detect any inacouracy in the readings
given by Tregelles.

* Having looked at the passages
where xaraxaley has to be translated
in nearly all the old Latin versions I
find that cremare is employed in ¢ of

358 THE JOURNAL OF PHILOLOGY.

There is however a third solution of the problem which
seems to me to remove nearly all difficulty. It is to suppose
that the original word wupwOncerat gradually became less
and less legible: so that the first scribe who undertook to
copy the epistle found only .yp. eHcetai—the first and fourth
letters having perisht—and wrote down a verb which occurred
to him as containing all the letters he could see. A second
copyist was too late to see yp but in time to see 6; so he
wrote agavucOnoetas (or -covra:)®. A third when only the
termination remained visible strayed a little further from the
original word but returned exactly to the sense with xara-
KanoeTat.

Very similar is the triple variation in Heb. xi. 13; where
Tregelles reads xouicapevos with & P 17, 39: A has wpoodefa-
pevor: the common reading is Aafovrec. Everyone must I
think feel that here xopsoduevos is the right word. The
sower going forth may be said mpoodéxyeoOas or (ver. 17) avadé-
xecOae the promise of the harvest: but it is not till he
brings his sheaves home that he xopiferat rao érrayyeXiac.
We may here say almost with certainty that one copyist
coming first or seeing clearest preserved xoprodpevoe; another
(whence A) saw the end of the word but made a bad guess
at the beginning; others gave up finding out the letters—
perhaps by this time their remains were past searching for—
and supplied the sensible but commonplace Aaféprrec.

These instances of gradual obliteration occur just where
we should expect to meet with them, in epistles which for a
long time were not generally known, and therefore very possi-
bly not transcribed. And in the case of the Epistle to the
Hebrews there seems to be a further reason for assuming that
its existence depended for many years on a single copy. For
the absence of any clear tradition as to its author cannot to
my mind be in any way accounted for except by supposing that

Mat. xiii. 830 and df of Heb. xiii. 11;
exurere in e of Mat. xiii. 40; ardere in
d f of 1 Cor. iii. 15; with these excep-
tions always comburere.

> It must be noticed that the plural
dgancbjcorra: is a slight objection to

my hypothesis: but I think a very
slight one. For there is but little dif-
ference between ceTa! and COTA!:
or—perhaps more probably—the change
to the plural may have been the result
of a later transcription.