Greenwood professor of hellenistic greek and indo-european philology in the victoria university of manchester tutor in new testament language and literature


Download 3.25 Mb.
Size3.25 Mb.
1   ...   15   16   17   18   19   20   21   22   ...   33
are fasting,"

Jn 25 o!ti a}n le

The aorist, being future by virtue of its mood, punctiliar by

its tense, and consequently describing complete action, gets a

future-perfect sense in this class of sentence; and it will be

found most important to note this before we admit the less

rigid translation. Thus Mt 521 o!j a}n foneu

has committed murder," 547 e]a>n a]spa

saluted," Mk 918 o!pou e]a>n au]to>n katala

seized him:" the cast of the sentence allows us to abbreviate

the future-perfect in these cases. Mt 531 at first sight raises

some difficulty, but a]polu

into effect as the determination. We may quote a passage

from the Meidias of Demosthenes (p. 525) which exhibits

the difference of present and aorist in this connexion very

neatly: xrh> de> o!tan me>n tiqh?sqe tou>j no

skopei?n, e]peida>n de> qh?sqe, fula xrh?sqai—tiqh?sqe

applies to bills, qh?sqe to acts.

The part which the Subjunctive plays in the scheme of

the Conditional Sentences demands a few lines here, though

any systematic treatment of this large subject must be left

for our second volume. The difference between ei] and

Conditional e]a
Sentences, istic as compared with earlier Greek. We

Simple, have seen that e]a
General and cative; while (as rarely in classical Greek)

Future. ei] can be found with the subjunctive. The

latter occurs only in 1 Co 145, where the peculiar phrase

accounts for it: cf the inscription cited by Deissmann

(BS 118), e]kto>j ei] mh> e]a>n1 . . . qelh

care to build much on Rev 115. In Lk 913 and Phil 311f. we

probably have deliberative subjunctive, "unless we are to go

and buy," "if after all I am to attain . . . to apprehend."

The subjunctive with ei] is rare in early papyri: cf OP 496

(ii/A.D.) ei] de> h#n (=^#) o[ gamw?n pro

ktl. The differentiation of construction remains at present

stereotyped: ei] goes with indicative, is used exclusively when

past tenses come in (e.g. Mk 326), and uses ou] as its negative;

while e]a

almost invariably, unless the practically synonymous future

indicative is used. ]Ea

express future conditions. This is not only the case with ei]

c. fut.—in which the NT does not preserve the "minatory or

monitory" connotation2 which Gildersleeve discovered for

classical Greek--but even with ei] c. pres. in such documents

as BU 326, quoted above, p. 59. The immense majority

of conditional sentences in the NT belong to these heads.

We deal with the unfulfilled condition below, pp. 200 f., and

with the relics of ei] c. opt., p. 196.

Some Uses of Leaving the Dependent Clauses for sub-

the Negatives :— sequent treatment, let us turn now to some

Ou] mh< aspects of the negative mh< mainly though

not exclusively concerning the Subjunctive.

Into the vexed question of the origin of the ou] mh< con-

struction we must not enter with any detail. The classical

discussion of it in Goodwin MT 389 ff. leaves some very

serious difficulties, though it has advanced our knowledge.

Goodwin's insistence that denial and prohibition must be
1 Cf above (p. 169), on ei] mh2 But 1 Co 314f. cf Hb P 59 (iii/B.C.).


dealt with together touches a weak spot in Prof. Sonnen-

schein's otherwise very attractive account of the prohibitory

use, in a paper already quoted (CR xvi 165 ff.). Sonnen-

schein would make ou] mh> poih

prohibition mh> poih

Similarly in Latin quin noli facere? is "why not refuse to

do?" The theory is greatly weakened by its having no

obvious application to denial. Gildersleeve (AJP iii. 202 ff.)

suggests that the ou] may be separate: ou@: mh> skw

don't jeer, ou@: mh> gea Brugmann

(Gram.3 502) practically follows Goodwin, whom he does not

name. We start from mh< in cautious assertion, to which we

must return presently: mh> geit may perchance happen,

mh> skwyou will perhaps jeer, mh> e]rei?j tou?to = you will

perhaps say this. Then the ou] negatives the whole, so that

ou] mh< becomes, as Brugmann says, "certainly not." Non

nostrum est tantas componere lites: these questions go back

upon origins, and we are dealing with the language in a late

development, in which it is antecedently possible enough that

the rationale of the usage may have been totally obscured.

The use of ou] mh< in the Greek Bible calls for special com-

ment, and we may take for our text some remarks of Gilder-

sleeve's from the brief article just cited. "This emphatic

form of negative (ou] mh<) is far more common in the LXX and

the NT than it is in the classic Greek. This tendency to

exaggeration in the use of an adopted language is natural."

And again, "The combination has evidently worked its way

up from familiar language. So it occurs in the mouth of

the Scythian archer, Ar. Thesmoph. 1108 ou]ki> mh> lalh?si

su<;" Our previous inquiries have prepared us for some

modifications of this statement. "The NT" is not a phrase

we can allow; nor will "adopted language" pass muster

without qualification. In Exp T xiv. 429 n. the writer

ventured on a preliminary note suggested by NP 51,

a Christian letter about coeval with x and B, in which

Mt 1042 or Mk 941 is loosely cited from memory and ours

a]polli?, (sic) substituted for ou] mh> a]poleDidache 15

quoting Mt 526. Ou] mh< is rare, and very emphatic, in

the non-literary papyri. On the other hand, we find it

13 times in OT citations in NT, and abundantly in the

a See D. 249.


Gospels, almost exclusively in Logia. In all of these we have

certain or probable Semitic originals. Apart from these, and

the special case of Rev, it occurs only four times in Paul and

once in 2 Pet. It will be seen therefore that if "translation

Greek" is put aside, we have no difference between papyri

and NT. Paul's few exx. are eminently capable of bearing

emphasis in the classical manner. The frequency of ou] mh< in

Rev may partly be accounted for by recalling the extent to

which Semitic material probably underlies the Book; but the

unlettered character of most of the papyrus quotations, coupled

with Gildersleeve's remark on Aristophanes' Scythian, suggests

that elementary Greek culture may be partially responsible

here, as in the rough translations on which Mt and Lk had

to work for their reproduction of the words of Jesus. The

question then arises whether in places outside the free Greek

of Paul we are to regard ou] mh< as bearing any special

emphasis. The analysis of W. G. Ballantine (AJP xviii.

453 ff.), seems to show that it is impossible to assert this. In

the LXX, xlo is translated ou] or ou] mh< indifferently within a

single verse, as in Is 527. The Revisers have made it emphatic

in a good many passages in which the AV had an ordinary

negative; but they have left over fifty places unaltered, and

do not seem to have discovered any general principle to

guide their decision. Prof. Ballantine seems to be justified in

claiming (1) that it is not natural for a form of special

emphasis to be used in the majority of places where a negative

prediction occurs, and (2) that in relative clauses, and questions

which amount to positive assertions, an emphatic negative is

wholly out of place: he instances Mk 132 and Jn 1811—Mt

259 is decidedly more striking. In commenting on this article,

Gildersleeve cites other examples of the "blunting . . .

of pointed idioms in the transfer from classic Greek": he

mentions the disproportionate use of " the more pungent

aorist" as against the "quieter present imperative"—the

tendency of Josephus to "overdo the participle"—the con-

spicuous appearance in narrative of the "articular infinitive,

which belongs to argument." So here, he says, "the stress"

of ou] mh< "has been lost by over-familiarity." One is inclined

to call in the survival among uneducated people of the older

English double negatives—"He didn't say nothing to nobody,"

and the like—which resemble ou] mh< in so far as they are old

forms preserved by the unlearned, mainly perhaps because

they give the emphasis that is beloved, in season and out of

season, by people whose style lacks restraint. But this parallel

does not take us very far, and in particular does not illustrate

the fact that ou] mh< was capable of being used by a cultured

writer like Paul with its full classical emphasis.1

Let us now tabulate NT statistics. In WH text, ou] mh<

occurs in all 96 times. Of these 71 exx. are with aor. subj.

in 2, the verb is ambiguous, ending in -w; and 15 more, ending

in –ei]j (-ei) or -^j (-^), might be regarded as equally indetermin-

ate, as far as the evidence of the MSS readings is concerned.

There remain 8 futures. Four of these—Mt 1622 e@stai, with

Lk 2133 and Rev 96 1814 (see below)—are unambiguous: the

rest only involve the change of o to w, or at worst that of ou

to w, to make them aor. subj. The passages are:—Mt 2635

(-somai, xBCD) = Mk 1431 (-somai ABCD, against x and the

mob). (The attestation in Mt is a strong confirmation of the

future for the Petrine tradition in its earliest Greek form.)

Lk 2133 (-sontai xBDL) answers to the Marcan ou] pareleu<-

sontai (1331 BD: the insertion of mh< by xACL etc. means

a mere assimilation to Lk), while Mt has ou] mh> pare

(2435): it is at least possible that our Lucan text is only

a fusion of Mk and Mt. In Jn 105 ABD al. support

a]kolouqh17 (from LXX) we have the


emended to mnhsqw? (following the LXX) in correctors of x

and D and all the later MSS. There remains eu[rh

in Rev 96 (AP eu!rwsin, against xB2) 1814. We need

not hesitate to accept the future as a possible, though

moribund, construction: the later MSS in trying to get rid

of it bear witness to the levelling tendency. There is no

apparent difference in meaning. We may pass on to note

1 Winer (p. 634) refers to "the prevailing opinion of philologers" in his own

time (and later), that of ou] mh> poihfear that he

will do it." It is advisable therefore to note that this view has been abandoned

by modern philology. To give full reasons would detain us too long. But it

may be observed that the dropping out of the vital word for fearing needs

explanation, which has not been forthcoming; while the theory, suiting denials

well enough, gives no natural account of prohibitions.


the distribution of ou] mh< in NT. It occurs 13 times in

LXX citations. Apart from these, there are no exx. in Ac,

Heb, or the "General Epp", except 2 Pet 110. Rev has it

16 times. Paul's use is limited to 1 Th 415 (v. infr.) 53, 1 Co

813, Gal 516. Only 21 exx. in all come from these sources,

leaving 64 for the Gospels. Of the latter 57 are from actual

words of Christ (Mt 17, Mk 8 [Mk] 1, Lk 17, Jn 14): of

the remaining 7, Mt 1622 and 2635 (= Mk 1431), Jn 138

2025 have most obvious emphasis, and so may Lk 115 (from the

special nativity-sources) and Jn 1156. That the locution was

very much at home in translations, and unfamiliar in original

Greek, is by this time abundantly clear. But we may attempt

a further analysis, by way of contribution to the minutia of

the Synoptic problem. If we go through the exx. of ou] mh< in

Mk, we find that Mt has faithfully taken over every one, 8 in

all. Lk has 5 of these logia, once (Mk 132 = Lk 216) dropping

the mh<. Mt introduces ou] mh< into Mk 712, and Lk into Mk 422

and 1029, both Mt and Lk into Mk 1331 (see above).2 Turning

to "Q", so far as we can deduce it from logia common to

Mt and Lk, we find only two places (Mt 526 = Lk 1259, Mt

2339 Lk 1335) in which the evangelists agree in using ou] mh<.

Mt uses it in 518 (Lk 2133 has a certain resemblance, but

1617 is the parallel), and Lk in 637 bis (contrast Mt 71).

Finally, in the logia peculiar to Mt or Lk, the presence of

which in "Q" is therefore a matter of speculation, we find of

mh< 4 times in Mt and 7 in Lk. When the testimony of Jn

is added, we see that this negative is impartially distributed

over all our sources for the words of Christ, without special

prominence in any one evangelist or any one of the documents

which they seem to have used. Going outside the Gospels,

we find ou] mh< in the fragment of Aristion (?) ([Mk] 1618); in

1 Th 415 (regarded by Ropes, DB v. 345, as an Agraphon); and

in the Oxyrhynchus "Sayings"—no. 2 of the first series, and

the preface of the second. The coincidence of all these separate

1 It comes from the LXX of 1 Sam 111, if A is right there, with pichanged to the aor. subj. But A of course may show a reading conformed to

the NT.

2 As to Mk 411, note that in the doublet from "Q" neither Mt (1026) nor Lk

(122) has ou] mh<: the new Oxyrhynchus "Saying," no. 4, has also simple ou].

witnesses certainly is suggestive. Moreover in Rev, the only

NT Book outside the Gospels which has ou] mh<; with any fre-

quency, 4 exx. are from the Epp. to the Churches, where

Christ is speaker; and all of the rest, except 1814 (which is

very emphatic), are strongly reminiscent of the OT, though

not according to the LXX except in 1822 ( = Ezek 2613). It

follows that ou] mh< is quite as rare in the NT as it is in the

papyri, when we have put aside (a) passages coming from the

OT, and (b) sayings of Christ, these two classes accounting

for nearly 90 per cent. of the whole. Since these are just

the two elements which made up "Scripture" in the first age

of Christianity, one is tempted to put it down to the same

cause in both a feeling that inspired language was fitly

rendered by words of a peculiarly decisive tone.

Mh< in Cautious In connexion with this use of negatives,

Assertions. we may well pursue here the later develop-

ments of that construction of mh< from which

the use of ou] mh<; originally sprang, according to the theory

that for the present holds the field. It is obvious, whatever

be its antecedent history, that mh< is often equivalent to our

"perhaps." A well-known sentence from Plato's Apology

will illustrate it as well as anything: Socrates says (p. 39A)

a]lla> mh> ou] tou?t ] ^# xalepo

is not this which is hard, to escape death." This is exactly

like Mt 259 as it stands in xALZ: the ou] mh< which replaces

ou] in BCD does not affect the principle. The subjunctive

has its futuristic sense, it would seem, and starts most

naturally in Greek from the use of mh< in questions: how

this developed from the original use of mh< in prohibition

(whence comes the final sentence), and how far we are to

call in the sentences of fearing, which are certainly not

widely separable, it would not be relevant for us to discuss

in this treatise. Mh> tou?t ] ^# xalepo

meant "will this possibly be difficult?" So in the indicative,

as Plato Protag. 312A a]ll ] a@ra mh> ou]x u[polamba

perhaps then you do not suppose " (Riddell 140). We have

both these forms abundantly before us in the NT:—thus

Lk 1135 sko

to> fw?j . . . sko

the light . . . is darkness"; Col 28 bele

sulagwgw?n, "Take heed! perhaps there will be someone who


. . . " (cf Heb 312); Gal 411 fobou?mai u[ma?j mh< pwj ei]kh?


vain." So in the papyri, as Par P 49 (ii/B.C.) a]gwni

a]rrwstei? to> paida

a@ra e]nqwskwn e@laqen u!dati, "I suspect he may have jumped

into the water unnoticed": so Tb P 333 (216 A.D.) u[forw?mai

ou#n mh> e@paqa . In all these cases the prohibi-

tive force of mh< is more or less latent, producing a strong

deprecatory tone, just as in a direct question mh< either

demands the answer No (as Mt 79 etc.), or puts a suggestion

in the most tentative and hesitating way (Jn 429). The

fineness of the distinction between this category and the

purpose clause may be illustrated by 2 Co 27, where the

paratactic original might equally well be "Perhaps he will

be overwhelmed" or "Let him not be overwhelmed." In

Gal 22 the purpose clause (if such it be), goes back to the

former type--"Can it be that I am running, or ran, in

vain?"1 So 1 Th 35. The warning of Ac 539 might similarly

start from either "Perhaps you will be found," or "Do not

be found": the former suits the pote< better. It will be

seen that the uses in question have mostly become hypotactic,

but that no real change in the tone of the sentence is

introduced by the governing word. The case is the same

as with prohibitions introduced by o!ra, ble

etc.: see above, p. 124. One very difficult case under this

head should be mentioned here, that of 2 Tim 225. We have

already (p. 55) expressed the conviction that dwh is really

dw>^, subjunctive. Not only would the optative clash with


syntactic rule. The difficulty felt by WH (App2 175), that

"its use for two different moods in the same Epistle would

be strange," really comes to very little; and the survival of

the epic dw<^ is better supported than they suggest. There

is an apparent case of gnw<^ subj. in Clement Paed. iii. 1,

e[auto>n gan gnw<^. A respectable number

of quotations for dw<^ is given from early Christian litera-

1 TreThis interpretation as a whole has to reckon with the alternative rendering,

"Am I running (said I), or have I run, in vain?"—a decidedly simpler and

more probable view: see Findlay in Exp B p. 104; Thess. (in CGT) p. 69.


ture in Reinhold 90 f. Phrynichus (Rutherford NP 429,

456) may fairly be called as evidence not only for the

Hellenistic d&

Share with your friends:
1   ...   15   16   17   18   19   20   21   22   ...   33

The database is protected by copyright © 2019
send message

    Main page