110 LANGUAGE OF THE NEW TESTAMENT.
Greek MS. may be trusted. We can hardly decide
whether it was beginning to come in in N". T. days,
or was introduced by early scribes : the scholarly
scribes of the middle ages, certainly, eliminated such
forms where they found them.
In hypothetical sentences, the constr. is generally
in accordance with classical rules. In Luke ix. 13
(et fJLrjTL
. . .
dyo/oacrco/xei/), 1 Thess. v. 10 (etre yprjyo-
pwjjii/ LT Ka0v$u>iJiv) the use of i c. subj. is not an
irregularity but a refinement. In the former place,
the subj. is deliberative : the sense is " unless we are
to buy," or perhaps "unless are we to buy?" in
the latter, the verbs in the protasis are, quite cor-
rectly, attracted into the mood of the apodosis. But
we note that here too the opt. has all but passed out
of use :
except for the phrase cl rvyoi twice in St.
Paul (1 Cor. xiv. 10, xv. 37) the only instances are
in the last chapters of Acts and in 1 Pet. iii. 14, 17
(true text).
Hypothetical sentences are, however, only a parti-
cular case of rekitive sentences, and come under the
same rule with them, that the subj. will be used if
the relative (pron. or particle) that introduces the
protasis has av with it, and not otherwise. And this
rule is broken in the N. T. rarely in hypothetical
sentences, oftener in temporal, perhaps only once
(James ii. 10) with a rel. pron. The use of et c. subj.
in 1 Cor. xiv. 5 has parallels even in Attic, and is
quite exceptional in the N. T., as in Attic : but the
lav oiSa/xei/ of 1 John v. 15 would be impossible in
classical Greek, while in the N. T. it is only the
extremest case of four or five. In the other cases,
however (Matt, xviii. 19 [?],.Luke xix. 40, Acts viii. 31,