NEGATIVES: et or, et /w}. 185
x. 37,. Acts xxv. 11 (ovSeV), 1 Cor. vii. 9, xi. 6, xvi.
22, 2 Thess. iii. 10, 14.
Can we trace any principle here ? If not, it might
be worth while to remark that about half the
instances of et ^ are in a single writer : and we
might say that et OVK is the rule, and et^ the ex-
ception, in all N. T. writers but St. John. But on
examining the instances, we shall see that in all
the places where^ is used except the last, it is used
with a past tense of the indie, of an unrealised sup-
position
: in all where OVK is used, either the verb is
in a primary tense of the indie., or the sense is "if,
as was the fact " (Rom. xi. 21, Heb. xii. 25, 2 Pet. ii.
4), or one of the reasons stated above applies (Luke
xvi. 11, 12).
This then appears to be the rule of N. T. usage
that et with the indie, almost always takes ov, except
with a past tense in the sense specified. 'Eav however
always takes /AT; c. pres. as well as c. aor. subj. And
though ct OVK is a deviation from classical usage, it
may admit of justification on the principles of the
classical language. If we resolve the conditional
particle into a relative one, et will be " in the case in
which,"
. . . and lav " in any case in which " . . . :
and of these relative sentences, the one would regu-
larly take OVK and the other py. In practice, how-
ever, we must not expect always to find an assignable
difference of meaning between et OVK and lav p,rj, any
more than between the simple et and lav : compare
Matt. vi. 15, lav pi] dc^re, with Mark xi. 26, et Se
{yxets OVK a</>tere which, though not part of the
genuine text of St. Mark, belongs to the oldest form
of the " Western Text," and shows what were the