PRONOMINAL USE OF NUMERALS 71
vii. 17) in an indirect question a survival paralleled
in its limits by that of the etymologically identical
" whether " in English.
The various uses of the indef. pron., the unaccented
TIS, contain no deviation from classical usage. But
there are one or two words, not strictly pronouns,
which may be noted here as having unclassical quasi-
pronominal uses. Els approaches, as in late Hebrew
and Aramaic, the sense of a mere indef. art. in a
few passages of the Gospels and Apoc. Matt. viii.
19, ix. 18 (?), xxvi. 69 ; Rev. viii. 13, xviii. 21, xix. 17.
In Matt, xviii. 24, John vi. 9 (?) the word no doubt
has a distinctive meaning: but in Matt. xix. 16=
Mark x. 17, it seems to be merely= ns. Intermediate
are cases like Matt. xvi. 14= Mark vi. 15, Matt,
xviii. 28; or again Mark ix. 17, Luke v. 12, 17, etc.,
where ets is followed by a gen. or the prep. e, and
thus, though there is no emphatic insistence on
singularity (as there is e.g., in Matt. x. 29, xviii. 6,
10, 12), there is a certain amount of antithesis between
the individual and the class out of which he is
selected. If we cannot say that the use of ei9 c.
gen. in Matt. xxvi. 14, etc., is unclassical, we may
say it is found more frequently and used more freely
than in classical Greek.
In antithetical sentences such as Matt. xx. 21,
xxiv. 40, xxvii. 48, it can hardly be said that the
use of et is unclassical, for the sense is not so much
"the one . . . the other" as " one . . . and one"
of the two (or in Matt. xvii. 4, and parallels the three)
already mentioned or referred to.* But there appear
to be no exact classical precedents for the opposition
* It is otherwise in Mark iv. 8, 20, if we there read &>.