340 Table of Greek Correlatives, with Explanations. 
ive, primarily i. g. Gr. 5re or Lat. guod, and correctly derived by Pas- 
sow from the old personal pronoun @s, ?.—8. Here ¢ is lost in e (==<a7;) 
comp. Sansc. i-tas, thence; Lat. i-nde, thence.—9. Here 7 is lost ine 
(=<ai ;) comp. Sansc. i-ha, here; Zend. i-dha, here; Lat. 7-bz, there.— 
10. Here 7 is lost in e (—ai;) comp. ?v%«.—11. The form 4%, either, 
or, (as if in this way, in that way,) probably for 7j with subscript Tota, 
an obsolete dative from the pronominal root % The form «i 7f, (as if 
in this way, under these circumstances,) probably with form of the 
dative, from the same pronominal root. The vowel 7 in both forms 
is lost in e (=ai.) For examples of the derivation of the conditional 
conjunction from pronouns; comp. Sansc. yadi, if, from yas, who; 
it, if, neuter of the pronoun i; Goth. ith, if, for ita, neuter of the 
pronoun is; Germ. wenn, if, from wer, who.—12. The forms Ett, 
yet, 10é and 754, and, (comp. Sansc. i-ti, also,) we place here, although 
with some hesitation.—13. No form extant.—14. No form extant.— 
15. The form icéxes from ioos needs no illustration.—It ought here to 
be observed that ?yder, %v Ie, and probably the other forms in the Sth, 
9th, and 10th series, are capable of the relative signification; comp. 
Germ. der and Eng. that, which are both demonstratives and relatives, 
the ground of which is worthy of further investigation. 
em. 1. The lengthened forms, as évted%ev, (Lon. 2vSstter,) évtad- 
Be, (lon. évPatre,) évtavdor, (Lon. éyPavtoi,) are more emphatic than 
the others. 
Rem. 2. Some of the forms in this column are strengthened or ret- 
dered emphatic by the addition of 04; as évdévde, éyGdde. Add also 
éyteviert, évyFadi, iyFavt. 
e I. column is not quite complete. It is, however, regular, beau- 
tiful, and worthy of admiration. It consists of proper demonstratives, 
an important and interesting class of words. They have sometimes 
been called redditives, as answering or responding to the interroga- 
tives. But this name is inappropriate, as it would seem to imply an 
origin posterior to that of the interrogatives. ‘The correlative ele- 
ment is expressed throughout by t, a sound which is justly believed 
to have a natural appropriateness to perform this office. Hence it 1s 
found with remarkable uniformity in different families of the Indo- 
European stock of languages, as we shall hereafter see. The irregu- 
Jarity in the masculine and feminine forms of the pure pronoun is only 
apparent, for the form of the oblique cases, 100, tis, Tod; TH, TH» T?> 
etc. plainly show that the nominative was originally 74, 7%; 74.—It 
ought here to be observed in respect to the pure pronoun, that 111 
sometimes used asa relative, (comp. the remark on @vdsy and tra 
above ;) and, in the genitive and dative singular, also as an interrog@- 
tive and indefinite; and that although its primary sense is that of a 
