Babyloniaca, études de philologie assyro-babylonienne.

226

O. E. iAVN

were dropped in very early times, the final syllables might readily
become mere consonants; then we should better understand, how the
Hebrew nmajim appears as mi-ma, Samajim as sa-mu-ma, susim as
zu-u-si-ma, and the like (see the list in ~ 37). Concerning the chapter
on the nouns, I offer a few criticisms. The word abM etc. had already
in the time of Hanmurabi (~ 20 b) a long 4, as it is proved by a-huu-urn, Code 12.40. The mimation occurs very seldom, and mostly with
the sign turn, but this has to be read -tu, in those cases (see su-tum-u
16. 38 and ~ 1 d). Elsewhere I have tried to explain the loss of the
rimation by Assyrian influence; the mimation presumably never existed in Assyrian1.
On the case-endings DR. B6HL has some remarks on p. 33; he fails
to find rules of declension which are everywhere valid; yet the rule
that i in the construct gen. sing. is dropped by dissimilation2, is
observed here also. The construct and absolute states must be studied
separately; in the second case the regularity is far greater than in
the first, where the endings have been dropped or confused to a
large extent, - already in the time of Hammurabi.
I finally call attention to an interesting note on Greek loan-words
(p. 20) and close with thanks to the author for his very careful and
suggestive work.
1. Om nominernes bbjning, p. 103 f.
2. ana arrisu, but ana harran sarrim.