The Oxyrhynchus Papyri, Part 13

Leaving undecided the question whether the original language of Tobit was Greek or Semitic, he thinks that the book was composed in Egypt not long before 170 B.C., and that the recension of א is the nearest approach to the original, while that of BA did not reach its present form until about A.D. 180, and the third recension was later still.

The conditions of the problem are somewhat altered by the discovery of 1594, which is on the whole much nearer to BA than to א or the third recension, here fortunately extant. In vv. 14–17, where the two main recensions do not greatly differ, 1594 agrees with BA against א in the insertion of ek from (l. 3), hagion holy ones (l. 3; hagion angelon holy angels BA; angelon angels א), prosanapherousin offer up (l. 3; add. tas proseuchas ton hagion the prayers of the holy ones BA; parestekasin stand by א), the omission of auton their (l. 8), and the insertion of estai it shall be (l. 12); against these can be set only the agreements with א in the form epesan they fell (l. 8), the insertion of hapanta all things in l. 13 (panta all א; om. BA), and kai and for BA’s hoti that in l. 9. In vv. 18–19, where the text of א is longer than that of BA and differently arranged, the new fragment agrees with BA in having emautou my own, not eme my, in l. 15 and in constructing pasas tas hemeras all the days with optanomen hymin I appeared to you (ll. 18–19), whereas א connects the first phrase with the preceding eulogeite bless or with an added repetition of it, auto hymneite sing praise to him. Against this must be set the partial agreements between 1594 and א as to the verb in l. 16 (emen meth' hymon I was with you: om. B; elthon I came A), and the occurrence in 1594. 20 of etheoreite me you were beholding me (cf. Old Latin videbatis me) corresponding to א’s theoreite me you behold me. With the peculiar readings of the third Greek recension 1594 agrees against the other two in respect of the omission of Sarran Sarah in l. 2, and of angelon angels in l. 3, the insertion of epi ten gen upon the earth in l. 9, and the reading theou God in l. 6 (theou without tou megalou the great the cursives; cf. Dei Old Lat.). But elsewhere the third Greek recension follows א rather than 1594, and is shorter even than BA in v. 19.

The new recension has also a number of peculiar readings, such as the constant use of kai and as a connecting particle, where BA vary the monotony by de but/and (l. 12; om. א) or hothen whence (l. 17; om. א) or the absence of connexion (l. 19), and especially the new arrangement of vv. 18–19, which avoids both the obvious omission in B and the redundancy of א at this point. On the whole 1594, while belonging to the BA type of text, is distinctly better. Is this superiority to be explained as resulting from a revision of the BA text in the light of א, or from the priority and greater purity of the text illustrated by 1594, of which BA is a later form? The second hypothesis seems to us much the more probable for several reasons. In the first place 1594 is an older MS. than B or A. Secondly, the constant use of kai and in 1594 points to a more archaic text than that of BA. Thirdly, the text of BA, where in comparison with that of 1594 it is markedly inferior, as in vv. 15 and 18, seems to have arisen out of the text of 1594.