ADDENDA ET CORRIGENDA.
Page [19] ff. For further information the reader may be referred to Miss Clarke's Sidelights on Teutonic History during the Migration Period (Cambridge, 1911), which contains a very clear and interesting account of the various characters mentioned in the heroic poems.
Page [43], ll. 5-7. This suggestion can hardly be maintained. The true name of Theodberht's son was probably Theodwald.
Page [46], l. 3 ff. The consideration of this difficult question has recently been somewhat facilitated by Richter's Chronologische Studien zur ags. Literatur (Halle, 1910). The general effect of Dr Richter's investigations is to confirm the view put forward by Prof. Sarazzin (Engl. Stud., XVIII 170 ff.) as to the antiquity of Genesis A. Unfortunately I fear that the statistics are not complete and, further, that the evidence is not always treated with strict impartiality. Thus in Beowulf such half-verses as to widan feore are regarded as proofs of shortening (through loss of h) and reckoned in the final statistics (pp. [9], [85]); but in Genesis A the metrically equivalent on fyore lifde is not so reckoned (pp. [24], [89]). The half-verse geseon meahton is cited in Beowulf without qualification (p. [15]), but in Exodus as doubtful (p. [18]). In Beowulf -wundor seon is taken to be –×– (pp. [13], [15]), but in Daniel A it is treated as doubtful (p. [32]). Dr Richter's conclusion that Genesis A is an earlier poem than Beowulf is certainly not substantiated by the treatment of feore or of postconsonantal r, l, m, n[670], or again by that of frea, don, gan, sie[670]; while compounds such as þreanyd obviously do not stand on the same footing as case-forms like þrea. It is only in the treatment of intervocalic h that Genesis A apparently shows a more archaic character than Beowulf. In the former poem Dr Richter cites only one case of contraction (p. [28]), and even this is doubtful; but the same remark applies to at least 18 of the 24 (genuine) cases which he cites for Beowulf (p. [15]). Out of the five or six probable cases of contraction in this poem three occur in practically the same phrase—in (on, to) sele þam hean—while two of the others occur in consecutive verses (910 f.). The conclusion to which the evidence seems to me to point is that both Genesis A and Beowulf (even in its Christianised form) date from the seventh century, but that the former has been somewhat better preserved than the latter. As Genesis A is doubtless of monastic origin, we may reasonably expect that it was committed to writing at an earlier date. In the intervening period the text of Beowulf may have suffered many changes (such as the insertion of the article) at the hands of minstrels.
The further question raised by Prof. Sarazzin (op. cit., p. 192 ff.) as to the relationship of Genesis A to Caedmon cannot be discussed here. But in view of the evidence brought forward I do think it would be worth while to examine and compare the characteristics of the various 'Caedmonic' poems from all points of view. Apart from the extremely improbable dating of certain sound-changes proposed by Prof. Morsbach (cf. p. [66] ff. above) and accepted both by Sarazzin and Richter, I see no reason for supposing that Caedmon's poems have entirely perished.