402
NOTES AND QUERIES. [11 s. vm. NOV. 22. wia
THE HERULI IN 'WIDSITH.'
The activity of the Heruli in the fifth century was so well known and so widely felt that a list of the Germanic tribes of that period which does not contain their name cannot be regarded as complete. The recognition of this has caused various students of 'Widsith.' who could not find the name of the Heruli in that poem, to endeavour to introduce it either by constructive criticism or by emendation of supposed scribal error. Among these attempts the following are the most prominent:—
I. The Heruli are the Harlungs: so Jacob Grimm, 'Geschichte der deutschen Sprache,' 1880, p. 47 ; Matthaei, Zeitschrift für deutsches Alterthum, xliii.; 319; but Mr. R. W. Chambers says No (' W.,' p. 31).
II. The Heruli are the Eolas of 'Widsith,' 1. 216, the r-stroke having been omitted by the scribe. Hence Eorlas: Jacob Grimm, 'G. D. S.,' p. 598; Möller, Anzeiger für deutsches Alterthum, xxii. 152, 160; Mr. Chambers concurs ('W.,' p. 216).
III. The Heruli are the Heathobeards: Müllenhoff, 'Beovulf,' pp. 29, 32; Much, Paul und Braune's Beitäge, xvii. 201; Heinzel, A. für d. A., xvi. 271; Mr. Chambers dissents ('W.,' p. 206).
IV. The Heruli are the Earule of ' Wid- sith.' 1. 70, where we get eatule in the MS. Cp. ante, p. 261.
Of these propositions No. I. is an un- critical guess : the Harlungs were a fifth- century family, whereas the Heruli appear in history in the third century. No. II. depends upon the statements that we do not know who the Eolas were, that the name cannot be identified, and that it is probably corrupt. See Mr. Chambers's remarks, ad vocem. I have nowhere found warrant for supposing that the great critics of ' Widsith ' have either asked themselves wherein the probable corruption lies, or have decided what is the nom. sing, of this dat. pi. form. " Eolum " is dat. pi. of eolh. The h dropped out after I in oblique cases, and compensatory lengthening of the breaking So into eo followed.
This is quite regular ; cp. Wright, ' O. E. Grammar,' 337. The form eolh is West Saxon and South Mercian. In Anglian it is flh, elc, or elch, without breaking. If the poem were really Anglian, we ought to find elum in this place. " Ic wses mid Eolum " signifies that Widsith visited the descend- ants of some chief whose name had elh, eolh for its prototheme. Cp. " Elcwold," King
of East Anglia ; " Elcbertus," Archbishop
of York. The name is not corrupt, and
Mr. Chambers is in error when he says that
it cannot be identified. Daniel 'Haigh
'The A.-S. Sagas,' 1861, p. Ill, identified
it with the village in Hunts called " a?t
Eolum/' mentioned in grant No. DXCIX.
in Kemble's ' Cod. Diplom.' Another point
urged by critics who have not parsed the
name " Eolum " is that u is not organic in
Heruli. I must return to this presently.
III. The constructive criticism which identifies the Heruli with the Heathobeards is not logical. The argument proceeds thus : the Wicingas are the Heathobeards. Now the Danes drove the Wicingas away. They also drove the Heruli away. Therefore Heruli == Wicingas qui et Heathobeards.
IV. The emendation of eatule suggested by myself in 'N. & Q.,' u.s., is warranted by four considerations : First, the admitted necessity to find the Heruli in ' Widsith.' Secondly, by the instances of t/r confusion in other MSS. which I indicated. Thirdly, by particular instances from the Exeter- Book itself, which I am about to give. And fourthly, by the form of the written word.
To take the last case first, we are assured that "Eatule ^ = Eotol. Mr. Chambers says it is certain (' W.,' p. 211). What we- really have is an instance of the illogical identification of resultants : ol > ol and ul> ol, . ' . ul=al / This is quite erroneous. Ital-ia is correctly represented by Eotol in- Mercian, and by *Eatol in Northumbrian; but " Eatule " has no connexion with Ital-.
I said that t/r confusion is rare. In ' Widsith ' we get a difficult passage which runs (1. 88) :
" Ond ic waes mid Eormanrice | ealle j>rage, | >ser me Gotena cyning \ gode dohte."
This is much disputed, and the usual para- phernalia of faultfinding is imposed upon the passage, namely, faulty connexion with what goes before something presumably lost, something interpolated. Now "|>ser' r shows t/r confusion. It should be ]>set ; cp. lucis er pads, ' Exeter Book.' ed. Gollancz, E.E.T.S., 1895, p. 240. If we read " Ic wses mid Eormanrice ealle frage ]>set me Gotena cyning gode dohte," there is no difficulty : I was with Eormanric all the time that the king of the Gotas treated me benevolently." When the uncertain temper of Eormanric revealed itself, Widsith left his Court and travelled over the e]>el Gotena.
The only question left is that connected with the assertion that u in Heruli, Eruli,
