tories, regardless of linguistic genealogy (cf. Sapir, l.c., chapter IX). Thus, Slavic and Finnic on the one hand, and Celtic (Romance) and Basque on the other have important common elements. Before we can come to a hypothesis of the origin of those agreements, the facts themselves must be compiled-a work that practically amounts to a new 'Brugmann's Grundriß' on a geographical instead of a genealogical basis.

BRYN MAWR COLLEGE

THE DIDACTIC PURPOSE OF SOME EDDIC LAYS

THE

BY LEE M. HOLLANDER

HERE have been a number of attempts, recently, to establish ritual origin for a number of Eddic Lays. In a recent article 1 I examined in some detail Miss Phillpotts' well-reasoned endeavor 2 to discover, in modern practices and games, remnants of ancient fertility rites centering about the figure of Helgi: as in most special pleadings, the less favorable aspects of the case are touched upon lightly, if at all. After all is said, the fact remains that we have not one case of direct mention, and not any incontestable evidence, of there ever having been, in ancient Scandinavia, anything which may be called even vestigial drama; even though cult rites such as those inferred by Magnus Olsen, in his various studies on Frey worship, might have furnished an impetus similar to the Christmas, Easter, and Pentecost festivals of the Catholic Church which gave rise to the medieval tropes. Certainly, Scandinavia, no more than any other part of Europe, was devoid of such beginnings. But it is difficult to discount the utterly negative evidence of the bulky and many-sided literature that has come down to us.

This seems to me an insuperable objection to the assumption of even the simplest dramatic action. To be sure, Miss Phillpotts seeks to forestall this objection by urging that in Iceland, the home of this great literature, there were no ritual customs and games, such as May Pole dances, mock battles between Winter and Summer, midsummer bonfires, etc., such as did prevail in the old country. However, this argument ex silentio cannot be invoked now, and disregarded right after: in the first place, we are by no means sure that rites of this nature actually existed in old Norway. It would be strange if such a striking difference had escaped the sharply observing eyes of the saga writers. There was constant and copious intercourse

1 Scandinavian Notes, 1924, p. 108.

2 The Elder Edda and Scandinavian Drama, Cambridge, 1920.

with the motherland. Even granting that the first times of settlement were not propitious for the continuance of social rites and games, they would without fail have been reintroduced, at least partially. Or will any one contend that the colony had a radically different mentality, that people there were of a less social temperament, or that, perchance, living conditions differed materially? Certainly, there is no dearth of allusions -in fact, there are many copious descriptions, of the knattleikr (a ball-game), the hestathing (horse-fighting), taf (chess or draughts), of the social intercourse at the Althing, etc. For that matter, we are told of certain religious practices connected with the stalli (heathen altar). Moreover it is to be considered that, if anything, the Icelander was distinctly conservative in cultural matters-the adoption of Christianity by act of parliament notwithstanding. If any proof be desired, the persistence and popularity of Eddic and Skaldic poetry precisely in Iceland may serve as 'crown-witness.'

Then, of course, Miss Phillpotts' theory stands and falls with the attempted proof of Norwegian origin of all 'genuine' Eddic poems in chant metre (ljóthaháttr), which measure she regards as a Norwegian product. To me, her elaborate and ingenious argumentation seems to beg the question from the outset. Lacking incontrovertible criteria for the age and provenience of a single Eddic poem,3 it will not do to label-on the authority of Mogk and Finnur Jónsson, or of any other scholars -the chant metre poems as Norwegian and heathen, and then proceed to infer that, therefore, chant metre is Norwegian and heathen; branding as 'imitations cast in the old heathen form' poems that offer difficulties, viz. Svipdagsmál, Hrimgerthamál, Alvissmál. To me, it is incredible that the simple and effective stanza form could not be handled or 'imitated' by the skalds.

But if the theory of dramatic origin seems thus more than doubtful, for this class of poems, another originating cause is certain and has always been recognized-the master passion of the Icelandic mind for mythological lore and genealogicalhistorical information. Without exaggeration it may be said that the Landnámabók is a phenomenon unique for the stage of • Barring, of course, Gripisspé and Atlamó̟l.

4

cultural development represented by medieval Iceland. This leaning is abundantly evident in the Edda, too-more than is generally recognized. It is admitted that "almost all the chant metre poems are careful to impart mythological information"; " but hardly, that precisely this didacticism is the governing intention of these, and perhaps other, lays: it is a question of emphasis.

Chant metre is the verse of the following poems: Hávamál (mainly), Vafthrúthnismál, Grimnismál, Skirnismál, Lokasenna, Alvissmál, Helgakvitha Hjorvarthssonar 12-30 (Hrimgertharmó̟l), portions of Reginsmál and Fáfnismál, Sigrdrífumál, Svipdagsmál; to which we may add Sólarljóth and Getspeki Heithreks.

Scrutinizing the list, the didactic-gnomic purpose of most of these poems is evident at first glance. However skilfully this purpose is veiled it cannot be doubted that the prime intention of especially Vafthrúthnismál, Grimnismál, Alvíssmál is to convey a knowledge of mythological facts and names. The writer yields to no one in his admiration for the virtuosity of the poet of Vafthrúthnismál in dramatically organizing this lore and interweaving it with the action 5-both Othin and the giant staking their heads on the superior extent of their knowledge. Yet one cannot for a moment forget that the framework is, and is meant to be, subordinate to the lore to be imparted. The same is true of Grimnismál, notwithstanding the fact that here the framework contains genuine old folklore material which, in part, has survived to this day; and of Alvissmál, by Gering fittingly called "a versified chapter of skaldic poetics.' In other words, to make myself perfectly clear: the gnomicdidactic stanzas are not an unwarranted 'expansion' of the material; but, on the contrary, the framework—whether invented ad hoc, as in the case of Vafthrúthnismál (and Alvíssmél?), or adapted-is designed for the express purpose of introducing the information to be conveyed. If I may say so, these lays are so many glorified mnemotechnic devices.

4L.c., p. 47.

6

'Though one may not agree with Boer's evaluation (ed. II, p. 60) that it is the sublimest of all Eddic poems.

In his translation, p. 81 note.

Misled by overweening confidence in his philological method, Boer has outdone Müllenhoff and condemned more than twothirds of the poem; and, indeed, by this heroic treatment achieves, as he states with satisfaction, "ein nicht nur großartiges, sondern auch klares und schön komponiertes Gedicht!" Too bad that, e.g., the text of Wolfram's Parzival is too well authenticated to permit of a similar trimming with the hedge-shears-it would be an easy matter to achieve a much better proportioned poem, one happily divested of the additions of elende Stümper and törichte Abschreiber! That we do not have Grimnismál in its original shape, that no doubt there were accretions, may be conceded without deluding ourselves that this is capable of proof. Instar omnium: the last stanza does, indeed, spoil a very effective ending, as the earliest editors were quick to see; but how may we be sure it is 'interpolated'? Detter-Heinzel' suggest, in fact, that there may be a hidden meaning in thus adding still other Othin names, even after the god has revealed himself. At any rate, and this seems to have been overlooked, the present tense- -Óþinn nú

heitik-is noteworthy. It seems in place only after a long series of names has gone before; i.e., at, or near, the end. The stanza cannot stand anywhere else, and certainly not after 50,10 since the revelation of Othin's identity comes only in 53.

The didactic purpose is not so clear in the group of poems comprising Reginsmol, Fáfnismó̟l, Sigrdrífumél, and Svipdagsmál which are generally classed as heroic poems dealing with the fates of Sigurth. And as such they were interpreted by the Collector who naively links them into the heroic cycle. Nevertheless, an examination shows that, here too, the didactic purpose is to the fore; nor is the story, as we would fain believe, furthered a whit by the speeches of the various agents. It is, of course, tempting, and flattering to our critical insight, to declare the many gnomic stanzas contained in them to be 'interpolated'-without stopping to think that in so doing we ▾ Arkiv för nordisk filologi, XXII, 133.

8 Edition, p. 60.

• Edition, II, p. 193.

10 Gering's edition (as in subsequent quotations).