The inquirer is certainly quite in the dark as to a conservative view of the processes by which the poem reached its present form. The elaborate patchwork theory of ten Brink,' who distinguishes in the piece four separate lays, not including introductory and connecting material, has never been adequately criticised and refuted, although the general weakness of his method is apparently coming more and more to be recognized. One feels that the truth must lie somewhere between this and the view of Dr. Guest, for example, who accepted practically the whole poem as the work of one man, "soon after the age of eighty," the reference to Alexander the Great being "the only instance in which he has referred to one not a contemporary." But a careful examination of the problem from the point of view of construction is still lacking. Few men have thrown as much light upon these perplexing problems as Heinzel has done, both directly and indi-y rectly, yet we have no detailed study of the poem from his pen, while much of his most illuminating criticism is to be found in articles dealing with other subjects, which may be overlooked in collecting bibliography especially with reference to Widsith. In short, some of the most important questions in regard to the piece as a whole, not to mention many details, must be regarded as still awaiting solution. It is, indeed, too much to hope to gain the whole truth in regard to the baffling old poem. Many matters connected with it must remain undetermined. The illusion that analytic criticism can find out almost everything worth knowing is rather less common nowadays than it used to be. Yet it seems unwise to go too far in the direction of the caution that takes refuge in the impossibility of gaining further knowledge. At all events, the need of a thorough re-examination of Widsith, in the light of modern knowledge of ethnology and saga, and of a careful review and comparison of earlier theories, is perhaps sufficient excuse for rushing in where angels have feared to tread, or have trodden unsuccessfully. A good deal has been written which may safely be pronounced untenable, as, for example, Möller's attempt to force the 1 Paul's Grundriss, Vol. II, pp. 538 ff. References to the Grundriss in this paper are to the earlier edition. 2 History of English Rhythms, ed. Skeat, pp. 371 ff. 2 entire poem into the Procrustean bed of the strophic form,' or Michel's notion that it reflects the medieval conception of the seventy-two peoples inhabiting the earth. Apart from articles exploiting special hypotheses, however, there are various suggestive criticisms of detailed points which must be taken into consideration, some of which have a most important bearing upon the interpretation of the poem as a whole. Any consistent interpretation must, indeed, rest to a very large extent upon these details. The difficulty of securing critical unanimity as to their significance is one of the stumbling-blocks to the acceptance of even the most conservative view as to the evolution of the poem. But the effort to clear up these matters is certainly worth while, in view of the importance of the piece, even if the only result were to stimulate renewed discussion. The principal object of the present investigation, then, is, as the title indicates, to study the various processes in the evolution of the poem, and the interpretation of certain significant portions, which may lead to a decision as to the approximate date and provenience of the material, rather than to enter minutely into questions of ethnology, history, and saga. I Upon a hasty reading, the poem makes the impression of a jumble of heterogeneous material. A more careful examination shows that it falls into certain rather definite groups, and that the interest of the main narrative seems to be of two kinds, the details of personal experience, and the enumeration of peoples and rulers, with some historical, or avowedly historical, information added. The whole is introduced by a short prologue: WIDSIÐ MAÐOLADE, wordhord onleac, sē pe monna mæst mægþa ofer eorpan, 1 Das altenglische Volksepos in der ursprünglichen strophischen Form, pp. 1 ff. In certain parts of the poem it is quite possible that strophic structure is to be assumed, as for instance ll. 15 ff., but to extend the principle as far as Möller wished to do, and reprint the whole with stanzaic divisions, cannot be regarded as otherwise than highly dangerousindeed, the wide application which Möller made of his general theory to AS. heroic verse is generally discredited today. Cf. Heinzel, Anz. f. d. Alt., Vol. X, and note how little such strophic manipulations are likely to produce unanimity; ten Brink, Paul's Grundriss, Vol. II, p. 542, thinks that Möller's four-line strophes would form six-line divisions equally well. 2 Paul-Braune, Beiträge, Vol. XV, p. 377; refuted by Bojunga, Beiträge, Vol. XVI, p. 545. folca geondferde: oft he on flette gepah fælre freopuwebban, forman sipe Hredcyninges hām gesōhte, eastan of Ongle, Eormanrices, wrāpes wärlogan. Ongon pā worn sprecan:1 5 Autobiographical matter does not follow, however. The conventional formula ic . . . . gefrægn (1. 10), which, so far as it implies anything, means that the poet got his information by hearsay, introduces, after the valuable observation that virtue is necessary to a successful monarch, a long list of peoples and princes. Obviously, however, there is no personal note here—these are not the ones that Widsith visited, or supposedly visited. The information is not even conveyed in the first person, but in the third. Ætla weold Hūnum, Eormanric Gotum; 18 Becca Bäningum, Burgendum Gifica. This forms a contrast to the names introduced by the phrase ic was mid, later on. The mention of Eormanric seems rather superfluous, after the prologue. Offa, king of the Angles, and Hrothwulf and Hrothgar get a longer mention, closing the somewhat incongruous collection beginning with Alexander. The whole passage (11. 10-49) is a kind of rhymed summary of historical information. It constitutes a division of the poem by itself, the basis of it perhaps being, as ten Brink suggested, the "uralte versus memoriales" (ll. 18-34)." 1 The text follows that in the Grein-Walker Bibliothek der ags. Poesie, Vol. I, p. 1, with the addition of marking the quantity of the vowels. The punctuation has in some cases been changed. Elsewhere than in quotations, the spelling of the word Widsid has, for the sake of convenience, been modernized, and the marks of length omitted. 2 Ten Brink was no doubt right in setting this down as a mnemonic catalogue, and one of considerable antiquity. He looked upon 11. 35-44 as a later addition made among the Angles; 11. 45-49 as having been added in Mercia, while ll. 10-13 was assigned still a different origin. Into these details it does not seem possible to venture with any certainty. If, as is likely, it constitutes one of the oldest portions of the poem, we may have to take the changes of oral transmission into account. It represents a collection of facts and traditions thought worthy of perpetuation, and so committed to verse to assist the memory. The process outlined by ten Brink is not unreasonable, but it is improbable that it is correct, since there is but such slender evidence upon which to base it. It is worth noting that there are some interesting parallels in Old Norse. The editors of the Corpus Poeticum Boreale call attention to the opening lines of the Lay of Hlod and The singer then takes a fresh start, this time in the first Here, in place of the formal ic gefrægn stand the direct and personal ic geondferde, and ic cunnade. More cataloguing follows, but up to the end of the narrative (l. 134) it is sustained in the first person, whether the phrase ic was mid or ic sōhte be the one used. It is noticeable that in the passage immediately following this second introduction, certain lines, and those the Angantheow, remarking that they "look like a bit of a separate song, parallel to the English Traveller's Lay, ll. 15-35." (C. P. B., Vol. I, p. 565.) "Ár kvódo Humla Húnom ráða, The short enumerative pieces which the editors call "Heroic Muster Rolls" (Vol. I, p. 353) are stated to be "manifestly the echoes of genuine older verse, and may probably contain passages borrowed from them "- which suggests a process not unlike what we may believe to have taken place in parts of Widsith. Manifestly, the lines in Widsith are similar to such verse as this: "Alfr ok Atli, Eymundr trani, Steinn ok Kári, Styrr ok Ali" (etc., etc.). 1 This statement "So I traversed many foreign lands,” etc., following a passage which has no personal element in it, has given pause to various commentators. Müllenhoff remarks (Haupt's Zeitschrift, Vol. XI, p. 285): "Der zweite abschnitt wird mit vv. 50-56 eingeleitet. nach hrn Greins und der frühern herausgeber interpunction, wenn man abtheilt Sva ic geondferde fela fremdra londa geond ginne grund; gôdes and yfles þær ic cunnade u. s. w., muss man den ersten satz und das 'svâ' auf das was vorhergeht beziehen; es würde daraus folgen dass der sånger auch alle die fürsten die er eben aufgezählt besucht habe. . . . . (In 11. 18-49) zeigt der vielgereiste sånger seine erfahrung und sagenkunde; hätte er aber dort alle von ihm genannten könige besucht und selbst gesehen, was in aller welt sollte da noch das zweite, ziemlich abweichende verzeichniss von v. 57 an von völkern und zum theil auch von königen mit der ausdrücklichen bemerkung dass er bei diesen war? v.51 muss darnach anders interpungiert und das semicolon in ein komma verwandelt werden. wir würden jetzt die unterordnung oder das verhältnis der gedanken schärfer ausdrücken als es zu einer zeit geschah wo der satzbau noch wesentlich parataktisch war. aber die folge der gedanken ist doch ganz dentlich: ice habe so-wie folgt-viele fremde länder durchreist, gutes und übles erfuhr ich da, deswegen kann ich singen und sagen u. s. w." Möller notes (Vol. I, p. 34) that this interpretation of swă is supported by Beow., 1. 2144, although he is inclined to think that there is contamination in the Beowulf passage itself. He regards the swā as an interpolation here, saying that it is "ein beliebtes interpolatorenwort." Ten Brink, too, changes suã to Hwat. It seems well to remember that if 11. 18-49 or ll. 14-49 is an ones which contain the baldest enumerations, stand out prominently as awkward and hypermetric,' while others which introduce additional detail, mainly that of matters which have affected the singer personally, are of the normal length. Contrast, for instance, Mid Froncum ic was and mid Frysum and mid Frumtingum. Mid Rūgum ic was and mid Glommum and mid Rūmwālum. with the lines immediately following, Swylce ic was on Eatule mid Elfwine, 70 This distinction is perhaps not without significance in the evolution of the poem, as will be seen later. The mention of Guthhere, or Gunther, king of the Burgundians (11. 65 ff.) is also of especial interest. Earlier in the poem (1. 19) Gifica is represented as ruling the Burgundians. The curious combination of the Greeks and the Finns and Cæsar, already found in 1. 20, is repeated in 1. 76. The strangest collection of all is the passage 11. 79-87. The Picts and Scots, the Israelites and the Assyrians and the Egyptians jostle the Medes and Persians, the "Mofdings" and the "Amothings." Surprising, too, is the statement that Widsith has been with the Myrgings, his own people, and "ongend Myrgingum," after all these travels! The mention of Eormanric introduces a section of very different character. Here at last something the sort of tale promised by the prologue is realized. In striking contrast to insertion, something may very well have been cut out to make room for it, which would have made the usual meaning of swa quite in place here. But the adherents of the ballad theory were always loth to admit losses in practice, however willing they may have been to do so in principle. In the second place, the logical connection of the particle swā appears to have been less close than we are inclined to suppose nowadays. Müllenhoff's comment points in this direction. In an earlier article, I have shown this in regard to the adverb forpon (Jour. Germ. Philol., Vol. IV, pp. 463 tf.). If swā is "ein beliebtes interpolatorenwort," it is certainly also a favorite word for introducing a new sentence in poetry where no contamination can be held to be present, and is sometimes used, like modern English "so," or German "also," as a loose connective in narrative, not necessarily denoting a close logical connection between what precedes and what follows. In short, then, there seems to be no need to regard it with suspicion here, even if no interpolation exists. 1 Cf. ten Brink, Grundriss, Vol. II, pp. 540, 541. |