Bearing these facts in mind we can hardly doubt that the information used by Beowulf was acquired before the end of the sixth century—in all probability we may say considerably before that date.[58] Next we have to notice that we have practically no trustworthy information regarding the history of the English kingdoms before the middle of the sixth century, and I think it will be the opinion of any attentive student of early English history that even the best informed persons of Bede's time were not much better off in this respect than we ourselves are. How then are we to account for the preservation of detailed information regarding the early kings of the Danes and Swedes? The only answer to this question, so far as I can see, is that the doings of such persons must have become embodied in stories which were preserved by recitation in a more or less fixed form of words. Such recitative pieces may have consisted of poetry alone or of poetry mixed with prose, like some of the pieces contained in the Older Edda. If we may trust the analogy of what appear to be the oldest pieces in this collection, such as Völundarkviða or Helgakviða Hundingsbana II, the speeches would be given in metre, while the connecting narrative might be partly or wholly in prose and quite brief. We have no evidence for believing that the early Teutonic peoples ever used entirely prose narratives, like the Icelandic and Irish sagas, for such purposes.
At all events it seems to me that if Beowulf is no older than the middle of the seventh century we are bound to assume the existence of earlier poems or narratives on the same subject. Such pieces may of course have been quite short, and it is likely enough that our epic has made use of more than one of them. One perhaps may have dealt with the hero's exploits at the Danish court and another with his last adventure, while in the scene between Beowulf and Hygelac it is possible that an older poem has been incorporated, more or less complete, in the text.[59]
But we have yet to take account of what is perhaps the most striking feature of the poem, namely the fact that, though it abounds in expressions of Christian sentiment, yet the customs and ceremonies to which it alludes are uniformly heathen. Among these we may mention the funeral ship in v. 27 ff., the offerings at the shrines in v. 175 f., the observation of the omens in v. 204 and the curious reference to hanging in v. 2444 ff. (cf. v. 2939 ff.), probably also the use of the boar on helmets (vv. 303 f., 1111 f., 1286, 1451 ff., 2152) and the burial of the treasure (v. 2233 ff.), together with the curse imprecated on the person who should disturb it (v. 3069 ff.). But most important of all are the descriptions of the disposal of the dead by cremation in vv. 1108 ff., 2124 ff., 3137 ff. In the long account of Beowulf's obsequies—beginning with the dying king's injunction (v. 2802 ff.) to construct for him a lofty barrow on the edge of the cliff, and ending with the scene of the twelve princes riding round the barrow, proclaiming the dead man's exploits—we have the most detailed description of an early Teutonic funeral which has come down to us, and one of which the accuracy is confirmed in every point by archaeological or contemporary literary evidence[60]. Such an account must have been composed within living memory of a time when ceremonies of this kind were still actually in use.
The significance of these passages seems to me to have been altogether misapprehended by recent writers. If the poem preserves its original form and is the work of a Christian, it is difficult to see why the poet should go out of his way in v. 175 ff. to represent the Danes as offering heathen sacrifices; for not long before he has introduced a song of the Creation at the Danish court, and in the sequel Hrothgar is constantly giving utterance to Christian sentiments. Again why should he lay Beowulf himself to rest with heathen obsequies, described in all possible detail, when in his dying speeches (vv. 2739 ff., 2794 ff.[61]) the hero has been made to express his faith and gratitude to the Almighty? On the other hand if the poem was originally a heathen work these inconsistencies are perfectly natural. If it was to retain its place after the change of faith and to be recited in the presence of bishops or clergy, all references to actual heathen worship or belief would of necessity have to be either accompanied by censure—as is the case in the homiletic verses following v. 175 ff.—or else suppressed altogether, and their place taken by expressions in accordance with Christian doctrine. Hence it seems to me probable that such expressions are frequently in the nature of substitutions for objectionable matter, rather than gratuitous additions; and in the same way I would account for the occasional survival of ideas which appear to be essentially heathen[62], though they are cloaked in Christian phraseology. But references to practices such as cremation which, though heathen, had long ago passed out of use, would not excite the same repugnance and consequently might be allowed to stand.
It may be urged[63] that cremation seems to have lingered on among the Old Saxons of the Continent until late in the eighth century. True: but it is quite incredible that a Christian poet should borrow from this quarter a method of funeral for his Christianised heroes. If the description of Beowulf's obsequies stood alone a bare possibility might be conceded to the suggestion that it had once formed a poem by itself, unconnected with Beowulf, and based upon a traveller's story. But cremation is clearly regarded as the normal rite throughout the poem, apart from the legendary story of Scyld. We have another description of it in the episode dealing with Finn (v. 1108 ff.), and above all there is the purely incidental reference in v. 2124 ff.: "Yet when morning came the knights of the Danes could not burn his (Aeschere's) lifeless form with fire, nor lay the man they loved on the pyre. She had carried the body away," etc. Here the poet realises the significance[64] of the rite quite clearly and consequently notes that the inability of the Danes to carry it out added materially to their sorrow. In such a case the possibility of Christian authorship seems to me to be definitely excluded.
On the hypothesis that these descriptions had come down from the days of English heathenism all is easily explicable. At the time when the poem was Christianised it may very well not have been known that the rite of cremation was still practised among the heathen of the Continent, and in later days the verses of the old poet would be handed on in parrot fashion without their significance being generally understood. Well informed persons however, like Alcuin, who had travelled abroad, perceived clearly enough that, however much coated over with Christian phraseology, the heroic poems were in reality of an essentially heathen character.
Now cremation was widely prevalent in this country during the early days of the Saxon invasion—a fact attested by numerous cemeteries especially in the northern and midland counties, including the valley of the Thames. But it appears to have become a thing of the past when the Roman missionaries arrived here; otherwise it is difficult to account for the absence of any reference to the custom in the records which have come down to us. Indeed we may say with safety that it had passed out of general use, at least in the southern half of England, quite a generation before this time; for there are scarcely any traces of it to be found in those western districts which appear to have been conquered during the latter half of the sixth century. Consequently, if we are justified in believing that the descriptions of cremation ceremonies contained in Beowulf date from a time when the practice was still remembered, we must conclude that they were composed not later than the third or fourth decade of the seventh century[65].
But it is not contended, so far as I am aware, by any scholar that the account of Beowulf's obsequies belongs to the earlier parts of the poem. It is the final scene of the story, it is not contained in any speech, and further it is of a thoroughly epic character and would be quite out of place in a short lay. Hence, if the line of argument which we have been following is legitimate, we shall be forced to admit that though the poem has undergone a fairly thorough revision in early Christian times, it must in the main have been in existence some time before the conversion. I do not mean to suggest that the 'revision' was entirely limited to the religious element. Other changes and additions may have been made about the same time[66]. What I do mean is that the great bulk of the poem must have been in existence—not merely as a collection of lays or stories, but in full epic form—an appreciable time before the middle of the seventh century.
The other heroic poems do not furnish us with any similar criteria for estimating the date of their composition, but there seems to be no valid reason for doubting that they are quite as early. Two of them, Deor and Widsith, are expressed in the first person and lay claim to being of a remote antiquity. Deor says that he had been the bard of the Heodeningas and that he had been displaced by a skilful minstrel named Heorrenda. Since in old Norse literature Hiaðningar (i.e. Heodeningas) means 'Heðinn and his men,'[67] and since Heorrenda can hardly be separated from the minstrel Hôrand in Kûdrûn, it would seem that the poet claims to have been a contemporary of Heðinn and Högni, with whose story we have dealt briefly above (pp. [8] f., [16]).