In the course of the Heroic Age many of the Teutonic peoples were converted to Christianity. The change of faith began among the Goths soon after the middle of the fourth century and must have spread very quickly to the Vandals. The Gepidae and Langobardi seem to have followed the example of these peoples in the course of the following century. At the time of Justinian's accession the Heruli were probably the only Teutonic people in eastern central Europe who remained heathen. In the west the Burgundians accepted Christianity apparently about the beginning of the fifth century, and the Franks before its close. The conversion of England took place in the seventh century; that of the Frisians and Old Saxons for the most part in the eighth. The Northern Kingdoms in general were little affected by the change until towards the close of the tenth century, though the first missionary efforts in Denmark and Sweden began before the middle of the ninth. In parts of Sweden the heathen religion lingered on until late in the eleventh century.

In the Nibelungenlied it is clearly recognised, perhaps through scholastic influence, that the multitude assembled at Attila's court included both Christians and heathens; but no such distinction is drawn in the English and Norse poems. In the former all the characters are made to speak as Christians, though they observe heathen rites; in the latter no indication is given that any of the characters were Christians. In point of fact there can be little doubt that most of the persons who figure in the heroic stories were heathens. In all probability such was the case with the earlier Goths, Eormenric and his contemporaries, as well as with all the characters of the Danish cycles. On the other hand the later Goths, Theodric and his contemporaries, were certainly Christians, and so also were the Burgundians, Guthhere and his brothers, as well as Alboin, king of the Langobardi.

With the Christian religion we are not concerned here; for, greatly as it influenced the Teutonic peoples, it was in no sense native. It is to the religion which Christianity displaced that we must give our attention. Unfortunately however the records which have come down to us from the Heroic Age itself are entirely of foreign authorship, and on the whole they give us extremely little information on this subject. We are bound therefore to base our account of Teutonic religion upon the comparatively abundant evidence preserved in Scandinavian literature, though we must not assume that the religion of the Heroic Age possessed the characteristics which we find in the North some five centuries later. When we have given a brief summary of the chief features of this later religion we shall have to discuss in somewhat more detail the small amount of information available for the earlier period. This is rendered all the more necessary by the fact that in works dealing with the subject the religion of the Heroic Age has not generally been distinguished from that of the Germans of Tacitus' time.

Now the feature which will probably strike any one most forcibly from a careful study of Northern religion is an extraordinary discrepancy between the mythical stories contained in the Edda and elsewhere on the one hand and references to actual religious observances on the other. In the former we find the gods grouped together in an organised community, of which Othin is the recognised head. Frigg is his wife, Thor and many of the other gods his sons. Most of the mythical stories deal with Othin's exploits and adventures, and serve to illustrate his power and wisdom. On the other hand the references to religious rites point in quite a different direction. In Iceland, for which our records are most full, there is practically no evidence for the worship of Othin. Thor is by far the most prominent figure, and after him Frey; occasionally also we hear of NiÖrÐr[586]. References to the worship of other supernatural beings, elves and landvÆttir (genii locorum), are not unfrequent. In notices referring to Norway the evidence is not very different. We do indeed sometimes hear of worship paid to Othin, especially in legendary stories, relating to early times; but in references to what may be called the historical period—the tenth and eleventh centuries—Thor and Frey are distinctly more prominent.

Two explanations have been given of this curious phenomenon. One is that the cult of Othin was introduced into the North at a comparatively late period and that it had not yet obtained a real hold at the time when Iceland was settled. This explanation has no foundation in tradition. Indeed the evidence of the stories points to an entirely opposite conclusion. Moreover it is worth noting that according to Procopius (Goth. II 15) the inhabitants of 'Thule' (i.e. Scandinavia) worshipped 'Ares' more than any other god. Since Othin is essentially a god of war it is natural to suppose that he is the deity meant, rather than the somewhat obscure TÝr.

The other explanation is that the cults of Othin and Thor belonged to two different classes of the community, the former to princely families and their retinues, the latter to the country people, more especially the (non-official) landowners. This explanation seems to be in complete accordance with the facts. There is no evidence for the worship of Othin either in early or late times except by princes or persons attached to their courts, while there are very few instances of the worship of Thor by such persons. Further we may note that while names compounded with ThÓr- (e.g. ThÓrkell, ThÓrolfr) are about the commonest type of all among the ordinary free population, both in Norway and Iceland—and such names are significant since they denote that the persons who bore them were dedicated to the god—they are practically unknown in royal families It will be convenient now to give a short sketch of the two deities and their cults.

Thor is represented as a middle-aged man of immense bodily strength. He is well disposed towards the human race and looked upon as their protector against harmful demons, to whom he is an implacable foe. In the poems ThrymskviÐa and HÝmiskviÐa and in a number of prose stories we have descriptions of Thor's adventures with giants, in which he is generally represented as breaking their skulls with his hammer. He uses no weapon except the hammer, and when he travels he either walks or drives in a car drawn by goats. When he comes to the assembly of the gods he is said to wade through certain rivers on the way. His escort never consists of more than three persons; very often he goes alone. The picture which the stories give us is clearly that of an idealised Norwegian countryman of primitive times. There are scarcely any traces of his original connection with the thunder, though in Sweden it was clearly remembered.

The portraiture of Othin offers the greatest possible contrast to that of Thor. He is represented as an old man, generally with one eye, and he gains his ends not by bravery or physical strength but by wisdom and cunning. Sometimes we find him coming, usually in disguise, to giants or witches, in order to gain from them some magical power or knowledge of the future; sometimes he imparts his knowledge, again generally magical, to men. He presents his favourites with weapons and instructs them in the art of war. Above all he is the god who gives victory in battle.

Othin's chief dwelling is called ValhÖll (the 'hall of the slain'), and all persons who fall in battle were believed to go to him there. Hence we find such expressions as 'to go to lodge with Othin' or 'to go to Valhalla' used as euphemisms for 'to be killed.' Before joining battle it is said to have been customary to throw a javelin over the enemy with the words 'Othin has you all.' After a battle prisoners were commonly sacrificed to Othin, and on such occasions, and indeed at all human sacrifices, the formula regularly used was: 'I give thee to Othin.' The usual method of sacrifice was by hanging or stabbing or a combination of both. With this practice we may probably connect a somewhat obscure myth recorded in HÁvamÁl, str. 138, according to which Othin was sacrificed to himself, by hanging and stabbing, on the world-tree. Certainly it is to be noted that the sacrifices to Othin seem to have been invariably human. They were clearly rites of quite a different character from the sacrificial feasts frequently mentioned in the sagas, where the victims consisted of horses, oxen and other edible animals, part of which was offered to the gods, while the rest was consumed by the worshippers. We do sometimes hear of horses being sacrificed with men, but on such occasions dogs and hawks are also mentioned, and there is no evidence that any of the victims were eaten. There are very few records of human sacrifices to any god except Othin.

The picture of Valhalla presented to us in the poems is a glorified copy of a military king's court[587]. The vast number of slain warriors assembled there in Othin's service spend their days in single combats and their evenings in feasting. Beside them we find the Valkyriur ('choosers of the slain'), Othin's adopted daughters, who distribute ale to the feasters. These also are sent out by Othin to decide the issue of battles and to select warriors for Valhalla. It is noteworthy that the term Valkyriur seems to be applied both to supernatural beings—what may perhaps be called minor divinities—and also to living women endowed with supernatural powers, such as that of flying. Thus both Brynhildr and SigrÚn, the wife of Helgi Hundingsbani, are called Valkyries; and it was for deciding a fight contrary to Othin's command that the former was punished with perpetual sleep.

In Ynglinga Saga, cap. 8, Othin is said to have ordained "that all dead men should be burnt and brought on to the pyre with their property. He said that every dead man should come to ValhÖll with such property as he had on the pyre.... But the ashes were to be cast out into the sea or buried down in the earth." Valhalla seems to be represented as a spirit world somewhat far away and not connected at all with the burial place. This observation brings us to another remarkable discrepancy between the traditions and the customs which we find actually prevailing in the North. We know both from descriptions in the sagas and from discoveries made in modern times that in the last few centuries before the adoption of Christianity it was customary to bury the dead in their ships or in elaborately constructed wooden chambers—the whole being covered with a barrow of considerable size. In the Prologue to Snorri's Heimskringla this custom is said to be of later date than the one attributed to Othin; first was the age of burning, then the age of barrows. Now there is evidence both from the discoveries and from the sagas themselves that the barrows were regarded as sacred and that the spirits of the dead were believed to dwell either within them or in the immediate neighbourhood. Not unfrequently we hear of persons coming to a barrow to consult the spirit. Sometimes the ghost, embodied in the corpse, even defends his property against grave-robbers. The activities of the dead are often represented as injurious; but this is by no means always the case. On one occasion we hear of a dispute between several different districts for the possession of the body of a king whose reign had been distinguished by great prosperity.

On the whole then it is clear that the cult of the dead was practised in the North very much as in most other parts of the world. Yet modern discoveries have brought to light abundant evidence for cremation in the early iron age—sometimes in spots which are marked by no external monument—so that the statements of Ynglinga Saga may be regarded as based on good tradition. We are driven to conclude therefore that in their conception of immortality, as in their theology, the inhabitants of the North held two wholly inconsistent views—or, perhaps it would be more correct to say, two entirely opposite views as to the desirability of retaining the souls of the dead. In Iceland the practice of cremation seems to have been extremely rare, but when it was resorted to the object is said to have been to get rid of a troublesome ghost. The one view of immortality was by no means so closely bound up with the cult of Thor as the other was with that of Othin. But it certainly prevailed among Thor's worshippers.

The next most important deity after Othin and Thor was Frey. His cult was widely spread in Norway and Iceland; yet according to tradition its true home was Sweden[588]. The Swedish royal family and nobility traced their descent from Frey, and Upsala, their capital and the chief sanctuary of the North, was believed to have been founded by him. In Ynglinga Saga, cap. 12 f., we have an account of him which is worth quoting as an illustration of Northern manes-worship. Frey is here represented as a prince whose reign was characterised by unparalleled prosperity. His death was concealed for three years. But when it became known, the Swedes would not burn him; for they believed that prosperity and peace would last as long as Frey was in Sweden. They made a great barrow for him therefore and poured into it the tribute which they had been wont to pay him; and they worshipped him for prosperity and peace ever afterwards. A very similar account is given of the Danish king FrÓÐi the Peaceful—from which we may infer that in Frey we have to deal not with a deified man but with a mythical character—a 'king of the golden age.' His name originally seems to have meant 'prince' or 'lord' (Ang.-Sax. frea, cf. ??a?); very probably it was at one time a title of the Swedish kings[589].

Frey appears to be regarded as a youthful god. The blessings for which he was worshipped were peace and fertility, both of the crops and livestock, as well as of the human race. His power of controlling the weather may be accounted for by his association with the Swedish kings (cf. p. 367); but it is clear that his character contains elements drawn from more than one source. His father NiÖrÐr, who is sometimes associated with him, possesses much the same characteristics, though he appears to be more particularly connected with the sea. There can be little doubt however that both he and his son have inherited the attributes of an ancient earth-goddess. Although there has been a change of sex, NiÖrÐr's name is identical with that of Nerthus (id est Terra Mater), a deity who according to Tacitus, Germ. 40, was worshipped on 'an island in the ocean'—in all probability SjÆlland. NiÖrÐr also has a daughter called Freyia (i.e. ??assa, ??sp???a), who is represented as a female counterpart of Frey. It is worth noting that she is sometimes associated with the next world. According to GrÍmnismÁl, str. 14, she shares the slain equally with Othin.

The deities with which we have just been dealing were collectively known as Vanir. They were held to be of a quite different stock from the Aesir, to whom Othin and Thor belonged, and according to the mythology had been given to the latter as hostages. Of the other deities those who figure most prominently in mythical stories are Frigg (Othin's wife), Ullr, Hoenir, TÝr, Heimdallr, IÐun, Gefion and Balder; but we seldom hear of worship paid to any of these.

In the Edda all the gods together form a regularly organised community. Their home is called ÁsgarÐr, and they hold their meetings beside the 'world-tree,' Yggdrasill's Ash. It is to be observed that ÁsgarÐr is a totally different conception from Valhalla[590]; it is not an abode of the slain. Indeed in this connection Othin himself does not appear to be represented as a god of the dead. But apart from ÁsgarÐr each god has a special abode of his own—Thor at ThrÚÐheimr, Ullr at Ýdalir, NiÖrÐr at NoatÚn, Balder at BreiÐablik, etc. All these localities are mythical—or at all events incapable of identification. It is a striking characteristic of Northern mythology that the gods are not associated with any known localities. Practically the only exceptions are Frey and Gefion, who are connected by tradition with Upsala and SjÆlland respectively; and neither of these connections is preserved in the poems of the Edda. In order to understand this feature we must of course bear in mind the fact that our mythological records are almost entirely derived from Iceland, which lies far away from the old national sanctuaries.

It is probably due to the same cause that we hear but little of special cults. In Iceland the only noteworthy exception is that, beside the more usually prevailing cult of Thor, we find a number of persons who are devoted to the service of Frey. Certain chiefs bear the title FreysgoÐi ('priest of Frey'); in one case a whole family bore the surname FreysgyÐlingar. Temples apparently sometimes contained the figures of a number of gods, though Thor's or Frey's is usually the only one mentioned by name. In Norway however the case is somewhat different. We hear frequently of temples and statues of Thor, occasionally also of those of Frey. But in addition to these there are notices of sanctuaries belonging to other deities—though not to Othin. In FriÐÞiÓfs Saga, cap. 1 (and passim), mention is made of a temple and image of Balder in the district of Sogn. It is the fashion to treat this incident as a product of antiquarian speculation; but there is little in the story itself to justify such a view, and the fact that the worship of Balder is not found elsewhere proves nothing. More important however is the fact that in a number of records we hear of statues and temples of ThÓrgerÐr HÖlgabrÚÐr, with whom her sister Irpa is sometimes associated. There can be no doubt that under the rule of Earl Haakon of Lade the cult of ThÓrgerÐr was more prominent than that of any other deity, at least in the district of Trondhjem. This fact is the more remarkable because ThÓrgerÐr and Irpa are never associated in any way with the rest of the gods; in the poems of the Edda and even in Gylfaginning their existence is ignored.

A very interesting illustration of the practice of special cults occurs in NiÁls Saga, cap. 88, which describes a temple owned in common by Earl Haakon and GuÐbrandr, a powerful hersir (hereditary local chief) in the highlands. This temple contained figures of ThÓrgerÐr and Irpa and also of Thor in his car[591]. We know from other sources that the cult of Thor was hereditary in the family of GuÐbrandr. Indeed it appears to be generally true that families adhered to the same cult from generation to generation[592], though in one case we do hear of an Icelander bearing the title FreysgoÐi, who belonged to a family distinguished for its service to Thor.

The relations between the worshipper and his deity were of a personal and intimate character; he regarded the latter as friend, counsellor and protector. Where the two are of different sexes the relationship is apt to take a conjugal form. Thus in the FlateyiarbÓk, I p. 107 f., Olafr Tryggvason, after robbing one of Earl Haakon's temples, and carrying off the image, calls out in derision: "Who wants to buy a wife? I think ThÓrkell and I are now responsible for this woman, since she has had the misfortune to lose her husband who was exceedingly dear to her." One of the bystanders then addresses the image: "How is it, ThÓrgerÐr, that thou art now so humiliated and stripped in unseemly wise of the splendid apparel wherewith Earl Haakon had thee clothed when he loved thee?" So in the poem HyndluliÓÐ Freyia speaks of her devoted worshipper, Óttarr the son of Innsteinn, as her husband. We may compare with this the fact that in the FlateyiarbÓk, I 337 f., the priestess in charge of Frey's temple in Sweden is said to have been called his wife. I see no reason therefore for supposing that Snorri was giving rein to his imagination when he stated (Yngl. Saga, cap. 5) that Gefion was the wife of SkiÖldr who, though a mythical character (cf. p. 131 f.), was not a god.

Sometimes again we meet with a definitely hostile attitude towards a deity—generally Othin—and it must not be supposed that such ideas first arose after the introduction of Christianity. In Saxo's translation of the lost BiarkamÁl the hero suspects that Othin is among the enemy and expresses his eagerness to attack him. If once he can catch sight of him, he says, the god will not escape from Leire unharmed[593]. Such ideas can only be explained by a vivid anthropomorphic conception of the deities.

The same attitude appears elsewhere. In Gautreks Saga, cap. 7—a story which contains many archaic features—we find the destiny of a man being determined by Othin and Thor, the former of whom is friendly to him, the latter hostile. In the introduction to GrÍmnismÁl as the result of a disagreement with Othin Frigg plays a trick upon him which leads him into serious trouble. Nor is the married life of NiÖrÐr and SkaÐi as happy as might be wished. But the chief cause of discord among the gods is the malicious Loki. In the poem Lokasenna he charges most of the chief goddesses with unfaithfulness or unchastity, while at the same time he reproaches the gods with unseemly conduct or with being involved in humiliating positions. The picture of the divine community which the poem presents to us is anything but pleasant. No doubt Loki is representing every circumstance in the most unfavourable light possible; but there appears to be a definite mythical foundation for most of his charges.

Loki serves as a connecting link between the gods and the iÖtnar ('giants'), a class of beings who are represented as generally hostile to both gods and men. Yet there are exceptions to this rule; and some of the gods, e.g. NiÖrÐr and Frey, have wives from the iÖtnar. Next to them we must mention the dwarfs, who are distinguished for their cunning and skill in metallurgy. Neither of these classes however can properly be regarded as objects of worship. Elves were certainly worshipped, but only collectively, as far as we know. In early records they are scarcely ever spoken of as individuals. Most probably their origin is to be sought in animistic conceptions, connected with the cult of the dead. On this last subject enough has been said above; we need only add that the formal deification of dead men was not unknown[594]. Sacred trees and groves also figure as prominently as in other parts of Europe.

Thus far we have been dealing with the religion of the Viking Age, primarily as we know of it in Iceland and Norway. But we have seen that the actual records of religion in Iceland agree in no way with the theology of the Edda. Nor can it truly be said that the evidence for Norway shows a better case. Here too we find the worship of Thor and Frey. But for the worship of Othin, NiÖrÐr, Freyia and Balder the evidence is slight and generally doubtful. For that of the rest of the gods there is no evidence at all. On the other hand we find that the deity who after Thor figures most prominently of all in these records is one who is entirely unknown to the theology of the Edda. The only conclusion which it seems to me legitimate to draw from these facts is that the mythology of the Edda is not a true reflection of Norwegian religion, at all events as it existed in the Viking Age.

Now let us consider the various deities individually. It will be convenient to begin with those whom we know to have been worshipped in Norway or Iceland. There is no question that Thor was known not only in Sweden and Denmark but also in Germany and England, under the forms Donar and Thunor respectively. Apart from local nomenclature and the use of his name (as a translation of dies Iouis) in the fifth day of the week, there are a few direct references to worship of him—e.g. in the inscription on a brooch found at Nordendorf in Bavaria and in a Low German renunciation formula for the use of converts. His cult goes back without doubt to the Heroic Age and probably much earlier, though he is apparently not mentioned by Tacitus.

The cult of Frey was believed to have come from Sweden, as we have seen. How old it was there we do not know; but there is some reason for believing that it was not originally confined to that country. The Slavonic inhabitants of eastern Holstein worshipped a deity of the same name[595]; and the presumption is that they found the cult in existence when they occupied that district—not later than the seventh century. But the name Yngvi has a much longer history and can be traced in various records back to the time of Tacitus. From what is said of Ing in the Anglo-Saxon Runic Poem it is clear that he was a perfectly definite, though doubtless mythical, personality[596].

We have already noticed that NiÖrÐr can be traced back to a goddess Nerthus, who was worshipped in the first century by the Angli and other peoples in the south-western part of the Baltic. When the change of sex took place we do not know. The feminine form of the deity is probably preserved in Freyia, who under the name Skialf seems to have her roots in early Swedish tradition.

With Othin we shall have to deal presently. There is abundant evidence that he was known not only in Sweden and Denmark but also in England and at least the greater part of Germany. In the two latter countries he bore the names Woden and Wodan respectively. Even in Tacitus' time he appears (under the name Mercurius) as the chief god.

Balder's history is not so clear. From Saxo's account (p. 70 ff.) there can be little doubt that he was known in Denmark. The question whether he was recognised in Germany[597] depends practically upon the interpretation of the (second) Merseburg charm, to which we shall have to refer again shortly.

Now let us take the deities who are known to us only from the mythology. Both Frigg and TÝr were certainly known in England and Germany. Their names are preserved in the sixth and third days of the week. Frigg (Frea) also figures, as the wife of Wodan, in the Langobardic story quoted above (p. 115), while TÝr (Mars) is mentioned more than once by Tacitus.

Of Ullr traces are preserved in local nomenclature both in Denmark and Sweden. From Saxo (p. 81 f.) it appears that he was remembered in Danish tradition. Gefion's association with SjÆlland (cf. p. 400) is recorded by Bragi Boddason, the earliest Scandinavian poet of whom anything has been preserved. Both her name and that of IÐun can be traced in local nomenclature in the same island[598].

There remain of course a large number of less important deities who cannot be traced outside the mythology of the Edda. Many scholars hold that these were invented by Norwegian or Icelandic poets during the Viking Age; but it is at least equally possible that our inability to trace them elsewhere is due in part to the extreme poverty of our information. One piece of evidence which tells in favour of the latter view is that the Merseburg charm preserves the name of one of the least prominent of these deities—Fulla, the handmaid of Frigg. The fact too that these poets made no attempt to incorporate ThÓrgerÐr in the pantheon seems to show that in their time[599] the theological system of the Edda was more or less crystallised. At all events it is clear that, with the exception of ThÓrgerÐr and Irpa, all the deities whose worship is attested were known beyond Norway, and that most of them can be traced back to the Heroic Age or still earlier times.

There is a further reason for doubting whether the theology of the Edda was a product of late Norwegian poetry. Perhaps the most striking conception in this theology is that of the 'world-tree,' Yggdrasill's Ash. I have pointed out elsewhere[600] that this conception is largely derived from a tree-sanctuary and that a fairly close parallel to it is furnished by the description of the Upsala sanctuary given in Adam of Bremen's history (IV 26 f.) and the annexed scholia. Similar sanctuaries may have existed in Norway; but we have no record of one which possessed the same characteristics, and it is extremely improbable that any of them ever attained an importance comparable with that possessed by the Swedish capital. Again, there are features in the picture of the 'world-tree'—I would allude especially to the presence of snakes—for which no parallels can be found in any Scandinavian sanctuary of which we have record. Yet such features do occur in the tree-sanctuaries of more primitive peoples, especially among the Prussians and Lithuanians. From this it appears to me highly probable that the conception of the world-tree dates from a comparatively early period. The idea of universality which it embodied cannot be held to prove the contrary; for this idea was possessed also by the Irminsul[601], the sacred pillar of the Old Saxons. We have no reason for doubting that a philosophical conception such as this was possible before the Viking Age.

I am inclined therefore to think that the theological system of the Edda in its main features dates from times anterior to the Viking Age. From earlier sources—the works of Tacitus and various German and English authorities—we know altogether the names of about a score of deities, half of whom belong to either sex. It is probable however that a much larger number have been lost. At all events there can be no doubt that the religion of the Heroic Age was a highly developed polytheism. Procopius (Goth. II 15), speaking of the inhabitants of 'Thule' (Scandinavia), says that they worship many gods and demons (da???a?), both in the heavens and in the atmosphere, in the earth and in the sea, besides certain other spirits (da????a) which are said to be in the waters of springs and rivers. Again, in the preceding chapter he states that the Heruli of central Europe worshipped a great crowd of gods (p???? t??a ??????te? ?e?? ?????), whom they thought it right to appease even with human sacrifices. Procopius' evidence is important not only because it is almost contemporary but also because he clearly distinguishes between the religion of the Teutonic peoples and that of the Slavs. Of the latter he says (ib. III 14) that "they consider one god, the creator of the lightning, to have sole control over all things, and they sacrifice to him oxen and offerings of all kinds.... Yet they also reverence rivers and nymphs and some other spirits (da????a), and sacrifice to them all, using divination in these sacrifices." It will be seen that this type of religion is not very far removed from what we find among the Thor-worshippers of Iceland.

In an earlier chapter (p. 255) we quoted from the Life of St Ansgar the story of a man who claimed to have been present at an assembly of the gods. From this story it is clear that in Sweden not very long after the beginning of the Viking Age the gods were believed to form an organised community. For earlier times no such explicit information is to be found; but we can scarcely doubt that some similar belief prevailed during the Heroic Age. In the Langobardic story quoted above (p. 115) Fria (Frigg) is the wife of Wodan, as in the Edda. In the (second) Merseburg charm we find a number of deities taking part in an incantation. Of the goddesses Sunna is said to be the sister of Sinthgunt, and Volla the sister of Fria. Evidence to the same effect is furnished by a letter of Bishop Daniel of Winchester to St Boniface, in which the writer speaks of a genealogy of the gods and advises his correspondent to put awkward questions to the heathen regarding the origin, numbers and relationships of their deities[602]. It may be noted that the earlier Anglo-Saxon genealogies, which go back probably to the seventh century, trace Woden's ancestry back for five generations.

Of course it is not to be denied that some of the deities of whom we hear may have been recognised only locally or by certain nations or confederacies. Such an explanation is very likely in the case of more than one deity mentioned by Tacitus, whom we cannot identify with any probability. In later times the same may be true of the god Fosite[603], to whom an island in the North Sea—identified with Heligoland by Adam of Bremen—was wholly dedicated. It is quite possible too that the god Seaxneat (Saxnote), who is mentioned in the Renunciation Formula and from whom the kings of Essex claimed descent, was worshipped only by the Saxons.

But even if such evidence was a good deal stronger than it actually is we should not be justified in inferring from it that the religion of the Heroic Age was of an essentially national rather than universal character. It is not only in Northern records of the Viking Age or the Christian period that we hear of families which were supposed to be descended from Othin (Woden). Out of the eight royal genealogies of the English kingdoms which have come down to us seven are traced back to the same deity; and it is highly probable that most of these date from before the conversion[604]—i.e. from within a century of the Heroic Age. But Woden was not a national but a universal deity.

Moreover what little we do know of this god from English and German sources is in full conformity with the character which he bears in Northern records. In the Anglo-Saxon poem on the Nine Herbs he is skilled in magic; in the Merseburg poem he is an expert in incantations. In the Langobardic story we find him represented as the giver of victory. In Tacitus' time he was already worshipped above the other gods; and human victims were offered to him. The same author (Ann. XIII 57) records the custom of dedicating a hostile army to Mars and Mercurius—a vow which entailed the total destruction of everything belonging to the enemy. The great deposits of antiquities which have been found at ThorsbjÆrg, Nydam, Vi and elsewhere are commonly believed to be relics of such dedicatory spoils. Finally, later popular belief often placed Woden at the head of the Wild Hunt or ghostly army. For the existence of a conception corresponding to Valhalla we have no explicit evidence[605]. But such a doctrine would clearly be in full accord with all that we know of the cult.

In funeral rites both inhumation and cremation were practised. The latter custom however seems to have died out almost everywhere before the introduction of Christianity—in England about the middle of the sixth century, among the Franks and Alamanni much earlier. Only among the Old Saxons it lingered apparently until towards the close of the eighth century, when it was rigorously put down after their subjugation[606]. How far the two practices were associated with different conceptions of immortality, as in the North, it is impossible to tell. In Beowulf (cf. p. 54) cremation is regarded as a pious duty owed to the dead; but all heathen references to the destiny of the soul hereafter have been removed from the poem. On the other hand there is evidence from later times for offerings at the grave, necromancy and all other practices usually associated with the cult of the dead[607].

The most important piece of evidence however on this subject is furnished by Procopius' account of the Heruli (Goth. II 14). He states that with them it was not lawful for a man to die of old age or disease. When he felt himself to be dying he had to request his relatives to make away with him as soon as possible. They had then to construct a huge pyre and set the dying man in the highest part of it. A compatriot, though not a relative (cf. p. 346), is then sent up to stab the man, and on his return the wood is immediately kindled. When the fire is burnt out the remains are collected and buried forthwith, and the widow is required to strangle herself at the tomb. Such rites as this are commonly ascribed to the desire to set the soul free while in possession of its faculties. But in view of the fact that in Northern tradition cremation is bound up with the doctrine of Valhalla[608]—a doctrine which is in no way inconsistent with this explanation—it is certainly significant that the two rites should be associated here, more especially since the Heruli were an essentially military people. On the whole the evidence of this passage is distinctly favourable to the view that a belief closely approximating to the doctrine of Valhalla was prevalent during the Heroic Age.

Valkyries (walcyrgan) are not unfrequently mentioned in Anglo-Saxon literature[609], and it is clear that similar beings were known in Germany, though this word does not occur in extant records. In England, as in the North, both human and supernatural beings were included under this term, though they are not always clearly distinguished from witches. But, more than this, the poetic description of valkyries which we find in the Edda[610] can likewise be traced in Anglo-Saxon poetry. In a charm against sudden pains we hear of mighty women who rode over the hill, mustered their host and cast their spears. The idea that sudden pains were due to the agency of such beings[611] comes doubtless from popular belief; but the description cannot be accounted for in this way. Again, in the (first) Merseburg charm we find supernatural women (idisi) taking part in a battle; and it is to be remembered that the word walcyrge can hardly mean anything else than 'chooser of the slain.' Certainly we have no evidence to prove—or disprove—that the valkyries were associated with Woden in early times. But the features noted here again point clearly to the existence of a conception akin to Valhalla and, what is more, to the poetic treatment of such a conception.

In the course of this discussion[612] I have endeavoured to point out that the theological system of the Edda cannot properly be regarded as an invention of (Norwegian-Icelandic) poets of the Viking Age—that, on the contrary, it is derived in great measure from much earlier times. I do not mean to deny that the growth of the system has been very largely influenced by poetry. But the evidence seems to me to show that the poetic treatment of the subject had begun—and probably more than begun—in the Heroic Age and among many of the Teutonic peoples. It is perfectly true that the notices of Teutonic religion contained in Tacitus' works convey the impression that religion was regarded as a very serious matter and that the general attitude towards the gods was highly reverential. The same impression is conveyed by Alcuin's account of Fositesland; and probably no one will deny that the euthanasia of the Heruli was based upon a very real conception of immortality. But to compare such records with the poetry of the Edda would manifestly be absurd. For analogies to them we must turn to notices relating to actual religion, and here we shall find evidence that the people of the Viking Age were no less religious than those of earlier times. We may instance the reverence shown by ThÓrolfr of Mostr to his holy hill and Earl Haakon's devotion to ThÓrgerÐr HÖlgabrÚÐr.

On the other hand the attitude towards the gods shown in the Edda finds an exact analogy in the only record of 'theological' poetry which has survived from the Heroic Age. In the Langobardic story (cf. p. 115) the anthropomorphisation of the deities is already complete; and the chief god[613] is duped by his wife. We could scarcely wish for a better parallel to the account given in the introduction to GrÍmnismÁl. In view of this story it is scarcely possible to doubt that familiarity, not to say levity, in the treatment of the gods characterised the poetry of the Heroic Age, just as much as that of the Viking Age.

It would be well to hesitate however before assuming that the gods of Tacitus' time were treated in the same way. His account shows that Teutonic theology had then passed beyond the purely tribal stage, and that certain deities were worshipped by a number of peoples, if not universally. But it does not suggest the existence of a highly anthropomorphic conception of the gods. Further we have to bear in mind that Tacitus is separated only by a century and a half from Caesar. The account of German religion given by the latter (B. Gall. VI 21) is difficult to account for by any explanation. But unless we are to believe that Caesar was thoroughly imposed upon we must conclude that nothing in the nature of a developed polytheism can have existed in his day. To the theology of the Heroic Age his account of the Gaulish gods (ib. VI 17) would be far more applicable than what he says regarding the worship of the Germans.

The question we have been discussing appears to throw some light upon the rapidity with which most of the Teutonic peoples accepted Christianity. The facts which we know with regard to the conversion are as follows: (1) that it almost invariably began in the king's court; (2) that violent opposition was offered only in kingless communities, as among the Old Saxons, or in defiance of the king's authority, as in Norway; (3) that after the conversion the gods (in general) disappear at once and for good; (4) that magical practices and the belief in spirits and even in certain female agricultural deities ('Erce,' Holda, Berhta, etc.) lasted among the country people for many centuries. From (4) we may probably infer that the religion of the country people was chiefly animistic—similar no doubt to what we find in Iceland, with the exception that we have little evidence for the cult of the thunder-god. Again, the explanation of (3) hangs together with (1); for the statements of ecclesiastical writers render it clear that the religion of the courts was essentially theistic. But it is plain from the discussion in the Northumbrian council recorded by Bede (H. E. II 13)—the only discussion of this kind of which we have any detailed account—that here at least this religion retained little vital force[614]. This fact is fully explained if, as I have endeavoured to point out, theology had largely passed from the realm of dogma into that of poetry.

The conclusion then to which we are brought is that Teutonic religion, at all events in the courts, underwent a profound change in the course of the Heroic Age. It is to be observed that in the earlier part of that age—as in the earlier part of the Viking Age—we find, especially among the more northern peoples, a fanatical devotion to warfare for its own sake, accompanied by lust for destruction and apparently also by a vivid conception of a life hereafter. In the latter part of the Heroic Age these phenomena disappear, except among the Heruli who, according to Procopius, differed from all the rest of mankind. The ideal which the princes of the later period set before themselves may be gathered both from Beowulf and from Roman authorities; it was to enjoy wealth and splendour in this life and to have their fame celebrated by future generations. For their attitude towards a future life the speech of the Northumbrian councillor recorded by Bede (l.c.) may probably be regarded as typical[615].

In a work such as this it is scarcely necessary to give even a brief summary of the characteristics of Greek religion, since the main outlines of the subject are probably much more familiar than those of even the later religion of the North. At the same time the amount of information which has been preserved is so great and the unsolved problems presented by the subject so numerous that it is clearly better left in the hands of experts. I shall attempt no more therefore than to call attention to the salient points in which the religion of the Homeric poems differs from that of later times and to the chief characteristics in which the former resembles or differs from the religion of the Teutonic Heroic Age and the Viking Age.

The various objects of worship recognised in Greece belong in general to much the same categories as those which we have noticed above. They may be classified roughly as gods, genii locorum and manes. In the last class we may perhaps include the 'heroes,' though the position occupied by them is somewhat peculiar. They were for the most part characters of the Heroic Age, and sometimes we find the cult of the same hero recognised in a number of different states. In general the worship of the gods took a different form from that paid to the manes and heroes, though occasionally the cult of a deity seems to have been associated with, or superimposed upon, that of a hero.

In the Homeric poems the gods figure much more prominently than the other classes. The most frequently mentioned of them are Zeus, Hera, Poseidon, Athene, Apollo, Artemis, Hephaistos, Aphrodite, Ares and Hermes. Zeus, as head of the divine community, corresponds to Othin (Woden); but in other respects there is little resemblance between the two. As god of the thunder his affinities are rather with Thor. Hera, as wife of the chief god, may be compared with Frigg, while Poseidon, as god of the sea, has an element in common with NiÖrÐr; but he is also an earthquake god, which the latter is not. His characterisation also is much more clearly marked. Apollo and Artemis, as a pair of young deities, brother and sister, with certain characteristics in common, bear some resemblance to Frey and Freyia; but the sexual element, so prominent in the Northern deities, is wanting in the Homeric poems. On the whole the translator of the Icelandic New Testament seems to have been happily inspired in rendering Diana (Artemis) by Gefion[616], while Freyia's true counterpart is rather to be found in Aphrodite[617]. Hephaistos, the smith of the divine community, has no Northern god corresponding to him; as the maker of heroes' armour and other metal objects he plays the same part as Weland. Ares, as god of war, has an element in common with TÝr. The Homeric Hermes bears no resemblance to Othin; his duties are to a certain extent discharged by the valkyries in Northern mythology. Athene is a character totally foreign to Northern theology.

All the above deities, together with a number of others less important, form a regularly organised community, like the Aesir. Their home is located on Mount Olympos in the north of Thessaly—a conception probably more primitive than ÁsgarÐr, which is never represented as a place known to the human race. But, though Olympos is the home of the gods collectively, most of them (like the Northern deities) have also one or more dwelling-places of their own, often in distant localities. Thus Poseidon's home is at Aigai and Apollo's at Delphoi, though he is also connected with several localities on the eastern side of the Aegean—Chryse, Cille, Tenedos, etc. Athene has a home at Athens, while Hephaistos is connected with Lemnos, Ares with Thrace and Aphrodite with Paphos in Cyprus. Zeus himself, apart from Olympos, has abodes at Dodona and Ida.

The belief that the gods had homes of their own in various localities is clearly to be taken in connection with the fact that they are said to have sanctuaries in the same places. Thus Aigai is mentioned, together with Helice, as a place where sacrifices were offered to Poseidon (Il. VIII 203 f.). Zeus' home at Dodona is his sanctuary (ib., XVI 233 ff.), and it is to her sanctuary at Paphos that Aphrodite goes (Od. VIII 362 f.). We need not doubt therefore that, as in later times, the cults of the various deities were largely of a local character. It is a different question of course whether the cults—or rather the deities themselves—were of local or tribal origin; but in certain cases such an explanation appears to be by no means improbable. Many scholars believe that Ares was originally a Thracian deity; and the fact that he is represented as the father of Boeotian heroes can scarcely be regarded as conclusive evidence to the contrary. In Il. IV 51 f. Hera states that three cities, Argos, Mycenae and Sparta, are specially dear to her; and there is little evidence that her cult was ever prominent in any other part of the Greek mainland. If the same explanation is true of such deities as Apollo and Aphrodite we must suppose either that their cults have spread from one locality to another or that deities belonging to different localities have been identified. The local origin of river gods, such as Spercheios and Scamandros, is of course clear enough; but these are little more than genii locorum.

But the really important feature in Homeric theology is that the various deities, whether of local origin or not, are all brought together as members of one community, or rather family, and that as such they are represented as meeting with universal recognition. Indeed, they are not regarded even as peculiar to the Greek race; the Trojans recognise the same divinities, and Poseidon goes to receive sacrifices even from the Ethiopians. A similar, though less important, community is that of the marine deities, one of whom, Thetis, figures prominently in the Iliad.

The spirit in which the gods are treated is in general very similar to what we find in the Edda, and in no way more reverential. Sometimes the treatment is humorous, as in ThrymskviÐa; sometimes again the deities are represented in a very unfavourable light. For the trick played upon Wodan by Fria in the Langobardic story (cf. p. 115) we have an interesting parallel in Il. XIV 292-353, where Hera distracts Zeus' attention from the war and coaxes him to sleep. But this is only one of a number of incidents which give a generally unpleasant picture of the domestic life of this pair, much more so than is the case with Othin and Frigg in the Edda. Again, the story of Ares and Aphrodite, which forms the subject of Demodocos' lay (Od. VIII 266-366), is very much what we should expect to find in a poem dealing at length with the scandalous charges brought against the goddesses in Lokasenna. Ares' speech in v. 292 ff.—as also Hermes' remarks in v. 339 ff.—may be compared with those of Frey in SkÍrnismÁl. In the Iliad the same two deities are more than once treated contemptuously.

The same spirit is shown in the treatment of the relations of gods with men. It is frequently recognised that deities ought to show gratitude to their worshippers for the sacrifices offered to them, as Freyia does in HyndluliÓÐ, though at the same time they are at liberty to refuse a petition, as in Il. VI 311—a case which may be compared with Frey's conduct in VÍga-GlÚms Saga (cf. p. 253). In the Iliad we find several deities taking an active interest in the fortunes of the war—Poseidon, Hera and Athene on the side of the Achaeans, and Apollo, Ares and Aphrodite on that of the Trojans. Poseidon more than once comes to rouse the Achaeans, when he thinks they are becoming slack. Apollo and Athene interfere in the struggle in a manner which strikes the reader as unfair; indeed the latter is frequently guilty of flagrantly mean and dishonourable conduct towards her opponents. The attitude of the heroes themselves to the gods is quite in keeping with the way in which the latter treat them. Achilles openly abuses Apollo for deceiving him (Il. XXII 15 ff.); and Diomedes attacks and wounds both Ares and Aphrodite, but only after assurances or assistance from Athene. In this respect the Homeric princes are not quite so bold as the hero of BiarkamÁl (cf. p. 402 f.).

But it is by no means only towards communities that the deities display their favour or hatred. Indeed their attitude towards the contending forces at Troy seems to be largely determined by their relations with certain individuals, especially Helen and Paris. In the Odyssey Poseidon is represented as persecuting Odysseus, one of the Achaean leaders, for many years on account of the injury done by him to Polyphemos. On the other hand both this hero and Diomedes enjoy the special favour of Athene. It is worth noting that Diomedes appears to have inherited this favour from his father (cf. Il. V 800 ff.), and the same may be true in the case of Odysseus (cf. Od. XXIV 367 ff., 517 ff.); at all events it is continued towards his son. In neither of these cases is any hint given of conjugal relations, such as we find in the North; indeed the sexual element seems to be practically absent from Athene. Elsewhere however such relations are clearly involved, e.g. in the cases of Aineias and Achilles, who are the sons of Aphrodite and Thetis respectively. In the Odyssey (V 119 ff.) the principle is stated plainly; and the hero himself has conjugal relations with Circe and Calypso, both of whom are described as goddesses.

Of other mythical beings the genii locorum are perhaps the most prominent. The distinction between them and the class with which we have been dealing is by no means so clearly drawn as in the north of Europe (cf. p. 407 f.). Many of them are even described as gods (?e??). River gods, such as Spercheios and Axios, are represented as the fathers or ancestors of several heroes. Among other, less important, beings of the same type we may mention the nymphs[618] to whom certain caves and springs were sacred (e.g. Od. XIII 347 ff., XVII 205 ff., 240 ff.).

References to chthonic deities are not very frequent. The Erinyes are mentioned several times, especially in curses; but they are not individualised, though the singular is occasionally used. The most important person in the under-world appears to be the queen, Persephoneia; but she is never actually brought upon the scene. Hades himself is seldom more than a name.

The sacrifices to the gods mentioned in the Homeric poems are as a rule similar to the sacrificial feasts of which we hear in the North.[619] A portion of the victims, usually bulls or rams, was offered to the deity, while the rest was consumed by the worshippers. On the other hand we hear occasionally of victims which were given wholly to the gods. Such appears to have been the case with the victims sacrificed on the occasion of a solemn oath; thus in Il. XIX 267 f. the boar is thrown into the sea[620]. It is perhaps worth noting that both here and in III 276 ff. the oath refers to chthonic deities[621], as well as to Zeus, the Sun and Earth. Victims sacrificed to the dead likewise seem to be offered entire (e.g. Od. XI 44 ff.; Il. XXIII 166 ff.). Moreover it is only in connection with funeral rites that we hear of human sacrifices, namely when Achilles puts to death twelve Trojan youths at the pyre of Patroclos (ib., 175 f.). Horses and dogs are also sacrificed on this occasion, as well as oxen and sheep, and the whole scene is in accordance with Northern custom. No mention is made in the poems of sacrifices such as that of Polyxene, which form so favourite a theme with the dramatists; but in view of the Northern evidence[622] we are entitled to doubt whether they are altogether inventions of later times.

The theology of the Homeric poems received formal recognition in Greece down to the acceptance of Christianity; but there is no evidence that the system as a whole possessed any vital force even in the earliest times of which we have record. From the sixth century onwards poets and philosophers began to regard Zeus as much more than the chief of the gods; but even in popular religion it appears that each state honoured certain deities, while the rest were largely or altogether neglected. Thus at Athens more prominence seems to have been given to Athene, and perhaps also to Poseidon, than to Zeus, while Hera was predominant at Argos and Samos, and Apollo at Delphoi, etc. This is a feature for which the Homeric poems themselves give evidence, as we have seen, and there is no reason whatever for supposing it to be of later growth. Sometimes too we find prominence given to deities, such as Demeter and Dionysos, who seem to be of little consequence in the Homeric poems, while other cults, such as that of Serapis, were introduced from abroad in comparatively late times.

Chthonic deities are more prominent in later literature than in the Homeric poems; and many authors describe the worship paid to them down to a comparatively late period. Yet the rites seem to have been of a primitive character. But perhaps the most striking element in the religion of classical Greece was the worship of 'heroes.' Every city possessed shrines (???a), at which sacrifices were offered to heroes with rites similar in general to those used in the worship of chthonic deities. Originally these heroes seem to have been local persons and the shrines their tombs; but characters prominent in heroic poetry sometimes received worship in many different states.

In the Homeric poems there is no evidence for this hero-worship. Indeed these poems contain few traces of a cult of the manes at all, except in funeral ceremonies and in connection with the necromantic sacrifice of Odysseus. This fact has been connected with the Homeric doctrine of immortality[623], and in view of the Northern evidence (cf. p. 397 ff.) there can be little doubt that the true explanation is to be found herein. The method used in the disposal of the dead is cremation, and, as in the North, it was believed that this sent the spirit away from the body to a place of the dead—not a separate place for each particular family or community but a common home for the souls of the whole Greek race. In Il. XXIII 65 ff. the spirit of Patroclos comes and exhorts Achilles to pay him the last rites: "Never again shall I return from Hades, when ye have allotted me the due of fire" (V. 75 f.). So also when Odysseus visits the home of Hades the first spirit he meets is that of his follower Elpenor, who reproaches him with not attending to his obsequies and begs him, when he returns to Circe's island, to "burn him up with his arms, all that he possesses." The idea is clearly the same as in Ynglinga Saga, cap. 8. The honourable way therefore to treat a fallen foe is not to strip him but to burn him with his armour, as in Il. VI 417 ff.[624] Even at the beginning of the historical period this belief seems not to have entirely died out. According to the story told by Herodotus (v. 92), when Periandros sent to consult the spirit of his wife Melissa, she complained that the clothes which he had given her at her funeral were of no use because they had not been burnt.

It has already been mentioned (p. 261) that this story seems to show that at one time the home of the dead was located in a definite, though probably not very well known, region; and the vague indications given as to the hero's wanderings in the Odyssey can hardly be regarded as evidence to the contrary. If this is correct the home of Hades is in one respect probably a more primitive conception than Valhalla, just as Olympos is more primitive than ÁsgarÐr. A reasonable explanation would be offered if we had evidence that part of the population of Greece was believed to have come from that region[625]. An idea of this kind was certainly in Snorri's mind when he wrote the Ynglinga Saga; but unfortunately the account which he has given is obviously, at least to a large extent, of scholastic origin.

In other respects however the Homeric conception of immortality appears to be less simple than the doctrine of Valhalla, as we find it in certain Northern records. The use of the verbs ??pte?? and ta???s??s? is probably capable of a different explanation; but there are certain other words and expressions, e.g. ?at???e?, ??e???, ?e?? ?ata??????? (Hades; cf. also Il. XX 61 ff.), which seem to point to a belief that the world of the dead was beneath the earth. That is the conception implied in certain local beliefs which we meet with in later times, at Hermione and elsewhere, where a deep cavern or lake was supposed to lead thither. More important however is the fact that the home of Hades is totally devoid of the attractiveness of Valhalla; the gloomy picture which is drawn of it accords rather with the Northern home of Hel. The burning of the body, together with armour and funeral sacrifices, appears to procure little real advantage to the soul; it is regarded, as in Beowulf, rather in the nature of an honour due to the deceased. Yet in view of the parallels which we possess, both from the North and elsewhere[626], we can scarcely doubt that the custom had its origin in a much stronger motive, the force of which was scarcely appreciated in later times. For a change of feeling, if not of actual faith, positive evidence is supplied by the reprobation expressed in Il. XXIII 176—a passage as significant in its way as the homiletic verses which follow the reference to heathen sacrifices in Beow. 175 ff. (cf. p. 53).

In the course of this chapter we have noticed a considerable number of features common to Greek and Teutonic religion. Many of them are to be found in other religions also, and these we need not discuss further. Here we are concerned only to determine what may be regarded as characteristic of the Heroic Age. In this category we may probably include the following features.

1. The religion was predominantly a worship of gods, rather than of spirits. Herein lies the chief contrast with the religion of later times in both regions[627]. In Classical Greece chthonic worship and hero worship seem on the whole to be more prominent than that of the gods. The same is true of Scandinavian countries in the latest heathen period, though Thor is still prominent and hero worship is scarcely distinguished from manes worship. In Germany and England, where Christianity was adopted during the Heroic Age or soon afterwards, the gods disappear at once, while forms of chthonic worship survive for centuries.

2. The same gods were, to a large extent at least, recognised everywhere. Whether by borrowing or by identification of cults they had ceased to be merely tribal deities. How far back this feature goes in Greece we cannot tell. Among the Teutonic peoples we can trace it in part back to Tacitus' time; but it was probably intensified during the Heroic Age.

3. The conception of the gods was definitely anthropomorphic. For the Teutonic Heroic Age this is made clear by the Langobardic story. There is no absolutely conclusive proof that the gods in general were regarded as forming a regularly organised community, as in the Edda and the Homeric poems; but all the evidence which we have (cf. p. 407 f.) points in this direction.

4. The relations between gods and human beings are of a somewhat peculiar character both in Teutonic poetry and in the Homeric poems; but they are almost identical in the two cases. The gods are not treated with any very great reverence. The conduct attributed to them is not unfrequently reprehensible, their purposes can often be thwarted by the help of other gods, and the bravest warriors are sometimes even ready to attack them openly. Yet the human and the divine are not confused; a man is not a god, though many heroes, both Teutonic and Greek, are sprung from gods. It should be observed that the deity from whom most English princes claimed descent is Woden, a universal and not a tribal god. This belief must be regarded as an anti-tribal force.

5. Both in Northern tradition and in the Homeric poems the practice of cremation was associated with the belief in a common home for the souls of the dead. This practice does not appear to have been common even in the earlier part of the Viking Age; and consequently the tradition probably comes down from the Heroic Age, at which time we know that cremation was widely prevalent. The cheerless home of Hades[628] differs considerably from Valhalla, though there is some ground for suspecting that the Greeks of the Heroic Age had once cherished a belief endowed with greater vitality. But both conceptions possess certain essential features in common, namely the removal of the soul to a distant place—a belief really incompatible with the local worship of heroes or manes—and the fact that this distant place of souls was a universal home and not reserved for the souls of one tribe. This belief again was doubtless an anti-tribal force of considerable importance.

Briefly then we may define as the predominant characteristic of heroic religion, both Greek and Teutonic, the subordination of chthonic and tribal cults, which as a rule go together, to the worship of a number of universally recognised and highly anthropomorphic deities—coupled with the belief in a common and distant land of souls. These characteristics, at all events among the Teutonic peoples, seem properly to have belonged only to the religion of the royal and military classes. Hence, when the royal families are converted to a new faith, as in England, or when kingless states grow up, as in Iceland, we find in all cases more or less of a reversion to the more primitive forms of religion. It is on the same principle that I would account for the differences in religion between heroic and historical Greece.

[586] These are the three gods mentioned in the solemn oath which, according to LandnÁmabÓk, IV 7 (HauksbÓk), had to be sworn on the sacred bracelet at all legal proceedings: hialpi mÉr svÁ Freyr ok NiÖrÐr ok hinn almÁttki Áss, etc. In the later MelabÓk (a compilation of the seventeenth century) it is suggested that Áss here means Othin; but I do not think this explanation is generally accepted. It is scarcely credible that Thor should be ignored on such an occasion.