The reverse process cannot be traced so clearly. The division of a kingdom between brothers or other relatives does not seem as a rule to have led to a permanent partition. Very often indeed it was apparently no more than a temporary distribution of estates and spheres of jurisdiction, not necessarily in solid blocks[557]. In such cases the kingdom was still regarded as one property, of which the kings were joint possessors. But there can be no question that many kingdoms established on alien soil, e.g. in Britain, were offshoots from other kingdoms.

This consideration brings us to the much debated question of the relationship between kingdom and nation. It has been assumed by many scholars that among the Teutonic peoples the kingdom was a comparatively late outgrowth from the nation or tribe. In reality this problem seems to me to have much in common with that of the hen and the egg. With the earliest kingdoms of all we are not concerned here; it will be enough to mention that our earliest historical notices testify to the prevalence of kingship, though not always to monarchy in the strict sense of the term. In the Heroic Age however we certainly find kingdoms springing up where no nation or tribe, properly speaking, can be said to have existed previously. We may cite the case of Odoacer, who in 476 made himself king in Italy with the help of his troops. In principle we may regard him as the princeps of a comitatus. Then we have to consider the various Anglo-Saxon kingdoms. The Mercian royal family traced their descent from Offa, the ancient king of Angel, while the West Saxon dynasty claimed to be sprung from that Wig, the son of Freawine, who was earl of Slesvig under Offa's father, according to the story preserved by Saxo. But to the origin of the rest the genealogies give us no clue. If they were all of royal origin—and apparently they did claim divine descent—the Angli must have possessed a numerous royal class; and we are scarcely justified in denying that this may have been the case[558]. On the other hand it is by no means impossible that some of them were sprung from foreign peoples, such as the Danes, Swedes or Warni. But what we may regard as practically certain is that the individual kingdoms did not rest upon a national or tribal foundation. There is not the slightest ground for supposing that (e.g.) the East Anglians as a people belonged to a different nation or tribe from the Northumbrians. It is scarcely credible that the first kings were anything else than principes in command of comitatus, whether they set out from the homeland in this position or established themselves at a later date by severing their allegiance from other kings in Britain itself.

Nor is there any reason for supposing that this phenomenon was peculiar to Britain. The story of Waldhere tells of the presence of small Teutonic communities in eastern Gaul, each under a royal family of its own. And not only heroic stories but also historical works relating to the earlier part of the Heroic Age frequently refer to comparatively small bands of warriors—such as that led by the Goth Sarus in Stilicho's time—in various parts of the Roman empire, and even beyond its borders[559]. Such bands may very well have produced communities like the one ruled by Waldhere's father[560]; but it would be absurd to speak of them as nations or tribes. They have clearly far more in common with the military kingship established by Odoacer. The peculiarity of his position indeed lies only in the magnitude of the power to which he attained.

In brief we have to distinguish between two classes of kingdoms in the Heroic Age. In the new kingdoms, settled on foreign soil, we find an essentially military kingship, an imperium vested in a particular family. These kings either established themselves in Roman cities, such as Ravenna, Langres or York, or moved about from one royal estate to another. Of national assemblies we have frequently no trace at all, while the council is identical with the comitatus and consists of relatives and nominees of the king. Such kingdoms often rest on no national or tribal foundations; the king and his comitatus form the nucleus of the organism. On the other hand the older kingdoms, especially in the North, retained many features of a more primitive constitution. The king's position had a religious significance, and his capital, e.g. at Leire or Gamla Upsala, was the chief national sanctuary, at which assemblies, primarily religious but possessing considerable political influence, took place from time to time. It is likely too that the councils here were originally permanent bodies with more or less fixed prerogatives—essentially religious, but yet by no means without political power. Between these two types of kingdoms we find others of an intermediate character, especially in nations which had migrated en masse; and there can be little doubt that during the Heroic Age even the most conservative of the older kingdoms were influenced by the newer type. It is the newer type of course which we must regard as truly characteristic of the Heroic Age.

In post-heroic times again we find a reversion to the national idea of a kingdom, though on a much larger scale. In English history this tendency can be traced from the seventh century onwards. In Bede's works it is clear that such an expression as Merciorum gens (Myrcna maegþ) had come to mean something more than the royal family of the Mercians with their property and dependents. By the ninth and tenth centuries however this feeling is much more clearly perceptible. We may cite King Alfred's will, where it is clearly recognised that the kingdom should not be divided up as a family property. But it is not until the time of Aethelred II that the full sense of the king's responsibility to the nation finds expression in definite terms.

The form of government which we find depicted in the Homeric poems seems to be not unlike that which we have discussed above. Here too kingship is universal—apparently also without any recognised constitutional limitations to the royal authority. The murder or expulsion of a prince is not unknown; but such cases are due to strife within the royal family. Any differences which we can detect between the authority of Homeric kings and that wielded by early Teutonic rulers may be ascribed partly to the much smaller size of the kingdoms and partly to a social feature noted in the last chapter (p. [363] f.), namely that in many Greek communities kingly or princely rank seems to have been claimed by a number of different families.

The last consideration is especially prominent in the Odyssey[561]. The throne of Ithaca has been in the possession of one family for three generations. Yet in I 394 ff. Telemachos says that there are many kings of the Achaeans, both young and old, in the island, and he expects that one of them will take the sovreignty, now that Odysseus is dead. In Scheria also we hear of twelve sceptre-bearing kings under Alcinoos; but to this case we shall have to return shortly. That a king was not necessarily a person of great magnificence may be inferred also from one of the scenes depicted on Achilles' shield (Il. XVIII 556 ff.), where we find a king in the harvest-field watching the work of the reapers and feasting on the spot. We are reminded here of the story of the Norwegian king Sigurðr Sýr who was summoned from the harvest-field to greet his stepson, St Olaf, and whose state-robes had to be sent to him there in order to enable him to make a suitable appearance[562].

In this connection it is perhaps worth noting that according to the Catalogue of Ships the contingents supplied by several communities were under a number of different princes. Thus the Epeioi have four leaders and the Boeotians five, without counting those from Orchomenos and Aspledon. The troops from Argolis (exclusive of Agamemnon's dominions) are led by three princes, all of whom according to later authorities were related—two of them, Diomedes and Sthenelos, being sons of Adrastos' daughters, while the third, Euryalos, was his brother's son[563]. Diomedes is said to be the commander-in-chief, but as Sthenelos is his charioteer the relations between them are evidently of an intimate character. It is not stated whether all these princes were actually reigning kings, or merely leaders selected for the expedition; but Diomedes and Sthenelos at least have no fathers living.

There is no evidence, so far as I am aware, for any form of election for kings. In the case of Bellerophon we are told that "the Lycians apportioned him a demesne" (τέμενος τάμον); but it was the king who granted him half the royal rights. And similarly in all other cases the kingly power seems to have been obtained from some relative by blood or marriage. This renders it more easily intelligible that the plural kingship—if such it was—of which we have spoken above, may be due ultimately to family arrangements. It is scarcely necessary to suppose that in such cases the kingdom was always divided into geographical halves and quarters. As to the relationship between the various kings under such an arrangement—e.g. whether the phrase συμπάντων ἡγεῖτο applied to Diomedes in Il. II 567 means a formal recognition of lordship on the part of his colleagues—we have apparently no precise information.