1. Hamlet, I.v.7–8; Eyrbyggja saga, ch. 27, in 4 Íslenzk fornrit, Sveinsson, Einar Ó., ed. (Reykjavík, 1935)Google Scholar [hereinafter cited as Eyrbyggja]; English translation: Eyrbyggja Saga, trans, by Pálsson, H. and Edwards, P. (Toronto, 1973)Google Scholar. References to sagas are by saga name and chapter. For convenience I have identified an accessible translation where available. All translations from Old Norse and Old English are mine.

2. Early Icelandic law made a distinction between murder (morð) and killing (víg). Murder was an unacknowledged, secret, or concealed killing. The law of the medieval Icelandic Commonwealth goes by the name Grágás, literally ‘Grey Goose.’ The laws mostly date from the 12th and 13th centuries. See generally Ó. Lárusson, ‘On Grágás—The Oldest Icelandic Code of Law,’ Third Viking Congress (1958) 77–89. There are three main manuscripts of Grágás: Grágás: Codex Regius, Finsen, V., ed. (Copenhagen, 1852)Google Scholar [hereinafter cited as 1 Grágás]; Grágás: Staðarhólsbók, Finsen, V., ed. (Copenhagen, 1879)Google Scholar [hereinafter cited as 2 Grágás]; Grágás: Skálholtsbók, Finsen, V., ed. (Copenhagen, 1883)Google Scholar [hereinafter cited as 3 Grágás]. A translation of §§ 1–116 of 1 Grágás has recently appeared with other volumes scheduled for future publication; see Dennis, Andrew, Foote, Peter and Perkins, Richard, Laws of Early Iceland: Grágás (Winnipeg, 1980)Google Scholar. For the definition of murder, see 1 Grágás, § 88 and 2 Grágás, § 315.

Early English and Norman laws also made the distinction between killing and murder. See Pollock, Sir Frederick and Maitland, F.W., The History of the English Law, 2 vols., 2d ed. (London, 1898) ii, 486Google Scholar; and II Canute c.5.§ 1 in The Laws of the Kings of England England from Edmund to Henry I, Robertson, A.J., ed. (Cambridge, 1925)Google Scholar. See also Larson, L.M., The Earliest Norwegian Laws (New York, 1935) 422Google Scholar.

3. Early Icelandic law provided very detailed and complicated procedures for publishing killings and other actionable offenses. 1 Grágás, supra note 2, § 87 and 2 Grágás, §§ 278–81.

4. For the effect of ecology on feud, see Black-Michaud, Jacob, Cohesive Force: Feud in the Mediterranean and the Middle East (New York, 1975) 125–26Google Scholar, 162–72; Peters, E.L., ‘Some Structural Aspects of the Feud Among the Camel-Herding Bedouin of Cyrenaica,’ Africa XXXVII (1967) 261–82CrossRefGoogle Scholar.

5. For a discussion of the distinction between feud and war see Black-Michaud, Cohesive Force, supra note 4, 1–8, 22–23, 27–32, and works cited therein.

6. Brennu-Njáls saga, chs. 129–47, in 12 Íslenzk fornrit, Sveinsson, Einar Ó., ed. (Reykjavík, 1954)Google Scholar [hereinafter cited as Njál] English translation: Njal's Saga, trans, by Magnusson, M. and Pálsson, H. (London, 1960)Google Scholar.

7. The rule of equivalence raises some interesting questions. If you kill my brother, does equivalence require me to kill you or your brother? It is tempting to see a very strange manifestaton of the rule in Icelandic wergeld law, which provided for compensation payments to be made from each member of the killer's kin group out to fourth cousins to the corresponding member of the victim's kin group. 1 Grágás, supra note 2, § 113. The killer, balanced against the victim, was not included in the compensation schedule. It has been suggested that the reason lies in the law having assumed the exile of the killer and the confiscation of his goods. See Phillpotts, Dame Bertha, Kindred and Clan (Cambridge, 1913) 13Google Scholar. Rather than hypothesizing a successful outlawry action, I prefer to explain the killer's lack of liability to the lack of a receiver corresponding to him. No doubt the law also recognized that the killer's kin would make sure the killer subsidized their obligatory payments.

8. There is an extensive literature on the proper definition of ‘law’ and the ‘legal.’ It is a slough to be avoided. Useful discussions of the issues and a review of the literature can be found in Abel, Richard L., ‘A Comparative Theory of Dispute Institutions in Society,’ Law and Society Review VIII (1973) 217–347Google Scholar at 221–39 and Roberts, Simon, Order and Dispute: An Introduction to Legal Anthropology (New York, 1979) 17–29Google Scholar, 184–206.

The status of the bloodfeud as a juridical mechanism has also produced a small feud of its own in the anthropological literature. The position taken by various writers depends on their position in the dispute noted in the preceding paragraph. I do not undertake to supply a complete bibliography here, but the following works contain useful and enlightening discussions of the bloodfeud as an institution in its own right, having at times a distinctly juridical look to it. Evans-Pritchard, E.E., The Nuer (Oxford, 1940) 150Google Scholar 62; Hasluck, Margaret, The Unwritten Law in Albania (Cambridge, 1954)Google Scholar; Peters, ‘Structural Aspects,’ supra note 4; Black-Michaud, Cohesive Force, supra note 4. And among historians see especially Wallace-Hadrill, J.M., ‘The Bloodfeud of the Franks,’ in The Long-Haired Kings and Other Studies in Frankish History (London, 1962) 121–47Google Scholar; also Bloch, Marc, Feudal Society (Chicago, 1964) 128Google Scholar; Davies, R.R., ‘The Survival of the Bloodfeud in Medieval Wales,’ History LIV (1969) 338–57CrossRefGoogle Scholar.

Representative of the anthropologists whose definition of law leads them to discuss the feud as subculture violence, war, or anarchy are Bohannan, Paul, Social Anthropology (New York, 1963) 290–91Google Scholar; Pospisil, Leopold, Anthropology of Law: A Comparative Theory (New York, 1971) 2–10Google Scholar; and Radcliffe-Brown, A., Structure and Function in Primitive Society (New York, 1965) 208Google Scholar.

9. See, e.g., 1 Grágás, supra note 2, §§ 86, 90; 2 Grágás, § 275; 3 Grágás, 695, s.v. vígt; also Leges Henrici Primi, Downer, L., ed. (Oxford, 1972)Google Scholar c.36. lc, c.70.12.

10. The traditional view that the state and royal justice were incompatible with kin solidarity and bloodfeud has been qualified by Wallace-Hadrill, ‘Bloodfeud of the Franks,’supra note 8, 129: ‘when we come to inquire what it was that made the composition tariffs of Lex Salica work and why wergelds and lesser compositions were in fact paid, the answer is, not fear of local royal officials but fear of feud; or rather it was both.’ So too in Scotland where royal and private justice coexisted and were mutually dependent. See Wormald, Jenny, ‘Bloodfeud, Kindred and Government in Early Modern Scotland,’ Past and Present LXXXVII (1981) 54–97Google Scholar. See also Goebel, Julius Jr., Felony and Misdemeanor (Philadelphia, 1976) 17–25.Google Scholar

11. See Whitelock, Dorothy, The Beginnings of English Society, 2d ed. (Harmondsworth, 1974) 39–40Google Scholar; see, e.g., 1 Grágás, supra note 2, § 110; and Njál, supra note 6, chs. 75–77. See generally Hrafnkels saga Freysgoða, in 11 Íslenzk fornrit, Jóhannesson, Jón, ed. (Reykjavík, 1950)Google Scholar [hereinafter cited as Hrafnkel]; English translation: Hrafnkel's Saga, trans, by Pálsson, H. (Harmondsworth, 1971)Google Scholar.

On the difficulty of enforcing a judgment of outlawry, consider the quip made to a successful plaintiff: ‘I guess you got this much out of your lawsuit: you may call him an outlaw.’ Hrafnkel, ch. 4 (English trans., ch. 12).

12. See the important and influential article by Colson, E., ‘Social Control and Vengeance in Plateau Tonga Society,’ Africa XXIII (1953) 199–212CrossRefGoogle Scholar; Gluckman, M., ‘The Peace in the Feud,’ Past and Present VII (1955) 1–14CrossRefGoogle Scholar.

13. The burning of Njál and its consequences is a case in point.

14. For Iceland, see 1 Grágás, supra note 2, §§ 113, 118; 2 Grágás, § 56; Barlau, S.B., ‘Old Icelandic Kinship Terminology: An Anomaly,’ Ethnology XX (1981) 191–202CrossRefGoogle Scholar. For England, see generally Phillpotts, Kindred and Clan, supra note 7, 211, 224; Lancaster, Lorraine, ‘Kinship in Anglo-Saxon Society,’ British Journal of Sociology IX (1958) 234–50Google Scholar, 359-77; and Loyn, H.R., ‘Kinship in Anglo-Saxon Society,’ Anglo-Saxon England III (1974) 197–209Google Scholar.

The nature of Germanic kinship has been hotly debated for over a century, and the statement in the text to which this footnote is appended is only a little less controversial now than it was in 1900. The issue is whether early Germanic kin were agnatic clans—patrilineal descent groups—or whether the kin set was bilateral—cognatic. The debate has been characterized by remarkably able performances on both sides. Some early spokesmen for patrilineality were Sir Maine, Henry, Ancient Law (Everyman ed. 1917) 88–89Google Scholar; and Sir Vinogradoff, Paul, The Growth of the Manor, 2d ed. (New York, 1911) 135–36Google Scholar. Spokespeople on the side of cognation were Pollock and Maitland, History of English Law, supra note 2, ii, 240–45; Phillpotts, Kindred and Clan, supra note 7.

The more recent writers on both sides have greatly benefitted from the work of anthropologists who have shown that patrilineal descent does not preclude the recognition of cognatic kin, nor does a bilateral kinship system preclude lineal biases in certain areas. See, e.g., Fox, Robin, Kinship and Marriage (Harmondsworth, 1967) 50–52, 153Google Scholar; Radcliffe-Brown, Structure and Function, supra note 8, 23. In other words, the issue now is whether Germanic kinship is cognatic admitting patrilineal and patrilateral biases, or whether it is patrilineal admitting cognation. Fox cautions against making such distinctions, but the question has produced the following useful and interesting discussions. For patrilineality see Meinhard, H.H., ‘The Patrilineal Principal in Early Teutonic Kinship,’ in Beattie, J. and Lienhardt, R., eds., Studies in Social Anthropology: Essays in Memory of E.E. Evans-Pritchard (Oxford, 1975) 1–29Google Scholar; Charles-Edwards, T.M., ‘Kinship, Status, and the Origins of the Hide,’ Past and Present LVI (1972) 3–33CrossRefGoogle Scholar. For the bilateral kin set, see Bullough, D.A., ‘Early Medieval Social Groupings: The Terminology of Kinship,’ Past and Present XLV (1969) 3–18CrossRefGoogle Scholar; Lancaster, ‘Kinship in Anglo-Saxon Society,’ supra.

15. Pollock and Maitland, History of English Law, supra note 2, ii, 242.

16. The plight of the person bound to opposing sides is a favorite theme of Germanic literature, depicted in mythic grandeur in the Nibelung cycle and in more homely fashion in the actions of Styr þorgimsson. Eyrbyggja, supra note 1, ch. 45. Styr fought in a battle on the side of his first cousins once removed who were arrayed against his daughter's husband and his grandson. He killed a member of his son-in-law's party and then switched sides and evened the score by killing a follower of his cousins.

17. See Gluckman, ‘Peace in the Feud,’ supra note 12. In Gluckman's view, cross-cutting ties will inhibit the feud, or at least work as a deterrent to violence within the feud. Exogamous marriage, friendship, residence patterns, fostering, among other things, will inevitably cause some people to have close ties with both sides. These people ‘have an interest in bringing about a settlement of quarrels.’ Ibid. 8. In most situations, there will be someone so situated and should there not be, kin groups will set about establishing cross-cutting ties by strategic marriage arrangements. The sagas and Beowulf offer many examples of such marriages. See, e.g., Beowulf, Klaeber, F., ed., 3d ed. (London, 1951) 11. 1071–1159Google Scholar; Njál, supra note 6, chs. 93, 159. The frequency of the practice is also attested to by the Old English word freoðuwebbe, literally ‘peaceweaver,’ meaning ‘woman.’

How effective cross-cutting ties actually were in preventing violence is another matter. Consider Beowulf's words: ‘Seldom where a prince has fallen does the avenging spear lie still, even though the bride may be good.’ Ibid. 11. 2029–31. The person with ties to both sides need not necessarily be a force for peace. See, e.g., Styr's actions, in the preceding note. He resolved his conflict by killing on behalf of both parties. Yet despite Beowulf's cynicism and Styr's violent temperament, people acted as if they agreed with Gluckman. They still sought to create cross-cutting ties. They were well aware that such bonds would prevent violence in few or maybe even no cases, but they were convinced that peace stood a better chance if there were people on both sides with conflicting loyalties. See also J. Black-Michaud, Cohesive Force, supra note 4, 70–72.

18. See Fox, Kinship and Marriage, supra note 14, 50–52; Bullough, ‘Early Medieval Social Groupings,’ supra note 14, 15.

19. The English sources are remarkably opaque as to the relative size of the compensation paying and receiving groups. The evidence is assembled and discussed in Phillpotts, Kindred and Clan, supra note 7, 205-12 and MacCormack, G., ‘Inheritance and Wergeld in Early Germanic Law,’ Irish Jurist VIII (1973) 143–63Google Scholar at 157–63.

20. The actual extent of the bilateral kindred, that is, exactly how distant a cousin can be before ego will cease to recognize him as his kin, is a complicated problem for both England and Iceland greatly beyond the scope of this introduction. For Iceland, cf. 1 Grágás, supra note 2, §§ 25, 35 where second cousins of litigants can be challenged for interest if they are judges or members of the panel of neighbors with 1 Grágás, § 113 where fourth cousins are included as wergeld receivers and payors. The saga evidence shows examples of kin recruitment out to second cousins once removed; see, e.g., Njál, supra note 6, chs. 73-74. See also Phillpotts, Kindred and Clan, supra note 7, 37–41. For England see generally the works cited supra note 14.

21. There are many competing factors governing the choice of expiator. The killer might not be available, having already been killed or outlawed, or it simply might be prudent to avoid him. There is often a very strong sense that the killer should suffer, but this is frequently at odds with the rule of equivalence. See supra note 7. The killer is not the best expiator when he is of considerably lesser rank or reputation than his victim. There might also be an urge to humiliate the other side by killing its best man. In this regard, the following Norwegian law introduced into Iceland about 1270 is instructive:

May it be known to all men that a barbaric custom has prevailed in our countries for a long time: when a man has been killed, his kinsmen want to remove the best man in the [killer's] family, even though he is ignorant and innocent of the slaying. And they do not want to avenge themselves on the real killer, even when they have an opportunity to do so ….

Quoted in Lars Lönnroth, Njáls saga: A Critical Introduction (Berkeley, 1976) 146Google Scholar. Some sense of accomplice liability also extended the class of possible expiators. In the sagas vengeance frequently fell on those who accompanied the killer or even those who laughed at or heard a libellous insult. See, e.g., Njál, supra note 6, chs. 45, 91–92. See also the remarkable conversation between Guðrún and Snorri over choosing a vengeance target in which the merits and drawbacks of a series of possible expiators are discussed. Laxdœla saga, ch. 59, in 5 Íslenzkfornrit, Sveinsson, Einar Ó., ed. (Reykjavík, 1934)Google Scholar [hereinafter cited as Laxdœla]; English translation: Laxdœla Saga, trans, by Magnusson, M. and Pálsson, H. (Baltimore, 1969)Google Scholar. I discuss rules of vicarious liability in the feud and factors governing the choice of expiator in my ‘Justifying Skarpheðinn: Of Pretext and Politics in the Icelandic Bloodfeud,’ Scandinavian Studies LV (1983) 316–44Google Scholar.

The heart of early legislation intended to limit the bloodfeud lay in efforts to limit the class of expiators to the killer. See, e.g., Robertson, Laws of the Kings of England, supra note 2, II Edmund c. 1; and the 13th century Icelandic law quoted in the previous paragraph. The fair inference from these laws is that it was common and acceptable practice for the vengeance target not to have been the killer.

22. See Robertson, Laws of the Kings of England, supra note 2, II Edmund c. 1.§ 1; in some cultures the troublemaker is killed by his kin. For examples among the Eskimo, see Hoebel, E.A., The Law of Primitive Man (Cambridge, 1954) 91Google Scholar. Repudiations could apparently be made subject to a condition precedent. In Njál, supra note 6, ch. 13, Hrút, the spokesman for the kin, stipulated his niece's husband could kill her foster-father with impunity should the foster-father visit them for more than three days without the husband's permission.

23. The miserable lot of the kinless man is a common and moving theme of much Anglo-Saxon poetry. See, e.g., Beowulf, supra note 17, 11. 2231–70. He is also a frequent subject of the the laws. See, e.g., Attenborough, F.L., The Laws of the Earliest English Kings (Cambridge, 1922)Google ScholarIne c.23.§ 1, Alfred c. 31; Robertson, Laws of the Kings of England, supra note 2, I Canute c.5.§ 2c.

24. See Attenborough, Laws of the Earliest English Kings, supra note 23, II Edward c. 6. Cf., Robertson, Laws of the Kings of England, supra note 2, II Edmund c. 1.§ 2.

25. The mercantile diction reflects the language and sensibility of the early English and Icelandic laws, not my own views on the matter. The Icelandic laws provide that in order for children to be capable of inheriting, they must be born of a mother who was ‘bought’ with requisite brideprice—mundi keypt. 1 Grágás, supra note 2, § 118; 2 Grágás, § 58. A wedding ceremony went by the name brúðkaup, literally ‘bride-purchase.’ The early English laws use the phrase ‘to buy a woman or wife’ to mean ‘to marry.’ See Attenborough, Laws of the Earliest English Kings, supra note 23, Ine c. 31 (‘Gif mon wif gebyccge…’). See also ibid., Æthelberht c. 77.

26. E.g., the vengeance Kári takes on behalf of Njál and his sons in Njál, supra note 6, chs. 146–47; the support Vermund gives to þórarin in Eyrbyggja, supra note 1, ch. 19; the aid both Grettir and Atli receive from Gamli þórhallsson, Grettis saga Ásmundarsonar, chs. 42–43, 48, in 7 Íslenzk fornrit, Jónsson, Guðni, ed. (Reykjavík, 1936)Google Scholar [hereinafter cited as Grettir]; English translation: Grettir's Saga, trans, by Fox, D. and Pálsson, H. (Toronto, 1974)Google Scholar. See also Hœnsa-þóris saga, chs. 12–14, in 3 Íslenzk fornrit, Jónsson, Guðni and Nordal, Sigurður, eds. (Reykjavík, 1938)Google Scholar where distant affines figure prominently in prosecuting the killing case and in vengeance-taking.

27. See Lancaster, ‘Kinship in Anglo-Saxon Society,’ supra note 14, 247–48.

28. E.g., Laxdœla, supra note 21, ch. 27; and also Gísla saga Sútrssonar, ch. 2, in 6 Íslenzk fornrit, þórólfsson, Björn K. and Jónsson, Guðni, eds. (Reykjavík, 1943Google Scholar) describing, however, events in Norway [hereinafter cited as Gísli]; English translation: The Saga of Gisli, trans, by Johnston, G. (Toronto, 1963)Google Scholar.

29. See 1 Grágás, supra note 2, § 90 and 2 Grágás, § 293, which provide that a man has a privilege to kill for sexual assaults on any of six women: his wife, daughter, mother, sister, foster-daughter, and foster-mother.

30. E.g., Njál, supra note 6, chs. 42, 45, 92, 132; Egils saga Skalla-Grímssonar, ch. 40, in 2 Íslenzk fornrit, Nordal, Sigurður, ed. (Reykjavík, 1933)Google Scholar [hereinafter cited as Egil;] English translation: Egil's Saga, trans, by Pálsson, H. and Edwards, P. (Harmondsworth, 1976)Google Scholar.

31. Lancaster, ‘Kinship in Anglo-Saxon Society,’ supra note 14, 239.

32. The Old Icelandic word fóstbróðir applies to two types of relations: first, where two male children are raised together because one of them is being fostered by the other's parents, and second, where two men swear to avenge each other. See infra note 35. It does not appear that an adult son of the fosterer was considered the foster-brother of his father's foster-son. For example, in Njál, the Njálssons are nowhere referred to as foster-brothers of Nja's foster-son, Höskuld. See Miller, ‘Justifying Skarpheðinn,’ supra note 21, 319.

33. Laxdœla, supra note 21, ch. 16; this type of transaction was known in the laws as arfsal, literally ‘inheritance transfer,’ and it looked very much like a commendation of sorts. þórð apparently does not lose his free status, but he clearly has bound himself for life as Höskuld's client. Elsewhere, the sagas preserve similar accounts of grants of inheritance in consideraton of future support. E.g., Eyrbyggja, supra note 1, ch. 31; Vápnfirðinga saga, ch. 7, in 11 Íslenzk fornrit, Jóhannesson, Jón, ed. (Reykjavík, 1950)Google Scholar [hereinafter cited as Vápnfirðinga]; English translation: The Weaponfirthers'Saga in Four Icelandic Sagas, trans, by Jones, Gwyn (Princeton, 1935)Google Scholar. See also 1 Grágás, supra note 2, § 127; 2 Grágás, § 67.

34. See, e.g., Laxdœla, supra note 21, ch. 27 where Óláf wishes to patch up the rift between his half-brother þorleik and himself by offering to foster þorleik's son because ‘it is said, always is the lesser man foster-father to the other's son.’ See also Snorri Sturluson, Haralds saga ins hárfagra, chs. 38–40, in 1 Heimskringla 94, in 26 Íslenzk fornrit, Aðalbjarnarson, Bjarni, ed. (Reykjavík, 1941)Google Scholar; English translation: Heimskringla, trans, by Hollander, L. M. (Austin, 1964) 92–93Google Scholar.

35. In Old Icelandic a blood-brother could be called fóstbróðir (see supra note 32) and also, variously, ‘sworn-brother'—svaribróðir—or ‘oath-brother’—eiðbróðir, the latter names clearly distinguishing the relationship from fostering. The parties confirmed the oaths of reciprocal vengeance by passing under raised strips of turf where they mingled their blood with the earth. Descriptions of the ceremony are found in Gísli, supra note 28, ch. 6; Fóstbrœðra saga, ch. 2, in 6 Íslenzk fornrit, supra note 28; English translation: The Sworn Brothers, trans, by Hollander, L.M. (Princeton, 1949)Google Scholar.

36. See Gísli, supra note 28, ch. 6.

37. See Lancaster, ‘Kinship in Anglo-Saxon Society,’ supra note 14, 239.

38. Droplaugarsona saga, ch. 10, in 11 Íslenzk fornrit, Jóhannesson, Jón, ed. (Reykjavík, 1950)Google Scholar [hereinafter cited as Droplaug]; English translation: Droplaugarsona Saga in Three Icelandic Sagas, trans, by Schlauch, Margaret (Princeton, 1950)Google Scholar.

39. Two of the Saxon Chronicles Parallel, 2 vols., Plummer, Charles, ed. (Oxford, 1892–1899) i, 76Google Scholar, anno 878 A.D.

40. Alfred may have thought that Guthrum's baptism would give Guthrum a surety acceptable to Alfred to whom Guthrum could swear his oaths—namely God. The Anglo-Saxon Chronicle, ibid., notes that Guthrum's promise to accept baptism was secured with hostages and great oaths. There is no mention of Alfred having retained hostages after Guthrum accepted baptism. And a later treaty between them was confirmed with oaths ‘for themselves and their followers, both living and unborn, who seek God's mercy and ours.’ Attenborough, Laws of the Earliest English Kings, supra note 23, Alfred and Guthrum, preamble.

41. Turner, Victor, ‘An Anthropological Approach to the Icelandic Sagas,’ in Beidelman, Thomas O., ed., The Translation of Culture: Essays to E.E. Evans-Pritchard (London, 1971) 349–74Google Scholar at 364. Turner makes a case for the excellence of the sagas as resources for historical anthropological study. His discussion of Njál as an ‘anthropological paradise’ is, unfortunately, marred by occasional errors of detail.

42. There are exceptions. See, e.g., Laxdœla, supra note 21, ch. 43 where Bolli lives with his wife's kin. See also Gísli, supra note 28, ch. 5; Njál, supra note 6, ch. 90.

43. E.g., Njál, supra note 6, chs. 11, 13, 27. The woman given in marriage as part of a settlement between feuding kin groups was probably more hostage than wife. See supra note 17. Her kin often took care to provide her with a retinue sufficient to protect her against the ruder types of mistreatment. See Beowulf, supra note 17, 11. 2032 61.

44. 1 Grágás, supra note 2, § 78; 2 Grágás, § 232.

45. Some were not. 1 Grágás, supra note 2, § 81; 2 Grágás, § 242, provide that a legal household exists if a man has land or if he has milk cows. This allowed tenant farmers to qualify as heads of households and hence be eligible to sit on a panel of neighbors. 1 Grágás, § 89; 2 Grágás, § 287.

46. Njál, supra note 6, ch. 128; Landnámabók 155, in 1 Íslenzk fornrit, Benediktsson, Jakob, ed. (Reykjavík, 1968)Google Scholar; English translation: The Book of Settlements, trans, by Pálsson, H. and Edwards, P. (Winnipeg, 1972) 59.Google Scholar

47. E.g., Laxdœla, supra note 21, chs. 48–49; Njál, supra note 6, chs. 126–28; Grettir, supra note 26, chs. 30, 43. Cf. Black-Michaud, Cohesive Force, supra note 4, 48.

48. Settlement patterns in Anglo-Saxon England and Iceland, however, were distinctly different. Farmhouses were not contiguous in Iceland as they were in England; there were no villages or hamlets. Moreover, the English picture changes markedly through the course of the Anglo-Saxon period, free peasant holdings giving way more and more to manorial organization. See generally Loyn, H.R., Anglo-Saxon England and the Norman Conquest (London, 1962) 156–70Google Scholar, 195–98. See also Charles-Edwards, ‘Kinship, Status and the Origin of the Hide,’ supra note 14, 3–15. A fair inference, assuming the Anglo-Saxon kin group did not extend much beyond the fourth degree coupled with the existence of villages, would be that ties of neighborhood and friendship played a major role in group recruitment for the bloodfeud.

The Icelandic word for friendship—vinátta—also has a quasi-juridical significance, indicating an agreement between the friends to take counsel together and not to act inconsistently with the interests of the other. For further discussion see Miller, ‘Justifying Skarpheðinn,’ supra note 21, 339–41 and Byock, Jesse L., Feud in the Icelandic Saga (Berkeley, 1982) 42, 95Google Scholar.

49. Beowulf, supra note 17, II. 2501–08, 1053–55.

50. Tacitus, De Origine et Situ Germanorum, Anderson, J.G.C., ed. (Oxford, 1938)Google Scholar ch. 14.

51. See The Battle of Maldon, Gordon, E.V., ed. (London, 1937) II. 202–325Google Scholar.

52. Bede, Ecclesiastical History of the English People, Colgrave, Bertram and Mynors, R.A.B., eds. (Oxford, 1969) ii. 9Google Scholar.

53. This point has recently been made in Woolf, R., ‘The Ideal of Men Dying with Their Lord in the Germania and in The Battle of Maldon,’ Anglo-Saxon England V (1976) 63–81Google Scholar. Ms. Woolf's thesis is that the ideal is peculiar to the sources mentioned in her title. A retainer's duty was to avenge his lord, not to acquiesce in a suicide-like death once his lord was killed. Cnut hoped to make life mirror art by legislating on the subject (c. 1020). The heriot was excused for the man who fell before his lord in battle and his heirs were confirmed to his succession. Robertson, Laws of the Kings of England, supra note 2, II Canute c.78. The enactment was less an inducement to be a hero than to be an heir, so Cnut also legislated severe penalties for the man who fled from his lord in battle—he forfeited all his possessions, his life, and his heirs lost all rights of succession. Ibid. c.77.

54. See, e.g., The Battle of Maldon, supra note 51, ll. 231–53; Beowulf, supra note 17, ll. 2650–56, 2864–91.

55. Attenborough, Laws of the Earliest English Kings, supra note 23, Alfred c. 42 §§ 5–6. This law is designed more to limit the feud than to induce retainer loyalty. The lord's feuds were to remain affairs of the lord's kin and followers; they were not to involve his followers' kinsmen.

56. See Plummer, Saxon Chronicles, supra note 39, i, 46–50, anno 755 A.D.; and Woolf, ‘Ideal of Men Dying with their Lord,’ supra note 53, 69–70.

57. Attenborough, Laws of the Earliest English Kings, supra note 23, II Æthelstan c.2.

58. When the Althing was founded in 930, thirty-six chieftaincies (goðorð) were recognized. The constitution was reformed in 965 when the country was divided into four Quarters. At this time three new chieftaincies were created in the North Quarter. The other Quarters were brought up to parity with the North c. 1005, making forty-eight chieftaincies all told. Three chieftains holding the ‘ancient,’ that is, pre-965, chieftaincies presided over each of the three district things in each Quarter. They hallowed the assembly ground and supervised the business of the thing. All the chieftains sat together as a Court of Legislature at the Althing. The holders of the ancient chieftaincies selected the judges for the courts, both those which met at the district things and for the Quarter courts which met at the Althing. All of the chieftains selected judges for the Fifth court, which served roughly as court of appeals from the Quarter courts. Chieftaincies were transferable and inheritable; they could also be shared. Although the chieftains had wergelds no different from the ordinary freeman, they in fact constituted a recognizable aristocratic class. For further discussion see generally Jóhannesson, Jón, A History of the Old Icelandic Commonwealth, trans, by Bessason, Haraldur (Winnipeg, 1974) 35–82Google Scholar.

59. 1 Grágás, supra note 2, §§ 78, 81, 83; 2 Grágás, §§ 242, 243.

60. 1 Grágás, supra note 2, §§ 81, 83; 2 Grágás, § 243.

61. This is a very representative description of the course of an Icelandic bloodfeud. Wallace-Hadrill similarly characterizes the Frankish bloodfeud as ‘drift[ing] from blood to arbitration and back again, without ever becoming what we would call legally clear.’ Wallace-Hadrill. ‘Bloodfeud of the Franks,’supra note 8, 142–43. For the view that feud is an interminable process characterized by different phases ranging from peace and composition to blood, see ibid. 146–47; Black-Michaud, Cohesive Force, supra note 4, 63–85, 109–18; Peters, ‘Structural Aspects,’ supra note 4, 268. This view of the feud is also implicit in Heusler, Andreas, Das Strafrecht der Isländersagas (Leipzig, 1911) 38–47Google Scholar, where legal action and arbitrated settlement are considered no less characteristic of the Icelandic bloodfeud than blood vengeance.

62. A chieftain who could not help his followers win lawsuits, or simply got out-lawyered, would lose his followers to more successful practitioners. See, e.g., Vápnfirðinga, supra note 33, ch. 11; Njál, supra note 6, ch. 107.

63. For instance, when the sons as heirs of their chieftain father divided his property, it was the son skilled in law who would take the chieftaincy, or at least exercise the office on behalf of his brothers. See Droplaug, supra note 38, ch. 2; Vatnsdœla saga, ch. 27, in 8 Íslenzk fornrit, Sveinsson, Einar Ó., ed. (Reykjavík, 1939)Google Scholar; English translation: The Vatnsdaler's Saga, trans, by Jones, Gwyn (New York, 1944)Google Scholar.

64. 1 Grágás, supra note 2, §§ 20, 57.

65. The laws provide for the possibility of disruption by allowing as few as six of the thirty-six judges to reassemble in a safer place to finish judging. Ibid. § 41.

66. E.g., Hrafnkel, supra note 11, ch. 4 (English trans., ch. 11). See also Eyrbyggja, supra note 1, ch. 56; Njál, supra note 6, ch. 145; and Vöðu-Brands þáttr, ch. 4, in Ljósvetninga saga in 10 Íslenzk fornrit, Sigfússon, Björn, ed. (Reykjavík, 1940)Google Scholar.

67. In actions in the district things the support of the litigant's own chieftain and his followers was usually sufficient. In actions in the Quarter courts at the Althing larger numbers were recruited and chieftains to whom the litigants were not attached were approached for support. See, e.g., Njál, supra note 6, chs. 119–20. If this broader based support was not forthcoming an action was as good as lost. Injustices could occur. See, e.g., Bandamanna saga, in 7 Íslenzk fornrit, Jónsson, Guðni, ed. (Reykjavík, 1936)Google Scholar; English translation: The Saga of the Eight Confederates in Schlauch, Three Icelandic Sagas, supra note 38. See also, e.g., Grettir, supra note 26, ch. 46. But often the failure to get support was the result of a process similar to a verdict of the neighbors. It was a judgment on the merits of the case. See, e.g., Droplaug, supra note 38, chs. 7–8, where Helgi Droplaugarson could muster no support beyond his close kin because the case was ‘unpopular.’ Helgi had killed his step-father.

68. E.g., Eyrbyggja, supra note 1, ch. 10; Njál, supra note 6, ch. 122.

69. See Njál, supra note 6, ch. 129.

70. See Droplaug, supra note 38, ch. 4. This example is especially difficult. The laws provide that a sister's husband (SiHu) is part of ego's wergeld paying and receiving group. 1 Grágás, supra note 2, § 113. A wife's brother (WiBr), however, is not included in either group. Yet, from SiHu's viewpoint, it is to his WiBr that he owes a duty, and it would seem he would expect some mutuality of obligation. The problem could be avoided if SiHu and WiBr were the same man, as would be the case if a brother and sister each married into another brother-sister set. But these kinds of marriages were not especially favored in Iceland. The dilemma in Droplaug shows a chain reaction effect thus:

Hrafnkel and Helgi were close kin; Hrafnkel asked Helgi for a half-share in the chieftaincy which had once belonged to their common paternal ancestor. Helgi refused, so Hrafnkel went to Hólmstein, his SiHu, for aid. Hólmstein said he, personally, could not help him because he would not go against his SiHu, Helgi Ásbjarnarson; he sent Hrafnkel for aid to Helgi Droplaugarson who remarked that Hólmstein should regard his marriage to Hrafnkel's sister as more important than his obligation to his SiHu. What we see in Hólmstein's actions is an admission that there was some reciprocity of obligation. Hólmstein will not go against his SiHu although as Helgi Droplaugarson says in strictly legal terms his duty is to his WiBr, i.e., to the man whose SiHu he is. The end of the matter was that Hólmstein broke up a fight at the thing between the two factions and was instrumental in bringing about a settlement.

71. E.g., Gísli to Véstein in Gísli, supra note 28, ch. 6; Bolli and Kjartan in Laxdœla, supra note 21, ch. 28.

72. See supra text accompanying note 15.

73. In Icelandic law, the monetary liability of a kinsman for wergeld diminishes as the degree of kinship from ego increases. See 1 Grágás, supra note 2, § 113. The English picture, though hazier, is in accord; see Phillpotts, ‘Kindred and Clan,’ supra note 7, 205–6 and passim; and MacCormack, ‘Inheritance and WergeId,’ supra note 19, 160–63.

74. Eyrbyggja, supra note 1, ch. 27: Hér er nú þat höfuð er eigi myndi undan teljask at mœla eptir þik, ef þess þyrfti við.’

75. I have discovered no nonliterary evidence of the ceremony for Iceland or England. Analogous rituals, however, have been recorded by anthropologists in the 20th century. Although the literary provenance of the ceremony leads to some problems of interpretation, they are not greater in degree than problems of interpretation peculiar to other types of sources. If there is a danger that the ceremony may be purely an author's invention, there is the advantage that the descriptions are the richer in circumstance and social context for being preserved in literary settings.

There are some thirty so-called Icelandic family sagas. Written mostly in the second half of the 13th century, they purport to describe the feuds and disputes of the Icelanders of the ‘Saga-Age,’ c. 930–1030. The 250-year time lapse between word and deed has occasioned much debate as to the historicity of the sagas. The earlier view was that they recorded actual disputes with little or no fictional accretion; the present view treats them as works of art that may also happen to preserve material of historical interest. Whether the sagas are historical or not in the sense this debate means historical, i.e. whether the sagas are accurate chronicles, is largely irrelevant for the social historian or anthropologist. The feuding and disputing process the author describes will, if the writer has any intention of being comprehensible to his audience, accord with the way things were done in his time or an earlier time whose procedures were still remembered. I consider the saga evidence to reflect mainly the practice of the 13th century, although there are cases where the saga writer consciously recalls an earlier state of law. See, e.g., Eyrbyggja, supra note 1, ch. 38. On the problem of saga historicity see generally Andersson, T.M., The Problems of Icelandic Saga Origins (New Haven, 1964) 41–55Google Scholar; Hallberg, Peter, The Icelandic Saga (Lincoln, Neb., 1962) 49–69Google Scholar; Nordal, S., The Historical Element in the Icelandic Family Sagas (Glasgow, 1957)Google Scholar. On the historicity of legal matters in Njál, see Lönnroth, Njáls saga, supra note 21, 241–48; and also see Lehmannand, Karlvon Carolsfeld, H.S., Die Njálssage insbesondere in ihren juristischen Bestandtheilen (Berlin, 1883)Google Scholar. The best work on the law and the family sagas remains Heusler's Strafrecht, supra note 61.

76. Eyrbyggja, supra note 1, ch. 27. The excuses proffered by Vigfús's kinsmen for not acting on their duty are interesting in themselves. Of the three kinsmen approached by þorgerð, one—a second cousin—excused himself because of a promise made to the killer not to sue him as long as there were others to take up the claim, and two—a second cousin and a second cousin once removed—saw no reason why the action should fall to them when there were other kinsmen equally, and in the case of the second cousin once removed, more closely related to Vigfús than they were. Apparently, the theory of corporate liability then as now could serve conveniently to deny individual accountability by locating it vaguely somewhere else.

77. Eyrbyggja, supra note 1, chs. 23, 26.

78. Fault was very much a part of the Icelandic legal system. As in Vigfús's case, a community assessment of a party's fault figured prominently in what kind of support a legal action would get. Court procedure also allowed issues of fault to arise by way of counter-claim, or more frequently by putting in issue whether the injured party had become ‘unhallowed’ (óhelgi) by his own actions in the incident. A finding of óhelgi deprived a man and his kin of standing to maintain an action on his behalf. See, e.g., 1 Grágás, supra note 2, §§ 86, 90, 113; 2 Grágás, §§ 270, 271, 293. See generally 3 Grágás, 656, s.v. óheilagr. For the role of accident in bloodfeud, see Black-Michaud, Cohesive Force, supra note 4, 19–20, 112; Peters, ‘Structural Aspects,’ supra note 4, 270.

79. 1 Grágás, supra note 2, §§ 94, 118; 2 Grágás, § 297. The term aðili referred generally to both the rightful prosecutor and the chief defendant. The former could more specifically be designated as sakar-aðili, the latter as varnar-aðili. See Cleasby, R. and Vigfusson, G., An Icelandic-English Dictionary, 2d ed., ed. Craigie, W. (Oxford, 1957)Google Scholar s.v. aðili.

The devolution of the right or duty to be the aðili in a killing case followed the inheritance law except that no killing case was allowed to fall to any male under sixteen or to a woman even though they may have been the heir. But at the time of Vigfús's death, the law read otherwise; women had not yet been disabled as plaintiffs in killing cases. The law was changed c. 993, ironically enough, because of later developments in the feud we are concerned with here. Arnkel was killed by a group led by Snorri goði; Arnkel's heirs were his sisters, and it fell to them to prosecute the action. The outcome was disastrous. Only one attacker received any penalty whatsoever and that was a mild one: three years banishment. ‘Because the killing case of so great a chieftain as Arnkel had been handled so embarrassingly a new law was enacted disabling women and men younger than sixteen winters from being killing-case-prosecutors (vigsakaraðili).’ Eyrbyggja, supra note 1, ch. 38.

Single women over twenty and widows, however, had the power to settle actions for injuries to themselves, but they could not accept lesser compensation than the law provided for such an injury. If a woman wished to sue, she had to transfer the action to a man, since women were unable to take part in legal procedures at the thing. 1 Grágás, supra note 2, § 94; 2 Grágás, §§ 336, 340. Still, the case was hers to transfer.

80. 1 Grágás, supra note 2, § 113 recognizes only one woman, the daughter of the killer or the victim, as a participant in wergeld payment and receipt. A conflicting provision gives the compensation in the killing case to the heirs, whether men or women. Ibid. § 95; 2 Grágás, § 324. The wife, who is not an heir, is excluded by both provisions. 1 Grágás, § 118.

81. Making the responsibility for the killing case fall to the male heir was more a procedural and jurisdictional measure than an attempt to limit the extent of the bloodfeud. It provided a specific person, or at most, a group of brothers, in relation to whom challenges of judges and panels of neighbors could be made. 1 Grágás, supra note 2, §§ 25, 35. It also saved the defendant in a killing case from multiple outlawry suits at the hands of different branches of the victim's kin group who might not be related to each other and hence would not consider themselves bound by the other's actions. It is important to recognize that the failure to win a killing case did not preclude recourse to blood vengeance and, in most cases, it did not absolve the killer or his kin from liability for money compensation; 1 Grágás, § 113. Even a finding that the killing was justified because the victim was óhelgi (see supra note 78) did not necessarily mean his death would not be compensated for pursuant to an arbitrated settlement; see Heusler, Strafrecht, supra note 61, 61–65, 115–23.

82. E.g., foster-sons and fathers, blood-brothers, and affines could be avengers of the blood but were not possible aðili.

83. Eyrbyggja, supra note 1, ch. 27.

84. See Pollock and Maitland, History of English Law, supra note 2, ii, 83, 85; Leage, R.W., Roman Private Law, 3d ed., ed. Pritchard, A.M. (New York, 1961) 332–33Google Scholar.

85. Njál, supra note 6, chs. 41, 44–45.

86. Ibid. ch. 45.

87. 1 Grágás, supra note 2, § 238. If a man in any way suggested another was effeminate or took the passive role in homosexual or bestial couplings he could be killed with impunity. Calling attention to Njál's beardlessness was such an insult, as was the implication of oral copulation with farm animals in the epithet for the Njálssons.

88. Corpse mutilation is susceptible to multiple meanings depending on who does it and why. When the killer mutilates the corpse, it is usually to humiliate it or its kin. Skarpheðin decapitated the dead Sigmund to return an insult for an insult, and perhaps also to horrify Hallgerð. Achilles avenged Patroclus less by killing Hector, than by mutilating his corpse afterwards. Iliad Bks. XXII–XXIV. Likewise, Beowulf decapitated Grendel's corpse to repay him for his ravages of Hrothgar's retainers. Beowulf, supra note 17, ll. 1577–84.

Exultant victors made trophies of their victims' heads and other members. Beowulf bore Grendel's head back to Hrothgar as a ‘token of glory.’ Beowulf, supra note 17, l. 1654. Jarl Sigurð of Orkney had his men tie the heads of the slaughtered Scots to their saddle bows ‘to show their victory.’ Orkneyinga saga, ch. 5, in 34 Íslenzk fornrit, Guðmundsson, Finnbogi, ed. (Reykjavík, 1965)Google Scholar. David carried Goliath's head with him back to Jerusalem as a sign of his feat. I Samuel 17:51–57; cf. I Samuel 18:25–27. No people are more famous for their head trophies than the Jibaro Indians who lavish great care on the heads of their victims as they skin and shrink them. See Karsten, R., ‘Blood Revenge and War Among the Jibaro Indians of Eastern Ecuador’ in Bohannan, Paul, ed., Law and Warfare (Garden City, NY, 1967) 303–25Google Scholar at 322–25.

More matter of fact, if not much less hostile, is the role mutilation plays in publishing or proving a killing. Grettir's corpse was decapitated to humiliate it but also to bear the head to the person who had put a price on it. Grettir, supra note 26, chs. 82–84. The Salian Franks announced their killings either by stripping the corpse and hanging it from a stake in plain view or by impaling the head on a stake. See Pactus Legis Salicae, Eckhardt, K.A., ed. (Göttingen, 1955)Google Scholar ch. XLI, § 11.b. A bloody towel was sent by Clotild to her brother Childebert, the Frankish king, as proof of her husband's mistreatment of her, which Childebert then avenged. Gregory of Tours, History of the Franks, trans, by Thorpe, L. (Harmondsworth, 1974) III. 10Google Scholar.

Some corpses were mutilated to prevent them from returning as ghosts or to lay to rest a ghost that had already returned. This appears to be why Grettir decapitated Glám's corpse and placed the head against its buttocks. Ibid. ch. 35 and cf. ch. 18. A collection of other examples is available in Boberg, Inger M., Motif Index of Early Icelandic Literature (Copenhagen, 1966)Google Scholar §§ E431.7, E431.7.1, E446.3.

Yet, as we have seen in the case of Vigfús's head, a corpse need not be mutilated by the enemy. Veneration of the corpse could lead to a dismemberment far more thorough than that carved by an enemy intent on humiliating it. The insatiable demand of the faithful for relics of saints could lead to a second martyrdom of sorts for the corpse or even worse. Thus, Umbrian peasants, c. 1000, wanted to kill St. Romuald in order to make sure they got his precious bones. Huizinga, J., The Waning of the Middle Ages (Garden City, NY, 1954) 167Google Scholar. The monks of Fossanuova decapitated, boiled, and preserved the corpse of St. Thomas Aquinas who had died in their monastery, for fear of losing the relics. Ibid. In 1231, during the lying in state of St. Elizabeth of Hungary, a crowd of worshippers cut off her hair, nails, and nipples. Ibid. The initial animus toward the corpse is about all we can say for sure differentiates relic worship from trophy display. Moreover, the faithful were not beyond punishing and humiliating the relics of saints who had been slow to answer their prayers. For a most interesting discussion of the 9th to 11th century ritual of relic humiliation, see Geary, P., ‘L'humiliation des saints,’ Annales, economies-sociétés-civilisalions XXXIV (1979) 27–42CrossRefGoogle Scholar.

89. See generally Heller, Rolf, Die Literarische Darstellung der Frau in den Isländersagas (Halle, 1958) 98–122Google Scholar; Wolf, A., Gestaltungskerne und Gestaltungsweisen in der altgermanischen Heldendichtung (München, 1965)Google Scholar; Sveinsson, Einar Ó., Um Njálu (Reykjavík, 1933) 130–39Google Scholar; Thomas, R., ‘Some Exceptional Women in the Sagas,’ Saga Book of the Viking Society XIII (1952–1953) 307–27Google Scholar.

90. Heller, Die Literarische Darstellung der Frau, supra note 89, 98, 154, lists fifty-one occasions in the family sagas where women incite men to take vengeance or perform some other risky action.

91. See, e.g., Guðrún's repeated efforts to get her brothers to kill Kjartan in Laxdœla, supra note 21, chs. 47-48; and þorgerð Egilsdóttir's attempts to get her sons to avenge Kjartan. Ibid. chs. 52–54.

92. Heller and the other critics (supra note 89) make no special distinction between those cases where a head or bloody token is used and those where goading or incitement is unaccompanied by such props. The critics treat all these incidents as literary commonplaces and discuss them in terms of literary influence. Heller, Die Literarische Darstellung der Frau, supra note 89, 107; Sveinsson, Um Njálu, supra note 89, 130. No doubt much in the saga's depiction of women as goaders is a matter of literary convention. Yet to make it completely so is to ignore the very real role forced on women in the bloodfeud. Making no distinction between those cases of incitement in which a bloody token is used and those in which it is not fails to account for Hallgerð having made such a distinction in her remark to the shepherd about the importance of the head. See supra text accompanying note 86. These critics have all ignored Hallgerð's statement when discussing these scenes. Heller, in fact, does not include it among his fifty-one cases, apparently because Hallgerð's goading was ineffective.

The distinction is confirmed by the existence of analogous rituals recorded in nonliterary sources and observed by anthropologists. Those rituals requiring a corpse or a bloody token of the victim can be initiated by men as well as women and tend to compel action rather than merely urge it. Among the northern Albanians of this century, for example, a bottle is filled with the dead man's blood and is watched by the kinsmen day and night. As soon as the blood ‘boils,’ or ferments, they rush out to take vengeance. If the blood does not ferment, the kinsmen are more inclined to accept compensation in lieu of blood. Hasluck, Unwritten Laws of Albania, supra note 8, 231. Other examples from bloodfeuding cultures of the Mediterranean and Middle East are assembled in Black-Michaud, Cohesive Force, supra note 4, 79–80. Among the Frisians of the early Middle Ages, the corpse itself was hung in the house of the kinsmen, not to be taken down until it had been avenged. Bloch, Feudal Society, supra note 8, 126. A similar ceremony was apparently known among the biblical Hebrews; see the case of the Levite's concubine in Judges 19:27 20:11.

Goading, on the other hand, is commonly the provenance of real women who would be surprised to learn that the goading of their Icelandic sisters was attributable uniquely to male authors reworking stereotypes from earlier literary sources. In Greece, Albania, and Corsica women improvised funerary dirges whose purpose was to incite the victim's kin to take vengeance:

These dirges express ferociously bloodthirsty sentiments…and are repeated by [the] womenfolk for years after the event to instill into the male heirs of the deceased, who may have been infants at the time of the killing, the necessity to bring vengeance when they grow old enough to bear arms.

Black-Michaud, Cohesive Force, supra note 4, 78–79.

93. Njál, supra note 6, chs. 111–16; this case is treated at length in Miller, ‘Justifying Skarpheðinn,’ supra note 21, 318–20, 328–42.

94. Flosi was Hildigunn's fastnandi, literally, the man who gave her in marriage. As such he bears some obligation to support actions on behalf of her husband and his kinsmen. The precise extent of that duty is unclear and is at issue here. See also Hœnsa-þóris saga, supra note 26, chs. 11–12, where the fastnandi, þórð, although obliged to support the legal action on behalf of his kinswoman's husband's father, does not feel this duty extends to vengeance-taking. Under certain circumstances the fastnandi also incurred duties of support for children born of the marriage. 1 Grágás, supra note 2, § 148; 2 Grágás, § 132.

95. Njál, supra note 6, ch. 116.

96. 1 Grágás, supra note 2, §§ 55, 110. An outlaw suffered loss of all status. His property was confiscated, and any assistance given to him was itself punishable by outlawry. Even if the outlaw managed to get away from Iceland he could still be killed wherever he was found. Ibid. §§ 48–49, 55.

97. Ibid. §§ 55, 60. Various forms of mitigation might include passage abroad and a limitation of outlawry to certain specified districts.

98. Of the 119 lawsuits in the family sagas recorded by Heusler, sixty ended in settlement, while fifty were taken to conclusion, and nine others were scuttled. Heusler, Strafrecht, supra note 61, 40.

99. For the view that compensation payments and the settlements based on them were merely stages in a continuing feuding process and not conclusions of the feud itself, see supra note 61. Black-Michaud claims that compensation payments are never a deterrent to future hostilities. Rather, because they tend to be paid over time in installments, they serve as mnemonic devices that perpetuate the memory of the feud. Black-Michaud, Cohesive Force, supra note 4, 116–18.

100. Saga characters speak disparagingly of those who would carry their kinsmen in their purse. See, e.g., Grettir, supra note 26, ch. 24; Vápnfirðinga, supra note 33, ch. 18. The Norwegian Gulathing law went so far as to provide that no ‘man or woman’ could accept wergeld more than three times ‘unless he has taken revenge in the meantime.’ Larson, Earliest Norwegian Laws, supra note 2, 140. In Denmark the party who accepted compensation rather than taking vengeance was entitled to a ‘levelling oath’ from his opponent that he, too, would accept compensation if he were in the same position. Foote, P.G. and Wilson, D.M., The Viking Achievement (London, 1970) 428Google Scholar.

101. See, e.g., the case of Lýting in Njál, supra note 6, chs. 93, 98 and alsoch. 106 (claim of illegitimate son); see also Gunnlaugs saga ormstungu, ch. 13, in 3 Íslenzk fornrit, Jónsson, Guðni and Nordal, Sigurður, eds. (Reykjavík, 1938)Google Scholar (brother takes his own revenge because he considered the two people his father had already killed in vengeance to be insufficient).

102. ‘Hefna mundi Höskuldr þin, efhann œtti eptir þik at mœla.’ Njál, supra note 6, ch. 116.

103. Ibid. ch. 116.

104. See Schach, P., ‘The Use of the Simile in the Old Icelandic Family Sagas,’ Scandinavian Studies XXIV (1952) 149–65Google Scholar.

105. The killing weapon figures as the bloody token in Óláfs saga helga, ch. 123, in 2 Heimskringla 211–13, in 27 Íslenzk fornrit, Aðalbjarnarson, Bjarni, ed. (Reykjavík, 1945)Google Scholar; English translation by Hollander, supra note 34, 393; Beowulf, supra note 17, ll. 1142–45. See infra Appendix no. 8 and cf. Appendix nos. 9, 10.

106. See Appendix under appropriate headings.

107. There were, of course, exceptions. The women who actually take blood vengeance with their own hands are considered distinctly deviant. Thus Grendel's mother, who avenged her son, is literally a monster woman (ides aglœwif). Beowulf, supra note 17, ll. 1259ff. In Laxdœla, supra note 21, ch. 35, both saga characters and writer join in mocking Breeches-Auð as a cross-dresser; she avenged her divorce by wounding her former husband. In 20th century Albania, a woman was allowed to take an active part in the bloodfeud only if there were no close male kinsmen available. If she chose to take vengeance, she was obliged to become a sworn virgin and wear male dress. Once having assumed the role, she was not allowed to forsake it, and she was henceforth treated as a man for purposes of the feud, i.e., she could kill and be killed legitimately. See Whitaker, I., ‘“A Sack for Carrying Things”: The Traditional Role of Women in Northern Albanian Society,’ Anthropological Quarterly LIV (1981) 146–56CrossRefGoogle Scholar at 151.

Gísli, supra note 28, chs. 34–37, offers another viewpoint. Gísli's wife, also named Auð, fought well on her husband's behalf during his last stand and was considered a paragon of wifely virtue. Gísli's sister, þórdís, tried to avenge her brother by stabbing his killer, Eyjólf. The saga writer makes no explicit judgment about the propriety of the action, unless his judgment be found in her failure to kill Eyjólf.

108. See Appendix under appropriate heading.

109. The phrase hvers manns niðingr figures directly in three of the Icelandic examples. (Appendix nos. 2,5,8). It is also strongly suggested by Hallgerð's hvers manns amœli. (Appendix no. 3) The phrase is not explicitly mentioned in the other examples although it is implied and, as I suppose, silently supplied by the parties anyway. The sanction is invoked both in cases where the corpse and person to be charged were kin (Appendix nos. 5,8) and where they were not related (Appendix no. 2).

110. See generally Foote and Wilson, Viking Achievement, supra note 100, 426. See also Ström, F., Nið, Ergi, and Old Norse Moral Attitudes (London, 1974)Google Scholar.

111. A truce breaker was branded a griðniðingr. 2 Grágás, supra note 2, §§ 284–86. See also Njál, supra note 6, ch. 68 and Grettir, supra note 26, ch. 72, which preserves the anathema describing the lot that awaits the griðniðingr.

112. In the medieval Norwegian Gulathing and Frostathing laws, an outlaw forfeited all his moveables but not his land; land, however, was forfeited for nithing crimes, which both the Gulathing and Frostathing laws specifically enumerate. Larson, Earliest Norwegian Laws, supra note 2, 137, 256–57; see also ibid. 198–99.

113. See Sir Stenton, Frank, Anglo-Saxon England, 3d ed. (Oxford, 1971) 430Google Scholar.

114. See Plummer, Saxon Chronicles, supra note 39, 171, anno 1049.

115. The perception of a present public insult better accounts for the immediate violent rage and discomfort of the person to be charged than assuming the rage to be a literary convention. See infra Appendix nos. 1,2,6,7,8. With one exception these ceremonies took place before witnesses. Interestingly enough, in the unwitnessed case the person to be charged showed no emotion at the sight of the bloody token. See Appendix no. 4. Publicity was crucial to the validity of legal ceremonies in Icelandic law. See generally 3 Grágás, supra note 2, 643 s.v. lýsa.

Our point is confirmed by an analogous Corsican ritual. A man slow to avenge the death of his kinsman faced having the funeral dirges that the women had composed for the dead kinsman (see supra note 92) sung to him by men. When this happened it was said that the singer had dato il rimbecco—‘levelled an insult’—at the man intended to hear the song.

The opprobrium suffered by a man who did not take vengeance after such a powerful provocation was at one period in Corsican history generally regarded as so unendurable that a jury would acquit any man charged with murder if it could be proved that he had been the object of a rimbecco.

Black-Michaud, Cohesive Force, supra note 4, 79.

116. Laxdœla, supra note 21, ch. 60.

117. Genesis 4:10. See also the examples cited from Albania, Frisia, and Judges, supra note 92, where it is the corpse or blood that does the charging.

118. See supra notes 74 and 102.

119. See supra note 79. See also 1 Grágás, supra note 2, § 77.

120. 2 Grágás, supra note 2, § 307; 1 Grágás, §§ 35, 75, 77.

121. Njál, supra note 6, ch. 135 and 2 Grágás, supra note 2, § 307; Holmes, O.W., The Common Law, Howe, M., ed. (Cambridge, Mass., 1973) 279–80Google Scholar. Holmes uses the episode in Nj'al to illustrate his theory that intervivos succession was patterned on succession after death. In his view, the transferee takes as a fictional heir of the transferor. In support he cites the words of the transferee as he accepts the suit: ‘thou handest over to me this suit to plead and to settle…as though I were the rightful next of kin.’ Unfortunately ‘next of kin’ is a mistranslation of aðili—i.e. prosecutor or chief defendant. The error is easy to account for since the aðili incurred his designation in a killing case because he was the victim's heir. But aðili itself has no etymological connection with kinship or heirship. See Cleasby-Vigfusson, Icelandic-English Dictionary, supra note 79, s.v. aðili. The mistranslation costs Holmes a fine example, but his general argument survives without it. Moreover, it suggests that we are not being too fanciful in finding some of the force of the bloody-token ceremony to lie in the notion of a transfer of a suit from a dead man to a living one.

122. 1 Grágás, supra note 2, §§ 87, 90, 107; 2 Grágás, §§ 280, 335; see also Heusler, Strafrecht, supra note 61, 114–23.

123. Njál, supra note 6, ch. 66; see also 1 Grágás, supra note 2, § 87. The passage shows that the procedure's wondrous elevation of form over substance even baffled Icelandic lawyers like Mörð. See also Víga-Glúms saga, chs. 9, 22, in 9 Íslenzk fornrit, Kristjánsson, Jónas, ed. (Reykjavík, 1956)Google Scholar; English translation: Víga-Glúms Saga, trans, by Hollander, L.M. (New York, 1972)Google Scholar; Heusler, Strafrecht, supra note 61, 108–9.

124. The killing of a person was thought of as more an injury to his kin than to him as an individual. Thus, the kin receives compensation for its member's death, not the victim's estate. But there exists a competing notion that the individual had an injury personal to himself. We must distinguish between the corporate liability underlying the feud and those discrete incidents within the feud, such as an assault or a wounding, which were dealt with by the legal system as injuries personal to the injured party. Grágás provided for the survival of legal actions arising from such incidents which then descended to the heir as aðili. See 1 Grágás, §§ 75, 77. This is also the implication of 2 Grágás, § 283, where the aðili in the killing case can make a separate case out of the attack that preceded the killing. He could merge them or not at his election.

125. See, e.g., Hrafnkel, supra note 11, ch. 3 (English translation, ch. 7); Gísli, supra note 28, ch. 6.

126. Laxdœla, supra note 21, ch. 60.

127. Similarly, see the Frisian ritual discussed supra note 92. See also Black-Michaud, Cohesive Force, supra note 4, 79–80. The Albanian ritual in which a bottle of the victim's blood is watched to see when it ferments is used to determine when vengeance will be taken or if it will be taken at all. See supra note 92. The oracular quality of this ritual differentiates it from the Icelandic one where the grievant decides, not the blood.

128. See supra text accompanying notes 14–15.

129. Njál, supra note 6, ch. 98.

130. Ibid. chs. 116, 124.

131. Allmikill níðingr, ibid. ch. 124.

132. Hildigunn's charge led to the burning of Njál and his sons and consequently, to a pitched battle at the Althing and eventually to the killing of more than twenty of Flosi's followers. Njál, supra note 6, chs. 129–59.

133. Vápnfirðinga, supra note 33, chs. 13–14.

134. Ibid. ch. 14 (English translation, ch. 13). The manuscript is badly damaged, and many of the details of the ceremony have been lost.

135. See supra note 100. Compare this interesting account in Gregory of Tours, History of the Franks, supra note 88, IX. 18:

After having murdered Chramnesind's relatives, Sichar formed a great friendship with him, and they became so devoted to each other that they often had meals together and even slept in the same bed. One day as twilight was falling Chramnesind ordered his supper to be prepared and then invited Sichar round to eat with him. He came, and they both sat down to table. Sichar drank far more wine than he could carry and began to boast at Chramnesind's expense. He is reported to have said: ‘Dear brother, you ought to be grateful to me for having killed off your relations. There is plenty of gold and silver in your house now that I have recompensed you for what I did to them. If it weren't for the fact that the fine which I've paid has restored your finances, you would still today be poor and destitute.’ When he heard Sichar's remarks, Chramnesind was sick at heart. ‘If I don't avenge my relatives,’ he said to himself, ‘they will say that I am as weak as a woman, for I no longer have the right to be called a man!’ Thereupon he blew the lights out and hacked Sichar's skull in two. Sichar uttered a low moan as life left his body, then he fell dead to the floor. The servants who had come with him lost no time in making off. Chramnesind stripped Sichar's corpse of its clothes and hung it from a post in his garden-fence.

Stripping the corpse and hanging it on a post was the Frankish equivalent of announcing the killing; not to have done so would have made the killing a murder. Cf. Pactus Legis Salicae, supra note 88, ch. XLI, § 11.b.

136. This fairly represents the situation in the cases infra Appendix nos. 2, 5, 7, 8. In case no. 4, the charge was issued within hours of the killing. Skarpheðin had not failed in his duty and had no intention of not fulfilling it. As noted, the charge to him was purely a matter of organizing the vengeance taking. In case no. 8, the charge also followed closely on the killing, but there the prior relation between the victim and the person to be charged was such as to suggest strongly that the latter had no intention of avenging his brother's son.

137. Beowulf survives in a single manuscript which can be dated by paleographic means to c. 1000 A.D. The date of the poem itself, as opposed to the manuscript, is still the subject of very active scholarly debate. Until two years ago most scholars accepted an 8th-century origin for the poem. But some recent studies have proposed later dates ranging from the 9th century to the early years of Cnut's reign. The arguments for any particular dating variously depend on linguistic, metrical, stylistic, paleographic, and codicologic evidence too detailed and technical to resume here. See generally Chase, Colin, ed., The Dating of Beowulf, 6 Toronto Old English Series (Toronto, 1981)Google Scholar and the controversial codicological and paleographic argument for an early 11th-century origin in Kiernan, Kevin S., Beowulf and the Beowulf Manuscript (New Brunswick, 1981)Google Scholar; see also Fulk, R.D., ‘Review Article: Dating Beowulf to the Viking Age,’ Philological Quarterly LXI (1982) 341–59Google Scholar, for an intelligent critical reading of both books.

The usefulness of Beowulf for describing aspects of the Anglo-Saxon bloodfeud is not vitiated any more by an early 11th century date of composition than a late 7th-century date. The bloodfeud was, after all, a continuous feature of Anglo-Saxon life for the entire period; see, e.g., Bede, Ecclesiastical History, supra note 52, iv, 21–22; and Vita Wulfstani, Darlington, R.R., ed. (London, 1928) 38Google Scholar. Although the poem does not speak specifically of Anglo-Saxon feuds, it is an Anglo-Saxon poem destined for Anglo-Saxon audiences. When it describes bloodfeud, the lord-retainer bond, and the various norms associated with them, it is as useful an indication as any we have of what the Anglo-Saxon audience's expectations were in such matters, even if we cannot confine that audience definitely to the 8th century. See also Whitelock, Dorothy, The Audience of Beowulf (Oxford, 1951) 12–19Google Scholar.

138. See Whitelock, Audience of Beowulf, supra note 137, 39–70.

139. Beowulf supra note 17, ll. 1066–1159. This digression is known as the Finn or Finnsburh Episode. The Episode is distinguished from another Old English poem known as The Fight at Finnsburh or the Finnsburh Fragment. As the latter name indicates, the 48 surviving lines of this poem are all that remain of a longer heroic lay. Taking the two together the larger outlines of the feud can be reconstructed. But no one agrees about the particulars, and the Episode and Fragment have been the subject of active scholarly debate for over a century. Klaeber edits the Fragment with notes and commentary in his Beowulf, supra note 17, 231–53. For a recent edition that also contains an extensive bibliography, see Fry, Donald K., Finnsburh: Fragment and Episode (London, 1974)Google Scholar. See also Chambers, R.W., Beowulf: An Introduction, 3d ed. (Cambridge, 1959) 245–89Google Scholar.

140. The attack is the substance of the Fragment; see Klaeber, Beowulf, supra note 17, 245–47.

141. Beowulf, supra note 17, ll. 1089–94. Because of some pronouns with uncertain antecedents, it is not clear who offers terms to whom. The terms of the settlement itself, however, are reasonably clear.

142. Ibid. ll. 1100–9. I follow the manuscript reading að (1. 1107), i.e., ‘oath,’ not Klaeber's unnecessary emendation ad, i.e., ‘pyre.’ On these lines see Whitelock, Audience of Beowulf, supra note 137, 18 and n.4.

143. On the problems of the peace settlement, see Moore, B., ‘The Relevance of the Finnsburh Episode,’ Journal of English and Germanic Philology LXXV (1976) 317–29Google Scholar at 318–21.

144. Beowulf, supra note 17, ll. 1138–39.

145. Ibid. ll. 1142–53. Again I follow the manuscript reading woroldrœdenne, ‘universal law or custom,’ rather than Klaeber's unnecessary emendation weorodrœdende, ‘troop-leader.’

146. The symbolic nature of Hunlafing's action has been recognized by critics. See, e.g., Donaldson, E. Talbot, Beowulf, Tuso, J., ed. (New York, 1975) 21Google Scholar n.1; Anderson, E., ‘Formulaic Typescene Survival: Finn, Ingeld, and the Nibelungenlied,’ English Studies LXI (1980) 293–301CrossRefGoogle Scholar; Ayres, H., ‘The Tragedy of Hengest in Beowulf,’ Journal of English and Germanic Philology XVI (1917) 282–295Google Scholar at 292; Moore, ‘Relevance of the Finnsburh Episode,’ supra note 143, 321. The resemblance of the scene to one of our saga examples has been noted. See Ayres, ‘Tragedy of Hengest,’ 292 n.17 and Appendix no. 6.

147. See supra note 139; Klaeber, Beowulf supra note 17, 246, 1. 16.

148. See Chambers, Beowulf, supra note 139, 252 n.2.

149. The ‘ing’ is a patronymic suffix; hence, Hunlafing is the son of Hunlaf.

150. The case for the sword belonging to Hnæf is especially well argued by Anderson, ‘Formulaic Typescene Survival,’supra note 146, 295.

151. The meaning and form of woroidrœdenne is the subject of thorough discussion in Bonjour, Adrien, ‘The Climax of the Finn Episode,’ in University of California Publications in English III (1943) 312–30Google Scholar. I have adopted his gloss ‘universal law,’ ‘universal obligation,’ which he takes to indicate the obligation to take vengeance for one's lord. Ibid. 329–30. This is probably the case. Yet it is possible that woroidrœdenne refers more precisely to vengeance taken in response to the ceremony Hunlafing performs, the ceremony itself being a universal custom. The word occurs nowhere else in Old English; the only context we have to judge its meaning is here. The syntax of the sentence supports the more particular meaning as well as the general one: ‘he did not refuse the universal law when Hunlafing laid the best of swords…in his lap.’ I only suggest the possibility.

The poet is, however, making an unambivalent and witty reference to the ceremony in his use of the word sweordbealo (1. 1146), literally, ‘sword-evil,’ ‘sword-bale,’ to describe how Finn came by his death. Sweordbealo, also, is unattested elsewhere in Old English. The poet, apparently, has invented a compound that describes both the efficient cause of Finn's death—a cruel sword thrust delivered most likely by Hengest— and the cruel sword-evil that was the efficient cause of the thrust—the ceremony of placing Hnæf's sword in Hengest's lap.

152. The poet is quite specific about what is placed on the pyre. Hnæf's corselet and helmet were mentioned, his sword was not. Beowulf, supra note 17, ll. 1111–12. See also Davidson, Hilda, The Sword in Anglo-Saxon England (Oxford, 1962) 10–13Google Scholar, where it is noted that swords or their remains are rare in cremation graves. Swords often escaped burnings or burial because they were valued as hereditable goods or were subject to death tax—the heriot.

The destruction of the corpse by fire may also explain the surrogate for the bloody token in a gruesome goading recorded in Guðmundar saga dýra, ch. 17, in Sturlunga saga, Jóhannesson, Jón, Finnbogason, Magnús and Eldjárn, Kristján, eds. (Reykjavík, 1946) 195Google Scholar. Önund had been burned to death by Guðmund in his own home. A year later Önund's daughter, Guðrún, serves her brothers and husband nothing but burned sheeps’ heads and feet for breakfast. The message is not lost on the men who ride out immediately thereafter to avenge Önund.

153. Elsewhere in Beowulf, the sight of a dead kinsman's sword is used as a mnemonic device to goad a young man to take vengeance in violation of a settlement. Ibid. ll. 2032-66. But in that case the sword is not part of a ceremonial charge. It had been appropriated as spoils by the other faction and so was not available to the grievant. Without it, much like Hallgerð without Sigmund's head, the grievant must resort to repeated eggings to move the would-be avenger to action.

In the sagas, the victim's own weapon figures frequently in avenging him without ever being formally handed to the avenger as part of a charging ceremony. The victim's weapon would normally pass to one of his heirs along with his other property, and it was often the heirs who took vengeance. See, e.g., Droplaug, supra note 38, ch. 13; Laxdœla, supra note 21, ch. 64; Njál, supra note 6, chs. 132, 146. In cases where the corpse was despoiled of its weapon, the avenger could even resort to trickery to obtain it. Grettir, supra note 26, ch. 86. There was, no doubt, a special sense of appropriateness in using the victim's weapon to avenge him. Besides the grim irony apparent to the expiator of dying virtually at the hands of the victim, it satisfied the rule of equivalence, the talion, with flourish.

A similar sense of appropriateness also made the killing weapon a desirable choice to do double service as the avenging weapon. Thus, the spear that killed Ásbjörn, stained with his blood, served not only as the bloody token in a formal charging ceremony but was also used to avenge him. Óláfs saga helga, supra note 105, chs. 133, 228. See also Gísli, supra note 28, chs. 13, 16. That saga mentions a custom nowhere else recorded, and not current during the author's time, that the responsibility to take vengeance fell to whoever removed the weapon from the corpse. The custom, if indeed preserving an earlier practice, bears some resemblance to the bloody-token ceremony at issue in this article.

154. Beowulf, supra note 17, ll. 2591–2610.

155. Ibid. ll. 2610–25. Unlike the Finnsburh Episode, this digression has received little critical attention. See Bonjour, Adrien, The Digressions in Beowulf (Oxford, 1950) 35–39Google Scholar; Kaske, Robert, ‘Weohstan's Sword,’ Modern Language Notes LXXV (1960) 465–68CrossRefGoogle Scholar.

156. Weohstan is said to bear the spoils to his magum. I follow Klaeber, Beowulf, supra note 17, 217, in taking his to refer to Eanmund, and in taking the dative plural magum, literally ‘to kinsmen,’ to be a generic term specifically referring to Onela as the head of the kin group to which both he and Eanmund belong.

157. Beowulf, supra note 17, ll. 2379–96.

158. E.g., like Weohstan, the Geatish warrior Eofor despoiled the corpse of his victim Ogentheow and carried Ogentheow's helmet, sword, and mailshirt back to his lord Hygelac. Beowulf, supra note 17, ll. 2985-88. Beowulf gave Hygelac the gifts he received from King Hrothgar for vanquishing Grendel. Ibid. ll. 2155–62. He gave Hrothgar the sword-hilt he took from the Grendel kin. Ibid. ll. 1677–79. The Frankish warrior Dæghrefn tried to despoil Hygelac's corpse in order to take the armor back to his lord, but was killed by Beowulf in the process. Ibid. ll. 2501–06. Examples from other sources are available in Whitelock, Beginnings of English Society, supra note 11, 30–31.

159. See Ibid. 35–36. ‘Heriot’ is a later form of Old English heregeatu, a compound meaning generally ‘army-equipment,’and specifically the exaction later known as ‘heriot;’ see Bosworth, J. and Toller, T.N., An Anglo-Saxon Dictionary (Oxford, 1898)Google Scholar and Toller, and Campbell's, A.Supplement (Oxford, 1921)Google Scholar s.v. heregeatu.

160. See Beowulf, supra note 17, ll. 2379–90.

161. Ibid. ll. 452–55.

162. The episode of Sichar described supra note 135 is especially instructive in this regard.

163. The ceremonial gift-giving from lord to retainer was the central ritual of the hall-life of the comitatus. The ritual symbolized and reconfirmed the bond between lord and man. Its importance is attested by Old English synonyms for lord like sincgifa, ‘treasure-giver,’ and beahgifa, ‘ring-giver.’ Bosworth-Toller, Anglo-Saxon Dictionary, supra note 159. For an interesting discussion of the politics of gift exchange in Beowulf, see Berger, H. Jr., and Leicester, H. Jr., ‘Social Structure as Doom: the limits of heroism in Beowulf,’ Burlin, R. and Irving, E. Jr., eds., Old English Studies in Honour of John C. Pope (Toronto, 1974) 37–79Google Scholar at 44–50. See also Beowulf, supra note 17, ll. 168, 2633–45, 3010–21 and Whitelock, Beginnings of English Society, supra note 11, 30–31.

164. Understatement, litotes in the terminology of rhetoric, is a characteristic feature of the Beowulf-poet's style. In his commentary to these lines, Klaeber notes the presence of the trope. Beowulf, supra note 17, 218.

165. See 1. 2616 and cf. 1. 2623 in the passage quoted in the text supra accompanying note 155.

166. Matthew 6:12 in Anglo-Saxon Gospels quoted in The Oxford English Dictionary s.v. forgive, v. 3.b. Forgyf us ure gyltas renders the Vulgate Latin dimitte nobis debita nostra; see Bosworth-Toller, Anglo-Saxon Dictionary, supra note 159, s.v. forgifan.

167. See Berger and Leicester, ‘Social Structure as Doom,’ supra note 163, 44–50.

168. Onela's release bound only himself, not his kin. Eanmund was survived by a brother, Eadgils, who had not forgiven Weohstan the feud. Eadgils later succeeded Onela as king of the Swedes, and it was perhaps for this reason that Weohstan is said to have given Wiglaf Eanmund's sword and armor among the Geats. There are strong hints in the poem that Eadgils and the Swedes attack Wiglaf and the Geats to avenge Eanmund. See Bonjour, Digressions in Beowulf, supra note 155, 36-39.

169. The poet's use of geaf to describe the transfer from Weohstan to Wiglaf contrasts with his use of forgeaf to describe the Onela-Weohstan transfer. The use of geaf some seven lines after forgeaf can only be intended to point out the distinction between the two transfers since there are no compelling metrical reasons requiring the choice of either verb.

170. Peters, ‘Structural Aspects,’ supra note 4, 270, records that among the Bedouin the killer might present himself to the ‘owner of the blood,’ accompanied by a holy man, and make an offer of his life: ‘Here is my knife, my life is yours.’ Peters notes that ‘no Bedouin would agree for a moment to undertake this journey if he thought the offer might be accepted.’

171. þorsteins saga hvíta, ch. 7, in 11 Íslenzk fornrit, Jóhannesson, Jón, ed. (Reykjavík, 1950)Google Scholar; English translation: Thorstein the White's Saga, ch. 8, in Four Icelandic Sagas, trans, by Jones, Gwyn (Princeton, 1935)Google Scholar; and also Hrafns saga Sveinbjarnarsonar, ch. 14, in Sturlunga saga, supra note 152, 218.