. 6
( 55 .)


the Lord your God is giving to you, and you dispossess them and
settle in their cities and homes, 2 you shall set apart three cities in
your land, which the Lord your God is giving you to inherit. 3 You
shall determine the distance and divide the limits of your land, which
the Lord your God has allotted to you, into three, so that any slayer
may ¬‚ee there. 4 This is the type of slayer who may ¬‚ee there and live:
whoever slays his fellow without intent and was not hostile to him in
the past. 5 Whoever came with his fellow into the forest to cut wood:
as his hand swings the ax to cut down the tree, the ax-head falls off
the handle and hits the other so that he dies “ that man shall ¬‚ee to
one of these cities and live, 6 lest the blood avenger pursuing him in
his hot anger, overtakes him and slays him because the distance is too
great, yet he was not liable to the death penalty because he was not
hostile to him in the past. 7 Therefore, I command you to set apart
three cities. 8 If the Lord your God extends your boundaries, as he
swore to your fathers, and gives you all the land which he promised
to them, 9 because you observe the instruction that I command you
this day, to love the Lord your God and to walk in his ways at all
times, you shall add three more cities to these three. 10 The blood
of the innocent shall not be shed in the land which the Lord your
God is giving to you, imputing bloodguilt upon you. 11 If a person
is hostile to another and lies in wait and strikes him mortally so
that he dies, and ¬‚ees to one of these towns, 12 the elders of his
town shall send and take him back from there and deliver him to the
blood avenger so that he dies. 13 You shall not have pity on him, but
shall make expiation of the blood of the innocent, and it will be well
with you.

The threat of the family™s agent is assumed by the statute in the Book of the
Covenant, Exod 21:12“14:
Whoever fatally strikes a man shall surely be put to death. 13 If [the
killer] did not lie in wait,6 but God caused it to meet his hand, then
I will appoint for you a place to which he may ¬‚ee. 14 But if a man
willfully attacks a man to kill him treacherously, you shall take him
from my altar to be put to death.
In biblical Israel, the victim™s family assumed primary responsibility for
ensuring that the slayer was held accountable for his offense. A member of
the family had the right and responsibility to kill the slayer and could do
so with impunity under certain conditions. This agent of justice was called
!dh lag, generally translated as “the blood avenger.”
It is best to understand this process as self-redress or blood feud, when
the victim™s kin avenges the killing by killing the slayer. This label allows us
to link this process to two essential characteristics of blood feud: It is local in
nature, and it is rule-bound.7 These characteristics are interrelated because
blood feud is a legal mechanism that both assures the redress of wrongs
and controls the violence to a level tolerable in a community. Blood feud
6 The root h-r-x is a synonym for the more common root ·b-r-a. However, it stresses the aspect
of planning and watching for an opportunity to entrap, rather than the aspect of hiding. Cf.
Lam 4:18; 1 Sam 24:12; Num 35:20, 22. (Zeph 3:6 is derived from a homophonous root
meaning “to lay waste”.) Menahem, according to Rashi, associates it with hunting.
7 Max Gluckman, “The Peace in the Feud,” Past and Present 8 (1955), 1“14. An alternate

version of this article is found in the author™s Custom and Con¬‚ict in Africa (Oxford: Basil
Blackwell, 1965), 1“26. The term feud has been used to characterize a bewildering variety
of phenomena, from an individual™s single act of retaliation, both lethal and nonlethal, for
murder, injury, and insult to continuous acts of full-scale aggression between large groups. Cf.
E. E. Evans-Pritchard, The Nuer: A Description of Their Modes of Livelihood and Political
Institutions of a Nilotic People (Oxford: Clarendon, 1940), 151“160; Gluckman, “The Peace
in the Feud,” 6“9; Gluckman, Politics, Law and Ritual in Tribal Society (Oxford: Basil
Blackwell, 1977), 111“112; Marc Bloch, Feudal Society (trans. L. A. Manyon; Chicago:
University of Chicago Press, 1974 [1961]), 138; Jenny Wormald, “The Blood Feud in Early
Modern Scotland,” in Disputes and Settlements: Law and Human Relations in the West
(ed. John Bossy; Cambridge: Cambridge University Press, 1983), 113, 115“116; Jacob
Black-Michaud, Cohesive Force: Feud in the Mediterranean and the Middle East (Oxford:
Basil Blackwell, 1975), 23; J. M. Wallace-Hadrill, “The Blood Feud of the Franks,” in
The Long-Haired Kings (Medieval Academy Reprints for Teaching 11; Toronto: Univer-
sity of Toronto Press, 1982 [1962]), 143; Norbert Rouland, Legal Anthropology (trans.
Phillippe G. Planel; Stanford: Stanford University Press, 1994), 277; Bernice Calmes Caudill,
Pioneers of Eastern Kentucky: Their Feuds and Settlements (Cincinnati, Ohio: Privately
printed, 1969); Keith F. Otterbein and Charlotte Swanson Otterbein, “An Eye for an Eye,
A Tooth for a Tooth: A Cross-Cultural Study of Feuding,” American Anthropologist 67
(1965), 1470“1482; J. K. Campbell, Honour, Family, and Patronage: A Study of Institutions
and Moral Values in a Greek Mountain Community (Oxford: Clarendon, 1964), 97, 173,
194, 196“197, 264; Joseph Ginat, Blood Disputes Among Bedouin and Rural Arabs in
Israel (Pittsburgh: University of Pittsburgh Press, 1987), 21“27, 40“59; Thomas M. Kiefer,

is not a paroxysm of rage, careening out of control. The biblical texts that
deal with homicide assume that there are constraints on the power of the
victim™s family to effect vengeance. The actions of the blood avenger were
to be channeled into certain options: His actions were not unfettered. They
were not wild justice or a step outside the law.
The rule-boundedness of blood feud was manifest in the limitation on
which individuals were involved. Only the slayer was subject to action, not
anyone else, whether having a connection to him or not. Only a speci¬c
member of the victim™s family, !dh lag, had the right and responsibility to kill
the slayer with impunity.
Another major restriction on the actions of the avenger in the Hebrew
Bible was the existence of a place of sanctuary for the killer.8 From the

The Tausug: Violence and Law in a Philippine Moslem Society (Case Studies in Cultural
Anthropology; New York: Holt, Rinehart and Winston, Inc., 1972).
Objections have been made to subsuming such varied phenomena under the rubric of feud.
First, the limited violence that occurs in many societies when a murder has occurred has led
to reservations about calling such events manifestations of feud. Second, some observers have
hesitated to identify feud as law because of the lack of an authority imposing a settlement. Cf.
Leopold Posp´sil, Anthropology of Law: A Comparative Theory (New Haven, Connecticut:
HRAF Press, 1974), 4“5, 8“9; E. Adamson Hoebel, The Law of Primitive Man: A Study in
Comparative Legal Dynamics (Cambridge, Massachusetts: Harvard University Press, 1954),
Formulating a clear distinction between a chain of revenge and a single act of retribution
executed on the offender is useful in highlighting the varying magnitudes of violence. However,
it must be noted that prolonged violence in general is rare and, therefore, using this de¬nition
of feud removes it, at least on a semantic level, from most traditional, preindustrial societies.
In fact, the word “feud” continues to be used by most anthropologists for self-redress because
of the potential threat of violence without the emphasis on prolonged violence. This appears
correct in my judgment.
8 Even though I have cast this in terms of restriction, it is incorrect to posit the existence of

completely unfettered blood feud. Biblical scholars have argued that originally, vengeance could
be taken of any killler, whether intentional or unintentional, and was only later restricted to
the intentional offender in the development of restrictions on blood vengeance in ancient Israel
(cf. Brevard Childs, The Book of Exodus [OTL; Philadelphia: Westminster, 1974], 457, 470;
Baruch Halpern, “Jerusalem and the Lineages in the Seventh Century BCE: Kinship and the
Rise of Individual Moral Responsibility,” in Law and Ideology in Monarchic Israel [JSOTSup
124; Shef¬eld: Shef¬eld Academic Press, 1991], 11“107; Henry McKeating, “Development of
the Law of Homicide in Ancient Israel,” VT 25 [1975], 46“47, and Milgrom, Numbers, 291).
But such a stage in social development is based on theoretical assumptions about the growth
of primitive societies, a type of inquiry generally abandoned by contemporary anthropologists.
Most recent anthropologists have shied away from producing evolutionary theories and have
concentrated on the synchronic analysis of the societies they study. Cf. Laura Nader, “The
Anthropological Study of Law,” in Law and Anthropology (ed. Peter Sack and Jonathan Aleck;
The International Library of Essays in Law and Legal Theory; New York: New York University
Press, 1992), 3“32; June Starr and Jane F. Collier, “Historical Studies of Legal Change,” in Law
and Anthropology, 105“110; Norman Yoffee, “Too Many Chiefs? (or, Safe Texts for the 90s),”
in Archaeological Theory: Who Sets the Agenda? (ed. Norman Yoffee and Andrew Sherratt;
Cambridge: Cambridge University Press: 1993), 60“78. The contemporary study of disputes

moment of the killing itself, the slayer was threatened by the blood avenger.
The family™s assumption of the initiative in remedying the wrong was the
direct motivation for the existence of sanctuary and was tempered by the
existence of sanctuary. The fugitive™s arrival in a place of refuge put a hold
on the actions of the blood avenger. He could not lay a hand on the fugitive.
By having impunity to kill the slayer whatever the circumstances from the
time of the killing until the slayer reached a place of sanctuary, the agent of
the victim™s family ensured that the slayer would go to a refuge and thereupon
be judged for his actions. The violence of self-redress acted as a threat: It
was an impetus to a formal trial and away from violence.
The slayer™s intention was not taken into account at this stage of the
process, and the only way he could attempt to claim mitigating circumstances
was to seek refuge in a place of sanctuary and acknowledge that he was the
killer. Ironically, only by going to a sanctuary and thereby identifying himself
as the killer without any denial or dissembling on his part could the killer
claim that the death was accidental.
Another major restriction was the interference of others in judging
whether the accused killer was guilty. Once the slayer entered the city of
refuge, he was subject to trial to determine whether he was an intentional
or an accidental slayer (Num 35:24; Deut 19:12). This decision limited the
ability of !dh lag to effect vengeance because if the slayer was judged to be
an accidental killer, he was permitted to stay in the city of refuge safe from
the avenger. Only if the slayer was determined to be an intentional killer was
he handed over to the avenger for execution. This procedure introduces an
element of objectivity into the process. Other people who are not the victim™s
kin determine the level of culpability the accused possesses in the death. The
omniscient narrator in Genesis 4 can conveniently produce an omniscient
Deity to judge Cain, but human beings do not have the talent of an infallible
ability to determine fault and, therefore, certain procedures to make such a
determination as objective as possible must be designed.
In all the legal sources, the avenger, !dh lag, acted as executioner. Although
it may appear that his role was reduced to nothing more than carrying out
the judgment of the court,9 it was his assumption in the ¬rst place of the
responsibility to avenge the killing by killing the killer that forced the killer
to seek refuge. Blood vengeance was the basis of the entire process for pros-
ecuting and punishing a killer. It was rule-bound, with a safe haven for the

has also presented a challenge to the self-help theory. The idea that ¬ghting precedes talking
in the evolution of societies is contradicted by the appearance of ¬ghting and talking side by
side in the same culture. Violence does not give way to negotiation at a certain level of social
development. Rather, both are present, albeit realized in different ways in different societies.
Cf. Simon Roberts, “The Study of Dispute: Anthropological Perspectives,” in Disputes and
Settlements: Law and Human Relations in the West (ed. John Bossy; Cambridge: Cambridge
University Press, 1983), 8“9.
9 Cf. Milgrom, Numbers, 217.

killer and procedures for determining his guilt. The places of refuge acted
as a check on the right of !dh lag to kill the slayer with impunity. He could
not kill a slayer while the slayer remained within the city of refuge. Courts
of various constitutions determined whether the killing was intentional or
accidental. The intentional killer was handed over to !dh lag, whereas the
accidental killer was sheltered from him.
What is important to recognize is that, unlike the modern Western crim-
inal court system, which has specialized personnel for identifying and ar-
resting, prosecuting, judging, and punishing offenders, including killers, the
legal system of ancient Israel was responsible for regulating the right of the
victim™s family to effect a remedy, not for initiating the prosecution of a killer
or for remedying the killing, a radically different concept.10 In such a legal
system, a relative of the victim, !dh lag, initiates the process and ensures that
punishment takes place.
The legal system re¬‚ects the singular role of the family in ancient Israel™s
social structure. The family consisted of a family per se, ba tyb, which in turn
was part of a hjp`m, a lineage or protective association of extended families,
that operated when the family was unable to solve a problem on its own.11
In fact, the term hjp`m is a term of relationship expressing kinship, real or ¬c-
tional, unconnected to ownership of land.12 This understanding emphasizes
the communal association of the members of a hjp`m for their socioeconomic
bene¬t, rather than merely a matter of blood ties. A feeling of kinship may be
based on other factors in addition to genealogy. Territoriality, for example,
was an important aspect of hjp`m.13 Certain towns were assimilated into
lists of twjp`m.14 Names of villages were identi¬ed with ancestors (Mic 5:2;

10 Even in the case when the victim could not be identi¬ed (and presumably his family had not
come forward searching for him), a local body representing the local community, the elders of
a town, not a state mechanism, would come forward on an ad hoc basis to address the problem
(Deut 21:1“13).
11 Norman K. Gottwald, in The Tribes of Yahweh: A Sociology of the Religion of Liberated

Israel, 1250“1050 B.C.E. (Maryknoll, New York: Orbis, 1979), 267, 298“302. His primary
proof that the members of a hjp`m were not related by blood is that taboos of sexual rela-
tionships between cousins that operated for a kinship group are absent in the rules regulating
relationships, such as in Leviticus 18: a kinship group of the next higher order above a family,
one that consisted of related families, was therefore exogamous, whereas a hjp`m was endog-
amous. Within a hjp`m, permitting endogamy means that the members of a hjp`m were not
related. Shunya Bendor correctly recognizes that this factor did not necessarily prevent kinship
(The Social Structure of Ancient Israel [Jerusalem Biblical Studies 7; Jerusalem: Simor, 1996],
82“86). Bendor also makes a distinction between compelling endogamy and allowing endogamy
by noting that Numbers 36, by prescribing endogomy, presumed that exogamy was an alternate
possibility under normal circumstances.
12 Levine, Numbers 21“36, 334.
13 C. J. H. Wright, “Family,” ABD 2.762; Elizabeth Bellefontaine, “Customary Law and

Chieftainship: Judicial Aspects of 2 Samuel 14:4“21,” JSOT 38 (1987), 50.
14 For example, the towns of Shechem, Tirzah, and Hepher (Gen 34; Josh 12:17, 24) were

included in the twjp`m of Manasseh (Josh 17:26; Num 26:30“33).

1 Chr 2:5 versus 4:4). Land was allotted according to twjp`m (Josh 13:15;
Num 33:54). Degrees of kinship were left inexact.15 All further degrees of
kinship beyond father™s brother were merged; no distinctions were made be-
tween cousins.16 Kinship was not determined in ¬nite steps of genealogy.
The term hjp`m, then, refers to a group that has assumed kinship ties, even
if technically a blood relationship was tenuous, even nonexistent.
Family ties in biblical Israel were not so much a matter of genealogical
relations as of responsibilities. What is critical to understand about kinship is
that it is a way of talking about rights and duties, about the claims individuals
make on others, and about how obligations are accepted or denied, not
simply a matter of blood ties. The blood avenger as a member of the victim™s
family takes on a duty on behalf of the victim that obviously the victim
The responsibility of the lineage in the adjudication of homicide was not
anomalous. A lineage had other responsibilities in the legal arena. A lag,
redeemer, was a close male relative who was obligated to reclaim land sold
by a member of his lineage (Lev 25:25; Jer 32:7“8; Ruth 3:12, 4:3“4) and to
redeem a relative sold into slavery (Lev 25:47“49). He acted on behalf of a
powerless person in the restoration of lost property.
In biblical law, the victim™s family assumed the primary responsibility
for responding to the slaying of one of its members. By contrast, the mem-
bers of the victim™s family did not have to assume that responsibility in
Mesopotamian law. They had the right to make a claim on the slayer, but
the slayer was not in mortal danger from a blood avenger waiting to strike
him down. There were, of course, angry Mesopotamians who would have
wanted to strike down the killer, but they did not have the legal right to do
so with impunity. In some cases, the victim™s family might have played a role
in determining the penalty, but it must be emphasized that the members of
the victim™s family were not otherwise involved in the remedy.
Indeed, in Mesopotamian law, those outside the victim™s family ensured


. 6
( 55 .)