HBES Berlin, July 2003, Ancestral Names, Steadman and Palmer

Ancestral Names

Lyle Steadm= an, Craig Palmer

Departments= of Anthropology

Arizona Sta= te University and University of Colorado at Colorado Springs

Human Behav= ior and Evolution Society Meeting

Berlin, Ger= many

July 18, 2004

= ; The question we try to answer in this short paper is why humans everywhere are identified by one or more ancestral names – for example, surnames in English, which identify individuals with a line of their ancestors. For more than a century anthropologists have argued that in tribal societies such names identify individuals with a social group, usually called a clan or lineage. However, as the anthropologist Har= old Scheffler (1976) and a few others (e.g., Roger) have pointed out, clans, lineages and so on are not social groups but only categories of individuals dispersed throughout the tribe. One reason for this is that the categories are usually exogamous: individuals sharing the same ancestral name cannot marry. Consequently, the mother and fathe= r in each family must be identified by different names. For more than a hundred years, anthropologists have attempted to explain ancestral names as identifying so= cial groups, even though they know they don’t. Because the use of such names is universal, an evolutionary explanation is called for.

Cooperation and altruism are crucial to evolutionary theory. In every tribal society (and hence= in human societies for thousands of years) virtually all cooperation and altru= ism occur between kinsmen, including distant kinsmen. This cooperation not only includes helping one another in various ways, but also mutual defense and trade -- t= he basis of the standard of living. This paper proposes that the evolutionary function of ancestr= al names, the effect that has led to their ubiquity, is that they are used to identify living co-descendants. Co-descendants are individuals who share the same ancestor and who, therefore, are kinsmen of one another.&nbs= p; The use of ancestral names to identify distant kin is the basis of t= he wide-ranging cooperation found in all tribal societies.

Identifying Distant Kin

As William Hamilton pointed out, all kinsmen are not alike. The most significant difference am= ong kinsmen is their genealogical distance from one another, and, as evolutiona= ry theory would predict, humans in all societies are concerned to identify this distance. In every society ev= er studied, close kin are identified by a set of kin terms, such as fat= her, aunt, cousin, grandchild etc. While such terms may indicate the sex or generation of the kinsman, they always, when used literally, precisely identify kinship distance. However, because they imply knowle= dge of the precise number of birth links separating individual kinsman, when used literally their range is limited to probably less than 100 kinsmen. Therefore, in addition to kin term= s, available evidence indicates that in every society individuals use another verbal system to identify more distant kinsmen: ancestral names. For example, in all European, and = European-based, societies, the family name of individuals identify them with a line of their male ancestors: their F, FF, FFF etc., back to the first ancestor so named. Even today in these countries, where many relationships are not based on kinship, when individu= als do share the same ancestral name they may be thought to be related. For example, when two individuals = with the same family name are mentioned in a newspaper article, when they are not related, the article almost always points that out. While most societies, like those i= n the West, use patrilineal names – names identifying individuals with a li= ne of male ancestors -- many use matrilineal names, which identify individuals with a line of female ancestresses: mother, mother’s mother, mother’s mother’s mother etc.&= nbsp; As we shall see, a few use both matrilineal and patrilineal names, and a few have names that identify individuals with ancestors throug= h either the mother or the father. In = every society, apparently, individuals use ancestral names to identify living co-descendants.

Identifying Genealogical Distance

While ancestral names may identify large numbers of co-descendants, they do not necessarily identify relative genealogical distance among those co-descendants; for example, both a close and a distant cousin may be identified by one’s own ancestral name. To understand how ancestral names = can identify genealogical distance, three points must be kept in mind. First, in all societies, an individual identified as a kinsman will always be considered closer than si= mply a fellow member of one’s tribe or society not so identified. However, this does not mean that everyone with the same ancestral name is considered closer than everyone without that name. This is be= cause of the second point: ancestral names are used to identify many kin w= ho do not personally share the same ancestral name. One identifies as kin all of bo= th parent’s identified kin. For example, individuals with patrilineal names also identify as kin those who = bear their mother’s patrilineal ancestral name. Indeed, kin are identified by the ancestral name of at least each of one’s four grandparents. In addition, all offspring of any = of these categories will be identifiable as co-descendants, regardless of their own ancestral name. Third<= /i>, in a great number of societies individuals are personally identified by more than one ancestral name.

For example, in the famous Trobriand Islands of Melanesia, everyone is identifi= ed by both a closer matrilineal ancestress (called by anthropologists a sub-cl= an name) and a more distant matrilineal ancestress (called a clan name) (see Malinowski, The Sexual Life of Savages).&n= bsp; This allows everyone to distinguish, and thus favor, a closer set of co-descendants (identified by the sub-clan name) within a more distant one = (the clan), and this applies as well to one’s father’s two matriline= al categories, and even to those of the grandparents.   Thus, ancestral names serve = to distinguish relative genealogical distance among sets of co-descendants.

     = ;            &n= bsp;      A more extreme (and equally famous) example of identification with multiple ancestors in the same line, is the Nuer, of the Sudan, who identify individuals with 5 or 6 patrilineal ancestors, ranked from closest to most distant. (Such a system has been labeled by anthropologists, focusing on so= cial groups, a segmentary lineage system.) Like in the Trobriands, the absolute distance to the common ancestor= is usually unknown, but one’s closer ancestor is used to identify closer co-descendants within a larger set of co-descendants. The largest set, which may number = nine or ten thousand individuals, is identified by the most distant ancestor, ca= lled the clan ancestor. And the th= ree or four intermediary ancestors are used to further discriminate genealogical distance within the set of “clan” co-descendants (Evans-Pritchard, 1940, The Nuer). By being identified wi= th a line of ancestors that are distinguished by their relative distance from one another, such names effectively discriminate closer from more distant co-de= scendants, and thus permit quite fine-tuned favoritism of genealogically closer co-descendants over more distant ones.&nbs= p; Only those ancestors who have left living identifiable descendants to the present time are identified; other ancestors are “irrelevant̶= 1; (Evans-Pritchard 1940:200). T= he same, of course, holds true for the 5 or 6 names of one’s mother̵= 7;s patrilineal ancestors and living co-descendants.   Although Evans-Pritchard, li= ke other anthropologists, tries to explain the significance of such names as identifying social groups, he writes that, while “Every Nuer village = is associated with a lineage [i.e., an ancestral name]… the members of it often constitute only a small proportion of the village population” (= p. 203). That is, while lineage ancestral names may be used to identify villages, which are their most important social groups, a village obviously is not a lineage.

            We now examine the unique ancestral naming systems of the Dobe !Kung, living in the Kalahari desert of southwestern Africa, and the Australian Aborigines. Neither of these = peoples traditionally practiced agriculture; each are hunters and gatherers and are assumed to exhibit the most ancient traditions extant of our species.

            The Dobe !Kung are peoples of the Kalahari desert formerly known as Bushmen. Richard Lee, who studied the !Kung, makes the typical observation: “As in all prestate societies, the cen= tral organizing principle of !Kung life is kinship” (1984:57). While Lee writes that the !Kung ha= ve no surnames, only personal names (1984:67), such names in fact identify ancest= ors: all those sharing the same name claim descent from the ancestor who first b= ore the name (1984:67). Thus, such names identify living co-descendants. These names “…are inherited from ancestors [but never a parent] according to a fairly strict set of rules”: A first-born son takes the name of= his father’s father; a second-born son takes the name of his mother’= ;s father. Likewise, a first-born daughter takes the name of her father’s mother; a second-born daughter takes the name of her mother’s mother. Additional children are named usual= ly after a sibling of their mother or father (Lee, 1984:66-7). With a population of 466 in 1964, = there were 36 men’s names and 32 women’s names. “Each man’s name may be inherited and shared by up to 25 other men, and each woman’s name by = up to 25 women” (1984:67). Furthermore, everyone bearing one’s father’s or mother’s name is referred to, metaphorically, as a “father̶= 1; or “mother”. All = those with have the same name as one’s father’s or mother’s sib= ling are called “uncle” or “aunt”, and so on.

Because such names are, in fact, ancestral, not personal, and used to identify co-descendants, the !Kung, as we might expect, use what Lee calls “nicknames” extensively to distinguish individuals (1984:67). Why does Lee call the ancestral na= mes “personal”? Proba= bly because he could not see how they could be called clans, which anthropologi= sts assume are social groups, for each !Kung family consists of individuals eac= h of whom have a different ancestral name. The important point about these names= is this: all those sharing the same name claim descent from the same ancestor = (Lee 1984:67-8), and therefore identify themselves as co-descendants of one anot= her. Thus, each Dobe !Kung can identify many co-descendants whom, like people in tribal societies everywhere, are called by a “close” kin term, = used metaphorically, such as “grandparent”, “father”, “mother”= ;, “aunt”, “brother”, “nephew”, “grandchild”, etc.

            The Australian tribes exhibit an equally unique naming system. In perhaps every Australian tribe studied, individuals are identified by both patrilineal and matrilin= eal ancestral names. Each individ= ual has a patrilineal (clan) name and a matrilineal name (usually referred to a= s a moiety, section or sub-section). These names are used to identify co-descendants even beyond one̵= 7;s own tribe (of five or six hundred people), which can be the basis of some t= rust and cooperation, including trade. Lee’s comment about all social relationships being based on kinship applies as well to the Australian Aborigines.   An understanding of the significance of the section and sub-section names, which has involved anthropologists in much controversy, presents a fascinating scientific challenge, but here we will not attempt their elucidation (see Steadman). What can be said is that both sect= ion and sub-section names are ancestral names transmitted matrilineally. However, what is peculiar is that = one does not acquire one’s own mother’s ancestral section na= me, but instead her mother’s ancestral section name. With sub-section names, one acquires the name not from one’s mother or mother’s mother, but from one’s mother’s mother’s mother’s mother. In the Australian tribes that use section names, everyone in the tribe is identified by one of four section names. The four are: first, one’s own; second, one’s mother’s; third, one’s father’s and, fourth, one’s father’s mother’s, in w= hich one finds one’s spouse. In tribes using sub-section names, everyone is in one of eight categories. The important point here: each Australian identifies many categories of co-descendants using both patrilin= eal and matrilineal names.

            A.R.Radcliffe-Brown, perhaps the most widely respected Australianist, makes clear the function of these Aboriginal ancestral names:

        &= nbsp;   The relationships between one person and another in the kinship system are individual relationships. In deciding what they are, appeal is always made to actual genealogical connection… [As] to the suitability of a proposed marriage it is the genealogical connection between the two persons that is considered…[W= ]hen the genealogical connection is too remote to be traced the natives fall bac= k on a consideration of the [matrilineal] section or sub-section or the [patrilineal] clan to which an individual belongs, but…in the minds of the natives themselves they are dealing, throughout all the ramifications of the kinship system, with real genealogical relations of parent and child or sibling and sibling. (1931:436)

&= nbsp; Finally, we use the example of Samoa, based on the work of Richard Gilson and Lowell Holmes (1974) to show an even more complex way that multiple ancestral names are used not only to identify kinship distance but serve as the basis of a highly structured society. Ea= ch Samoan is personally identified with many patrilineal ancestors, including those through each of one’s grandparents, and these ancestors are use= d to identify many co-descendants. Their first and most ancient ancestor, a God called Tagaloa, identifies all Samoa= ns as co-descendants. Tagaloa is= said to have created the Samoan Islands and put one or few pairs of ancestral hu= mans on each island, and who are used to distinguish various sets of co-descendants. Between Tagal= oa and parents, everyone identifies many ancestors, each of whom may be used to identify a set of living co-descendants.&n= bsp; Each set of ancestrally identified co-descendants is headed by an elected leader (called a matai; in the past, the leader of the large= st categories, who represent an island’s original couple, was called a <= i>tui, usually translated as king). = All these leaders, and the set of co-descendants they represent, are ranked vis= a vis one another, according to the genealogical position of their ancestor represent by their leader. For example, a matai who represents an ancestor who was the father of the ances= tor represented by another matai, outranks that second matai. Each individual is a member of eve= ry set of co-descendants, called aiga, from whose founding, or apical, ancestor he= or she can trace descent, which can number in the dozens. And an individual may be a leader = of several of these aiga. Wide-ranging cooperation among large numbers of Samoans, based on th= eir identification with distant ancestors, not only is the basis of the coopera= tive hierarchy among the leaders, but has led to a great number of highly respec= ted specializations, such as master boat-builder, master navigator, master house builder, etc. Nevertheless, genealogical distance, based on the closeness of the common ancestor, remai= ns the basis of favoritism.

The Evolutionary Significance of Ancestral Names

When considering= the evolutionary significance of ancestral names, it is crucial to keep two thi= ngs in mind. First, even t= hough most of these ancestral names are either patrilineal or matrilineal, they identify far more than one’s patrilineal or matrilineal relatives. In the previously discussed Nuer of Africa, for example, one can identify through one’s grandparents̵= 7; clans, perhaps 40,000 co-descendants, far more than the 10,000 or so bearing one’s own ancestral “clan” name. Second, humans don’t identify genealogical distance pointlessly; they adjust their behavior to t= his distance, exhibiting greater altruism and trust toward the closer kin. If the identification of genealogi= cal distance beyond first cousins were not accompanied by corresponding degrees= of altruism this aspect of human kinship would not be of great interest to evolutionists. However,= there is just such a correlation. For example, so-called segmentary lineage systems are accompanied by what Sahlins refers to as “segmentary sociability”: The closer = the kin relation, the greater the sociability . . . the more distant, or more nearly unrelated, the less” (1961:331).

&= nbsp;

&nbs= p; When these two points are kept in mind, ancestral naming systems in all tribal societies can be seen as the basis of wide-spread human altruism. Sahlins observed that “segmentation and complementary opposition are very widespread--nearly universal--features of human social organization . . . [in societies] that virtually run the evolutionary gamut from simple tribes to proto-statesR= 21; (1961:322). Sahlins continues: “segmentary sociability is particularly marked if segmentation is organized genealogically because kinship itself connotes sociability . . . The closer . . . the greater the sociability . . . t= he more distant . . . the less” (Sahlins 1961:331).

Apparently,= all tribal societies use ancestral names to identify kin, including those exhibiting the most ancient traditions of humans: those of the Australian Aborigines and the !Kung of the Kalahari.&= nbsp; This suggests that the use of ancestral names was part of the strate= gy of early Homo sapiens when they left Africa, some fifty or sixty thousand y= ears ago. Indeed, the use of these names, when combined with ancestor worship, which seems equally universal a= nd ancient (Steadman and Palmer 19XX), by identifying distant kin and promoting cooperation among them, may have provided the evolutionary edge that led to= the replacement of all other Homo species (such as Homo erectus and the Neanderthals). That is, the identification of distant ancestors and rituals aimed at worshipping them, promoted cooperation, which included common defense, trade and specializati= on, giving them a decided advantage over others who did not have such traditions. Within a few mill= ennia after contact, the other kinds of humans vanished, and modern humans rapidly spread around the world.

&= nbsp;

Conclusion

&= nbsp;

There is a simple test to determine whether the function of ancestral names fundamentally is to identify genealogical connection, as we are arguing, or social groups, such as a sub-clan or clan.= We predict, for example, that everywhere first cousins will be regul= arly favored over second cousins regardless of their clan identification. Nowhere will second cousins in one’s clan be favored regularly over first cousins not in one’s clan. This would demonstrate = that the fundamental significance of ancestral names is that they identify kinsm= en, not groups, and that genealogical distance, even among distant kin, is alwa= ys of great importance.

An evolutio= nary explanation of human altruism and cooperation must account for ancestral na= mes and the patterns of behavior based on them. Thus, we end with a suggestion for future research: We must real= ize that the traditional nature of ancestral names means that an understanding of an individual’s evolved psychological mechanisms is = not sufficient to account for the existence of such names. Evolutionary models of human behav= ior must be expanded to include the influence of selection on cultural traditio= ns.