Article Review II
Palmstierna, Markel, Anna Frangou, Anna Wallette, and Robin Dunbar. 2017. "Family Counts: Deciding When To Murder Among The Icelandic Vikings". Evolution And Human Behavior 38 (2): 175-180. doi:10.1016/j.evolhumbehav.2016.09.001.
1. What is the central question addressed by the authors? 10pts
The authors are addressing the question of the existence of a correlation between the number of human kin of a victim of murder in comparison to the number of kin of the killer in Viking communities. They also looked at whether it mattered if this kin group was made up of more biologically-related kin or affinal(in-laws) kin. The hypothesis was that killers targeted victims with smaller kin groups than themselves.
2. What methods did they employ to address the central question? 15pts
Anthropologists used 3 historical Viking family sagas to gather accounts of murders, family lineage and major events. A value of r was used for each level of kinship, with parents and siblings having r=.5, uncles, aunts, grandparents, half siblings having r=.25, and parent's cousins having r=.125. The sagas were then used to document 1891 social events and 1020 individuals, all of which branching in a single pedigree besides a few people, which were mostly slaves. Concubines were counted as wives, and their children as half-siblings of legitimate offspring. They then collected all accounts of non-accidental deaths from these sagas beside those during large battles, and documented who was the victim and killer. Sagas were cross-referenced to check the accuracy of accounts. In some situations there was more than one killer, which was also noted. Both victims and killers were then analyzed for their amount to kin out to cousins, and those r values were added up to a total "kin value". For the statistical comparisons, values of total r, degrees of freedom, chi square analysis, p and the Z test were given when needed to legitimize data. Netdraw software was used to plot the kinship network among the involved people, creating a web showing the relationship between killers, victims and their kin.
3. Were their methods appropriate and adequate to address the central question? 10pts
The method of gathering data from sagas, while not entirely reliable, is most likely the best source of information because of the time period being so long ago and the apparent thoroughness of the family histories. The report recognizes the author biases of these sagas as possibly affecting data, but states that it can still be used as a legitimate data source as a historical document because it resembles natural human social networks, parts can be confirmed from other historical sources and archaeological evidence, some people are included in multiple sagas as overlapping evidence, and the fact that Norse culture valued blood money or revenge murder as the proper response to a murder, failure to obtain this would be dangerous as the family would be considered weak and vulnerable. The statistical analysis mainly focused on the number of kin and connected murders. Hamilton's rule of relatedness and the concept of r-relatedness was used in classifying the relationship of kin within the population. Kin was measured by the sum of the r-values based on the level of blood relation, then affines and foster kin(which was rare). The logic was that the larger the r sum, the larger the family and the greater the chance of violent retribution should one of their members be killed. However, there most likely was present multiple factors that could not be measured by number of kin, including the family's affluence, the specific role the victim played in the family(possibly relating to their value to their kin), and the characteristic makeup of the family during that time, as a large intimidating male population may seem more dangerous than a family with many elderly people, small children or women. However, the results would most likely not be majorly affected even if those characteristics were known to the point where the conclusion would be unsupported.
4. What were the results? 10pts
Overall, 18% of all adult males in the sample died as a result of murder. Anthropologists found that killers had significantly more kin than their victims in 80% of the cases found in the sagas. First, there were 153 murders recorded by a total of 66 people. Almost half of these killers were involved in multiple murders, many times having victims from the same kinship group. The mean number of kin of killers was 16.4(r), compared to victims having on average 5.8. Many of the victims did not have any recorded kin(32%), while most killers did(86%). None of the victim/killer relationships were close kin, aside from one man killing his sister-in-law. Biological kin alone was also tested, but still showed that 65% of murders involved killers with more kin than their victims. Six people were responsible for almost a third of the reported murders, and they all had not only a significant difference in number of kin compared to their victims, but also had a large amount of kin in general. In contrast, "Only those who killed once did not have significantly more kin than their victims, though on average even they had slightly more kin, perhaps because their murders were less calculated and more often spontaneous responses to insults or casual attacks." In tackling the possible data bias of killers having longer life periods, and therefore more opportunity to sire children, offspring were omitted for testing purposes and yielded a still large margin between killer and victim kin, 10.98 vs 50.4 on average. The overall relatedness of the included population was mapped, and anthropologists found that of a total of 607 kinship ties, 162 were victim-killer relations, a high percentage. Most who had high numbers of kin ties also had high number of killer ties, meaning that killers were often related to one another. More of these connections were affinal rather than biological. Many victims are also related to one another, possibly as the result of a feud between families. Although foster kin were considered a cultural norm, they did not seem to be considered in the comparison between kin by killers, and were the most often of-kin death that occurred with 3 total.
5. How did the results answer the central question? 10pts
The central question was if the amount of kin between the potential killer and victim affected murder in a sampling of Viking sagas. The hypothesis was that a larger kin group of the killer compared to that of the victim would lessen the risk of retaliation by the victim's family, and therefore be carried out. The data from the 3 Viking sagas supports this hypothesis in that it shows a higher number of kin for the killers compared to their victims for a majority of instances. These murders were between families, not among them, bringing about the subject of Hamilton's rule in that there is present a sense of familial protection for the continuation of their genes in offspring. Because of possible skewing, specific variables such as number of offspring were temporarily disregarded to test for their effect on the data, which was overall very little. Tests were also done on whether the matriarchal vs. patriarchal side of kin was more prevalent in the difference in kin between parties, which was found to not be majorly different. However, looking at kin ties within the community found that many killers had more affinal ties than the victims did, which could possibly be evidence for the linking of families by marriage in an effort to strengthen and expand their power and influence. They would also share an interest in the survival and protection of the offspring of these unions. A very interesting piece of evidence is that the larger gaps between killer and victim kin numbers occurred with killers who struck more than once, up to 19 people in one case. Scientists concluded that the most logical reason for this was that these killers took more time plotting and rationalizing their actions as to if they would be in danger of reciprocity by the victim's family and therefore needed a larger family as support and protection. Logically, one would not kill a multitude of people from a large, influencing family without proper protection from the effects of the deed. Many of the victims of these "multi-killers" were related to one another, suggesting a specific family feud between killer and victims. Those who only killed once were least likely to have a significant gap in kin between them and their victim, suggesting more spontaneous and event-driven murder. The study mentions Lanchester's Law as being identifiable in these conclusions, which states that people(and also chimps) are more likely to engage in a risky action when the odds of success are very much to their favor. In this instance, the odds of success depend on the amount of people on your side in comparison to the enemy's, or the victim. These murders therefore were less likely to be carried out by someone with a significantly smaller kinship group than of whom they plan to kill.
6. What would you do differently to address the central question/issue? 20pts
The source of data of this experiment is a little more subject to biases and vague on details than something I would have liked to have used to answer this central question. However, as there is no time machine readily available, it seems to be one of the only sources of information on this particular subject. Although 3 closely-related sagas is a comfortable amount of data, more sets of sagas from another time period or area of the realm could have increased the ability for comparisons and diminished data-affecting variables. The authors of the study mention that the culturally expected response to a murder was to demand a revenge murder or blood money. However, there was little discussion as to how this possibly could have affected results. I believe that not only could kinship be a deciding factor on these murderous actions, but also financial gain. For families that are larger, and therefore could probably have more resources, they may not be as concerned with receiving blood money, and choose to take the life of one of the killer's family members. On the other side, someone whose kin had just been killed by a larger, wealthier family may demand blood money for their own financial gain instead of a revenge murder, which would in no way be recorded in the saga according to the information given. While it is not clear if that would significantly change data, as there is not any information as to which was more common, revenge murder or blood money, it is still an important point that I would try to discuss. There is also no data as to if a murder was done in revenge to a previous one, which type of kin was most often chosen, if any. There was no example given of a murderer becoming a victim, which would seem to be the most pure example of a revenge killing. This point could be discussed to answer a subset of questions concerning the roles of kin within the family in these types of situations. Finally, since these sagas seem to be historical storybooks of the happenings during the time of these families, it would be interesting to pair characteristics of people within their kinship to their role in murder activity. Comparing the leadership roles of the victims and killers might add another dimension to the data, possibly to discover if there was a larger difference in kinship size between a killer and a victim who was of an important familial leadership role versus a victim who was of a lower status, younger age, or a woman. In short, a more holistic approach to the concept by identifying these variables and discussing their possible effect on the data would improve this study.
7. What is the significance of the paper to biological anthropology, i.e., why should we care? 25pts
rticle Review IIeffects of kinship wledge of the positive points of social societies. ans but for other socially-inclined prima
This study provides ancient documentation evidence for both Hamilton's rule and Lanchester's Law, which are used in the study of human and primate social interactions and their relationship to the quest for individual and genetic survival. This tie between kin and violence shows a correlation between the protection that comes from a committed group and the amount of violence that it will meet from other groups. A high relatedness, r, could therefore be seen to correlate to a high dedication to protection of those who are closest to you, as shown by Hamilton's rule. This dedication would be intimidating and dangerous to those who seek to do harm on them, and leads to many deciding the risk is too great, proving Lanchester's law. As this study shows these principles of primate evolution and relatedness, it is therefore significant in the link between biological fitness and our actions toward one another, reflected in cultures across the globe in various time periods. It also supports the notion that there is generally safety in numbers of those we can trust, and power comes from the number of those alliances. Further research of this kind, since it states it is the first for this particular concept, should be done in other areas of the world and time periods, not only for humans but for other socially-inclined primate species to compare and expand knowledge of the effects of kinship in social societies.
...(download the rest of the essay above)