In a late Neolithic culture of Northeastern France (ca. 4100 BCE to 3500 BCE), ancient DNA analysis of remains and consideration of their burial nature and placement reveals that there were fairly frequent human sacrifices of people with a different DNA profile than the local population which was buried in a respectful and traditional way. Some sacrifice victims showed visible indications of violent deaths and of dismemberment with body parts buried in locations other than their natural location relative to the rest of the body in the grave.
The location and character of graves in the cemetery also suggest that these were "pure" human sacrifices as opposed to "accompaniment" burials (e.g. of a spouse or subordinate in association with a principal person's death to accompany the principal decedent in the afterlife) or a disposition of the bodies as mere waste. As the authors explain this possibility:
According to the sacrifice hypothesis, i.e., the ritual murder of individuals outside of any funerary framework, the human victim is the subject and the vehicle of a supernatural transaction that establishes a direct link between men and some deities or spirits (as regularly encountered in different past civilizations such as the Aztecs, Mayas, Ibos, and Dayaks).This hypothesis was previously explored in the paper:
Lefranc P, Denaire A, Jeunesse C. Dismembering bodies and atypical human deposits of the 4th millennium in the Upper-Rhine valley: part of sacrificial practices?. In: Sibbesson E, Bickle P, editors. Neolithic bodies, Neolithic Studies Group annual conference, British Museum, London, 3rd November 2014: Neolithic Studies Group Seminar papers, Oxbow books; in press.
This new culture integrated more women from a Western European hunter-gather background than earlier Neolithic groups in the same region.
Taken together, our data highlighted a major genetic break between Middle Neolithic LBK-derived groups in Alsace and the subsequent Late Neolithic Michelsberg groups. This genetic discontinuity appeared to be linked to new affinities with Middle Neolithic farmers from the Paris Basin, correlated to a new and important H-G legacy. All of the evidence gathered is in favor of a close genetic connection of the Michelsberg group with farmer communities from the Paris Basin region. When chronologically and geographically replaced, the data gathered suggest that the diffusion of the Michelsberg culture was linked to a movement of human groups from West (Paris Basin) to East (Rhine region). Since an admixture between H-G and farmers maternal components has been demonstrated in the Middle Neolithic groups of the Paris Basin (Gurgy "les Noisats", 5,000 to 4,000 cal. BC), we can propose that the important frequencies of H-G haplogroups measured in the MICH group may have been indirectly inherited from the MNF group in the Paris Basin (see below). The occidental origin (in the Paris Basin) of the Michelsberg culture has been proposed for a long time, based on archaeological data. This hypothesis considered that the Michelsberg culture did not appear in the Rhine region as alternatively supposed, but emerged in the Paris Basin from the interactions between the Chasséen culture, itself originating from southern France, and the Noyen culture, itself deriving from the interaction between Cerny (Paris Basin) and Chasséen. The major changes observed in pottery traditions between the Michelsberg culture and anterior cultures in Alsace and Germany, combined with the diffusion of specific ceramic features from southwestern traditions in the Paris Basin, led the author to consider that migrations may have been involved in such important transformations. Consistently, the paleogenetic data provided by our study supported the involvement of farmer group migration to explain such a global cultural rupture. The groups involved in the Michelsberg diffusion originated in the Paris Basin and carried in Alsace few LBK-derived maternal lineages but a significant number of Paris Basin farmers and H-G maternal legacy.
The exact provenance of the outsiders is not entirely clear. They came from nearby, but could either have been members of a slave caste or people kidnapped from neighboring communities with a different culture. Hunter-gatherer mtDNA, for example, is much more frequent in the sacrifice victims than in those conventionally buried.
Both the CV and NCV groups showed lineages originating from the (south)western regions as described above. Whereas all "H-G lineages" (of potential western origin) were found concentrated in the NCV group, the CV group contained a strong proportion of haplogroups H (H, H1 and H3) and X, which were more common in southern European and Paris Basin farmers. A similar phylogeographic signature could then be observed in both the CV and NCV groups, indicating a unique cultural and biological group at the local level but differently treated in death. Interestingly, isotope analyses demonstrated diet homogeneity between both groups and, as such, supported a shared geographical origin.
If we highlighted that the Gougenheim group represented a phylogeographically consistent biological group, we note differences in maternal gene pools of sub-groups CV and NCV. Although the small number of individuals in each group (and the very important mitochondrial diversity measured in each group, see below) did not allow any satisfactory statistical testing of differentiation, we could see that haplogroups W and X were specific to group CV, whereas haplogroups N1a, U5 and T were specific to group NCV. Only two HVR-I haplotypes were found to be shared by both the CV and NCV groups: haplotypes J1_16069–16126 and H_CRS. Nevertheless, these haplotypes are very frequent in Neolithic populations (found respectively at 6.7% and 14.3% in the compiled database) and thus represent a poor indicator of group affinities or maternal kinship. We consider that the distinct maternal gene pools characterized for both the CV and NCV groups, submitted to distinct mortuary treatment, points to a relative genetic isolation of the two groups from each other. This genetic isolation may have been related to the social stratification of the community, with combined endogamy in each social group and the genetic isolation of distinct social stratum and/or to some geographic distance between separate communities. These assumptions nicely echo the hypothesis of sacrifices retained for the Gougenheim site. Ethnological and archaeological examples indeed show that the victims of human sacrifice are often slaves, sometimes specifically bought for the occasion, but more regularly abducted during ambushes or razzia in some enemy villages of the neighborhood or taken into captivity during warlike events. In such circumstances, a genetic differentiation between distinct social groups corresponding to master vs. slave groups or between adversarial neighbor groups (allied communities are characterized by the exchange of women and genes) can be expected. In consequence, the obtained paleogenetic data fit well with the sacrifice hypothesis, even if they cannot be conclusive. However, the “peripheral accompaniment” assumption—archaeologically unlikely in Gougenheim—involving the same social stratum of subjugated people, is theoretically possible when considering only DNA data.
Both sub-groups CV and NCV presented an important maternal diversity (0.83 +/- 0.13 and 0.92 +/- 0.09, respectively), in total accordance with mitochondrial diversities generally measured for the Neolithic period in Europe. As already proposed for other Neolithic farmer groups, this maternal diversity could be an indicator of a patrilocal social system (although Y chromosome data are necessary to directly test this hypothesis). Interestingly, each multiple structure also showed a striking maternal diversity. For instance, the structures 1076 and 1077, grouping 4 NCV individuals each, presented a total of 6 distinct mitochondrial sequences over the 7 individuals genotyped. Moreover, when considering all of the structures with aDNA data, no grouping or spatial structure could be highlighted according to maternal lineages. Finally, concerning the individuals found in multiple “non-conventional” deposits, no shared haplotype could be detected. Only two individuals found in the multiple “non-conventional” structure 1077 shared the haplotype K_16224–16311, but once again, this very common haplotype in Neolithic groups (2.02%) may not be indicative of a maternal kinship. Considered together, the mtDNA data compiled for the Gougenheim site project the image of the random sampling of individuals inside both sub-groups CV and NCV. The biologically and socially differentiated sub-groups, submitted to very different mortuary treatments, must have been large enough to present such maternal diversities. Even if the presented arguments cannot be conclusive, the picture depicted at the genetic level is in total accordance with the idea of the random kidnapping of the individuals submitted to sacrifice. Whether these individuals were neighbors randomly abducted for religious needs or were slaves remains an open question.
The new open access paper is:
Hat tip to Bell Beaker Blogger.