Sampling
First in any study of skeletal samples is the evaluation of bone preservation. Counts of juvenile skeletons from prehistoric and historic cemeteries are often very low, introducing a bias into skeletal samples (Guy et al., 1997; Jackes, 1992; Lewis, 2002; Saunders and Katzenberg, 1992). Several authors have identified several factors as responsible for the under-representation of immature individuals. These factors are cultural beliefs about infants and children influencing mortuary behavior, the effects of biological (intrinsic) and environmental (extrinsic) processes causing differential preservation of immature bones, and incomplete archaeological recovery because of biased excavation techniques.1
Considerable evidence exists for differential burial practices that will frequently bias against
'Others divide all influences on preservation into just two categories, intrinsic and extrinsic. Intrinsic factors refer to the resistance of the element (e.g., bone) to postmortem damage based on its biomechanical properties, and extrinsic factors refer to the geographic and geologic environment ofa site, the nature of the local flora and fauna, and the activities of humans (which includes human cultural choices) (Henderson, 1987).
infants and children or alter the proportions of subadult and adult skeletons from a cemetery. The practice of infanticide has been and still is widespread and relatively common among many human cultures (Lewis, 2007; Scrimshaw, 1984). Both the deliberate killing of babies and "passive infanticide" in the form of neglect, by increasing the risks to infant survival, decrease the likelihood that some deceased infants will receive formal cemetery burial. In addition, the definition of life after birth is usually dependent on a cultural definition of when life begins. Some groups acknowledge infant life several days after birth, whereas others do not consider children fully human for several years (Saunders and Barrans, 1999). Lewis (2007) has recently summarized several examples where biological anthropologists have interpreted archaeological discoveries of concentrations of neonatal burials as cases of infanticide. She points out that the evidence is equivocal since results do not meet expectations. Ancient DNA studies do not confirm the prevalence of female deaths at some of these sites, which would be expected if, as is common, males are more valued (although mortality samples may have more males because of their physiological susceptibility to mortality). In addition, observations of unusually high concentrations of newborn skeletons may be affected by the aging methods used, and the interpretation of the absence of the neonatal line in tooth enamel as indicating infanticide shortly after birth ignores the fact that burials may be normal stillbirths. The archaeological context of burials is still important when making inferences about infanticide (Smith and Kahila, 1992).
Infants who die of natural causes are frequently buried far away from cemeteries, in house floors, and entryways or in other contexts. Taran is a well-known term in Gaelic referring to the ghosts of unbaptized babies buried outside of normal cemeteries. Ethnohistoric accounts by the Jesuits of the burial practices of seventeenth-century Iroquoians of southern
Ontario refer to the burial of infants along pathways so that their souls might reenter the womb of a passing woman (Thwaites, 1896). Modern excavations of Iroquoian village sites have uncovered high proportions of newborn infants buried in long house floors (Saunders and Spence, 1986; Spence, 1986), and it seems that the choice of the central corridor of the long house often represented the "path" along which an appropriate woman might walk (Saunders and Fitzgerald, 1988).
It is also believed widely that the bones of infants and children, because they are small and fragile, do not preserve and therefore are often lost to the excavator (Johnston and Zimmerman, 1989). Some years ago Gordon and Buikstra (1981) found that soil acidity levels are correlated significantly with bone preservation in both adult and children's skeletons, with juvenile bone durability declining rapidly with decreasing soil pH. Certainly intrinsically, the bones of young individuals have high organic and low mineral content and are less dense than the bones of adults, which makes them more susceptible to decay (Currey and Butler, 1975; Specker et al., 1987). Research on living individuals has shown that the bone mineral density or BMD (the mass of inorganic mineral matter per unit volume) in a normal infant skeleton decreases in the first year after birth and then increases through childhood and adolescence (Rauch and Schoenau, 2001). However, lower than normal bone density can occur because of problems of maternal health as well as of several other pathological factors after birth (see Stodder, this volume). The question is whether the bioarch-aeologist can devise investigations that could detect such phenomena in skeletal mortality samples.
Some researchers have compared observed skeletal samples with expectations from documentary data to judge relative preservation of different sex and age groups. Walker et al. (1988) examined mortality profiles derived from an analysis of burial records and skeletal collections from a nineteenth-century
Franciscan mission cemetery in California. The baptismal and death records of the mission indicated that most people buried in the cemetery were either infants or elderly adults, which fits the typical U-shaped distribution of mortality observed in earlier populations. However, the skeletal sample contains mainly young adults. Even though the skeletal sample size is small (only 2% of all people buried in the cemetery), a random sample of burials should not deviate so much from the known age distribution of people buried in the cemetery. The authors concluded that age-specific differences in preservation account for the missing children and elderly adults. Later, Guy et al. (1997) claimed that the proportion of infants (under one year) in archaeological cemeteries generally fluctuates around 5% or 6%, whereas the proportion of infants dying in pre-vaccination populations recorded by historical demographers should never fall below 25% of live births. They identify taphonomic variability (susceptible bones in poor burial environments) as the main cause of low percentages of juvenile bones in skeletal samples, arguing therefore that paleodemographic reconstruction is restricted uniformly. The same explanation was offered by Jones and Ubelaker (2001) who compared an excavated skeletal sample from a nineteenth-century Swiss-German church cemetery in Pennsylvania to parish registers listing the burials. Of those burials listed in the registers as infants or children less than five years of age, only 50% preserved as skeletal remains.
On the other hand, others have argued (Lewis, 2007; Sundick, 1978) that even the smallest growth centers and the most fragile of the flat bones of the vault of the skull or the scapula of subadult skeletons can be as well preserved as the bones of the robust adult skeleton. Some attribute incompleteness to the degree of skill of the excavators and the need to be able to recognize juvenile bones. Few would care to admit that excavation methods at a site had been less than perfect, but obviously deficient training may contribute to inadequate recovery of skeletal remains.
More recently, Bello et al. (2006) have proposed a new method for evaluating bone preservation that takes into account the different aspects of what represents "well-preserved," "present," and "complete." Skeletons from an excavation may be missing whole elements, some of the elements that are present may be broken, and the cortical surfaces of intact or broken bones may be damaged. These authors evaluated preservation in three different ways. They recorded an anatomical preservation index (the percentage of bones preserved out of the total anatomical number of bones of the skeleton per individual) and a bone representation index (the ratio of the actual number of bones recovered at excavation to the total number of elements of the skeleton expected). Lastly, they calculated a qualitative bone index or the ratio between sound cortical surfaces and damaged cortical surfaces for each bone. They evaluated skeletal samples from three historical cemeteries for these indices. They found juvenile bones to be more poorly preserved and less well represented than adult bones, especially for the 0-4-year group. They also observed that female immature skeletons did not survive as well as male immature skeletons in the Christ Church, Spitalfields sample (London, U.K.). An evaluation of extrinsic factors such as burial location within a site and burial depth (graves of infants and children tend to be shallower) found no evidence that these factors strongly influenced the state of preservation of the bones from these three sites. However, intrinsic anatomical properties of the bones themselves (i.e., bone mineral density) do seem to be important (as observed by the negative correlation between preservation and age), and extrinsic factors, in some situations, may serve to exacerbate this general phenomenon. But these authors did not present general demographic statistics to evaluate the overall proportions of juveniles in their samples compared with adults. This comparison can help to detect human influences such as differential mortuary practices for infants and children.
Thus, we should expect that explanations for juvenile sample bias in skeletal collections are diverse, multifactorial, and often unique to the cultural and archaeological context of the site (see, for example, Hoppa and Gruspier, 1996; Hoppa and Saunders, 1998; Nawrocki, 1995; Paine and Harpending, 1998; Saunders, 1992). On the other hand, Guy et al. (1997) reject cultural explanations for the lack of infant skeletons, citing an example from an historic French cemetery where the proportion of infants represents a fifth of the recorded proportion in the parish registers. They note that this population punished infanticide and practiced the baptism of neonates, presuming then that most infant skeletons should have been found. Yet, several examples exist from historic cemeteries where it is clear that spatial locations of burials are age-specific (Lilley et al., 1994; McWhirr et al., 1982; Molleson and Cox, 1993) or where substantial proportions of immature individuals, even infants, are present (Cook and Buikstra, 1979; Farwell and Molleson, 1993; Hutchins, 1998; Lewis, 2007; Owsley and Jantz, 1985; Saunders et al., 1993; Saunders et al., 1995a; Sperduti et al., 1997).
Post a comment