1“The mother who refuses to breastfeed her newborn child” is a figure often included among the sinners in the Byzantine frescoes depicting scenes of Hell and the punishments of the damned. Is this artistic representation a direct reference to breast milk as the infant’s ideal food, as already attested in the written sources? Indeed, written sources, such as medical texts and vitae of saints, have been traditionally used for the reconstruction of infant feeding practices in Byzantine times, but to what extent the recommendations cited in these documents were actually followed cannot be determined by their study alone (for an overview of documentary evidence, see Bourbou, Garvie-Lok, 2009; Bourbou, 2010; Bourbou et al., 2013).

2Although isotopic signatures can only be used as indicators of weaning age, rather than an absolute measurement, the application of stable isotope analysis is a powerful tool that can add considerably to our knowledge of infant feeding in the past (i.e., Katzenberg, 2000). Up-to-date isotopic data for Byzantine (6^th-15^th centuries AD) weaning practices in Greece includes the analysis of bone samples taken from 61 non-adult human skeletons, by combining juvenile values from an assortment of eight sites from various geographical locations in the country (fig. 1), and standardizing them to mean adult female values for its site to produce an aggregate sample (Bourbou et al., 2013).

3In this study a general contrast between a higher proportion of elevated δ¹³C and δ¹⁵N values in non-adults aged three years or less and a lower proportion seen in older children suggests that weaning during the Byzantine period in Greece was complete by the fourth year, a pattern consistent with the documentary evidence. Present evidence allows arguing only that infants were weaned sometime before their fourth year but the elevation of δ¹³C and δ¹⁵N values in some children aged around 2.5-3 years indicates that breast milk formed a substantial part of some infants’ diets well into the third year of life (fig. 2, fig. 3).

4The above study has also paved the future for comparing isotopic data on weaning practices among Byzantine and Western medieval populations from the Mediterranean area and Western Europe, dating from the Roman to the late medieval period (for a review see Bourbou et al., 2013, p. 3910-3911). Byzantine populations seem to have maintained traditions carried over from the Roman period, following later weaning ages as did Roman and early medieval populations from the Mediterranean region and Western Europe. It seems, however, that this so far uniform pattern is in contrast with the isotopic data obtained from the later medieval period in Western Europe. The two available studies suggest variation with one site indicating that cessation of breastfeeding was more abrupt and occurred at an earlier age, at or shortly before two years of age (i.e., at Wharram Percy in Britain, Fuller et al. 2006), while the other indicates that later weaning was still the preferable practice (i.e., at Saint-Laurent de Grenoble in France, Herrscher, 2003). Chryssi Bourbou and Sandra Garvie-Lok (2009) have also attempted to determine whether a difference in weaning practices existed between the varied populations of late medieval Greece. In this period, invasions and migrations brought communities of Western European (i.e., of Frankish origin) to Greece (Lock, 1995). While few skeletal remains of those communities have been examined, the δ¹⁵N values of the small number of infant remains analyzed are consistent with an earlier weaning age similar to that of Wharram Percy.

5After birth, breast milk provides the newborn with immunological protection and a supply of balanced nutrients, as it contains over 100 different constituents that contribute to the health, growth and development of the newborn (Lewis, 2007, 97-99). The onset of weaning represents a crucial period in a child’s life; this offers a plausible explanation for the frequent clustering of deaths around the weaning age in non-adult mortality patterns (i.e., Katzenberg et al., 1996; Fuller et al., 2006; Wheeler, 2012). Furthermore, the introduction of solid foods can be related to the development of specific pathological conditions, such as infections, metabolic and hematopoietic conditions (i.e., Fairgrieve, Molto, 2000). Additional support for the close association between weaning practices and the development of pathological conditions (i.e., cribra orbitalia, porotic hyperostosis and scurvy), come from the study of 174 non-adult individuals in which these conditions tend to develop after the age of 3 to 4 years – a time interval which coincides with the onset of weaning, as suggested by present isotopic data and documentary evidence (Bourbou, 2010; for the association between scurvy and weaning, see Bourbou, 2014).

6Byzantine physicians of the 4^th-7^th centuries AD, suggested that a child may be brought up on milk until the age of 2 years; afterwards its diet may be changed to cereal food, while the particular choice of honey and goat’s milk in infant feeding is also documented in the written sources (Paul of Aegina, ed. Heiberg, I.5; Oribasius, Coll. Med., ed. Raeder, V.5; Aetios of Amida, Lib. Med., ed. Olivieri, IV.28). If such recommendations were broadly followed, they could have also contributed to the development of specific pathological conditions. The use of goat’s milk as a dietary staple for infants can result in serious complications, as shortly after birth develop severe megaloblastic anemia due to folic acid deficiency (Chanarin, 1990). Similarly, honey is often contaminated with the spores of Clostridium botulinum, causing botulism, a severe and often fatal form of food poisoning (Nevas et al., 2005).

7The onset of weaning and the types of foods introduced after cessation of breastfeeding, as well as a number of additional variables (e.g., harsh and unhygienic living conditions), play an important role, and the possibility that some of these factors were at work in Greek Byzantine non-adult populations deserves further investigation. As weaning process is linked to an increased risk of infant mortality and the development of pathological conditions associated with weaning stress, future isotopic studies together with bioarchaeological analyses of non-adult populations will eventually shed more light on infant feeding practices and health complications.