The Agora Bone Well

The Agora Bone Well presents the study of more than 460 humans (mostly newborn infants) and 150 dogs, along with the artifacts deposited in a well near the Athenian Agora. Dorothy B. Thompson excavated the well more than 80 years ago. It has remained largely forgotten, except for an occasional suggestion that its assemblage represented plague, human sacrifice, or large-scale infanticide. The nuanced analysis of the evidence from this cold case tells a heartbreaking tale of all-too-ordinary life and death in ancient Greece.

The first chapter presents the well’s immediate context: its excavation, construction, and surroundings. It was constructed in the Classical period as part of a larger water system and was used by industrial workshops into the Hellenistic period. The bulk of the well’s assemblage was deposited after the abandonment of the surrounding structures in the early second century B.C.E. The well shaft was closed after 150 B.C.E. with a fill of earth and stones.

Chapter 2 presents the human skeletal material. The commingled newborn remains were analyzed collectively, but four older persons could be examined individually. The patterned skeletal lesions on the adult male (45–55 years old) suggest hereditary hemochromatosis, which leads to disability and death. Interestingly, the modern treatment is bloodletting, which he did not receive. The child (8–10 years old) also had extensive pathological lesions, indicative of a long-term disease. Brucellosis, an infection contracted from unpasteurized goat or cow milk, is the most likely cause.

The oldest infant (16–18 months old) suffered from numerous traumatic injuries. Many broken bones had healed, but a skull fracture with minimal healing potentially led to death. This infant is the oldest archaeological example of battered child syndrome, important because “the paucity of archaeological cases of abused children has led some scholars to speculate that physical abuse is a modern phenomenon” (36). The oversized cranium of the second oldest infant (6–8 months old) is indicative of hydrocephalus, which would have caused many problems and, ultimately, death. The infant’s survival for several months suggests the child “was cared for during a period when it would have become progressively more debilitated and more disturbing in appearance” (38).

The remaining 13,018 human specimens belonged to nearly 460 neonatal infants. The sex distribution is nearly equal, and the gestational age of the infants clusters near the end of full-term pregnancy. Very few individuals were older than a week. A large number of specimens showed signs of illness or other health issues that could have greatly increased mortality: preterm or undersized births (34%), cranial pathologies from likely infections (25%), and various developmental defects (e.g., cleft palate inhibiting normal feeding). The authors convincingly link these patterns with natural infant mortality.

Chapter 3 presents the faunal remains. A small quantity of butchering or food consumption refuse and industrial debris shows the well was a convenient dumping spot for local activities. Most of the animal remains derive from more than 150 dogs deposited in the well with no signs of butchery. That they were mostly fully grown indicates they do not represent natural deaths but were selected intentionally. The middling size of the dogs and the frequent signs of skeletal pathology due to malnutrition, illness, and trauma suggest they derived from “common urban mongrels . . .  that inhabited the environs of Athens” (59).

Chapter 4 presents the artifacts. Fragments from over 290 ceramic vessels were identified. About 180 of these vessels were represented by small fragments, suggesting local debris dumped into the well. The more intact vessels are plausibly linked with the deposition of infants. Likely, some were grave goods: a ceramic feeder, 10 mostly complete unguentaria, other intact fine ware vessels, and two canteens. A striking contrast with other, contemporary, ceramic assemblages is the number of large, open vessels (lekanai, large bowls, kraters, and mortars). These large vessels are credibly argued to be containers or coffins for the deceased.

The nonceramic finds were mostly local refuse, including more than 8 kg of corroded bronze. The authors argue in favor of Andrew Stewart's interpretation of a marble female herm identified as Eileithyia, the goddess of childbirth. They suggest that it was perhaps deposited in the well by a grieving parent or a midwife.

Chapter 5 ties the various strands together in a presentation of the archaeological and cultural context of the well. There are several parallels to the Agora Bone Well in the Hellenistic and Roman Mediterranean. Three wells from Eretria and Messene included both infants and dogs, and a sewer at Ashkelon included only infants. In all cases, the infant age ranges are similar to those from the Agora Bone Well, and Liston’s preliminary study at Eretria reveals similar endocranial pathologies. A review of intramural infant burials and infant cemeteries also shows age ranges paralleling natural infant mortality, the primary difference from the Agora Bone Well being the presence of additional older infants. The ritual disposal of dogs is a recurring feature in many of these mortuary contexts.

The exclusion of more dramatic explanations for the Agora Bone Well is largely based on the skeletal evidence. The age range, sex ratio, and high prevalence of health issues parallels natural infant mortality and is not indicative of plague or selective infanticide. The authors link the paucity of infants over a week old to the rite of Amphidromia, where infants were inspected for healthiness. It is plausible that infants who survived beyond this point were given a more formal burial. The number of infants suggests they derived from the population living in the northwestern region of Athens. Most likely a midwife (or a few) started using the well as an isolated location on an abandoned plot of land. The older individuals, with their debilitating ailments, were potentially introduced unceremoniously.

There are few quibbles to make with this excellent study. Most seriously, though, the conclusions depend on accurate age estimates, based on humerus and femur lengths. The authors note, “There has been considerable debate . . . concerning techniques for estimating fetal and infant age from long-bone lengths” (41). While two methods of age modeling are provided, the publication does not include the raw biometrical data with precise measurements. Similarly, the raw data for dog biometry and pathologies are not included. Data appendices would have made this study more valuable to other researchers.

A smaller point is the authors’ doubt of the reliability of chemical or DNA analysis due to the “chance of cross-contamination of the samples in the fetid mixture of organic material . . . below the waterline in the well” (43). Uncontaminated aDNA and isotopes have been successfully analyzed from waterlogged contexts (the authors reference an aDNA study of the infants in Ashkelon’s equally fetid sewer). It is unwarranted to preclude potential future analyses that could add insight to this excellent study.

The Agora Bone Well is a poignant text, effectively marrying scientific methods and humanistic inquiry. The rigorous analysis is relevant to scholars studying ancient childbirth, paleopathology, zooarchaeology, ceramics, mortuary ritual, Athenian topography, and more. Unlike most multidisciplinary volumes, this text weaves together a clear, concise, and coherent narrative. While the content is academic, the writing shines with an accessible empathy: “The Agora Bone Well may have offered a pragmatic solution to the all-too-common phenomenon of perinatal infant death, but its use does not necessarily suggest the absence of grief among the mothers and other members of the families who suffered these losses” (139).

W. Flint Dibble
Malcolm H. Wiener Laboratory for Archaeological Science
American School of Classical Studies at Athens
wfdibble@gmail.com