Urban health is a hot topic for the World Health Organization (WHO). This is not surprising since around 55% of the world’s population now lives in cities, with fast increasing rates.[1] But city life was also common in antiquity. It is estimated that between 10 and 30% of the Roman population lived in cities.[2] Living in confined spaces promotes the spread of both infectious and non-communicable diseases and general stress. WHO has been addressing this issue for years.[3] Bioarchaeologists are also now working on this topic.[4] The present work therefore fits well into this newer field of research.
The work discussed here is divided into nine chapters and concludes with an extensive bibliography.
Chapter 1 briefly summarizes the history and archaeology of Knossos-Colonia Iulia Nobilis Cnossus, re-founded after the Roman conquest in 67 BCE, and Chapter 2 places it in relation to Crete and the central Mediterranean region. Chapter 1 also presents the research question, examining and comparing mortality, nutrition, dental diseases, joint diseases and enthesopathies as well as postcranial skeletal measurements over a period of approximately 1100 years and in various stages of Knossos’ development. The aim is to determine the effects of urbanization on the health of the city’s inhabitants. The section on ‘Urbanism and Health’ (pp. 6–7) is surprisingly short, with little theoretical outline and no discussion of the theses by WHO or Betsinger and DeWitte.
Chapter 3 presents the material basis (data: Appendix A). Knossos is one of the longest continuously inhabited places in Europe — since around 7000 BCE (p. 1). Even though only relatively small areas of the Hellenistic-Roman city have been excavated (in contrast to the Minoan palace), the area has been explored by the ‘Knossos Urban Landscape Project’ (KULP) of 2005–2016. Moles’ maps of the location of the necropolises, individual graves and skeletal complexes examined in her work (Figs. 3.1–3.6) gives the reader a good insight into the shrinking extent of the ancient city and its necropolises over time. The human skeletal remains examined originate from excavations by the British School at Athens between 1930 and the 1970s (p. 19). The period under investigation is between the Hellenistic and Late Antiquity (c. 323 BCE–800 CE) (p. 19). The main types of tombs are presented and illustrated with selected examples (Figs. 3.7–3.28). A type chart would have been helpful here. Since many graves had been robbed or disturbed, the grave goods are generally incomplete and can hardly be used to estimate social status. Grave types were, therefore, used for this purpose: mausolea are considered the most elaborate graves, pit graves the simplest (p. 36–37).
Chapter 4 deals with mortality. A minimum of 537 from 109 graves were examined, of which 82 date to the Hellenistic period (61 graves), 231 to the Roman period (29 graves) and 223 to Late Antiquity (19 graves) (data in Appendix B). The sex ratio is slightly skewed towards males in the Roman period (F 36, M 39) and Late Antiquity (F 36, M 41), while the Hellenistic sample includes 31 females and only 16 males (p. 53). As in many other Hellenistic to Late Antique skeleton series, there is a clear deficit of children: in the Hellenistic period, 14.2% are subadults, in the Roman period 37.0% and in Late Antiquity 39.2% (Table 4.3). Newborns and infants are particularly underrepresented; the former are not present in Hellenistic burials examined. The highest adult mortality rate is found in the 20–30-years-old age group, as is typical for ancient to early Byzantine populations in the Mediterranean region and is also reflected in contemporary grave inscriptions. Their proportion declined slightly from the Hellenistic to the Late Antique period, while the percentage of 30–45-year-olds increased, and the proportion of those over 45 remained more or less stable and even increased in Roman times (Figs. 4.5–4.7). This suggests a slightly longer lifespan in adulthood, possibly caused by improved living conditions in Roman times. Todd Whitelaw and colleagues (2019, Fig. 17; 19) estimated the population in the different periods on the basis of changing city areas with 50 to 200 people per hectare. The corresponding values for the Hellenistic to Late Antique periods are given by Moles (Table 4.5), with the most likely figure being 100–150 inhabitants per hectare. According to this, the average population in the Hellenistic period was approximately 13,756, in the Early Roman period 11,121, and then halved by Late Antiquity. Moles also estimated the number of deaths in the various periods (Table 4.6). It remains unclear why she only used half of the reconstructed population figures (adults) for this. If the annual mortality rate of 42.1 per 1,000 people estimated by Roger Bagnall and Bruce Frier[5] based on census data from Roman Egypt is used, the figures are more than twice as high. Regardless of which estimate is applied, the number of 537 anthropologically examined individuals is extremely low compared to the estimated death figures.
Chapter 5 is devoted to stable isotope analysis for dietary reconstruction (data: Appendix C). Ninety-three individuals were analyzed, i.e. 17% of the total sample, exclusively adolescents and adults (p. 86) and predominantly sex-determined (Fig. 5.20). The diet of livestock and humans was based on so-called C3 plants, which are predominant in the Old World. C4 plants such as millet apparently played only a minor role (cf. Fig. 5.23). There are no sex differences in the average values (Table 5.6), although the ranges are wider for women, suggesting a more varied diet (p. 81). However, when the three time periods are considered separately, there are clear differences: the highest δ15N value and the largest range are found among Late Antique women (Table 5.7). Overall, the δ13C values decrease slightly from the Hellenistic to the Late Antique period (Fig. 5.21), while the δ15N values increase from 8.5‰ to 9.1‰ in Late Antiquity (Fig. 5.22). This indicates a plant-based diet with occasional consumption of animal protein (p. 100). Social differences in diet appear to have been rather small. The comparative table 5.12 is helpful, as it contains Bronze Age series from Knossos and then mainly Hellenistic to Late Antique series from the Mediterranean region. Further comparisons between the Late Antique samples from Knossos and the CIMA database[6] would have been possible here.
Chapter 6 studies the dental diseases caries, tartar, and intravital tooth loss. Periodontal diseases were not diagnosed due to the poor preservation of the jaws. The affected teeth or alveoli were each related to the total number of preserved teeth or alveoli. Unfortunately, no information on the number of individuals affected is available. Generally, both reference values should be provided. As no original data on the dental findings are available, it is not possible to determine the number of individuals affected retrospectively. Odontometric data were not collected. As expected, caries prevalence increases with age (Table 6.8; Fig. 6.10), which is particularly evident in cavities (Fig. 6.11). The same applies to tartar deposits (Fig. 6.27) and ante mortem tooth loss (Fig. 6.37; 6.39). Men are significantly more affected than women by all dental diseases (Fig. 6.12; 6.29; 6.40). There are different distribution patterns with regard to time periods and grave types. Moles primarily attributes the frequency of dental diseases to diet (p. 142). This is certainly true in most cases. However, poor dental hygiene can also contribute to an increase in dental diseases.
Chapter 7 examines activities based on degenerative changes in the body and vertebral joints as well as enthesopathies (basic data in Appendix D). Men show twice as many degenerative changes in the large body joints (Fig. 7.6) and vertebrae (Fig. 7.16) as women. The differences are smaller for enthesopathies (Fig. 7.27). It must be taken into account that only some of the affected individuals could be assigned to a sex (p. 151). There are clear sex differences in terms of stress on the cervical and thoracic vertebrae, while the lumbar vertebrae are subjected to the same stress (Fig. 7.17). The thoracic and cervical vertebrae are most severely affected. Overall, there is evidence of a slight increase in degenerative vertebral changes (Fig. 7.14) and in the large body joints (Fig. 7.13) from the Hellenistic to the Late Antique period, while a decrease can be observed in enthesopathies (Fig. 7.26). Different movement and stress patterns can probably be assumed here. Schmorl’s nodes were found in 52 individuals, increasing significantly from Roman times to Late Antiquity (p. 165). Ankylosing spondylitis (AS) is suspected in nine individuals (Table 7.12), with only two vertebrae connected in each case. Diffuse idiopathic skeletal hyperostosis (DISH) is assumed in three cases (Table 7.13), with two or three vertebral bodies connected; a definitive diagnosis requires at least four connected vertebral bodies[7]. In both AS and DISH, most cases were observed in osteothekai and chamber graves, which could indicate better nutrition due to connection to the middle class.
Chapter 8 is entitled ‘Development,’ but deals exclusively with postcranial osteometry in adults. While bone measurements of subadults are not considered, the chapter heading is misleading. First, large long bones are examined, followed by metacarpals and metatarsals. Since only relatively few large long bones were completely preserved, but numerous metacarpals and metatarsals, this approach is very useful. It is striking that the talus and calcanei are not included, as these are usually more or less well preserved. The most important statistical parameters are given for each skeletal element, followed by histograms with 1 mm class widths separated by sex, dating and grave type. These histograms are a substitute for the original measurements, which are not given. Adult stature was estimated on the basis of 1194 large long bones, mostly metacarpals and metatarsals. The average for women is 156.78 cm (n=505), and for men 171.85 (n=689) (p. 224). For both sexes, a slight increase in height can be observed from the Hellenistic to the Late Antique period (Table 8.31). Overall, the body heights from Knossos are in the upper third of the Greek comparison series (Table 8.1).
The concluding Chapter 9 successfully demonstrates that a diachronic study of the state of health and disease within the ancient city is a worthwhile endeavor. The equal consideration of mortality data, selected diseases, metric data and stable isotopes for nutritional reconstruction is a promising approach. It is encouraging that the author relates the above-mentioned data to the archaeological features, i.e. to dating and grave types as proxies for social classification.
What is missing? Interested readers will unfortunately not find sinusitis as a proxy to air pollution3, non-specific stress markers such as linear enamel hypoplasia, Harris’ lines or cribra orbitalia and porotic hyperostosis — due to bad preservation (p. 181).
Overall, this work is a very useful publication. It is characterized by a high degree of transparency, as much of the primary data is available online in Appendices A–D. It is also one of the few studies to undertake a diachronic comparison between the Hellenistic and Late Antique periods in terms of mortality, nutrition, selected diseases and osteometry. This work thus provides indispensable comparative material for the period in question. It belongs therefore in the library of every anthropologist and paleopathologist who deals with the period between 300 BCE and 800 CE in the Mediterranean.
Primary data from Moles’ work can be downloaded here:
Appendix A: Catalogue of Skeletal Units – https://doi.org/10.30861/9781407360355-AppendixA
Appendix B: Demography Results – https://doi.org/10.30861/9781407360355-AppendixB
Appendix C: Stable of Isotopes – https://doi.org/10.30861/9781407360355-AppendixC
Appendix D: Activity Results – https://doi.org/10.30861/9781407360355-AppendixD
Notes
[1] World Health Organization (ed.), Fact Sheet: Urban Health. https://www.who.int/news-room/fact-sheets/detail/urban-health (19.03.2025).
[2] J. W. Hanson and S. G. Ortman, “A systematic method for estimating the population of Greek and Roman settlements”, Journal of Roman Archaeology 30, 2017, 301–324, here p. 323. https://doi.org/10.1017/S1047759400074134
[3] World Health Organization (ed.), Global report on urban health (Geneva: WHO, 2016).
[4] E.g., T. K. Betsinger and S. N. DeWitte, “Toward a bioarchaeology of urbanization: Demography, health, and behavior in cities in the past”, Yearbook of Physical Anthropology 2021, 1–40. https://doi.org/10.1002/ajpa.24249; T. K. Betsinger and S. N. DeWitte (eds.), The Bioarchaeology of Urbanization (Cham: Springer, 2020).
[5] R. S. Bagnall and B. W. Frier, The demography of Roman Egypt (Cambridge: Cambridge University Press, 1994) 105.
[6] C. Cocozza et al., “Presenting the Compendium Isotoporum Medii Aevi, a Multi-Isotope Database for Medieval Europe”, Scientific Data 9, 2022, 354. https://doi.org/10.1038/s41597-022-01462-8
[7] D. Resnick et al., “Diffuse idiopathic skeletal hyperostosis (DISH) ankylosing hyperostosis of Forestier and Rotes-Querol”, Seminars in arthritis and rheumatism 7, 1978, 153–187, here 154.