Nutrition and Well-Being in the Roman World: The Evidence from Human Bones

On Sunday, I returned from Rome, where I participated in the one-day conference Nutrition and Well-Being in the Roman World: The Evidence from Human Bones.  The conference was organized by Kim Bowes, a classical archaeologist at U Penn who is also the professor-in-charge of the school of classical studies at the American Academy in Rome.  Bowes recognized the growing interest among classicists (especially classical historians) in the information that human skeletal remains can tell us, and she thought to get together most of the leading names in Roman bioarchaeology to discuss what we can and can't determine from bones.  It was fantastic to be in a room with dozens of people who care as much about the bones of dead Romans as I do -- there's not much opportunity for me as a North American to have these sorts of conversations in person.

Some of the papers were in Italian, so I couldn't live-tweet the conference, as I was concentrating on remembering all the Italian I've forgotten since I lived in Rome in 2007.  But I wanted to post a quick summary of each of the papers.

First up were Dominique Castex and Kevin Salesse of the Université Bordeaux talking about the Osteobiography of Epidemic Victims from the Catacomb of Saints Peter and Marcellinus in Rome (1st-3rd centuries AD).  The last time I talked to Kevin at this past year's AAPA meetings, he didn't have isotope results yet.  At the conference, he displayed some preliminary C, O, and N results, which were both similar to and different than my data from the same time period.  Interestingly, he found a high rate of migration using just O isotopes, but the question was raised as to whether his "local" range, which he drew from Tracy Prowse's work at Isola Sacra, was a valid range (a valid question that's been asked of my data in the past as well).  Kevin also did biodistance analysis on dental nonmetric traits, finding that one chamber or niche is distinctly different from the rest -- and yet the O data weren't different, which is an interesting result.  Castex presented on some of the pathologies from the catacomb.  I was thrilled to hear that their rates of disease were quite low -- linear enamel hypoplasia at 1.9% and cribra orbitalia at 0.7%, for example.  These data compare with my pathology frequencies from Rome, data that have been called into question by others because they're just so much lower than expected.  There may be a bit of inter-observer error at work, or it could just be that the Romans were quite heterogeneous in health.  Castex, Salesse, and I seem to favor this latter interpretation.  I can't wait to read Kevin's dissertation, to find out more about this really interesting population.

Then Paola Catalano and her colleagues at the archaeological superintendency presented on Le Condizioni di Vita Quotidiana a Roma: il Contributo dell'Analisi Antropologica dei Sepolcreti Scavati negli Ultimi Anni dalla Soprintendenza Speciale per i Beni Archeologici di Roma.  Demographic analysis of some of these recently excavated cemeteries shows that these were non-natural populations, which is to be expected in bioarchaeology, and that all of them had wildly different demographic make-ups (in terms of age-at-death mostly).  A couple new sites that I haven't read much about yet are Quarto Capella and Padre Semeria, although some preliminary data were presented in the Proceedings of the 4th International Congress on Science and Technology for the Safeguard of Cultural Heritage in the Mediterranean Basin in 2009. I was most intrigued by the carbon and nitrogen isotope study that Flavio DeAngelis did on some skeletons from Casal Malnome.  He found ranges from about -21 to -17‰ for carbon and about 8 to 14‰ for nitrogen.  These are well in line with the data I produced from Casal Bertone and Castellaccio, and that others have produced from sites around Rome -- the Roman diet was a little bit of everything.

Mauro Rubini and Paola Zaio then presented information on Nutrition and Health in Central Italy During the Roman Period, a summary of the work they've been doing for the past several years.  In the rural areas of Italy, they found that quite a number of subadults had Schmorl's nodes, suggesting to them that adolescents were doing hard labor from a young age.  Not terribly surprising if we consider what agricultural work involved even as recently as the 20th century in the U.S.  A lot of Rubini's recent work focuses on leprosy and tuberculosis, and particularly in the cross-immunity between them.  For example, he found two cases of TB from Rome and two from Herculaneum; these make sense, as TB thrives in communities that have high population density.  But he also found two cases of TB from Sardinia, where we wouldn't expect to find it.  His suggestion was that there may have been less leprosy in Sardinia, as the disease seems to have wended its way down the peninsula from the north, meaning no cross-immunity to mediate the Sardinians' exposure to the TB bacillus.  It's not a theory I'd heard before, but it does make some sense, and it would be great if we had more biochemical/PCR analyses of skeletons from the Italian peninsula to start tracking the ecology of diseases like leprosy and TB in ancient times.

After lunch, I spoke about Heterogeneity of Health in Imperial Rome.  Some of what I said expanded on the presentation I gave in Leiden a few months back.  Namely, we can look at percentages of disease at various cemeteries in Rome (or the peninsula or the Empire) until our eyes cross, but the ways we've been separating the population -- age-at-death, sex, geography -- aren't telling us the whole story of the diversity of the Roman population.  Heterogeneity within the population needs to be explored to a much greater degree than we are currently doing, because drawing conclusions about, for instance, malaria from simple percentages of cribra orbitalia is not capturing any of the variability within the population.  I argued that one of the easiest ways to delve further into variation within the Roman population is through isotope analysis to identify immigrants.  Many of these individuals likely varied in their health outcomes and in their susceptibility to diseases endemic to Rome.  One of the most interesting things I found while looking at my admittedly small dataset (see graph) is that porotic hyperostosis was evident only in immigrants who came from warmer, drier climates compared to Rome. Even more interesting, of those people with porotic hyperostosis, the majority of them had elevated lead levels.  So it's not so straightforward to ask whether something like anemia is the result of dietary deficiency, parasites or malaria endemic to a region, or exposure to toxins.  I'm going to attempt to refine these ideas, following helpful questions and comments from the audience, for the talk I'm giving at Cambridge University next month.

Francesca Candilio presented work she and her colleagues are doing on Assessing Roman Health through the Analysis of Dental Pathologies.  Whereas most of us in this conference were talking about one specific time period (the Empire), this presentation was more synthetic, discussing dental pathologies from the Neolithic to the Medieval periods.  As expected, carious lesion frequencies increase once people become settled agriculturalists, but the frequency actually drops after the Eneolithic and remains more or less steady through Medieval times. Interestingly, Candilio talked about how there is a huge variation in caries frequency between Lazio and Marche, a variation that gets obliterated if she simply lumps them together into the same population.  So she argued for the need to investigate heterogeneity along the lines of both geographic area and temporal differences.  The problem with wanting greater chronological control, as Kim Bowes brought up in the discussion after this paper, is that we don't have a lot of skeletal data from the countryside (it's mostly from large and medium-sized cities) or from time periods when people cremated their dead, like the Republic.  Candilio is also a proponent of open source data and hopes that at some point, all bioarchaeologists who work in Italy can agree on a set of standards and a platform for sharing data.  It would make the sort of synthetic work she's attempting a lot easier to do.

And then Oliver Craig introduced some issues with isotope analyses in Quantifying Roman Diets: Pushing the Limits of the Bone Stable Isotope Record.  Craig has worked at Velia on the C/N isotopes from that population, where he found that they had surprisingly low N values for people who lived on the coast.  In a forthcoming publication, Craig looks at isotope values from Herculaneum, specifically at the relationship between elevated d13C levels and differences in radiocarbon (14C) dates.  He and his colleagues found that some individuals at Herculaneum were up to 80-100 years off in 14C dates, likely as a result of eating more marine foods.  But where Craig really cautioned us was in the interpretation of dietary content from isotope results, that is, in the proportions of diet that come from protein and carbohydrates.  Current methods for reconstructing dietary content are insufficient, he argued, and we need to better understand human physiology, such as the synthesis and fractionation of amino acids.  So rather than measuring C and N isotopes from collagen, which is the current practice, Craig suggested it may be better to measure amino acids.  He helpfully provided a lot of references in his talk, so I need to go track those down and read them, as his presentation was incredibly interesting but also a bit too fast for me to take good notes.

Luca Bondioli spoke about skeletons from the Roman period in From Order to Chaos: Putting Morphology, Diet, Demography and Health into Perspective. His first few slides explained the importance of preservation bias in calculating frequencies of diseases.  Namely, he showed that as the percentage of bone present increased, the frequency of disease decreased.  So if you are studying a cemetery with very poor preservation, you're likely to overestimate the disease load of that population.  He also showed that explanations that work for one population don't necessarily hold for another population.  In the Velia cemetery, he and his colleagues found that males with auditory exostoses had a higher frequency of skull trauma than males without AE, suggesting males were deep-sea diving and hitting their heads on rocks in the process.  But at Isola Sacra, males without AE had a higher frequency of skull trauma, so the explanation cannot be the same.  Bondioli also discussed peaks in Wilson band and Harris line occurrence at several sites.  The common explanation for Harris lines in the tibia and linear enamel hypoplasia in the permanent dentition is physiological stress associated with weaning.  But Bondioli found peaks in Harris line frequency at 6-9 months (possibly weaning) and then again at 3 years old.  There were likely various stressful times in a Roman child's life, and we can't attribute it solely to weaning.  We need to ask more questions about what they were doing at other times in childhood; this is a segment of the population that we really don't know that much about.

To conclude the papers, Peter Garnsey presented on History Meets Anthropology: The Case of Herculaneum.  As an historian, Garnsey really led the charge to incorporate bioarchaeological data into a better understanding of Roman history, particularly in terms of diet and health.  He showed some fascinating slides on the people who died on the beach at Herculaneum and the people who were found in the boat houses.  Most interestingly, the demographics of the individual boat houses were very different in their proportions of males/females and age groups.  These demographics differed strikingly from the population found on the beach, which was mostly adult males.  A newly interpreted inscription from Herculaneum (which I heard all about in June from Luuk de Ligt at the conference in Leiden) is allowing Garnsey and his colleagues to reconstruct the demographics of the population -- for example, Garnsey suggested that one-quarter of the population were ex-slaves and 40% of the population were slaves.  Additionally, people over the age of 40 are underrepresented.  They've got a JRA article coming out soon on the inscription, and I hope we hear more about the skeletons from Herculaneum soon.

Lisa Fentress acted as the discussant for the day.  She summed up the papers, encouraged us to find a way to make our data open access, and noted that "heterogeneity is one of the fundamental questions about Rome."  Of course, I completely agree with that sentiment, and with the idea that we won't understand the bioarchaeological data if we don't share more of it.  For my part, I have published all of my isotope data in my dissertation -- every single C, N, O, and Sr value is there.  But having a place to put my stature data, for example, and my pathology frequencies would be fantastic, as would being able to access others' data for comparative purposes.  It's currently quite difficult to suss out exactly what the percentages of pathology mean -- if they're TPR or CPR (true/crude prevalence rates), if they're reported for just adults or the entire population, etc.  Greater transparency in the bioarchaeology of the Roman world is absolutely necessary going forward, and I'm happy that several of the presenters agreed on this.

All around, I had a great time in Rome.  I talked to old friends, met new potential colleagues, and got some ideas for future projects.  I am thankful that Kim Bowes invited me to be a part of this fascinating conference.


andrea said…
The numbers of slaves and percentage of non natural population in Italy at the end of republic it has been always questioned. My first reading was Tenney Frank's History of Rome and his essay on racial mixture in Rome. Well I have always thought that reputed American historian conclusions were more influenced by his concern for american racial mixture than discovering real truth on Rome yet this topic is still one of full interest to me. Very interesting post.

Popular Posts