September 15, 2011

Foreign Women in Imperial Rome: the Isotopic Evidence

Just a short time ago, I had a paper at the European Association of Archaeologists meeting in Oslo.  I unfortunately couldn't attend the conference, so Rob Tykot presented it.  The paper was fun to write, though, and lays out the bioarchaeological evidence (albeit sparse at the moment) for women who immigrated to Imperial Rome.  Following is the complete presentation.  Comments are always welcome!

Foreign women in Imperial Rome: the isotopic evidence

K. Killgrove, Vanderbilt University
R. Tykot, University of South Florida
J. Montgomery, Durham University

A significant amount of classical scholarship over the years has been dedicated to understanding the demographic make-up of the population of Imperial Rome.  Without a proper census, however, classical demographers lack several key pieces of information necessary for reconstructing the number of citizens, slaves, and foreigners at Rome (Noy 2000:16). 

Tombstones provide the most solid evidence of immigrants who died in Rome.  Here we have an example of the inscription on a tombstone of a soldier, noting he was from Noricum (Corpus Inscriptionum Latinarum vi 3225, translated in Noy 2011). For the most part, though, the epigraphic habit was largely the province of the wealthy, educated elite, leaving us with little information about the lower classes.  Demographic estimates of foreigners at Rome range from 5% to 35%, suggesting that as many as one out of every three people in Rome arrived there from elsewhere. Below is the inscription from a large tomb that a group of freed slaves built in Rome (Année Epigraphique 1972, 14, translated in Noy 2011).  They all appear to have belonged to the same household (as they share a name and the designation C.L., “freed slave of Gaius”) yet came to Rome from various places: Greece, Asia Minor, and north Africa. The practice of commemorating one’s homeland is rare, though, and it is unclear how many slaves and free immigrants came from Italy or from further afield in the Empire (Morley 1996, p. 39).  Finally, the epigraphical record of immigrants to Rome is gender-biased, as the vast majority of inscriptions that mention immigrants are those of males (Noy 2000, p. 60). Part of this bias is attributable to the commemoration of soldiers, but males outnumber females three to one even in civilian immigrant inscriptions (Noy 2000, p. 61, Table 2). 

In order to learn more about female immigrants to the Imperial capital, we undertook a bioarchaeological study of human skeletal remains from Rome.  Through a combination of isotope analyses, palaeopathology, and burial style, we identified previously unknown female immigrants in the archaeological record of Rome and were able to reconstruct key aspects of their life histories.

Our skeletal material comes from the cemetery of Casal Bertone, which was located just 2 km from the center of Rome and was in use from the 2nd-3rd centuries AD (Musco et al. 2008). The majority of the graves were located within a simple necropolis, which included unmarked pit burials as well as burials a cappuccina.  An above-ground mausoleum that slightly postdates the necropolis was found as well, and it may have held people of higher social status.  Out of the 138 burials, we chose a stratified sample of 30 adults to subject to strontium, oxygen, carbon, and nitrogen isotope analyses – 19 males and 11 females.

This graph shows the strontium and oxygen isotope results for the first molars of adults from Casal Bertone.  The approximate isotope range of Rome is represented by a box comprising the upper and lower bounds of expected Sr and O values.  No other Sr studies in the Italian peninsula have been done on human skeletal remains, so the local range was estimated conservatively using geochemical modeling that took into account the fact that Rome was supplied by aqueducts that drew water from sources with distinctly different geology than is found in the volcanic Alban Hills (Killgrove 2010a, 2010b).  By combining Sr with an O range from previously published human skeletal data (Prowse et al. 2007), however, it is easier to see nonlocals.  Here, females T82A and T39 are fairly clearly immigrants to Rome because of low/high O isotopes and rather low Sr.  T42, on the other hand, is a borderline case since measurement error could put her within the local O range for Rome.  Clearly, though, isotope analysis of human skeletal remains is a viable way to identify female immigrants in the bioarchaeological record of Rome, particularly those who were not commemorated as such on tombstones.

Three of the 11 females we tested (27%) were probably immigrants to Rome.  Out of the 19 males studied, 6 were immigrants (32%).  More interesting, though, is the sex ratio in the immigrant population.  Whereas the sex ratio in tombstones that commemorate immigrants at Rome is 78% male versus 22% female, the ratio of immigrants discovered through skeletal evidence is 66% male versus 33% female.  This is, granted, a small sample but suggests that the bias towards male immigrants may in the future be rectified by studying skeletal data.

Epigraphy does occasionally tell us a little about the lives of female immigrants.  The tombstone of freedwoman Valeria Lycisca, for example, specifically notes that she came to Rome at age 12 (Corpus Inscriptionum Latinarum vi 28228).  Isotope analysis of the skeleton can give us similar information, in that it can help narrow the window of time in which a person immigrated. Two of the Casal Bertone female immigrants – T39 and T82A – also had third molars that could be subjected to Sr isotope analysis.  Both produced M3 Sr values that were very close to their M1 values.  The difference between T39’s first and third molars is .00016, and the difference between T82A’s first and third molars is .00017.  Their M3 values still place them towards the low end of the calculated Sr range of Rome.  People in the low end were probably immigrants from an area with younger geology (such as the southern, volcanic areas of Italy); however, it is possible people in this range were locals who consumed a significant amount of Roman aqueduct water (roughly 90% of all water consumed) throughout childhood.  Oxygen isotope analysis on the M3s has not yet been done.  Based on the small differences between these women’s M1s and M3s, it is likely that both immigrated to Rome after the development of their M3s was complete.  Therefore, T39, a woman of about 15-17 years at the time of her death, likely died shortly after arrival in Rome.

Not only can isotope analysis help us find immigrants to Rome and the date at which they immigrated, it can help us learn more about their lives before and after migration.  This graph shows the carbon and nitrogen isotope values measured from the femora of women buried at Casal Bertone.  There is a surprising gap between two clusters of data – representing a C3- and terrestrial-plant-based diet towards the bottom left and a diet slightly higher in C4 plants and aquatic resources towards the upper right – but the three immigrant women do not show evidence of having a non-Roman diet.  They were likely consuming a diet similar to that of local women (Killgrove & Tykot, n.d.).

We also looked at the change in use of carbon-based resources through a comparison of enamel and bone apatite values, which represent the ages of 0-3 and the last few years before death, respectively.  T42 and T82A changed very little about their diet between childhood and their deaths at 30-40 years old, at least in terms of the carbohydrate portion.  Surprisingly, some of the local women dramatically changed their diets – F1A and F6E, for example, consumed more C3 resources as adults.  The average d13Cap value for locals is -12.5 permil with a stdev of .5, which means that T39 had an enamel apatite value more than three stdev higher than the local average.  Her bone apatite value, however, is only 1.5 stdev higher than the local average.  You’ll remember that, based on age-at-death and a third molar Sr value that was likely nonlocal, T39 was probably a fairly recent immigrant to Rome before her death at the age of 15-17.  We are cautiously interpreting the d13Cap data as showing that T39 had started acculturating to a Roman-style diet from a childhood diet that was higher in C4 resources, but may have died before bone turnover could fully demonstrate this.

Although T39 died young, her skeleton betrays no major illness as the proximate cause of death.  Her postcranial skeleton was fairly friable and incomplete, but her teeth were decent, with just one carious lesion and minor calculus.  Her fairly high O isotope signature suggests she was from an area with warmer, drier climate than Rome; her low Sr suggests young, volcanic geology.  Both of these are consistent with an origin in the southern part of the Italian peninsula, but she could have been from a number of places in the southern Mediterranean.  Upon arriving at Rome, she changed her diet, likely out of necessity rather than by choice.  Similarly, T42 had no evidence of skeletal pathology and only minor dental disease.  Her Sr and O values are less clearly non-local, and her C and N values are in line with a Roman diet, so she may have come from an area not too distant from Rome.

T82A, on the other hand, had a much lower O isotope signature than the local range, suggesting her origin was somewhere with a cooler, drier climate than Rome.  The low Sr value could have been from volcanic geology south of Rome or could reflect travertine in the central part of the peninsula.  Together, the Sr and O suggest this woman may have been from the Apennines or geologically similar area.  She was average in height, and palaeopathological analysis revealed she didn’t perform a lot of physical labor in her life, nor was she chronically ill.  She consumed wheat/barley as a child and as an adult, but in the later years of her life, she ate less fish and more terrestrial plants than other Romans.  For such a healthy skeleton, T82A has very poor dental health.  Carious lesions in her molars and premolars could indicate that she was eating sticky, sugary foods (perhaps dried fruits), and rampant periodontal disease indicates a probable inflammation of the gums (gingivitis) during life.  Both conditions, if left untreated, can lead to chronic halitosis. This woman had a relatively unstressed life as judged from her skeleton.  Further, unlike most of the people in the cemetery, T82A was buried with grave goods; excavators found a bronze ring and 5 hair pins arranged around her skull.  The poor dental health indicative of consumption of sweet, sticky food combined with the presence of grave goods suggests this woman was of higher social status than others buried at Casal Bertone. Regardless, she was buried in a simple, unmarked grave that did not, on the surface, differentiate her from others in the necropolis.

The question remains, why did these women immigrate to Rome?  Anthropological theories on ancient migration suggest four main types of physical mobility: local, circular, chain, and career (Tilly 1978, pp. 51-4).  Local immigrants to Rome would have come from other nearby areas of the Italian peninsula, perhaps to find a job or a marriage partner.  Circular migration, where people migrate then return to their homeland, also occurred in the form of students, travelers, soldiers, and itinerant occupations like poet and shepherd.  Chain migration, where migrants at the destination encourage additional people to immigrate, was also prevalent in Rome, particularly with the diaspora of Jews and the rise of Christianity.  Career migration – moving to Rome to practice medicine or to teach – was another way that foreigners would end up at Rome.  Finally, slavery brought millions of people to Rome during the course of the Empire.  Most of the voluntary forms of migration, though, were not available to the majority of women in the Empire.  There is little evidence of family migration, except in the case of soldiers, whose wives and sometimes mothers would accompany them on campaign (Noy 2000).  And there is only one extant tombstone mentioning a foreign woman at Rome with a secular job: Coelia Mascellina, a wine and oil merchant from Baetica (Spain) (Année Epigraphique (1973) 71). In short, there is no indication in the archaeological record of Casal Bertone why these three women immigrated to Rome.  Based on their demographics and possible homelands, T82A and T42 may have engaged in local migration, whereas T39 likely came from further away.  As a recent arrival at Rome during her teenage years, perhaps T39 came as a new wife, or perhaps she came as a domestic slave.  The wealth of bioarchaeological evidence on these female immigrants to Rome makes it tantalizing to consider their motivations to migrate, but with so little known about the structure of slavery or the number of slaves at Rome, it isn’t yet possible to completely understand the pivotal moment of migration within their life histories.

Immigrant Roman women comprise one of the most understudied populations in the ancient world, with sparse evidence of their existence and their daily lives coming from tombstones and other written accounts.  As part of a larger project on migration to Imperial Rome, we found bioarchaeological evidence of female immigrants through isotope analysis.  Contrary to assumptions that migration is age- and gender-selective, chemical analysis of skeletons showed that immigration to Rome was not the exclusive domain of men; however, the reasons for and structure of migration is much better understood for Roman males because of historical and epigraphical biases in their favor. This project was undertaken on a small skeletal sample, but we hope we have shown that there is great potential in taking a bioarchaeological approach to understanding the lives of foreign women in Imperial Rome.

Acknowledgments:  P. Catalano, Soprintendenza Speciale per i Beni Archeologici di Roma; D. Coleman and J. Inglis, Department of Geological Sciences, UNC Chapel Hill; NSF (BCS-0622452), Wenner-Gren, and the Research Labs of Archaeology for funding to KK.


K. Killgrove (2010a).  Migration and mobility in Imperial Rome.  PhD dissertation, UNC Chapel Hill. [PDF]
K. Killgrove (2010b). Identifying immigrants to Imperial Rome using strontium isotope analysis. In Roman Diasporas: Archaeological Approaches to Mobility and Diversity in the Roman Empire, ed. H. Eckardt. Journal of Roman Archaeology, Suppl 78, Ch. 9. [PDF]
K. Killgrove and R. Tykot. n.d.  Investigating diets of the lower classes of Imperial Rome through carbon and nitrogen isotope analyses. Unpublished manuscript.
S. Musco et al. (2008).  Le complexe archéologique de Casal Bertone. Les Dossiers d’Archéologie 330 (Nov/Dec):32-39. [PDF]
N. Morley (1996).  Metropolis and hinterland: the city of Rome and the Italian economy.  Cambridge University Press.
D. Noy (2000).  Foreigners at Rome.  Duckworth.
D. Noy (2010). Epigraphic evidence for immigrants at Rome and in Roman Britain. In Roman Diasporas: Archaeological Approaches to Mobility and Diversity in the Roman Empire, ed. H. Eckardt. Journal of Roman Archaeology, Suppl 78, Ch. 1.
T.L. Prowse, H.P. Schwarcz, P. Garnsey, M. Knyf, R. Macchiarelli, & L. Bondioli (2007). Isotopic evidence for age-related immigration to imperial Rome. American Journal of Physical Anthropology, 132 (4), 510-9. PMID: 17205550.
C. Tilly (1978).  The historical study of vital processes.  In Historical Studies of Changing Fertility, ed. C. Tilly.  Princeton, pp. 3-56.
This post was chosen as an Editor's Selection for


Twitter Delicious Facebook Digg Stumbleupon Favorites More

Design by Free WordPress Themes | Bloggerized by Lasantha