It is high time for a book-length epigraphic study of freedwomen in the Roman world. Tatjana Sandon’s monograph offers just that: an explicit focus on freedwomen through the epigraphic sources that make up the bulk of the evidence for this population group. With that, the book is a fitting addition to the Edinburgh studies in ancient slavery series. The results of the analysis provide important nuances to the current consensus on the position of Roman freedwomen in society.
Roman freedwomen were well served by Matthew Perry’s important monograph in 2014, but until now a systematic study based on epigraphic evidence did not exist.[1] Sandon has collected around 10,000 mostly funerary inscriptions in Latin that were set up by and for women with explicit mention of freed status. Thanks to the large dataset that underlies the monograph, Sandon’s epigraphic study successfully demystifies several persistent assumptions. Most importantly, perhaps, she demonstrates that manumission for the sake of marriage was not the main road to freedom for enslaved women, thereby providing a new starting point for investigations into other possible strategies available to enslaved women.[2] The numbers presented by Sandon also make a case against overstating Roman interest in the reproductive role of enslaved women.[3] In a similar way, Sandon indicates that imperial freedwomen generally married within the imperial household rather than being snatched away as interesting marriage partners for freeborn men.[4]
The 10,000 Latin inscriptions underlying this study are unevenly distributed over the Latin West. As is usually the case, roughly half the inscriptions originate from Rome and another 37% comes from the other regions of Roman Italy. In most samples for individual chapters, this bias is skewed even more towards Rome and away from the provinces.
The book itself is divided into two parts and is structured thematically: part one (chapters 1–4) focuses on the private life of freedwomen, part two (chapters 5–7) shifts to discuss their public presence. Thematic chapters build on a particular epigraphic subset, catalogued at the end of each chapter. Thus, chapter one starts from 1,900 inscriptions that record age at death for freedwomen. Chapters two and three analyse 1606 epitaphs indicating marriages, chapter four is based on 981 inscriptions with family relations. Part two, conversely, uses only 238 inscriptions in total. Most of the 10,000 inscriptions of the underlying dataset are therefore not discussed at all in the book beyond their geographic distribution (maps 1.1. and 1.2).
Chapters 1–4 deal with the private lives of freedwomen in the context of their personal family and familia, by charting the life cycle of freedwomen, marriage age and age at death (AAD), the identity of their spouses, and motherhood respectively. Sandon argues that many freedwomen conformed to the general life course of Roman women through marriage, childbirth and death. ‘Standardized formulas’ like feminae optimae and fidelissimae uxores became ‘an important personal achievement’ for freedwomen, more so than for freeborn women (105, quotation 215). Such a life was a realistic option for many libertae, Sandon suggests, because most women for whom AAD was recorded were manumitted at a young age, often well before the age of 30 (see the graphs on p. 25). That made marriage with children a viable option. The minimum age for manumission of 30 as stipulated by the Lex Aelia Sentia of 4 CE thus seems to have had little influence on freedwomen. With findings like these, a little more discussion is warranted. Manumission in Roman Egypt, for example, generally occurred later in the life of women.[5] How and why are the findings from Sandon’s epigraphic dataset different? As she notes, freedom ‘could be obtained at any age’ (28, note 20). Indeed, Sandon does not consider the reasons why AAD might have been included in the first place, e.g. in the case of premature deaths, death in childbirth or thwarted expectations. As with freedmen, moreover, it is possible that AAD was recorded more often when manumission did not conform to the Lex Aelia Sentia.
Over the chapters of part 1 hovers the question of choice and agency: to what extent did freedwomen have a choice in marriage and family formation?[6] Although Sandon acknowledges the issue, in her view marriage seems to be largely instrumental as a way for women to partake in the ‘typical life of a Roman woman’ (213), and to become ‘a proper Roman matrona and mater’ (123). An important finding about freedwomen’s marriage partners that Sandon presents here, however, is that they were enslaved, freed as well as freeborn men, generally from a different familia. Clearly, freedwomen’s legal status was deemed less important than their person, and what is more: only in select instances was a husband also a woman’s patron.
Chapter three consists of an analysis of the terminology used in spousal commemoration in the epitaphs. This terminology is as clear in its meaning (uxor, coniunx, contubernalis, etc. all indicate a life partner) as it is imprecise from a legal perspective. Where in legal sources the term coniunx has the implication of a matrimonium iustum (legally valid marriage), Sandon shows that it is used far more indiscriminately in epitaphs. Coniunx may well refer to men and women who were stricto sensu ineligible to the title because they lacked the legal right to marry. Sandon appears to suggest that its ornamental use in inscriptions regularly indicates the exact opposite, that is, a legally invalid marriage (93–94).
The analysis of the various terminologies here gets somewhat confusing. In this chapter, Sandon pays little attention to the fact that not all newly freed women automatically gained the ius connubii (the legal right to marry) or, alternatively, that their partners may not have had the ius. This would be the case for Iunian latins or partnerships with mixed legal statuses. Focusing instead on choice of words by dedicatees, the blanket conclusion seems to be that all terminology creates ‘a very positive picture’ of the women concerned (108). When double partnerships are recorded, they are not recognized in this volume as the potentially polyandrous relationships they are.[7]
In chapter four, Sandon considers motherhood as part of freedwomen’s desired life as a free Roman matron. It is therefore remarkable that Sandon finds the attested presence of only one or two children at most in freedwomen’s tomb inscriptions (126). About half of these were freeborn. One of the explanations for this finding is that slave-born children remained in the slaveholders’ household and may not have been included on the tombstones of their mothers. When they were included on the tombstone, Sandon points to the possibility of different economic strategies within the slaveholders’ household depending on household size. Childbirth was dangerous and detracted from working hours, and particularly in smaller or medium-sized households it was not always opportune to encourage pregnancies. This is an important addition to the debate on enslaved women’s reproductive roles, since the inscriptions seem to suggest that reproductive coercion was not the norm.[8] Sandon does not dwell upon what agency the women themselves might have had in such matters, however, nor does she relate her findings to the legal benefits that came with having freeborn children, the ius liberorum.[9]
Freedwomen’s families recorded on tombstones do regularly include a partner and child(ren). Accordingly, Sandon suggests a Roman preference for the nuclear family form for ‘all social and economic groups, including Rome’s enslaved population’ (127–128), in line with Richard Saller and Brent Shaw’s influential 1984 paper – without however taking into account the problems with this interpretation and, thus, the resulting implications for understanding freedwomen and their families.[10]
The shorter second part of the book is concerned with the public lives of Roman freedwomen.[11] Sandon deals with occupational titles in chapter 5, religious activities in chapter 6 and public honours in chapter 7 respectively, but the evidential basis for this part of the book is considerably smaller (N= 238).
Sandon acknowledges the presence of domestic work in addition to occupations outside the home (138). I fully concur with Sandon’s conclusion that freedwomen were fully integrated into the Roman economy, with no identifiable legal or gender barriers. More generally, the second part of the book shows that freedwomen were integrated into society and economy not just through unremunerated and remunerated work, but also through religious functions and civic benefactions. By the very examples Sandon provides throughout part 2, therefore, it does not follow that religion was ‘a good, and sometimes the only, way for individuals with a servile past to gain a recognized social position, especially women’ (181). Other ways are also apparent.
The main text is supplemented by maps, tables and graphs to keep track of the data. The body of the text is followed by an Appendix Graphica and an Appendix Epigraphica. The former provides detailed and helpful tables to accompany chapters one, five and six. It details inscriptions with ‘geographical datum’, ‘occupational titles’, and ‘deities attested with freedwomen’ respectively. The Appendix Epigraphica gives a selection of inscriptions referred to in the main body of the text, juxtaposing Latin text and English translation by the author.
Sandon’s monograph does an excellent job of demonstrating why we need more substantial datasets like these, and of what can be done with large numbers of inscriptions. In this case, the study leads to important additions to our understanding of the lives of Roman freedwomen. At the same time, it is also an illustration of the problems that come with such an analysis and how large numbers and general patterns actually take away from Sandon’s excellent conclusion, that ‘we must acknowledge a thousand different lives’ (213, her emphasis). The book would have benefited from more reflection on the implications of its findings, and whether additional evidence confirms the analysis or not. That challenge therefore remains to be taken up.
Notes
[1] Matthew J. Perry, Gender, Manumission, and the Roman Freedwoman. Cambridge University Press 2014.
[2] Chapter 2; Contra Perry 2014; see e.g. Lauren Caldwell, “From Household to Workshop: Women, Weaving, and the Peculium,” In: B. Longfellow and M. Swetnam-Burland eds., Women’s Lives, Women’s Voices: Roman Material Culture and Female Agency in the Bay of Naples, University of Texas Press 2021, 51–66.
[3] Chapter 4; cf Matthew J. Perry, “Control of Roman Slave Sexuality: Authority, Profit, and Resistance.” In: D. Kamen and C. W. Marshall eds., Slavery and Sexuality in Classical Antiquity, Madison 2021, 254–270.
[4] Chapter 2; Ingenui make up only 1.7% of imperial freedwomen’s partners, pp. 54, 67.
[5] Roger S. Bagnall and Bruce W. Frier, The Demography of Roman Egypt. Cambridge University Press, 2006, 71: ‘female slaves often remain slaves into their 40s’.
[6] Sandon discusses the well-known case of Acte (CIL 6. 20905) extensively on pp. 60–63, for which I recommend the fuller account of Katharine P. D. Huemoeller, “Freedom in Marriage? Manumission for Marriage in the Roman World.” JRS 110 (2020): 123–39.
[7] For the clearest discussion to date of these partnerships, see Carmen María Ruiz-Vivas, “Familial Lives in Concordia: Gender Roles, Social Dynamics and Emotions in Ancient Rome.” The History of the Family 30.3 (2025): 1–28.
[8] But see now Katharine P.D. Huemoeller, The Child Follows the Womb. Gender, Reproduction, and Roman Slavery. Yale University Press, 2026.
[9] Sheena Finnigan suggests that freedwomen’s occupations and the number of children born to them point to active strategies on the part of freedwomen towards having children or preventing pregnancy, “Reconsidering the Relationship between Motherhood, Non-motherhood, and Work among Urban Roman Freedwomen,” ICS 49.1–2 (2024): 114–141.
[10] Richard P. Saller and Brent D. Shaw, “Tombstones and Roman Family Relations in the Principate: Civilians, Soldiers and Slaves”, JRS 74 (1984): 124-156. Criticism by Dale B. Martin, “The Construction of the Ancient Family. Methodological Considerations.” JRS 86 (1996): 39–60 and Sabine R. Hübner, “Household Composition in the Ancient Mediterranean – What Do We Really Know?” In: B. Rawson (ed.), A Companion to Families in the Greek and Roman Worlds, Wiley-Blackwell, 2011, 73–91.
[11] In doing so, it ties in with the work of Emily Hemelrijk: Hidden Lives, Public Personae. Women and Civic Life in the Roman West. Oxford University Press, 2015.