Everyone takes risks, avoids risks, or is exposed to risks in daily life. This is a truism. Yet living through a global pandemic has made questions about how people conceive risk more pressing than ever. Why has trust in experts eroded so significantly? What made masks so politically charged? How is individual choice conceptualized in relation to public health? Choice, of course, is only part of the equation, not least because the very freedom to choose is so unequally distributed.

Our current moment is both unprecedented and eerily familiar, even for a Roman historian. And I am not referring to the parallels most often elaborated between the fall of the Roman republic (or contemptable Roman emperors) and present politics, apt though they may be. Rather, I have been struck by similarities with how Romans managed uncertainty. Using some material from my current project, this essay explores how they saw a world full of (unequal) risk and perceived interconnections and expertise in the face of it. What follows may offer up more questions than answers, but I will end with some ideas about how ancient worldviews can enrich our understanding of challenges we face today.

Perhaps you will think “risk” an unlikely topic for a historian of ancient cultures, but bear with me (or read Esther Eidinow’s work). I am currently writing a book that addresses how Romans understood and coped with the dangers of childbearing, a set of practices with some noteworthy similarities to the management of individual and communal risk today. The population of the Roman Empire (like premodern societies in general) experienced low life expectancy at birth and high rates of fertility. I want to know how this demographic pattern framed Romans’ conceptualization of the dangers of pregnancy and childbirth, and how they protected themselves and their loved ones. Broadly conceived, the project aims at a new understanding of ancient perceptions of uncertainty—of fraught, risky futures—in part as a mirror for our own, ongoing struggles with the same.

We can open up this world through an obvious, dire threat: death in childbirth. In terms of modern demography, this involves calculating a maternal mortality ratio (MMR), measuring maternal deaths per 1,000 or 100,000 live births, usually over a year. For the Roman population, precise data eludes us, but ballpark estimates suggest a range between 0.5 percent and 2 percent, or 5-20 maternal deaths per 1,000 live births.1To put this in perspective, I recommend even a cursory glance at global MMR data, which shows that maternal mortality burdens in this range continue to affect communities today. Of course, such an estimate traffics in broad generalizations. At its height, the Roman Empire encompassed the entire Mediterranean region, with vastly different climates, cultures, and population densities. Furthermore, this range derives from premodern comparative evidence (more on that below). All caveats aside, 2 percent ­is a devastating number for any population, in any period.

I am most interested in what these estimates teach us—and what they potentially obscure—about the social impact of such loss. How is a community shaped by a maternal mortality rate of 2 percent? What are the social and emotional stakes? How did Romans themselves (who thought in very different terms) explain why some women died and others survived? Or why certain healing strategies worked? Their universe pulsed with hidden forces, understood in one influential model as sympathetic connections immanent to bodies and the environment. (I suggest Brooke Holmes’s interview and from a different angle, Nandini Pandey’s recent article.) Such forces might be harnessed to help or hurt bodies. I deliberately avoid the term “magical” here, since sympathetic forces could operate in any branch of the inquiry into nature (including what we would call philosophy and science). What kind of picture emerges when we put these pieces together?

Let us begin at the level of communities. Here, comparative, premodern demographic evidence is especially helpful. For example, a model based on preindustrial English parish records suggests an MMR of just over 1 percent (on average) for women ages 15-49. In a hypothetical village of 1,000 people, with approximately 250 women in this age bracket, the community would witness one woman die in childbirth every three or so years.2This hypothetical and the parish data come from R. Schofield, “Did the Mothers Really Die? Three Centuries of Maternal Mortality in ‘The World We Have Lost,’” in The World We Have Gained: Histories of Population and Social Structure; Essays Presented to Peter Laslett on His Seventieth Birthday, ed. Lloyd Bonfield, Richard Smith, and Keith Wrightson (New York, NY: Blackwell, 1986), 258–59. Even if she was not a personal friend or family member, you would probably feel the ripple effects of grief: children missing a mother, parents losing a beloved daughter, a husband losing his wife. In a Roman context, we might imagine news of a woman’s death traveling through letters and word of mouth—in social spaces or at public events (e.g., market squares, religious festivals)—but also through channels less visible to the historian, for example, among enslaved people with friends or kin in different households.

For most women who died in childbirth, there would be some sort of funeral, burial, and time of mourning—a period of contagious sadness in a household, neighborhood, or village. There was no other reality Romans could imagine: this was the stuff of tragedy but it was also part of life. Read a few Roman tombstones and you will begin to grasp the variety of their expressions of grief. Occasionally, sparse biographical details compel us to imagine lives we glimpse all too rarely, especially those shaped by the conditions of bondage. From one tombstone, we meet Rhanis Sulpicia, formerly enslaved by a man named Trio; she died while giving birth, barely sixteen years old. The epitaph, in elegiac verse, tells us that the fetus died with her: “Rhanis, born so recently, never before subjected to childbirth. . . Now one heap of ashes holds a double funeral” (CIL 14, 2737 = CLE 1297).

Specialists’ expertise, the gods’ power, and immanent forces in the environment were available to people wishing to protect themselves and their loved ones. Medical knowledge was diffuse and contested, religious authority (textual or otherwise) decentralized—no one branch of knowledge production had a monopoly on truth. Instead, experts vied with one another in a marketplace. Competitors might emphasize their differences to shore up their own authority, all while sharing similar worldviews. If one had the means, healing and protection could entail a combination of ostensibly different tools—you could, for example, consult an oracle through an itinerant wizard, leave a votive at a local shrine, undertake a medical procedure by a learned physician (their tools might have resembled these), borrow amulets from family or friends. Risk management of this sort was material, ephemeral, and ubiquitous, access highly variable, based on location and means.

While pregnant, a wealthy woman might find herself subjected to a stringent regimen and diet to ensure she produced the healthiest, fittest babies. This could include massage, exercise, and carefully calibrated intake of specific foods, overseen by members of her household and perhaps a hired midwife. Such treatment prefigures the rigorous surveillance and regulation to which pregnant bodies are subjected today—now a process that extends long before conception, as Miranda Waggoner has demonstrated. I have argued elsewhere that ancient regimen and dietetics, as such, required enslaved labor and access to resources only available to elites, an observation that carries with it proto-eugenic implications. What of all the children born to women without access to these resources? Their children would be inherently “lesser” in body and soul, based on this devastating logic.

While regimen and dietetics were the province of elites, other tools for (self-)care, such as amulets, persisted in a wider range of social contexts. In fact, we have detailed accounts of how such technologies might be used—or at least how some authors thought they should be. “Quick-birther” amulets, for example, meant to expedite difficult labor, appear throughout ancient sources, including one from an understudied text called the Herbarius (c. third/fourth century), made from coriander seeds (103.2, CML 4.185):

So that a woman may give birth quickly (mulier ut cito pariat):

Coriander seed: 11 or 13 grains in a clean little linen cloth bound with string; a boy (puer) or a virgin girl (puella virgo), standing at her [i.e., a woman in labor] left thigh, should hold it near her groin, and as soon as the birth is over, one must remove the remedy quickly, lest her entrails follow.3The boy or girl holding the seeds is probably meant to evoke the child who will emerge, and the emphasis on female sexual purity may signal the newborn’s pure state at birth.

On one level, we can see how a community of sorts—both human and nonhuman—must act together to make the amulet work. Dig a little deeper, however, and we can begin to sense how even a humble amulet of this sort might rearticulate social hierarchy. Consider: whose labor, whose bodies are coopted in the service of others’ wellbeing? The amulet might have been prepared by a member of the household, who perhaps visited the kitchen garden to collect the seeds. In a more affluent home, this garden would have been tended by enslaved people. After acquiring the seeds, a child must be found to stand still through a long labor. Who is this naked child holding the amulet in the recipe? The child, too, might well have been enslaved. And what about the woman in labor, with an amulet held to her groin perhaps for hours on end? Did she have a choice?

Literary sources suggest that Roman birthing rooms could be crowded, filled with kin, experts, and attendants. These were largely female spaces, but under certain circumstances—when things went terribly wrong—male doctors might enter the scene, bringing their tools with them. Ann Ellis Hanson was the first to observe this pattern.4Ann Ellis Hanson, “A Division of Labor: Roles for Men in Greek and Roman Births,” Thamyris 1, no. 2 (1994): 157–202. Soranus (a Greek physician) and the Latin author Celsus both discuss surgical fetal extraction when a woman’s life is in danger, with male doctors taking over from midwives.5Soranus, Gynecology 4.7.1 (Ilberg); Celsus, On Medicine 7.29. Human flesh-boundaries were highly fraught, and ultimately the patriarchal authority of a male physician was deemed necessary to perform such interventions. In other words, the antidote to this severe form of risk was seen to require a directed form of male expertise, which was, in a sense, also a directed form of violence.

I often find myself asking if we perceive and respond to risks in ways that bring us closer to Romans than most would admit. We live in a time of contested expertise, when the most powerful individuals contradict their own experts—recommending hydroxychloroquine on Twitter, telling Americans the pandemic will “fade away.” So how do people determine whom to trust, whose authority? Where is the line between a truth-teller and conspiracy theorist? Between panacea and poison?

Maybe we are also too quick to separate ancient “dangers” from modern “risks.” The vision of a networked, sympathetic cosmos—shaped by interconnections that bind human and nonhuman bodies together—is a case in point. On a micro level, this view offers one explanation for how the coriander amulet worked, how a fetus could be affected by a palmful of seeds. Such objects and their networks were buffers against risk and loss, responses to danger shaped by tradition, ideology, and affect. Within this networked world, however, hierarchy constitutes a key organizing principle. In a way, the juncture of these planes resembles a central feature of our own globalized order. Today, the world is incalculably interconnected. Entanglements abound on the horizontal plane of global markets, creating new, proliferating forms of risk (a facet of “risk society”). Yet, these entanglements may reproduce—and be produced by—racist, colonialist, vertical logics. I am sensitive to the differences here; of course, Romans had no mechanisms for communication or surveillance approaching those of modern nation states or multinational corporations. But I wonder if we dwell too much on these differences at our own peril, and in the process, miss a bigger point about coercion, connection, and power.