Journal of Interdisciplinary History, LI:1 (Summer, 2020), 1–37. Cam Grey Risk and Vulnerability on the Campanian Plain: The Vesuvius Eruption of A.D. 472 The Somma- Vesuvius volcano erupted spectacularly in early November, 472. A mixture of lava, pumice, and volcanic debris was ejected into the air in a series of surges, or pulses. Pyroclastic flows, comprising a mixture of gases and volcanic rock collectively termed tephra, surged out of the crater. In the damp conditions of a late-autumn day, these flows mixed with loose, wet, soil and other debris to create a series of lahars, or volcanic mudslides, which spread through the Campanian plains, causing damage to the city of Nola, and extended 200 km2 over the plains to the volcano’s northeast. The inhabitants of Naples, meanwhile, watched nervously from the safety of the catacomb dedicated to Januarius, their patron saint (see Figure 1), as large quantities of ash were flung into the air and blown by favorable winds as far as Constantinople. The eruption came to the attention of Marcellinus Comes, a resident there. In his continuation of Eusebius’ Ecclesiastical History in the form of a chronicle covering the years 379 to 534, he wrote, “Under the Consuls Marcianus and Festus. Vesuvius, the burned mountain of Campania boiling by inner fires, spewed burned bowels obscuring the light of day and covered all the surface of Europe with fine dust. This terrible dust is remembered every year at Byzantium, the 6th day of November.” A number of much Cam Grey is Associate Professor of Classical Studies, University of Pennsylvania. He is the author of “Climate Change and Agrarian Change Between the Fourth and Sixth Centuries: Questions of Scale, Coincidence, and Causality,” in Jan Willem Drijvers and Noel Lenski (eds.), Shifting Frontiers 12: The Fifth Century (Bari, 2019), 35–48; “Landholding and Labour in the Rural Economy,” in Jonathan Arnold, Shane Bjornlie, and Kristina Sessa (eds.), A Companion to Ostrogothic Italy (Boston, 2016), 263–295; with J. Mathieu et al., “Familiarity, Repetition, and Quotidian Movement in Roman Tuscany,” Journal of Mediterranean Archaeology, 28, no. 2 (2015), 57–81. Initial research for this article was undertaken as part of a Master of Environmental Studies degree at the University of Pennsylvania, supported by a New Directions Fellowship from the Andrew Mellon Foundation. Especial gratitude is due to Yvette Bordeaux and Sally Willig, who kindly agreed to act as supervisors of that project. The author also thanks the anonymous reviewer for The Journal of Interdisciplinary History, whose many suggestions and comments strengthened the manuscript considerably. © 2020 by the Massachusetts Institute of Technology and The Journal of Interdisciplinary History, Inc., https://doi.org/10.1162/jinh_a_01518 2 | Fig. 1 C AM GR EY Januarius and Companion Martyrs, San Gennaro Catacomb, Naples (c) DEA / A DAGLI ORTI / De Agostini Editore / age footstock later sources state that the thickness of the ash that fell in Constantinople equaled “a palm’s breadth,” perhaps as much as 8 cm.1 Although the event clearly left a lasting impression upon distant observers and remained in the collective memory of communities well beyond the Italian peninsula, we possess no unmediated eyewitness account of the eruption and little in the way of textual evidence for its effects upon Campania. Notwithstanding the 1 Marcellinus Comes (ed. and trans. Brian Croke), Chronicon a. 472, 25. The San Gennaro Catacomb fresco has received relatively little scholarly attention, and dating remains controversial. For a brief account of the controversy and an assertion of a sixth-century date, see Thomas Granier, “Lieux de mémoire, lieux de culte à Naples aux Ve-Xe siècles: saint Janvier, saint Agrippin et le ‘souvenir des évêques,’” in Claude Carozzi and Huguette Taviani-Carozzi (eds.), Faire mémoire: Souvenir et commémoration au Moyen Âge (Aix-en-Provence, 1999), 77; for the controversy about the precise date of the eruption, Giuseppe Mastrolorenzo et al., “The 472 AD Pollena Eruption of Somma-Vesuvius (Italy) and Its Environmental Impact at the End of the Roman Empire,” Journal of Volcanology and Geothermal Research, CXIII (2002), 21; Richard B. Stothers and Michael R. Rampino, “Volcanic Eruptions in the Mediterranean before A.D. 630 from Written and Archaeological Sources,” Journal of Geophysical Research, LXXXVIII (1983), 6361–6362; for narratives and further references pertaining to this eruption, John Grattan, “Aspects of Armageddon: An Exploration of the Role of Volcanic Eruptions in Human History and Civilization,” Quaternary International, CLI (2006), 13–14; Mauro Antonio Di Vito et al., “Human Colonization and Volcanic Activity in the Eastern Campania Plain (Italy) between the Eneolithic and Late Roman Periods,” ibid., CCCIII (2013), 139; for Nola specifically, Claude Albore Livadie, Giuseppe Vecchio, and Giuseppe Mastrolorenzo, “Eruzioni pliniane del Somma-Vesuvio e siti archeologici dell’area nolana,” Atti del Congresso Archeologia e Volcanologia in Campanie (Naples, 1998); Livadie et al., “Effetti delle eruzioni del TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 3 notoriously thin written sources for late fifth-century Italy, it is odd that no mention of Vesuvius appears in the correspondence of Simplicius, bishop of Rome at the time of the eruption, nor in the letters of Gelasius, his successor, whose references to food shortages and pestilence prior to his episcopate cannot be correlated to this event with any confidence. Scholars have therefore been forced to reconstruct the eruption and its aftermath from deposits of ash, lava, and volcanic mud in both naturally exposed sections and excavated contexts, along with traces of human activities and structures preserved underneath those volcanic products.2 The present study begins by surveying the evidence for the eruption, before outlining a series of questions and propositions that owe their genesis to analytical principles drawn from the field of disaster studies. Its principal objective is methodological: The very tenuousness of the available evidence makes this particular event an ideal context for developing and elaborating a particular set of theoretical tools and interpretational hypotheses. The definition of disaster herein acknowledges and emphasizes the dialectical relationship between natural and societal elements, launching an exploration of the intimate connection between the rich resources of the region and the hazard that hovered at the edge of Somma-Vesuvio sul territorio di Nola,” in Tephras—chronologie et archéologie (Clermont Ferrand, 2001); for the catastrophic effects of lahars on settlements in the Campanian Plain, Giuseppe Mastrolorenzo et al., “The 472 AD Pollena Eruption of Somma-Vesuvius (Italy),” Journal of Volcanology and Geothermal Research, CXIII (2002) 19–36; with specific reference to a villa on the northern slope of Vesuvius, which is now thought to have been affected by this eruption, Annamaria Perrotta et al., “Burial of Emperor Augustus’ Villa at Somma Vesuviana (Italy) by Post-79 AD Vesuvius Eruptions and Reworked (Lahars and Stream Flow) Deposits,” ibid., CLVIII (2006), esp. 462. The reference to “a palm’s breadth” can be traced as far back as the late sixth-century Constantinopolitan author John Malalas (Chronography 553), although whether the palm is oriented flat or vertical is uncertain. Malalas is one of several sixth–eleventh-century Byzantine authors whose details about the lead-up to, and aftermath of, both the volcano and the ash-fall in Constantinople may or may not have derived from trustworthy sources. Note also the narrative of Sebastian le Nain de Tillemont, Histoire des empereurs et des autres princes qui ont régné durant les six premiers siècles de l’Église ( Venice, 1739), VI, 416–417; the detailed synthetic account of Giuseppe Rolandi, Rosalba Munno, and Immacolata Postiglione, “The A.D. 472 Eruption of the Somma Volcano,” Journal of Volcanology and Geothermal Research, CXXIX (2004), 291–319. 2 Gelasius (ed. Gilbert Pomarès, 172; 180), Adversus Andromachum 13; 23. Gelasius mentions plague and famine when arguing that suppressing the festival of the Lupercalia will not lead to an outbreak of pestilence. Some limited degree of temporal precision is found in his reference to the visit of the Emperor Anthemius to Rome, most likely in 467, and to a pestilentia that struck the city of Rome during the pontificate of Simplicius between 468 and 483. 4 | CAM GR EY consciousness for its inhabitants. This relationship serves as a foundation for developing definitions of risk and vulnerability, concepts that necessarily sit at the heart of any contribution to the field of historical disaster studies. In this article, these concepts and the relationships between them create a framework for generating new questions about the experiences of the inhabitants of Campania both before and after this disaster. The questions raised in this analysis can shed new light upon our understanding of the longerterm socio-economic, cultural, and political processes at play. THE POLLENA ERUPTION OF A.D. 472 Vesuvius is one component of the Campanian Volcanic Arc, a group of active, dormant, and extinct volcanoes centered around the Bay of Naples on the west coast of the Italian peninsula, which includes volcanoes at Roccamonfina and Campi Flegrei, and on the island of Ischia (Figure 2). The region is characterized by exceptionally fertile soils, first deposited in layers tens of meters thick by a massive eruption of Campi Flegrei 39,000 years ago. Subsequent volcanic events in the region have perpetuated that fertility, offering a powerful illustration of the intermeshed relationship between risk and reward. Archaeological data suggest that the area has been densely populated since at least the Early Bronze Age.3 The volcanological record for Vesuvius reveals a sequence of eruptions of varying intensity and periodicity punctuated by a series of especially destructive events—from the Pomice di Avellino event around 3,900 years ago through the famous one that buried 3 Studies of soil fertility emphasize the high levels of phosphorus, potassium, nitrogen, and exchangeable calcium in the region, as well as the soil’s superior water-retention properties. For the soils prior to the 79 eruption, see J. E. Foss et al., “Paleosols of the Pompeii Area,” in Wilhelmina Jashemski and Frederick G. Meyer (eds.), The Natural History of Pompeii (New York, 2002), 65–79; for the qualities of soils engendered by the eruption of 79, as well as a soil derived from the 1631 eruption, see Yadzuru Inoue et al., “Interpretation of pre-AD 472 Roman Soils from Physicochemical and Mineralogical Properties of Buried Tephric Paleosols at Somma Vesuviana Ruin, Southwest Italy,” Geoderma, CLII (2009), 243–251; for the influence of volcanic activity on the growth and characteristics of alluvial fans in the area, Giovanni Zanchetta, R. Sulpizio, and Di Vito, “The Role of Volcanic Activity and Climate in Alluvial Fan Growth at Volcanic Areas: An Example from Southern Campania (Italy),” Sedimentary Geology, CLXVIII (2004), 249–280; for the density of the population, Di Vito et al., “Human Colonization,” 132–134. See also Claudio Scarpati, Perrotta, and Girolamo Ferdinando De Simone, “Impact of Explosive Volcanic Eruptions around Vesuvius: A Story of Resilience in Roman Time,” Bulletin of Volcanology, LXXVIII (2016), available at doi:10.1007/s00445016-1017-4. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 Fig. 2 | 5 The Campanian Plain and Bay of Naples, Geological Schematic with Notable Places SOURCE Stefano Carlino, Renato Somma, Claudia Troise, and Giuseppi De Natale, “The Geothermal Exploration of Campanian Historical Volcanoes: Review and Future Development,” Renewable and Sustainable Energy Reviews, XVI (2012), 1005 (reprinted with permission from Elsevier, modified). Pompeii and Herculaneum in 79, to an eruption of 1631, which killed 3,000 residents of the Campanian plain when it ejected lava, toxic gases, and boiling water and triggered destructive and fastmoving lahars. Since that date, Vesuvius has been almost continuously active. Although the volcano last erupted in 1944 and currently appears quiescent, it continues to be closely monitored. The events of the years 79, 1631, and 1944 have been exhaustively studied, and their eruptive processes, lava flows, and ashfall patterns sketched, debated, and analyzed. Scholars have derived orders of 6 | C AM GR EY magnitude for the physical force and impacts of these eruptions, using the Volcanic Explosivity Index (VEI), a subjective, relative measure of volcanic explosiveness developed under the auspices of the United States Geological Service (USGS) in the early 1980s. This schema draws upon both quantitative evidence, such as the volume of ejecta, and the qualitative evidence provided by eyewitness accounts and subsequent volcanological and geological surveys. The eruption of 79—termed a plinian event for the Roman politician and litterateur Pliny the Younger (61–113), its observer and reporter—is generally granted a VEI number of 5. The eruption of 1631 has recently been characterized as sub-plinian I (VEI 3–4), and the event of 1944, which was monitored by a seismograph and thus more precisely, tends to be interpreted as a violent Strombolian event (VEI 2–3). In comparison with such eruptions as those of Mount St. Helens in Washington (1980, VEI 5–6), of Pinatubo, the Philippines (1991, VEI 6), or of Tambora, Indonesia (1825, VEI 7), and with reference to models and reconstructions based on excavated and observed ejecta, lava, and lahars in the Campanian landscape, the respective sizes ascribed to the Vesuvius eruptions appear to be reasonably accurate.4 4 For the eruptive history of Vesuvius, see Giuseppe Rolandi, Paola Petrosino, and J. McGeehin, “The Interplinian Activity at Somma–Vesuvius in the Last 3500 Years,” Journal of Volcanology and Geothermal Research, LXXXII (1998), 19–52; Claudia Principe et al., “Chronology of Vesuvius Activity from A.D. 79 to 1631 Based on Archeomagnetism of Lavas and Historical Sources,” Bulletin of Volcanology, LXVI (2004), 703–724; Raffaello Cioni et al., “Explosive Activity and Eruption Scenarios at Somma-Vesuvius (Italy): Towards a New Classification Scheme,” Journal of Volcanology and Geothermal Research, CLXXVIII (2008), 331–346; for a systematic attempt to determine the chronology of fifth- and sixth-century eruptions of Vesuvius, Eliodoro Savino, Campania Tardoantica (284–604 d. C.) (Bari, 2005), 316–321 (Appendix 9); for hazard assessments, Cioni et al., “Explosive Activity”; Giulio Zuccaro et al., “Impact of Explosive Eruption Scenarios at Vesuvius,” Journal of Volcanology and Geothermal Research, CLXXVIII (2008), 416–453; Rolandi, “Volcanic Hazard at Vesuvius: An Analysis for the Revision of the Current Emergency Plan,” ibid., CLXXXIX (2010), 347–362; Benedetto De Vivo and Rolandi, “Vesuvius: Volcanic Hazard and Civil Defense,” Rendiconti lincei: Scienze fisiche e naturali, XXIV (2013), 39–45; for the VEI, Christopher G. Newhall and Stephen Self, “The Volcanic Explosivity Index ( VEI): An Estimate of Explosive Magnitude for Historical Volcanism,” Journal of Geophysical Research, LXXXVII (C2) (1982), 1231–1238; for an account of the 1980 Mount St. Helens eruption with reference to the relevant literature, Vyacheslav M. Zobin, “An Overview of the Dynamics of the Volcanic Paroxysmal Explosive Activity, and Related Seismicity, at Andesitic and Dacitic Volcanoes (1960–2010),” Frontiers in Earth Science (May 3, 2018), 2–4, available at doi:10.3389/feart.2018.00046; for the magnitudes of Tambora and Pinatubo, Chris Newhall, Stephen Self, and Alan Robock, “Anticipating Future Volcanic Explosivity Index ( VEI) 7 Eruptions and their Chilling Impacts,” Geosphere, XIV (2018), 572–603. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 7 But quantification alone, even when combined with the qualitative data used in deriving values on the VEI scale, falls short of capturing the effects of eruptions upon a surrounding countryside and its inhabitants. It fails to grasp the location and speed of lava flows and lahars or the distribution of ash and tephra, which are products as much of topography, soil type, and weather conditions as of an eruption’s explosive energy. Pliny’s account of the eruption in 79, for example, lays heavy emphasis upon the contrary wind (contrarius ventus) that impeded his uncle’s efforts to help the inhabitants of Stabiae, at the southeast edge of the Bay of Naples. Maps of ashfall patterns of the various historical eruptions of Vesuvius reveal that because the prevailing winds in the region generally blow from the southwest, tephra and ash were usually distributed most heavily across the Campanian Plain, to the northeast of Vesuvius. In late August of 79, however, as Pliny records, the wind was blowing from the northwest, meaning that Pompeii, located to the southeast, bore the brunt of that eruption.5 In any event, the attention paid to the events of 79, 1631, and 1944 is due in no small part to the existence of detailed descriptions and records—the letters of Pliny, various seventeenth-century eyewitness accounts, and a combination of photographs and detailed narratives from World War II bomber pilots and other bystanders. The fact that the eruption that concerns this article does not appear in surviving contemporaneous written sources has contributed to its uneven scholarly attention. Consequently, its context, immediate impact, and lasting effects remain poorly understood. The bulk of the attention that it has received tends to focus upon reconstructing the mechanics, process, and chronology of the eruptive event itself via careful analysis of the stratigraphy of geological and archaeological sections, and of the chemical composition of their constituent parts. The eruption itself is usually regarded as sub-plinian, with a VEI comparable to the 1631 event, around 3 to 4. The fall of ash and pumice created lasting changes to the morphology of the region and appears to have disrupted drainage networks. The associated lahars likely advanced rapidly across the landscape, although the absence of bodies in archaeological 5 For the contrary wind, see Pliny, Epistulae VI.14; for ash-fall patterns, Di Vito et al., “Human Colonization,” 5; for the effect of Vesuvius’ asymmetry on the directionality and density of lava flows, Guido Ventura et al., “Geomorphological Map of the Somma-Vesuvius Volcanic Complex (Italy),” Journal of Maps, I (2005), 37. 8 | C AM GR EY contexts suggests that many of the inhabitants were able to flee, though their destination is difficult to determine. Archaeological evidence from Nola, however, reveals buildings with roofs that appear to have collapsed under the weight of ash, burying the individuals below. Later hagiographical literature also mentions groups taking refuge in the San Gennaro catacombs outside Naples.6 These glimpses of behavior in the face of catastrophic natural disturbance are tantalizing. But the precise strategies available to late Roman communities under such conditions, their recourse to patterns and structures of social support and interaction, and the experience of the residents of Campania in the extended aftermath of that event remain almost completely unexamined. Indeed, scholars have been reluctant to move beyond vague and general references to the impoverishment and vulnerability of the area as a result of years of barbarian depredation, and appeals to the obvious fact that the Western Roman Empire was destined to fall only four years later with the deposition of the Emperor Romulus Augustulus (r. 475–476) by the barbarian mercenary Odoacer (c. 433–493). Despite our full account of the natural disturbance 6 For the VEI of the 472 eruption, see Cioni et al., “Explosive Activity,” 347 (Table 1); Di Vito et al., “Human Colonization,” 139; for estimates of the ejecta, Scarpati, Perrotta, and Luongo, “The History of Eruptions on the Northern Slope of Somma-Vesuvius,” in Girolamo F. De Simone and Roger T. Macfarlane (eds.), Apolline Project. I. Studies on Vesuvius’ North Slope and the Bay of Naples (Rome, 2009), 274; for an attempt to model the tephra fallout patterns of the eruption mathematically, R. Bonasia et al., “Numerical Inversion and Analysis of Tephra Fallout Deposits from the 472 AD Sub-Plinian Eruption at Vesuvius (Italy) through a New Best-fit Procedure,” Journal of Volcanology and Geothermal Research, CLXXXIX (2010), 238–246. This study, like much of the scholarship, uses the eruption of 472 as a means of predicting future eruptions of Vesuvius. See also E. Zanella et al., “Deposition Temperature of the AD 472 Pollena Pyroclastic Density Current Deposits, Somma-Vesuvius, Italy,” Bulletin of Volcanology, LXX (2008),1237–1248. For drainage networks, see Giuliana Alessio et al., “Flood Hazard of the Somma-Vesuvius Region Based on Historical (19–20th Century) and Geomorphological Data,” Annals of Geophysics, LVI (2013), available at doi: 10.4401/ag-6440; Lina Davoli et al., “Natural and Anthropogenic Factors of Flood Hazards in the Somma-Vesuvius Area (Italy),” Géomorphologie: Relief, Processus, Environnement, 7, no. 3 (2001), 195–207; for archaeological evidence, Scarpati, Perrotta, and De Simone, “Impact of Explosive Volcanic Eruptions,” 4; Mario Pagano, “Archaeological Sites and Volcanic Activity at Vesuvius after the Great 79 A.D. Plinian Eruption,” in Massimo Cortini and De Vivo (eds.), Volcanism and Archaeology in Mediterranean Area (Trivandrum, 1997); for a more generalized, schematic account of the effects of tephra build-up on buildings and roofs in Nola and its environs, Mastrolorenzo et al., “472 AD Pollena Eruption,” 33; Cioni et al., “Explosive Activity,” 343; for debate about whether to attribute late antique volcanic damage to the 472 eruption or that of 505 or 512, Eliodoro Savino, “L’area vesuviana in età tardoantica: modalità insediative e strutture produttive,” in De Simone and Macfarlane (eds.), Studies on Vesuvius’ North Slope and the Bay of Naples, 243. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 9 that the eruption represents, the story of the disaster, or disasters, that it precipitated remains to be told.7 To be sure, our evidence for that story is patchy and limited. The elusiveness of the requisite information prompts legitimate questions about how the eruption affected the inhabitants of Campania that must be posed within the context of what is known or surmised about settlement and exploitation patterns, as well as trade and exchange networks, in the region during the fifth century: Was this eruption a disaster, and if so, for whom? The field of disaster studies offers a useful analytical framework for answers. STUDYING DISASTERS, HISTORICAL AND CONTEMPORARY What is a disaster? The question is neither rhetorical nor disingenuous, even though disaster has always occupied the collective social consciousness. Instinctively, we feel that we would certainly know one if we saw one—like the levees breaking in New Orleans in the wake of Hurricane Katrina, the scenes of devastation in Indonesia following the Aceh Tsunami, and the catastrophic earthquake that devastated not only Lisbon but also European notions of progress to their very core in 1755. Pliny’s evocative and compelling account of the Vesuvius eruption, which buried the cities of Pompeii and Herculaneum under meters of ash, tephra, and pumice, would seem to belong in the same category. These events are, arguably, recognizable disasters that might legitimately be invoked as convenient explanations or catalysts for phenomena that could otherwise seem ineffable, such as the collapse of civilizations; abrupt social, economic, or political changes; sudden reductions in population or settlement numbers, etc. However, to focus upon the moment when a hurricane hit, a tsunami engulfed the shore, an earthquake struck, or a volcano blew its top is to ignore the subtle and dynamic interplay of human and environmental factors that create, in certain circumstances, an intensely human tragedy. The field of disaster studies sits squarely at the heart of this interplay between societal and natural elements. 7 For general comments about Roman society’s “culture of risk,” see Jerry Toner, Roman Disasters (Malden, Mass., 2013), 87–107; for catastrophist narratives, Mastrolorenzo et al., “472 AD Pollena Eruption”; Pagano, “Archaeological Sites,” which assumes the area to have been abandoned due to the volcanic eruption; Takayuki Kaneko et al., “Determination of Burial Age of the ‘Augustus’ Villa (Italy),” Geochemical Journal, XXXIX (2005), 575. 10 | CAM GREY Scholars have long recognized the disruptive potential of earthquakes, volcanic eruptions, and other similar natural events, but the study of disasters in and of themselves emerged only in the decades following World War II. The earliest scholarship viewed those events as natural occurrences that exceeded the bounds of normalcy and were therefore unavoidable. However, since the 1970s, scholars have become increasingly sensitive to the complex of factors that combine to produce a disaster, focusing upon the webs of relationships between societal, environmental, and cultural elements. The emergent consensus is that a disaster entails a confluence of factors in a process that develops or unfolds over time. According to one influential definition, a disaster throws together “a potentially destructive agent/force from the natural, modified, or built environment and a population in a social or economically produced condition of vulnerability, resulting in a perceived disruption of the customary relative satisfactions of individual and social needs for physical survival, social order, and meaning.”8 For current purposes, the three important elements in this definition are the catalyzing force, the vulnerable population, and the ensuing disruption. The explanation provided for the catalyzing force is deliberately and explicitly broad, encompassing not only events that can be attributed to earth processes and climatic or weather fluctuations but also occurrences like the collapse of a building or the meltdown of a nuclear power plant. Importantly, the event itself is not a disaster; it is, simply, an agent or force that has the potential to be destructive—in other words, a hazard. In this regard, it is equivalent to other hazards in both the natural, modified, or built environment and in a community’s cultural 8 Anthony Oliver-Smith and Susanna M. Hoffman, “Introduction: Why Anthropologists Should Study Disasters,” in idem (eds.), Catastrophe and Culture: The Anthropology of Disaster (Santa Fe, 2002), 4. For scholarly trajectories in disaster studies, see Monica Juneja and Franz Mauelshagen, “Disasters and Pre-industrial Societies: Historiographic Trends and Comparative Perspectives,” Medieval History Journal, X (2007), 4–7; Gerrit Jasper Schenk, “Historical Disaster Research: State of Research, Concepts, Methods and Case Studies,” Historical Social Research, XXXII (2007), 9–31; Michael K. Lindell, “Disaster Studies,” Sociopedia.isa, available at www.sagepub.net/isa/resources/pdf/ Disaster%20Studies.pdf (accessed December 30, 2018); for a contemporary statement of the issues, Oliver-Smith, “‘What Is a Disaster?’: Anthropological Perspectives on a Persistent Question,” in idem and Hoffman (eds.), The Angry Earth: Disaster in Anthropological Perspective (New York, 1999), 29; Toner, Roman Disasters, 1–16. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 11 and economic practices, political or religious expectations, or collective willingness or capacity to evaluate risk.9 A disaster occurs where these various agents, forces, or hazards meet, namely, in a specific, vulnerable population. WilchesChaux identified eleven different forms of vulnerability—natural, physical, economic, social, political, technical, ideological, cultural, educational, ecological, and institutional. Wisner et al. define vulnerability as “the characteristics of a person or group in terms of their capacity to anticipate, cope with, resist and recover from the impact of a natural hazard.” Vulnerability, therefore, provides the necessary foundation for treatments of disasters as the result of an interplay of societal, infrastructural, and natural factors over time. When, for example, scholars describe the magnitude 7.7 earthquake in 1970 Peru as a “500-year earthquake,” they are depicting it as an event for which the proximate effects and aftermath can be understood only within the context of the cultural, social, economic, and political structures of the preceding halfmillennium.10 Vulnerability emerges in the scholarly literature as a multidimensional tool for description and analysis of the relationship between human populations and their environments. In an influential intervention in the theoretical debate, Oliver-Smith noted a common depiction of vulnerability as the space where societal and natural realms dialectically interact, thus forming “the conceptual nexus that links the relationship that people have with 9 Oliver-Smith and Hoffman define hazards as “the forces, conditions, or technologies that carry a potential for social, infrastructural, or environmental damage. A hazard can be a hurricane, earthquake, or avalanche; it can also be a nuclear facility or socioeconomic practice, such as using pesticides. The issue of hazard further incorporates the way a society perceives the danger or dangers, either environmental and/or technological, that it faces and the ways it allows the danger to enter its calculation of risk” (“Why Anthropologists Should Study Disasters,” 4). 10 Greg Bankoff, “Vulnerability as a Measure of Change in Society,” International Journal of Mass Emergencies and Disasters, XXI (2003), 5–30; Oliver-Smith, “Theorizing Vulnerability in a Globalized World: A Political Ecological Perspective,” in Bankoff, Georg Frerks, and Dorothea Hilhorst (eds.), Mapping Vulnerability: Disasters, Development and People (Sterling, Va., 2004), 10– 24; Gustavo Wilches-Chaux, Desastres, Ecologismo, y Formación Profesional (Popayán, 1989); Ben Wisner, Piers Blaikie, Terry Cannon, and Ian Davis, At Risk: Natural Hazards, People’s Vulnerability, and Disasters (New York, 2004; orig. pub. 1994), 11; Bankoff, “Time Is of the Essence: Disasters, Vulnerability and History,” International Journal of Mass Emergencies and Disasters, XXII (2004), 23–42; Oliver-Smith, “Peru’s Five-Hundred-Year Earthquake,” in idem and Hoffman, (eds.), Angry Earth, 31–48. 12 | CAM GREY their environment to social forces and institutions and the cultural values that sustain or contest them.” From that perspective, it is a connective tool “linking social and economic structures, cultural norms and values and environmental hazards (and disasters) in causative chains.” The concept does not just allow us to locate disasters at the fulcrum between nature and society; it positively demands that we do so.11 Oliver-Smith and others, however, have also leveled criticism at this construction of vulnerability on the grounds that it fails to account for the culturally specific factors that render a particular society’s relationship with the natural world unique. Extracting the consideration of vulnerability from a specific cultural context produces an etic or externally generated estimation of vulnerability derived from a modern, Western intellectual discourse that assumes ‘society” and “nature” to be distinct and largely oppositional. Scholarship on this nexus in non-Western contexts, as well as other time periods, has disputed that assumption. Indeed, many scholars now speak of the role that cultural constructions of nature play in the social production of disaster in specific contexts while strongly emphasizing the need for analytical tools that enable consideration of disasters in a broadly comparative, cross-cultural perspective. They envisage the experience of disasters in the past as encompassing both emic and etic perspectives, treating culturally, historically, or geographically specific elements of a society’s vulnerability within an interpretative framework broadly conceived and widely adopted across disciplinary, geographical, and historical boundaries. As we shall see below, the concept of vulnerability is apropos for such an undertaking.12 Conceptions of the relationship between “society” and “nature” in the late Roman world have largely remained unexplored in the scholarly literature, which also tends to bifurcate notions of “classical” and “medieval” without tracing the lines of complementarity, coincidence, or contradiction between them. That said, in a recent, thought-provoking “speculative analysis,” Macfarlane sought to unpack the cultural assumptions underpinning two 11 Oliver-Smith, “Theorizing Vulnerability,” 10–12. 12 Ibid.,” 12–14; Bankoff, Cultures of Disaster: Society and Natural Hazard in the Philippines (London, 2003), 2–3, 152–153, 182–183. For the need to produce more comparative historical disaster studies, see Juneja and Mauelshagen, “Pre-industrial Societies.” TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 13 complementary accounts of Vesuvian activity contained in the Paschale Campanum, an anonymous continuation of Victor of Aquitaine’s Easter table, beginning in 464. The earlier of the two, referring to an event of 505, records that Vesuvius “bellowed” (eructavit)—a term that suggests a personified or otherwise animated interpretation of the volcano, which other observations duplicate, as we shall see below. The second account concerns the year 512, pithily juxtaposing a solar eclipse and a small eruptive event in a chain that, in Macfarlane’s treatment, hints at assumptions about cause and effect between the physical and metaphysical worlds. Macfarlane’s approach to these accounts points toward strategies for extracting greater detail regarding the “culture of risk” that infused late fifth- and sixth-century communities in the immediate vicinity of Vesuvius. We consider that “culture of risk” below.13 For present purposes, employing vulnerability as an analytical tool for differentiating the experiences of the eruption among the inhabitants of the Campanian Plain necessarily involves recognizing the heterogeneous nature of communities and the diversity of individual expectations and experiences during a disaster. The concept of differential vulnerability permits a close investigation of the structures, practices, and patterns that underlie a community’s social fabric, and of the way that inequalities in gender, status, wealth, or religion might bring about variations in response to hazards and variable experiences of disasters. This conceptualization of vulnerability acknowledges the spatial aspects of socio-economic or political power through a population’s distribution across a landscape, as well as the regular, seasonal, or occasional influences of geographical, environmental, and climatological factors. In the process, phenomena that might, at first blush, appear to be merely descriptive assume explanatory force: Populations in certain physical locations may be especially vulnerable because of wind patterns and rainfall and/or other topographical and hydrological characteristics. Furthermore, the socio-economic networks available to more or less wealthy landowners probably affect their capacity to respond to a disaster over time. Children, the elderly, and the infirm are also likely 13 Macfarlane, “Vesuvian Narratives: Collisions and Collusions of Man and Volcano," in De Simone and idem (eds.), Studies on Vesuvius’ North Slope and the Bay of Naples, 103–121. 14 | CAM GREY to be disproportionately vulnerable, and so on. These points are all relevant to our investigation of vulnerability scenarios for different populations in the Campanian Plain, notwithstanding the scarcity of the data currently available to support or substantiate those scenarios. Vesuvius and the other volcanoes in the Campanian Volcanic Arc have had a huge influence on settlement patterns within the plain, both positively and negatively. The volcanic soils of the Campanian Plain have given the region a long reputation for fertility and productivity, which was well established by the time of the Expositio Totius Mundi, an anonymous geographical text written in Greek during the reign of Constantius II (r. 337–361), which has survived in a fifth- or sixth-century Latin translation. The author of that text described the region as “the storehouse of queen Rome (cellarium regnanti Romae).” Procopius (c. 500–c. 570)—scholar, historian, and courtier of the Emperor Justinian (r. 527–565)—provided the most detailed exegesis of the relationship between Vesuvius and the communities at its feet. He accompanied Belisarius, Justinian’s general, on his first campaign to recapture Italy from the control of its Ostrogothic rulers, at least notionally. The specific event to which he referred is generally dated to 536 and interpreted as an episode in which Vesuvius stirred and rumbled but did not erupt.14 Procopius explicitly connects volcanic activity to fertility, although not necessarily in the causative way that would satisfy a modern scholar. Nevertheless, his association of the two is suggestive: “This, however, they declare emphatically, namely that whenever Vesuvius belches forth these ashes, the country all around is bound to flourish with an abundance of all crops” VULNERABILITY AND RISK ON THE CAMPANIAN PLAIN 14 For the debate about whether Vesuvius erupted at this time or merely exhibited signs of unrest, see Stothers and Rampino, “Volcanic Eruptions,” 6362–6363; Cioni et al., “Explosive Activity,” 347 (Table 1), which offers succinct summaries of the volcano’s activity; Principe et al., “Chronology of Vesuvius Activity,” 705 (Table 1), which suggests that Vesuvius erupted on this date; for relations between the volcanoes of the Campanian Arc and human populations throughout history, Di Vito et al., “Human Colonization.” Publication of the relevant survey data remains patchy and incomplete, hindering a coherent and detailed analysis of these relationships. For a partial and not updated catalog of sites and excavation reports, see Pagano, “L’area vesuviana dopo l’eruzione del 79 d. C.,” Rivista di studi pompeiani, VII (1995), 35–44. Jean Rougé (ed.), Expositio Totius Mundi et Gentium (Paris, 1966), 54–55; Savino, Campania Tardoantica, 306–308 (Appendix 7). TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 15 (29). Against this positive connection, Procopius also cites the terror that the threat of an imminent eruption elicited among the local population, observing that an eruptive event’s consequences for people moving through the surrounding countryside could be deadly: “If anyone walking on the road is caught by this horror, he cannot possibly survive, and if it falls upon houses, they too collapse under the quantity of ashes” (25). Procopius’ anecdotal evidence finds support in the extensive archaeological and volcanological survey of the region, which reveals the effects of successive eruptions from 3900 B.P. through the 472 event and beyond. Indeed, we need only recall the burial of Pompeii and Herculaneum in 79 for evidence of this volcano’s catastrophic potential.15 The longer-term contours of settlement and exploitation in the region, however, attest to a continuing economic vitality past the fifth century, albeit with some changes in networks of exchange and the distribution of goods. Ceramic assemblages from three complementary contexts—a theater located in the city of Naples, a modest villa from the via Saccaccio on the outskirts of the town of Nola, and a villa/bath complex located at Pollena Trocchia on the northern slopes of Vesuvius—offer hints of the differences between urban and rural contexts immediately prior to the eruption of 472, most especially regarding access to imported wares. In the cities of Naples and Nola, for example, imported wares comprise around 32 percent of assemblages, whereas in the villa sites of the surrounding countryside, the proportion plummets to around 10 percent. Even the higher of these two figures represents a reduction in the quantity of imported vessels by comparison with the fourth century. Together with a marked drop in wares from the eastern Mediterranean, leading to a preponderance of African Red Slip (ARS) wares among imported vessels, this general decline in imported ceramics serves as evidence for a reduction in the economic independence of Campania, as it became increasingly enmeshed in a state-driven system of supply oriented around imports from North Africa.16 15 Procopius (ed. and trans. Anthony Kaldellis), De Bello Gothico II.4, 328; Martine Henry, “Procope en Italie: les notices sur le Vésuve,” Historia: Zeitschrift für Alte Geschichte, LVII (2008), 317–326. 16 De Simone et al., “Late Antique Connectivity: A Snapshot of Regional Trade in AD 472 Campania,” in Luca Bombardieri et al. (eds.), SOMA 2012: Identity and Connectivity: Proceedings of the 16th Symposium on Mediterranean Archaeology, Florence, Italy, 1–3 March 2012 (Oxford, 2013), 16 | CAM GREY This narrative is complicated, however, by the abundance and heterogeneity of local, Italian coarse wares, which neither imitate ARS forms—as in the preceding period—nor overlap in function with imported wares. Recent, detailed analysis of intraregional coarse wares in the area immediately surrounding the Somma– Vesuvius complex suggests that shorter-distance networks for the movement of goods remained robust. The heterogeneity of these local wares, not only in their forms but also in their fabrics, hints at the coexistence of multiple, complementary sites of ceramic production in the region. Moreover, the emergence, or new visibility, of connections with Apulia to the southeast and with sites in the Campanian Apennines to the northeast may well indicate a modest degree of interregional exchanges of goods.17 Alongside these signs of complex and complementary networks of exchange and distribution in the late fifth century are indications of economic exploitation. The archaeological evidence currently available does not permit detailed explication of the relative incidence or importance of agriculture; the raising of pigs, sheep, and goats; or the cultivation of grapes and olives. Nor does it present a straightforward narrative of the economic vitality of the region, compared to preceding periods. Nonetheless, all these activities coexisted to some degree in the area within enduring and complex agrarian regimes. Additionally, paleobotanical work has discovered that chestnut trees were harvested, most likely for firewood and construction.18 II, 972; De Simone and Caterina Serena Martucci, “Thirst for Wine? An Amphorae Assemblage from Vesuvius and the Problem of Self-Sufficiency in Late Antique Campania,” Rei Cretariae Romanae Fautorum Acta, XLIV (2016), 127–133. 17 Martucci et al., “Local Production and Trade Patterns in the Environs of Vesuvius: The Pottery from Pollena Trocchia and Nola,” in Natalia Poulou-Papadimitriou, Eleni Nodarou, and Vassilis Kilikoglou (eds.), LRCW 4: Late Roman Coarse Wares, Cooking Wares and Amphorae in the Mediterranean: Archaeology and Archaeometry: The Mediterranean: A Market Without Frontiers (Oxford, 2014), 51–61. 18 De Simone, Maria Rosaria Vairo, and Robin Jennifer Veal, “Lo sfruttamento delle risorse boschive nella Campania tardoantica: l’evidenza antracologica da due siti vesuviani,” Rivista di Studi Pompeiani, XXIV (2013), 71–78; Alessandra De Luca and De Simone, “Dati archeozoologici da Pollena Trocchia e alcune considerazioni sull’alimentazione nella Campania tardo antica,” ibid., XXIII (2012), 159–162. For chestnut cultivation, see Gaetano Di Pasquale et al., “Reworking the Idea of Chestnut (Castanea sativa Mill.) Cultivation in Roman Times: New Data from Ancient Campania,” Plant Biosystems, CXLIV (2010), 870–871. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 17 Specific information is difficult to extract from these data. We catch glimpses of the interests and management practices of large landowners in the letter collection of Symmachus (340–402), a late fourth-century senator and magistrate, but we should resist the urge to establish any detailed picture of land-ownership patterns from this source. Nonetheless, the continuing preponderance of small and medium-sized sites over larger villa sites in archaeological surveys hints at a continuing diversity and vitality in wealth and social standing among the inhabitants of the region. Imperial responsa from the first half of the fifth century offer tax relief to the inhabitants, suggesting that local elites retained the cohesion and collective identity necessary to mount a successful petition to the emperor (more below). The tangible influence of local landowners was probably due to a combination of the Campanian Plain’s economic fertility and the attraction of the Bay of Naples as an aristocratic seasonal or holiday destination, at least into the second half of the fifth century.19 The fates of the towns in the region throughout the fourth and fifth centuries continued to follow the larger patterns of economic connectivity suggested by the ceramic typologies and the aggregate of smaller-scale productive regimes of the countryside’s villas, estates, and farms. For example, Puteoli (modern Pozzuoli), west of Naples, appears to have functioned as an entrepôt for goods from Egypt, Africa, and the western Mediterranean, and the town of Suessa, northwest of Monte Massico, served as a focal point for the production and distribution of local goods and foodstuffs. The administrative reforms of Diocletian (r. 284–305) and his colleagues in the Tetrarchy elevated the town of Capua, on the northeastern edge of the Campanian Plain, to the position of provincial capital and boosted the area’s civic centers, at least in part. Furthermore, the gifts of land from Constantine (r. 306–337) to the Church created conditions favorable for the emergence of ecclesiastical sees at places such as Forum Popilii in the ager Falernus 19 For Symmachus, see Savino, Campania Tardoantica, 37–47; for continuing diversity, ibid., 46, 87–89; idem, “L’area vesuviana,” 241–242; for enduring attraction, John D’Arms, Romans on the Bay of Naples; A Social and Cultural Study of the Villas and Their Owners from 150 B.C. to A.D. 400 (Cambridge, Mass., 1970); for the fourth and fifth centuries, Savino, Campania Tardoantica, 75–79, 90, 218–224, which argues for a decline in the attractiveness of the region for the wealthy aristocracy during the fifth century; for petitions, Codex Theodosianus (CTh) XI.28.7 (413), XI.28.12 (418). 18 | CAM GREY and Atellana in the ager Campanus. Detailed pictures of population size, building activity, and civic life in towns such as Nola on the Campanian Plain or Sorrento and Stabiae on the Bay of Naples are difficult to capture with the current state of the archaeological evidence. Nonetheless, these centers appear to have endured into the late fifth century, even as their wealth, grandeur, and economic horizons waned according to the vagaries of trade networks and the aristocracy’s adoption of alternative means for expressing wealth and status, including the construction and provision of churches and shrines.20 Narratives of the economic degradation of Campania usually focus upon the Gothic and Vandalic incursions of the fifth century, as well as the effects of Belisarius’ war of reconquest in the first half of the sixth century. Although the northernmost parts of the region appear to have suffered little from attacks by Alaric, the Visigothic king, in the first half of the fifth century and by Vandal raiders in its middle decades, the cities of the Campanian Plain and the Bay of Naples were not so fortunate. Textual accounts of Campania’s experience in the Gothic Wars during the first half of the sixth century hint at how warfare affected the region’s urban and rural populations. Not surprisingly, Procopius celebrates the actions of Belisarius in delivering and later protecting the city of Naples from its Gothic overlords. His account contrasts strongly with that in the Liber Pontificalis, an ongoing compendium of biographies of the bishops in Rome, which notes the Byzantines’ indiscriminate slaughter of the urban population in 535.21 A total rejection of this evidence in favor of an overly optimistic interpretation of the economic trajectories of towns and their hinterlands during the fifth and sixth centuries would be rash, but caution is also necessary. The deposition of Romulus Augustulus by Odoacer, the ensuing Ostrogothic incursion under Theoderic, and the later conflict between Byzantine and Gothic 20 For towns, see Savino, Campania Tardoantica, 199–223; for investment in the Church, Savino, ibid., 26–37 (Tables 1–7), 88–89 (Table 2). 21 For economic degradation of the region, ibid., 48–66; Procopius (ed. and trans. Kaldellis), De Bello Gothico, I.9-10, 272–277; Louis Marie Olivier Duchesne (ed.), Liber Pontificalis (Paris, 1886), I.60, 290; for an exploration of contrasting narratives of Belisarius’ actions, Gianluca Del Mastro, “Belisarius’ Repopulation of Neapolis: Troccla in Landolfus Sagax’ Roman History,” in De Simone and Macfarlane (eds.), Studies on Vesuvius’ North Slope and the Bay of Naples, 254–262. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 19 forces do not comprise the only narrative for the fates of the population of Italy in the period. Nor, however, can the observed processes of economic reorganization and socio-political reorientation function on their own as metonyms for civic or societal vitality. On that score, the analytical principle of vulnerability is especially useful as a heuristic tool. The most reasonable explanation is that the military and political exigencies of the period, combined with the changes in economic networks alluded to above, resulted in a region characterized by differential vulnerabilities—whether to volcanic activity, flooding, food shortage, or any other hazard. Contrasts between the ceramic assemblages of Puteoli and Suessa can serve as proxies for Wilches-Chaux’s category of economic vulnerability—on the one hand, a civic community enmeshed in seaborne trade but only partially engaged with the hinterland to its east and, on the other, a denser and richer nexus of smaller-scale connections largely sequestered from extra-regional linkages. In light of the differential aristocratic investment in civic institutions noted above, these contrasting networks are likely to have had strengths and weaknesses that varied according to the nature of the hazard encountered. Likewise, collective expressions of identity as instantiated by religious festivals may be indications of ideological or cultural vulnerabilities. Cult sites, for example, played multiple roles in creating and linking communities in central Italy. According to the late fourth-century poet Prudentius, the cult site of St. Hippolytus just outside Rome drew devotees from Capua and Nola from rich and poor and urban and rural alike. In the first quarter of the fifth century, Paulinus of Nola celebrated the shrine of St. Felix at Cimitile, just outside Nola, as a focal point for both pastoral and agricultural populations. In each case, the notion of vulnerability as defined by Wisner and his collaborators raises leading questions about the robustness and flexibility of communal and individual strategies for anticipating and coping with hazards. As we shall see, an acknowledgment of the coexisting hazards, and the context in which they occurred, brings greater nuance and subtlety into the analysis of how the inhabitants of Campania managed their fate than previous studies could offer.22 22 Prudentius, Peristephanon, xi.199–202, 205–208; Paulinus of Nola, Carmina, xxi. 542–567; Grey, Constructing Communities in the Late Roman Countryside (New York, 2011), 55. 20 | CAM GREY When we move from positing vulnerabilities to exploring outcomes in the decades immediately following the eruption of 472, some evidence suggests that rapid resettlement occurred as early as the first decades of the sixth century—consistent with indications that resettlement of the Campanian plain had occurred almost immediately after previous eruptions of Vesuvius. The site of Pollena Trocchia shows modest building above the deposits laid down by the eruption. Shells of land snails present in the layers immediately above those deposits suggest that the soil underwent cultivation. Local coarse wares datable to the early sixth century were also found, preserved beneath material dated to an eruption of 512; after that date, occupation and use of the site appears to have ceased.23 Meanwhile, in the town of Nola, buildings were cleared and floors re-established atop lahars in the aftermath of the 472 event. Likewise, the so-called villa of the via Saccaccio, located west of the city, though apparently much reduced from its apogee, was reoccupied in the decades immediately following the eruption. As at Pollena Trocchia, the final caesura in occupation seems to have been the otherwise little-known eruption of 512. Given the short period of time between the two events, we may well be witness to the progressive effects of successive hazards on already vulnerable rural inhabitants. Nonetheless, these patterns of rapid resettlement were no doubt motivated, at least in part, by the favorable growing conditions and excellent returns that the land produced. This juxtaposition of potentially catastrophic effects and demonstrably fertile soil gives tangible expression to the essential calculus between benefits and rewards, on the one hand, and threats or hazards, on the other, that confronted residents of the region. This calculus amounts to an evaluation of risk.24 23 For Pollena Trocchia, see Emilia Allevato et al., “Persistence of the Cultural Landscape in Campania (Southern Italy) before the AD 472 Vesuvius Eruption: Archaeoenvironmental Data,” Journal of Archaeological Science, XXXIX (2011), 399–406; De Simone et al., “Apolline Project 2007: il sito romano di Pollena Trocchia in località Masseria De Carolis,” in idem and Macfarlane (eds.), Studies on Vesuvius’ North Slope and the Bay of Naples, 232–33; De Simone et al., “Episodi vulcanici e vulcanoclastici ( V–XVII secolo) che hanno sepolto un edificio romano a Pollena Trocchia (Italia),” Il Quaternario, XXII (2009), 53–60. 24 For Nola, see Di Vito et al., “Human Colonization,” 139; for the villa of the via Saccaccio, Monica Lubrano, Gaetana Boemio, and Santa Sannino, “Note preliminari sulla villa romana di via Saccaccio in Nola,” Annali: Università degli Studi di Napoli Suor Orsola Benincasa (2011/12), 219–243; for resettlement, Scarpati, Perrotta, and De Simone, “Impact of Explosive Volcanic TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 21 Risk, a complex and multidimensional phenomenon, has stimulated considerable debate in recent decades. In its broadest sense, it may be defined as “a situation or an event where something of human value (including humans themselves) is at stake and where the outcome is uncertain.” When coupled with a hazard, risk tends to take on a negative valence—although, in purely abstract terms, it should be understood as a value-neutral concept that entails the potential for both beneficial and detrimental outcomes. In a manner analogous to the concept of vulnerability, risk, too, can function as a tool for evaluating a society. We can imagine a complex collection of cultural, ideological, and behavioral characteristics combining with infrastructural, technological, and structural elements to produce a given society’s “culture of risk.” Identifying that “culture of risk” is an exercise in both determining the unique character of that society and constructing a set of analytical principles that enables comparison with other societies, other communities, and other “cultures of risk.”25 For our present analytical purposes, we can distinguish two linked but subtly different types of risk, coexisting in a state of mutual interdependence—(1) a pure or “objective” form of risk, an etic evaluation or assessment of vulnerability to hazard based on all available information and all possible factors and (2) a society’s emic experience of risk, or “perceived” risk, the quantity or quality of risk that an individual or community is prepared or able to countenance. We should expect considerable variation in perceived risk within a community between men and women, the Eruptions,” 4–6; for the possible variation in resettlement between the northern and southern slopes of Somma–Vesuvius, De Simone, “Pollena Trocchia: Archives and Field Survey,” in idem and Macfarlane (eds.), Studies on Vesuvius’ North Slope and the Bay of Naples, 201–202. Compare with Mastrolorenzo et al., “472 AD Pollena Eruption,” which cautiously identifies this event as a moment that curtailed settlement. A re-examination of the survey evidence seems in order; the data is published only in heavily abbreviated form. For growing conditions, see De Vivo and Rolandi, “Volcanic Hazard,” 39–40, which identifies other factors, including strategic location just to the south of Rome and overlooking the Bay of Naples. 25 For the definition of risk, see Eugene A. Rosa, “The Logical Structure of the Social Amplification of Risk Framework (SARF): Metatheoretical Foundations and Policy Implications,” in Nick Pidgeon, Roger E. Kasperson, and Paul Slovic (eds.), The Social Amplification of Risk (New York, 2003), 56. Compare the detailed discussion of vocabularies and concepts in Wisner et al., At Risk, 13–19. For similar arguments with reference to beachside communities in Portugal, see S. Costas, O. Ferreira, and G. Martinez, “Why Do We Decide to Live with Risk at the Coast?” Ocean & Coastal Management, CXVIII (2015), 1–11. 22 | CAM GREY old and young, and the haves and have-nots across different geographical areas. We can conceptualize both this variation, and the grey area between any individual or group’s perceived risk and its objective risk, as entailing a mechanism for measuring a society’s vulnerability.26 The identification of perception as a fundamental component in both the “culture of risk” and the vulnerability of an individual, community, or society is an acknowledgment once again that disasters are essentially social phenomena. The evaluation of risk as an exercise in balancing knowable benefits against known hazards, and facing a future with an unknowable combination of positive or negative outcomes, explicitly entails temporality as a fundamental and multiscalar dimension in our analysis. Finally, the inclusion of differentiation in both the perception of risk and the experience of a hazard within a community recasts such seemingly descriptive phenomena as topography, population density and distribution, and micro-climatic variations as analytical tools in our evaluation of a disaster. In the case under discussion, for example, we must account for the dense but uneven distribution of settlement and exploitation in the Campanian plain during the Roman and late antique period.27 In part, this uneven distribution may be an artifact of the gaps in our information; considerable survey and excavation work still remains to be done. But it also points to variation in the region’s topography and hydrology. The Campanian Plain is a graben, or depressed block of land flanked by upraised sections or horsts, which extends 3,000 km2 from the Tyrrhenian Sea to the Apennines (Figure 2). The area has been subject to subsidence in the 26 For objective risk, see Wisner et al. At Risk; Juan Carlos Gavilanes-Ruiz et al., “Exploring the Factors That Influence the Perception of Risk: The Case of Volcán de Colima, Mexico,” Journal of Volcanology and Geothermal Research, CLXXXVI (2009), 238–252; for perceived risk, Åsa Boholm, “Comparative Studies of Risk Perception: A Review of Twenty Years of Research,” Journal of Risk Research, I (1998), 135–163, a good survey of the literature to the late 1990s. See also Wisner et al., At Risk, 13–16, which distinguishes “real” and “constructed” elements of risk; Costas, Ferreira, and Martinez, “Risk at the Coast,” 8–9. For specific reference to Vesuvius, albeit in a modern context, see F. Barberi et al., “Volcanic Risk Perception in the Vesuvius Population,” Journal of Volcanology and Geothermal Research, CLXXII (2008), 244–258; for the particularities of risk in ancient Roman society, Toner, Roman Disasters; for an economic theory based on differential impacts of disasters in contemporary contexts and various vulnerabilities, Alex Julca, “Natural Disasters with Un-Natural Effects: Why?” Journal of Economic Issues, XLII (2012), 499–510. 27 Wisner et al., At Risk, 12. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 23 past 25,000 years, one important corollary being a gradual slope from the northwest to the southeast. As a result, the southeastern section of the Campanian Plain, which sits at the confluence of several valleys and drainage networks, is particularly prone to floods and marshes. It is also a receptacle for the transportation and redeposition of pyroclastic materials and the concentration of lahars and other flow events related to an eruption of Vesuvius.28 To the northwest, the town of Nola, an influential center of Christian pilgrimage and an episcopal presence since at least the late fourth century, rests on a plain crisscrossed by roads serving fields, hamlets, and villages, which links this region to the towns and communities that consumed its produce. This part of the plain is subject to pyroclastic flows, ash fall, and lahars, exacerbated by the prevailing winds, which, as already noted, come predominantly from the southwest. Finally, to the west sits Naples, a thriving, populous, culturally heterogeneous city nestled precariously between Vesuvius and the large caldera volcano known as the Campi Flegrei.29 In a modern context, the differential physical or geographical vulnerabilities across the region have been evaluated and mapped extensively in recent years, as part of ongoing hazard assessments of the Somma–Vesuvius complex that encompass topographical, demographic, and climatic conditions like prevailing wind directions 28 See the detailed account of G. Florio et al., “The Campanian Plain and Phlegrean Fields: Structural Setting from Potential Field Data,” Journal of Volcanology and Geothermal Research, XCI (1999), 361–379. For the potential effects of marshiness on agricultural (and therefore fiscal) productivity in the late and post-Roman period, see Grey, “Revisiting the “Problem” of Agri Deserti in the Late Roman Empire,” Journal of Roman Archaeology, XX (2007), 372, n. 96; for transportation and redeposition, see Di Vito et al., “Human Colonization,” 139; Mastrolorenzo et al., “472 AD Pollena Eruption,” 32. 29 For Nola, see Savino, Campania Tardoantica, 212–216; Livadie, Vecchio, and Mastrolorenzo, “Eruzioni pliniane del Somma-Vesuvio”; idem, “Effetti delle eruzioni”; for strategic considerations and their economic corollaries, Savino, Campania Tardoantica, 79–85; for decline in wealth, Savino, Campania Tardoantica, 213; for the organization of the rural territory surrounding Nola during the early and high empire, Aniello Parma, “L’organizzazione del territorio rurale di Nola in èta romana,” in De Simone and Macfarlane (eds.), Studies on Vesuvius’ North Slope and the Bay of Naples, 133–143; for the prevailing winds, Mastrolorenzo et al., “472 AD Pollena Eruption,” 20. The disastrous effects of the 79 eruption on Pompeii and Herculaneum were a result of atypical wind patterns, since those two cities lie to the southeast and southwest of the volcano, respectively. For Naples, see Savino, Campania Tardoantica, 222–224; Paul Arthur, Naples, from Roman Town to City-State: An Archaeological Perspective (Rome, 2002); Caroline Bruzelius and William Tronzo, Medieval Naples: An Architectural and Urban History 400–1400 (New York, 2011). 24 | CAM GREY and the history of the volcano. Importantly, evaluations resting upon an expectation of a larger number of medium and small volcanic events will vary markedly from those that anticipate a smaller number of much bigger events. Likewise, the profiles of hazards will differ according to whether volcanic activity has tended to cluster into intensive periods or to distribute more evenly in time. For what it is worth, recent studies of Vesuvius posit all these potential patterns at different times in the volcano’s history.30 The upshot is that it is difficult to recapture the immediate prehistory of the 472 disaster. Nonetheless, the question of how much warning Vesuvius might have given prior to this eruption is certainly worth asking, and informed speculation about how the local population might have reacted to it well worth undertaking. Was knowledge or fear of the volcano’s potential effects directly or indirectly responsible for the distribution of settlement in the region? According to some excavators, certain buildings in Nola and in the Campanian Plain were already in a state of disrepair when they were covered by eruptive materials in 472. Yet whether this disrepair is an indication of a more general economic malaise in the region, evidence of the longer-term effects of the volcano’s presence and unrest, or merely an accident of survival and excavation with little or no diagnostic or generalizing force is difficult to determine.31 If Procopius is to be believed, the volcano, at least since the 472 eruption, was restless, rumbling and belching clouds of ash at relatively regular intervals: “Although the flames as a rule merely twist and turn upon one another, giving the inhabitants of that region no trouble, yet, when the mountain makes a rumbling sound like bellowing, it usually sends up not long afterward a mass of ashes (24). . . . Formerly this rumbling took place, they say, once 30 Cioni et al., “Explosive Activity”; Barberi et al., “Volcanic Risk Perception”; Giovanni Macedonio, Antonio Costa, and Albert Folch, “Ash Fallout Scenarios at Vesuvius: Numerical Simulations and Implications for Hazard Assessment,” Journal of Volcanology and Geothermal Research, CLXXVIII (2008), 366–377; Rolandi, “Volcanic Hazard”; Zuccaro et al., “Explosive Eruption Scenarios”; Principe et al., “Chronology of Vesuvius Activity,” 719; Roberto Scandone and Lisetta Giacomelli, “Precursors of Eruptions at Vesuvius (Italy),” Journal of Volcanology and Geothermal Research, CLXXI (2008), 191–200. 31 For buildings already in disrepair, see Pagano, “L’area vesuviana”; Mastrolorenzo et al., “472 AD Pollena Eruption.” The current sample is small, however, and interpretation is circular. Di Vito et al., “Human Colonization,” calls for more work on this issue. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 25 every hundred years or more, but in later times it has happened much more frequently” (28). A letter written by Cassiodorus (c. 485–c. 585), an Italian senator, on behalf of Theoderic (r. 493–526), the Ostrogothic king, which was concerned with relieving the taxes of the farmers of Campania, provides further glimpses of volcanic activity. Although the date, nature, and precise identification of this activity are controversial, Cassiodorus notes several warning signs that preceded a Vesuvius eruption: “For some time before, the mountain groans with the strife of Nature going on inside it . . . . Then the air is darkened by its foul exhalations; hot ashes scudding along the sea, a shower of drops of dust upon the land, tell to all Italy, to the transmarine Provinces, to the world, from what calamity Campania is suffering. . . . And yet, even those sandy tracts of pumicestone which the mountain vomits forth, dry and burnt up as they appear, have their promise of fertility. There are germs within them which will one day spring to life, and re-clothe the mountain side which they have wasted.” The description of Cassiodorus differs in many respects from that of Procopius, although both of them inextricably link hazard and benefit. Both accounts may well have derived from informants who witnessed or experienced an event—perhaps in 505 or 512 or, in the case of Procopius, 536— or who relayed a description of the 472 event preserved in local or familial memory.32 But what memory or experience of eruptions did the residents of the region have in the second half of the fifth century, that is, prior to these events? The periodicity of Vesuviusʼ activity is of particular relevance, as is the evaluation of how long societal memories of such activity are likely to endure. Evidence suggests 32 Procopius (ed. and trans. Kaldellis), De Bello Gothico, II.4, 328, appears to describe what De Vivo and Rolandi, “Volcanic Hazard,” termed a “Vesuvius Cycle”—intra-crater activity during a period of repose between Plinian and sub-plinian eruptions. Cassiodorus (trans. Thomas Hodgkins), Variae Epistolae (London, 1886), IV.50, 261–262. For Cassiodorus’ literary objectives, see MacFarlane, “Vesuvian Narratives,” 109–111. Rolandi, Petrosino, and McGeehin, “Interplinian Activity,” suggests that Cassiodorus is describing an “interplinian event” that occurred in 512. For debate over the incidence and nature of the 512 event, see Perrotta et al., “Emperor Augustus’ Villa”; for the case that an eruption happened at this time, Cioni et al., “The 512 AD Eruption of Vesuvius: Complex Dynamics of a Small Scale Subplinian Event,” Bulletin of Volcanology, LXXIII (2011), 789–810; Di Vito et al., “Human Colonization”; for disagreement on the basis of the geological evidence, Principe et al., “Chronology of Vesuvius Activity,” 704, 707. 26 | CAM GREY episodic, low-level activity in the centuries following the massive eruption that buried Pompeii and Herculaneum, as well as a small Strombolian event of 203 described briefly by Dio Cassius, whose repose in Capua 50 km away it appears to have disturbed. In a treatise exhorting his readers to embrace a life of penance, Pacian, a late fourth-century Spanish ecclesiastical writer, described both Etna and Vesuvius as “boil[ing] with unwearied volumes of flame,” activity similar to that described by Procopius and Cassiodorus 150 years later. Moreover, Pacian draws from an enduring cultural construction of Vesuvius as perpetually active, which probably persisted into the late fifth century. Given the principle that large-scale events attract disproportionate attention, more activity was likely to have occurred than our textual sources reveal, but no evidence of any further activity is available until the late fifth century. At any rate, grumblings, bellowings, belchings, or even the occasional shower of ash that Procopius mentioned do not amount to an explosive eruption. Warnings of an eruption could occur as few as twentyfour to forty-eight hours in advance, or stretch into weeks, months, or even years. In the lead-up to both the 1631 and the 1944 events, eyewitnesses referred to seismic tremors, increasing activity within the cone, and fissuring prior to the actual eruption.33 33 Pacian (ed. Angel Anglanda Anfruns), De Paenitentibus ( Valencia, 1982), XI.6, 34. For low-level activity, see Cioni et al., “The 2nd to 4th Century Explosive Activity Of Vesuvius: New Data on the Timing of the Upward Migration of the Post-A.D. 79 Magma Chamber,” Annals of Geophysics, LVI (2013), 789–810; for the eruption of 203, Dio Cassius, Epit. Book, 77.2.1. See also Rolandi, Petrosino, and McGeehin, “Interplinian Activity,” 31; Principe et al., “Chronology of Vesuvius Activity,” 705 (Table 1), 706. Cioni, “Explosive Activity,” 343–344, categorizes and describes “violent Strombolian eruptions” of Somma-Vesuvius in detail but stops short of including this event within that group. For disproportionate attention, see Scarpati, Perrotta, and De Simone, “Impact of Explosive Volcanic Eruptions”; for precursors and warnings, De Vivo and Rolandi, “Volcanic Hazard,” 43. Note that an earthquake of 62 in Campania is generally regarded as a precursor to the 79 eruption, although it is difficult to ascertain much about it with any degree of confidence. Modern historical catalogs of earthquake activity in the Mediterranean contain no record of such an event in or around Campania prior to 472, but the evidence from the Western Mediterranean in this period is patchy. The most recent catalog is Nicholas Ambraseys, Earthquakes in the Mediterranean and Middle East: A Multidisciplinary Study of Seismicity up to 1900 (New York, 2009). See also the accounts of Emanuela Guidoboni, “Vesuvius: A Historical Approach to the 1631 Eruption ‘Cold Data’ from the Analysis of Three Contemporary Treatises,” Journal of Volcanology and Geothermal Research, CLXXVIII (2008), 347–358; Antonella Bertagnini et al., “Eruption Early Warning at Vesuvius: The A.D. 1631 Lesson,” Geophysical Research Letters, XXXIII (2006), available at doi.org/10.1029/2006GL027297. For the 1944 event, see Paul D. Cole and Scarpati, “The 1944 Eruption of Vesuvius, Italy: Combining Contemporary Accounts and Field Studies for a New Volcanological Reconstruction,” Geological Magazine, CXLVII (2010), 391–415, which provides details of eyewitness accounts. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 27 At the current state of knowledge, we can do no more than imagine that any level or type of seismic and/or volcanological activity that emerged prior to the eruption of 472 would have evinced a range of attitudes and responses, in both the long and the short term, intermingling incoherently within and between the communities of the region. These attitudes were likely generated by, or mutually implicated with, vulnerabilities of the various types that Wilches-Chaux identified, although the fit is by no means perfect. Procopius, Cassiodorus, Dio Cassius, and Pacian each reveal an underlying sense of wariness mixed with optimism regarding the volcano that dominated the Campanian Plain. Locals were hardly ignorant of the potential threat that Vesuvius posed, at least in the abstract. Yet the amount and quality of their information was certainly limited by educational and/or cultural factors, technologies of memory or memorialization, or other elements. 472: QUESTIONS AND PROPOSITIONS How can vulnerability be employed as a tool for evaluating the 472 eruption, and what new insights does such an approach present? A series of brief sketches based on the evidence in the preceding sections, together with comparisons with other communities managing volcanic risk—organized loosely around the WilchesChaux typology—can provide an answer. The objective is primarily to illustrate the analytical potential of further exploring phenomena that are generally regarded as end points in the analysis of the socio-economic, political, cultural, and physical contexts of late fifth-century Campania: the nature, extent, and relative robustness of trade networks; the presence or absence of an institutional power structure, whether the administrative apparatus of a state or of the Church and its agents; the location of habitation and resource exploitation within the topography of the region; and the influence of previous knowledge about volcanic activity on the perception and tolerance of the risk posed by Vesuvius. Each of these factors, and many others, can be treated as vulnerabilities, not only to Vesuvius but also to a raft of other hazards that characterized and constituted the lived experiences of the inhabitants of the Campanian Plain and its environs. Beginning with the nexus of ideological, cultural, and educational vulnerabilities that constitute knowledge and experience of volcanic eruptions in general, and Vesuvius in particular, what REASSESSING THE ERUPTION OF 28 | CAM GREY were the explanatory principles available to Roman and late Roman observers? Certainly, Procopius’ lengthy description of the 536 event—animated, at least in part, by his self-consciously Thucydidean, rationalist approach to the writing of history—presents chains of causes and effects that, though not necessarily “scientific,” are not completely inaccurate or implausible. But those explanatory principles might not have been available to everyone in the late antique world; they might have been restricted to certain segments of the population or to Procopius himself and/or his putative informant(s). Furthermore, if, as in the case of the 79, 1631, and 1944 eruptions, seismic activity preceded the eruptive event, it could have shaped the risk tolerance of the inhabitants of the region. In any case, what relationship existed between knowledge of volcanic hazard and perception of risk? Did familiarity, as it were, breed contempt—or paralysis?34 Comparative literature, albeit drawn largely from questionnaires and surveys of modern populations, suggests that, even in contexts where individuals or a community are acutely aware of the danger that they face, their willingness to prepare effectively for it can be low. What about the inhabitants of the Campanian Plain? Although the paucity of our data militates against systematic evaluation of their physical, economic, social, political, technological, or institutional vulnerabilities, they were unlikely to have perceived the volcano as a constant, imminent threat to life and livelihood. Instead, agriculturalists in the region were more likely to invoke volcanic activity as one element in a congeries of hazards, the most pressing of which were far more prosaic and immediate— fluctuations in rainfall and temperature, the predations and demands of the state and powerful patrons, and so on.35 34 For ancient interpretations of volcanic activity, see Harry M. Hine, “Seismology and Vulcanology in Antiquity?” in Christopher J. Tuplin and Tracey E. Rihll (eds.), Science and Mathematics in Ancient Greek Culture (New York, 2002), 56–75. See also the summary of the debates in contemporary literature about the effectiveness of local knowledge in Bankoff, “Historical Geography.” 35 For the relationship between knowledge of hazard and perception of risk, see the complementary constructions of Christopher J. L. Dibben, “Leaving the City for the Suburbs—the Dominance of “Ordinary” Decision Making over Volcanic Risk Perception in the Production of Volcanic Risk on Mt Etna, Sicily,” Journal of Volcanology and Geothermal Research, CLXXII (2008), 288–299; Franck Lavigne et al., “People’s Behaviour in the Face of Volcanic Hazards: Perspectives from Javanese Communities, Indonesia,” ibid., CLXXII (2008), 273–287; Jean-Christophe Gaillard, “Alternative Paradigms of Volcanic Risk Perception: The Case of Mt. Pinatubo in the Philippines,” ibid., CLXXII (2008), 315–328; Manga S. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 29 The threat of famine, which encompassed economic, social, political, technical, and cultural vulnerabilities, would have been even more pressing. Our textual evidence for food shortages in Italy and around Rome in the decades immediately prior to this event, usually due to military conflict, are scant and hardly conducive to detailed analysis. Nonetheless, famine, when further complicated by, say, an inflexible landowner or a rapacious tax collector, could have exacerbated the vulnerability of some farmers in Campania to the eruption of 472 by eating into their stored resources or damaging networks that otherwise might have helped— such as the network ostensibly attached to the shrine of St. Hippolytus. The abandonments of Pollena Trocchia and the villa of the via Saccaccio in the wake of two or three volcanic eruptions in quick succession may well hint at the region’s thin economic margins, especially after repeated or cascading hazards.36 Caution, however, is necessary. The historical record suggests that, among agricultural populations in particular, wholesale abandonment of land is almost always a last resort. The little evidence available about land-tenure patterns in Campania during the Ostrogothic period does not indicate the predominance of a few wealthy landowners with large operations staffed by slaves or tenants. Another letter of Cassiodorus censuring the seizure of property in Campania suggests that both large and small landowners either continued to coexist in the region in the decades following Njome et al., “Volcanic Risk Perception in Rural Communities along the Slopes of Mount Cameroon, West-Central Africa,” Journal of African Earth Sciences, LVIII (2010), 608–622. See also the chapters exploring the factors exacerbating risk in Pidgeon, Kasperson, and Slovic (eds.), Social Amplification. For the role of information, see Costas, Ferreira, and Martinez, “Risk at the Coast,” 9. 36 For the threat of famine, see Dionysios Ch. Stathakopoulos, Famine and Pestilence in the Late Roman and Early Byzantine Empire: A Systematic Survey of Subsistence Crises and Epidemics (Burlington, 2004), 243–244, which records evidence for food shortages in Italy datable to c. 467 and c. 471, although chains of cause and effect are difficult to untangle. Widespread predation is well attested in the power relations of the period. For further discussion, with fuller references to the relevant sources and scholarly literature, see Grey, Constructing Communities 10–12, 142–154. For precursors, see Perrotta et al., “Emperor Augustus’ Villa” (the current excavators have thus far been reluctant to attribute some of the damage done to this villa in this light); for cascading disasters, Gianluca Pescaroli and David Alexander, “A Definition of Cascading Disasters and Cascading Effects: Going Beyond the ‘Toppling Dominos’ Metaphor,” Planet@Risk Global Forum Davos, III (2015), 58–67, available at https://planet-risk. org/index.php/pr/article/view/208/355. On a grander scale, Kyle Harper enlists cascading disasters as an explanatory principle in The Fate of Rome: Climate, Disease, and the End of an Empire (Princeton, 2017). 30 | CAM GREY this eruption or reoccupied the countryside shortly afterward. We should resist the urge to ascribe responsibility for fundamental shifts in land ownership or socio-economic power relations in the region to this event—if, in fact, such shifts occurred, which remains the subject of considerable debate.37 Although such observations leave us no closer to identifying the behavior of the volcano in this period, they reveal how the concept of vulnerability can inflect and expand the types of question that we ask about this event, the circumstances that precipitated it, and its aftermath. Blanket interpretations of the region as experiencing widespread demoralization and economic depression should be scrupulously avoided. The archaeological evidence sketched and discussed briefly above intimates considerable variation in experiences between and within different micro-regions. Abandonment of rural buildings need not have entailed widespread and permanent flight from the region so much as a possible redistribution of settlement there, in the form of agglomeration not unlike that in other areas of Italy and Western Europe at the time, for which no volcanic eruption serves as a convenient explanation.38 The reasonable approach is to acknowledge the existence of multiple vulnerabilities prior to the eruption, and to posit multiple experiences of it as a result. Although our appreciation of these 37 Cassiodorus, Variae Epistolae, IV.10. For resistance to abandonment as a strategy, see Michael Baksh and Allen Johnson, “Insurance Policies among the Machiguenga: An Ethnographic Analysis of Risk Management in a Non-Western Society,” in Elizabeth Cashdan (ed.), Risk and Uncertainty in Tribal and Peasant Economies (Boulder, 1990), 217–218; Thomas W. Gallant, Risk and Survival in Ancient Greece: Reconstructing the Rural Domestic Economy (Stanford, 1991), 113–142; Grey, Constructing Communities, 61. 38 Rural socio-economic patterns across the western Mediterranean are characterized, above all, by heterogeneity. For a recent synthetic survey account, see Simon Esmonde Cleary, The Roman West, AD 200–500: An Archaeological Study (New York, 2013). Contrast the localized catastrophist explanations in Mastrolorenzo et al., “472 AD Pollena Eruption.” Excavators at the “villa of Augustus” point toward reorganization or reuse of the site on a lesser scale, noting evidence of disrepair in the period. See Perrotta et al., “Emperor Augustus’ Villa,” 462. By contrast, Livadie, Vecchio, and Mastrolorenzo, “Eruzioni pliniane del SommaVesuvio,” offers scattered evidence for resettlement in the years following the eruption. Again, a dearth of published evidence currently hampers definitive interpretation. For an archaeologically focused account of microregional diversity in the experiences of this eruption that complements the present study but came to the attention of the author too late to be fully incorporated, see Girolamo Ferdinando De Simone and Ben Russell, “The Late-Antique Eruption of Vesuvius in A.D. 472 and Its Impact from the Bay of Naples to Aeclanum,” Journal of Roman Archaeology, XXXII (2019), 359–388. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 31 experiences can only be speculative at this point, we can advance hypotheses and questions that can help to focus future treatments of this event within the socio-economic and cultural history of Campania. One scenario, already signaled, is the possibility that evidence for abandonment of settlements in the years leading up to the eruption is not necessarily evidence for the widespread impoverishment of the region prior to the event. Yet, if the distribution of the population altered in the period, what does that shift imply about the vulnerability of the town of Nola, or the city of Naples? Did population pressure, suburban sprawl, or the presence/absence of city walls affect the capacity of communities to respond to an event such as a volcanic eruption? How, to take another scenario, can we evaluate the extent to which the experiences of pastoralists and agriculturalists to a volcanic eruption differed, if at all? Osteological and ceramic discoveries from the region certainly reveal an abundance of sheep and goats in the period but not necessarily an increase in the importance of these animals in local foodways. Legal documents from the late fourth and early fifth centuries reveal farmers having a certain amount of distrust for shepherds, though, as already noted, the writings of Paulinus of Nola show agriculturalists and pastoralists participating as equal partners in communal celebrations of the feast of Saint Felix of Cimitile. Leaving aside the debate about the degree of socio-economic, or even familial, integration between the two parties, how might their respective socio-economic patterns have differentially prepared individuals to respond to a hazard?39 Pastoralists were likely to be more mobile, but they might have been unable or unwilling to abandon their flocks in the face of tephra, ash, and lahars. Agriculturalists were in danger of losing buildings and crops. The timing of the eruption of 472 in November is important in this respect. Pastoralists practicing any form of 39 For the presence of sheep/goats, see De Simone et al., “Pottery as a Proxy Indicator for Diet Change in Late Antique Campania,” in R. Gül Gürtekin-Demir et al. (eds.), Keramos, Ankara (Ankara, 2015), 218–228, 324–239; Paul Arthur, “Pots and Boundaries: On Cultural and Economic Areas between Late Antiquity and the Early Middle Ages,” in Michel Bonifay and Jean Christophe Tréglia (eds.), LRCW 2: Late Roman Coarse Wares, Cooking Wares and Amphorae in the Mediterranean: Archaeology and Archaeometry (Oxford, 2007), 15–27; for distrust, CTh, IX.30.2, 364 (Campania), 31.1, 409 (Italy); for participation in communal activities, Paulinus of Nola, Carm., xxi.542–567; for familial connections, Grey, Constructing Communities, 30. 32 | CAM GREY transhumance were likely to have been in the process of moving their flocks from upland to lowland pastures. If their flocks were located on the Campanian Plain, they could well have stood to lose more at this time than at other times of the year. Agriculturalists, however, were likely to have been relatively inactive in November, since the last plowing and sowing of the year will have happened months earlier. Nonetheless, any seedlings or young crops would have been especially vulnerable, and the loss of stored grain and other supplies for the winter could have been serious. The matter of food production and storage invites consideration of the medium- or longer-term ramifications of an eruption of this magnitude. Scholars have recently begun to explore with increasing granularity the climate-forcing impacts of larger eruptive events upon temperature and solar radiation, driven primarily by the quantities of sulfate and ash deposited in the atmosphere. To date, however, they have devoted little attention to response strategies among agriculturalists regarding those events. For example, some make a strong case for interpreting the 1257 eruption of Rinjani (Samalas), Indonesia (VEI 7), as having been instrumental in triggering the onset of the Little Ice Age. Comparable climatological effects lasting multiple years on a global scale have been ascribed to a cluster of eruptions datable to the period between 536 and 547, all of which appear to have been considerably larger than the 472 Pollena event. It is plausible that this cluster of events affected food production and storage strategies in various parts of the Mediterranean, although the impact of the 472 eruption is unlikely to have been as widespread or long-lasting.40 40 For Rinjani (Samalas), see Newhall, Self, and Robock, “Future Volcanic Explosivity Index ( VEI) 7 Eruptions,” 575–578; for large-scale effects of large-scale sixth-century eruptions, Michael Sigl et al., “Timing and Climate Forcing of Volcanic Eruptions for the Past 2,500 Years,” Nature, DXXIII (2015), 547–548; Ulf Büntgen et al., “Cooling and Societal Change during the Late Antique Little Ice Age from 536 to around 660 AD,” Nature Geoscience, IX (2016), 231–237; Matthew Toohey et al., “Climatic and Societal Impacts of a Volcanic Double Event at the Dawn of the Middle Ages,” Climatic Change, CXXXVI (2016), 401–412. Even large-scale events are likely to have engendered a range of responses in different contexts; we should not uncritically elide the distinction between effects and responses. For fuller discussion of this critical methodological distinction, see Grey, “Climate Change and Agrarian Change between the Fourth and Sixth Centuries: Questions of Scale, Coincidence, and Causality,” in Jan Willem Drijvers and Noel Lenski (eds.), Shifting Frontiers 12: The Fifth Century (Bari, 2019), 35–48. For volcanic activity and fertility of soil, see Perrotta et al., “Emperor Augustus’ Villa,” 462; De Simone et al., “Apolline Project 2007,” 232–233; more generally, Grattan, “Aspects of Armageddon.” TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 33 The letter of Cassiodorus cited above conveys something of the potential microregional repercussions of even a relatively small eruption in the years immediately following it. Any holdings damaged by eruptive material or secondary flows would have required a capital investment to make repairs or would even have forced an involuntary sale or abandonment in some circumstances. One mitigating factor might have been the fertility for which volcanic soils are justly renowned, as the presence of snails in the humified layer overlying volcanic deposits at Pollena Trocchia and at a nearby villa on the northern slopes of Vesuvius testifies. Again, this eruption fell on the smaller end of the scale; it did not have global consequences. Local as it was, however, its effects varied markedly, even on a microregional scale. To invoke Wilches-Chaux’s typology again, this inquiry also considers economic, social, political, and institutional vulnerabilities. What did long-term administrative and fiscal trends have to do with the vulnerability of the inhabitants of the region in the decades preceding and following this eruption? As noted above, to the west and south of Somma-Vesuvius, the Bay of Naples was renowned for its luxurious aristocratic villas. Even in the period under discussion, the region was witness to a continuing presence of influential aristocrats, and under the supervision of a governor who reported directly to the praetorian prefect of Italy. Moreover, following the fiscal and administrative reforms of, first, the Tetrarchy and, later, Constantine, Campania, like the rest of Italy, became liable for the payment of taxes. Significantly, Campania was heavily implicated when the grain from Egypt was redirected to the new capital Constantinople in the first half of the fourth century, thus reconfiguring the relationship between the city of Rome and its immediate hinterland. Some recent scholarship maintains that these developments brought a greater degree of centralized control to the trade that entered the region. Be that as it may, the region’s economic fertility clearly had fiscal implications for the western half of the Roman Empire.41 Although the early sixth-century letter of Cassiodorus cannot offer details about how those connected processes fared in the immediate aftermath of the eruption, it does suggest what the ruler’s 41 De Simone and Martucci, “Thirst for Wine?” 127–128; Savino, Campania Tardoantica, 47–48, 66–67. 34 | CAM GREY attitude toward local landowners would have been under such circumstances. In that letter, Cassiodorus directs the praetorian prefect of Italy to send assessors to the region to determine the extent of the damage and the proportion of tax relief required. The similarities to Suetonius’ description of the actions of the Emperor Titus in response to the eruption that buried Pompeii and Herculaneum almost 450 years earlier are striking. Neither author, however, has much to say about the precise mechanisms and effects of those initiatives, although both Cassiodorus and the Expositio Totius Mundi give us some idea of the greater local demands placed upon Campania to supply the city of Rome. A comparable sensibility is evident in a law passed by Emperor Honorius (r. 395–423), which devotes special attention to relieving the tax burden of local landowners after the Visigothic depredations of the 410s, offering more generous treatment to Campania than to other, neighboring regions of Italy.42 The fact that no comparable response to the 472 eruption appears in the legal sources might be due to the eruption coinciding with a four-month interregnum between the reigns of Olybrius (r. April–November, 472)—placed on the throne by Flavius Ricimer, magister militum of the western court—and Glycerius (r. March 473–June 474)—appointed by Gundobad, Ricimer’s successor. A vacuum at the very top of the Roman administration would not inevitably have engendered chaos throughout society, but it might have severely limited the opportunity to petition for a direct intervention in late 472 or early 473, in the manner that Cassiodorus’ letter reveals. Unfortunately, we have no information about the effectiveness of tax-extraction practices in the region throughout the last decades of the fifth century. Although we cannot infer that the late imperial and post-Roman fiscal administrations operated in exactly the same way, the Edictum Theoderici, which provides evidence for the fiscal system under the Ostrogothic Kingdom, seems to indicate that tax assessment and collection practices remained largely intact into the reign of Theoderic. If so, the tried and true tax principles, which included the role of the land as the basis for assessment and the role of local aristocrats as guarantors and collectors 42 Cassiodorus, Variae Epistolae, IV.50; Suetonius, Titus, 8.3; CTh, XI.28.12, 418; for a discussion of context and circumstances, Savino, Campania Tardoantica, 79–82. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 35 of the community’s tax burden, were at least nominally in place in the early 470s. What is not clear is whether (1) a weak centralized administration resulted in local aristocrats choosing not to fulfill their fiscal obligations following the eruption; (2) a lack of oversight and cohesion led to increased hardship (the absence of a ruling emperor removing any mechanism for appeal); or (3) local potentates usurped the claims of the state, collecting revenue from their less powerful neighbors, tenants, and dependents and keeping it for themselves.43 The possibility of elites absconding with the tax revenue raises the issue of how the Church conducted itself in the years just before and just after the disaster. How might its control of property and its power in local politics have affected the vulnerability—or, better, vulnerabilities—of the inhabitants of the Campanian Plain? The evidence suggests that, in the wake of Constantine’s donations of land to ecclesiastical institutions, the Church of Rome took ownership of a considerable proportion of the land in Campania; individual churches in Campania benefited as well. Furthermore, local aristocrats seem to have made private donations of land and resources to regional churches, which enabled members of ecclesiastical hierarchies to gain positions of no little influence as patrons and landlords during the period.44 To what extent did local churches function as focal points within their communities? Archaeological evidence of people sheltering in a church in the town of Nola, for example, points toward officers of that church possibly opening their doors to local residents fleeing the eruption. Was this an unanticipated act in the heat of the moment, or was the Church a refuge because it was acknowledged as a trusted partner of long standing? Scholars speak of improvisation during disasters, moments when accepted mechanisms for societal organization fail and individuals or groups emerge to improvise new strategies to replace them, at least temporarily. By this reading, the 472 eruption might have provided an opportunity for local ecclesiastical leaders to assume a more significant role in the community. 43 For Roman tax practices in the Ostrogothic kingdom, see Gerda Heydemann, “The Ostrogothic Kingdom: Ideologies and Transitions,” in Jonathan J. Arnold, M. Shane Bjornlie, and Kristina Sessa (eds.), A Companion to Ostrogothic Italy (Leiden, 2016), 25; Grey, “Landholding and Labour in the Rural Economy,” in ibid., 276. 44 Savino, Campania Tardoantica, 26–37 (Tables 1–7), 88–89 (Table 2). 36 | CAM GREY The fresco from the San Gennaro Catacomb evoked at the beginning of this paper, however, complicates this reading (Figure 1). The fresco presents the proto-martyr Stephen; Januarius, the patron saint of Naples, claimed to have been martyred during the Diocletianic Persecution (whose blood to this day is said to liquefy in a vial three times a year in Naples Cathedral); and (probably) Agrippinus, a third-century bishop of Naples long-associated with Januarius as a protector of the city. In the background, the eruption of Vesuvius is clearly visible. It is likely that the cult of Januarius was already well established in the region before the 472 eruption. If a recent sixth-century dating for the fresco is accurate, the painting was done shortly after the eruption, although the earliest textual celebrations of the saint delivering the city of Naples from this event is dated to the ninth century. Is the fresco a case of ecclesiastical opportunism or expected institutional response? Either interpretation is possible; more evidence and more detailed analysis is necessary. Yet framing the question in this way, and at this scale, points toward more finely grained micro-regional accounts of the rise of the Church’s influence in this period.45 Was the eruption of Vesuvius in November 472 a disaster? If so, for whom? These questions, posed at the start of this article, turn out not to be the most interesting ones to ask about this event and the longer-term processes that serve as its context. Instead, the discussion above used the evidence pertaining to this eruption to explore how historians can employ concepts drawn from disaster 45 Pagano, “Archaeological Sites”; Mastrolorenzo et al., “472 AD Pollena Eruption,” 33. For evidence of church leaders aiding or organizing their communities in the face of threats and disasters, see Ralph W. Mathisen, “‘Nature or Nurture?’: Some Perspectives on the Gallic Famine of ca ad 470,” The Ancient World, XXIV (1993), 91–105; Grey, Constructing Communities, 130; for theoretical treatments of improvisation, Lindell, “Disaster Studies,” 3–4, 7; The cult of Januarius’ first appearance is c. 431. See Carlo Ebanista, “Il culto Ianuariano a Nola,” Campania Sacra, XXXVII (2007), 275–276; Granier, “Lieux de mémoire,” 78, n. 77. However, textual attestations are more difficult to trace at this early date. It appears in the Miraculi sancti Agrippini (Bibliotheca hagiographica latina antiquae et mediae aetatis [BHL], 175–177, Acta Sanctorum, 65 [1925], Novembre IV, 118–128), which were written and supplemented between the eighth and eleventh centuries, but it is entirely absent from the sixth or early seventh-century Acta Bononiensia martyrii sancti Ianuarii—BHL, 4132, Acta Sanctorum, 46 (1970), Septembre VI, 870–871—as well as the eighth- or ninth-century Acta Vaticana martyrii sancti Ianuarii—BHL, 4115–4119, Acta Sanctorum, 46 (1970), Septembre VI, 866–870—which together constitute the extant accounts of the life and miracles of Januarius. TH E VE SU VI US ER U P TIO N O F A .D . 4 72 | 37 studies to tell nuanced, complex, and multidimensional stories about the inhabitants of Campania in the fifth and sixth centuries. The principal objectives of this study are twofold: The first is to address the challenges presented by the ancient evidence for this event. The points of entry are the textual allusions to an eruption in the late fifth century, but the narratives constructed about that eruption relied principally upon geological and archaeological data. Such an approach throws into high relief the potential contradiction between interpreting this eruption as an event and analyzing the disaster (or disasters) that followed it for decades or longer thereafter. The projects of social history, disaster studies, and archaeology deliberately tend to eschew events in favor of processes and structures that unfold gradually over time. Yet, this eruption presents tangible markers of a specific moment. Textual, geological, and stratigraphical evidence coalesce to force the conclusion that this eruption was indeed a disaster: In the decades preceding it, ceramic typologies looked different from how they looked immediately after it. However we interpret its intent, the striking fresco from the San Gennaro Catacomb leaves little doubt about the significance of the event that it commemorates; the writing of Marcellinus Comes is equally explicit. Nonetheless, the absence of local textual evidence and the strong hints of rapid re-occupation of the region or, at the very least, the re-exploitation of its rich and fertile soils equally suggest that, for some inhabitants of the region, the eruption was of only minimal impact. The second objective of this article is to employ the concepts of risk and vulnerability in order to create a new understanding of the experiences surrounding the eruption in the long and the short term, taking into account location, the nature and robustness of social and economic resources, the mechanisms for accessing and exploiting power, and so on. The many lacunae in our knowledge about this event and the contexts within which it occurred mean that some questions have to remain unanswered. Nonetheless, this perspective on the 472 eruption demonstrates that treating a disaster as a process that unfolds over an extended period can illuminate the social, political, cultural, and economic structures of the late Roman world.