How did Mithras-worship, a cult that rapidly attracted followers across the Roman Empire over the course of a few generations, come to an end? By most modern accounts, the rise of Christianity was largely to blame, albeit via a host of different mechanisms. Yet in the first major survey of late-antique Mithraism across the empire, David Walsh offers a far more nuanced and innovative account of the cult’s decline and archaeological disappearance. Rather than a sudden, violent, and Christian-driven end, Walsh’s Mithraism shrivels from within, the victim of endogenous transformations which made the cult unable to draw continued commitment and investment from worshippers in the changing world of late antiquity.
Chapter 1 lays the groundwork for this argument by identifying changes within the cult itself that Walsh later uses as explanations for the cult’s decline. What had been a relatively stable and coherent package of material culture and meaning in the second and early third centuries CE fissured into a much more diverse set of “cults of Mithras” (94). These changes included adaptations to the “standard” plan of mithraea; a wider range of terms to describe sanctuaries and even to name the god; a changing social composition of worshippers that included more senatorial elites (the late-antique shift that has been most discussed beyond the present work); and, most important for Walsh’s later arguments, two key transformations in ritual practice. First, Walsh suggests that the widespread deposition of low-denomination coinage in mithraea from the late third to the end of the fourth century reflects the opening of the cult to a much broader group of “casual” worshippers engaging in a depositional practice shared across cults and sanctuaries in this period. Second, he observes a restructuring of the initiation system and its grades, with initiations becoming less intense and the grades taking on new meanings. In short, for Walsh, the late antique version of the cult was a watered-down, vulgarized, and diversified iteration of its high-imperial form.
Chapter 2 observes the gradual drop in building and renovating mithraea across the empire from the late third century onwards, which is taken as indicative of the cult’s failure to persuade new investment from worshippers and thus the cult’s slow decline. The different chronological rhythm of construction and restoration of mithraea—as opposed to other forms of temples and public buildings—suggests that the decline of investment in Mithraic sanctuaries cannot be explained solely as the product of a wider decline of investment in urban infrastructure. Walsh also notes regional differences: mithraeum construction ceases slightly earlier in Gaul/Germany (early fourth century) than in Italy or the Danube (mid fourth century).
To explain the observed drop-off in constructing or renovating mithraea and their major furnishings, Walsh draws on correlations and sociological theory (Chapter 3). The shrinking size and change in the social composition of communities that might have engaged in the cult allowed fewer opportunities for the cult to spread through the types of tight social networks that had permitted it to spread so aggressively in the second and early third centuries. It is not surprising that a cult so closely tied to the particular social dynamics of the High Roman Empire1 should struggle in the face of the societal transformations of the fourth century; Walsh simply draws attention to some of the smaller-scale dynamics driving this apparent decline.
Walsh’s second set of explanations for the observed decline is both more exciting and more problematic: the changes in ritual practice observed in Chapter 1 created an environment that fostered less commitment from worshippers. With fewer or watered-down initiation rites, a worshipper was not bound to the mithraeum community or the cult by the same emotional experience. Because Walsh accepts Roger Beck’s problematic notion of all mithraea being miniature maps of a Neoplatonic cosmos,2 the changes in mithraeum design meant that the cult was no longer peddling unique cosmological secrets. The incorporation of more generalized cult practices like coin offerings made by a wider group of worshippers also meant that the cult no longer occupied a distinctive niche in a wider religious ecosystem. Together, then, changes in ritual practice created a cult that no longer conferred distinct religious capital, and was thus unable to sustain the support or engagement of worshippers.
The final chapter examines the archaeological evidence for the end of mithraea, most of which were seemingly abandoned by the end of the fourth century. Even though more mithraea surveyed show signs of damage (26 mithraea) rather than abandonment or cleaning-out (16 mithraea), Walsh rightly notes the very limited evidence for targeted Christian destruction (or even Christian communities active in the areas around destroyed mithraea). Instead, Walsh suggests a much wider range of explanations. These include the raids and conflicts of the fourth century that also sparked abandonment of nearby sanctuaries, intentional fragmentation of important images by worshippers, natural disasters, and accidents. Here, too, Walsh notes regional diversity: mithraea in frontier provinces like Gaul and Germany show more violent ends than “interior” ones like Italy and Dalmatia, leading him to put greater weight on destruction due to non-religious conflicts.
Walsh’s recognition of the diversity of the fates of mithraea, and his turn towards multiple and endogenous explanations for the cult’s disappearance, are welcome directional changes in discussions of Mithras-worship in the third and fourth centuries. Yet the thinness of the evidence upon which Walsh’s case is predicated leaves many of the strongest claims open to challenge and far different readings; the most novel arguments are built on the weakest evidentiary foundations.
Take, for example, the idea of changing ritual practice and the watering-down or abandonment of initiatory rites that Walsh sees as leading to weaker commitment from worshippers. The main evidence cited for this weakening is primarily that grades are almost never mentioned epigraphically (outside a set of inscriptions related to one senatorial family’s mithraeum at Rome); this is coupled with the supposition that senatorial elites would not deign to undergo any rites that were too strenuous. Epigraphic silence on grades, though, is hardly surprising. Almost all mentions of initiatory grades in the High Empire come from Rome, Ostia, and Dura-Europos (thanks to the preservation of graffiti on wall-plaster); these even show the same level of diversity that Walsh attributes to late antiquity, with “non-canonical” ranks of stereōtēs (strenthener) and melloleōn (wannabe lion) attested at Dura.3 Given the documented drop in Mithraic inscriptions of all kinds in the fourth century, would we expect anything other than near-silence? Indeed, one might instead be struck by the indices of continued ritual practices in late-antique mithraea; at Bornheim-Sechtem, a mid-fourth-century rebuilding of a mithraeum was marked by foundational rites that echoed (with some elaboration) similar rites used to found a mithraeum on the site three or more generations earlier. Might we not then expect the same continuity in any Mithraic initiatory rites, especially given that our main (but not unproblematic) textual sources describing such rites were composed in the late fourth century CE?
Similarly, it is often unclear whether the types of evidence Walsh cites are products of late-antique historical reality or the formation of the archaeological record, a difficulty Walsh himself often notes, but which does not stop him from deploying the material to support his claims. The empirical data upon which this book depends were generated, recorded, and interpreted by excavators at different periods, with different levels of training, and published to very different degrees. Nearly a third of the mithraea in Walsh’s catalogue were “excavated” in the nineteenth century; over three quarters were excavated before 1980; and not a single one excavated since has been published with a full, contextual report. Despite the difficulties arising from data created under such variegated conditions, each site in the catalogue, once re-described, is weighted equally as evidence for the wider claims Walsh makes. Yet one could easily come to rather different conclusions from the original excavation reports. For example, Walsh cites Jajce (a site he improbably identifies as both built and destroyed within the period 301-330 CE) as a case of targeted destruction; instead, the apparent fragmentation and dearth of roof tiles may suggest post-antique quarrying, while damage to the relief occurs at the points that project the furthest (Mithras’ head and hands), exactly where one might expect damage from collapse. Was the mithraeum at Jajce really violently destroyed in late antiquity, or simply pillaged long after its collapse?
Walsh’s arguments about the timing, nature, and causes of the end of Mithras-worship also rest heavily upon chronological correlations that are far less robust than Walsh’s text and distribution maps might suggest. Of the 37 mithraea he catalogues, the “ends” of their use (and, in many cases, their supposed late-antique foundations) are based primarily or entirely on coin evidence in 21 cases. Without firm archaeological context, these coins are indices not of a sanctuary’s foundation or abandonment date, but rather of one particular rite (coin deposition) and the supply of new coins to the areas in question, which—as Walsh acknowledges—coincidentally drops off sharply at the same time coin sequences in mithraea seem to end. At least four other mithraea have no published evidence that allows their end to be dated.
The brightest spot may be that the types of legacy data upon which Walsh largely relies are currently being supplemented by a host of new data coming from recent, scientific, contextual excavations that will allow a more robust testing of Walsh’s various hypotheses. One common thread from these excavations—published discussions of which are, in most cases, forthcoming4—is the relative vitality of Mithras-worship through the fourth and into the fifth century CE. The Mithraeum of the Colored Marbles at Ostia seems to have been constructed in the last quarter of the fourth century, and continued in use through the early fifth century, with worshippers engaging in archaeologically detectable rites reminiscent of those practiced centuries earlier; the mithraeum at Kempraten was substantially rebuilt after 388; the mithraeum at Angers not only seems to have continued in use through the early fifth century, but may have included foederati among its worshippers. This picture of continued investment in maintaining sanctuaries, and the degree to which we can recognize continuities or minor elaborations in ritual practice, might generate a rather different model of Mithras-cult in late antiquity than the one Walsh paints here.
Ultimately, though, whether Walsh’s boldest arguments will hold true as excavations publish their data may matter less than the way that Walsh successfully re-frames the questions surrounding the end of Mithras-worship. The disappearance of Mithras-worship and the closure of each Mithraic sanctuary can no longer be indiscriminately ascribed to the rise of Christianity. Instead, scholars and excavators will have a much richer set of causal mechanisms that they can explore and test thanks to Walsh’s work. His book charts new directions in the study of Mithras-worship, and subsequent work on the cult will benefit mightily from engaging with Walsh’s novel framework and ideas.
1. R. Gordon, "Mithraism and Roman Society," Religion 2 (1972): 92–114.
2. R. Beck, The Religion of the Mithras Cult in the Roman Empire: Mysteries of the Unconquered Sun (Oxford, 2006).
3. Indeed, Robert Turcan goes so far as to suggest that the grade system often seen as canonical in modern accounts of Mithraism was largely a localized phenomenon confined to Italy: “Hiérarchie sacerdotal et astrologie dans les mystères de Mithra,” in R. Gyselen (ed.), La science des cieux: sages, mages et astrologues (Leuven, 1999), 249-261.
4. M. McCarty and M. Egri (eds), The Archaeology of Mithraism: New Finds and Approaches to Mithras-Worship, BABesch Supplementary Series (Leuven, forthcoming).