Understanding the Status of the Cult of Mithras in the Tetrarchic Period: A Socio-Archaeological Approach

A number of inscriptions from the Tetrarchic period indicate that various Mithraic communities received the patronage of governors, duces and even the emperors themselves. Such support for a so-called mystery cult is striking given that in other cases the imperial government failed to restore the temples of more traditional cults that had fallen into decay. In this paper I will demonstrate that such anomalous building activity associated with mithraea is evident in proceeding generations, for while the construction and repair of temples generally declined significantly in the third century mithraea continued to be erected and maintained unabated throughout this period. I then go on to argue that there is evidence to indicate Mithraic congregations took the conscious decision to divide upon reaching a certain number followers, hence why newly constructed mithraea continued to remain small in size and are often to be found in close proximity to pre-existing mithraea. By applying sociological theory, it is clear that such circumstances provided the ideal situation in which to foster a particularly high level of commitment among the Mithraists, thus making them willing to continue contributing to the construction and repair of mithraea even when resources became scarce, while other cults that had a popular, but less committed following, saw their temples fall into disrepair. When the imperial government provided for support for various Mithraic communities in the Tetrarchic period this was not due to any religious reasons, but rather as a political move designed to channel this commitment into support for their own rule.


Introduction
In A.D. 307 the Tetrarchic emperors met for a conference at the town and legionary fortress of Carnuntum on the Danube frontier.During the course of their stay they restored a Mithraic temple and erected an altar to Mithras which referred to the deity, who was never part of the offi cial pantheon of Roman gods, as the 'Protector of the Empire' (CIMRM 1698).Around the same time, the fellow Pannonian town of Gorsium was reconstructed, having been destroyed in the late third century.Gorsium had been the centre of the Imperial Cult in the province, with a large temple precinct that had contained temples erected by various emperors (Fitz 1998a).Yet the Tetrarchs chose not to rebuild any of these temples, preferring instead to mine them for spolia to use in the construction of new secular buildings and, to date, no temple from this period of occupation at Gorsium has been found (Poulter 1992: 110-112).The contrasting attitudes expressed towards these cult buildings appear to have been repeated elsewhere during this period, with further inscriptions attesting to the patronage of Mithraic communities by senators, governors, and military commanders (CIMRM 140, 360, 1431(CIMRM 140, 360, , 1614(CIMRM 140, 360, , 2280)), while the evidence for the construction and restoration of other temples and shrines remains limited (Lavan 2011: 42-43).
In this paper I will attempt to demonstrate why certain elites came to act as patrons of a socalled mystery cult while seemingly ignoring more traditional alternatives during the Tetrarchic period.I will begin by providing a brief introduction to the cult of Mithras, after which I shall outline how the evidence for the construction and restoration of mithraea in various western provinces during the third century compares to the wider patterns exhibited by temples.I will then argue that Mithraists made the conscious decision to divide their congregations when they reached a certain number, rather than allowing them to continuously expand.It will then be demonstrated, by applying sociological theory, that by focusing on small congregations the cult of Mithras created an environment that fostered a signifi cant degree of commitment from its followers.I believe that this high level of commitment, at least in part, manifested itself in the continued construction and repair of mithraea in the third century and that this was recognised by the Tetrarchs and other elites who subsequently acted as patrons of Mithraic communities in order to channel this commitment into support for themselves.

The Mysteries of Mithras: A Brief Introduction
The Mysteries of Mithras was an all-male cult active in the Roman empire between the fi rst and early fi fth centuries A.D. Evidence for Mithraic worship has been found across the Roman world, from Syria to Hadrian's Wall, but the cult appears to have been far more common in the West than in the East or North Africa, with signifi cant concentrations of material uncovered on the Rhine-Danube Limes, as well as in Rome itself.From the fi rst to third centuries A.D., initiates appear to have consisted mostly of soldiers, customs-offi cials, and freedmen, with no evidence of elite participation until the fourth century (Clauss 1992).Little textual evidence regarding the cult survives, but archaeological and epigraphic records have yielded a signifi cant amount of information that allows for a partial reconstruction of what the cult of Mithras entailed.Members appear to have held one of seven grades, which in descending status consisted of: the pater, heliodromus, perses, leo, miles, nymphus and corax (Clauss 2000: 131-140).There is no evidence for any Mithraic festivals or processions taking place in public, with adherents appearing to have carried out all their rituals behind closed doors inside their temples (mithraea).These rituals included: initiation ceremonies, which involved blindfolding and binding the hands of the initiate who would then undergo various trials, such as having a torch, sword and a bow and arrow pointed at his face (Gordon 2009;Beck 2000: 152); blood sacrifi ces, which look to have mainly consisted of birds (Lentacker, Ervynck, and Van Neer 2004: 72); offerings of incense (Bird 2007); and cult feasts (Lentacker, Ervynck, and Van Neer 2004).Mithraea were frequently small and nearly all conform to the same plan (Fig. 1) featuring an unassuming entrance that did not align with the main axis of the structure, one or two anterooms, and a window-less inner-shrine.This inner-shrine consisted of a tripartite plan that included a central nave fl anked by two parallel benches that could usually accommodate around ten to fi fteen people (Volken 2004: 2), with the cult-relief placed on a podium or in a niche at the terminus of the nave (Clauss 2000: 42-61).

Mithraea and Wider Temple Building in the Third Century A.D.
In recent years, various studies have demonstrated that temple building in many western provinces went into decline during the third century.In Rome, only two new temples were founded after the reign of Severus Alexander (Richardson 1992: 456).Across the rest of Italy, 61 temples have provided evidence of building activity in the second century, but only 11 exhibit such evidence in the third century (Jouffroy 1986: 320), while in Gaul, just eight Romano-Celtic temples were built in the third century, compared to 38 in the second century (Fauduet 1993: 118-120).In Noricum and Pannonia, I have only been able to fi nd fi ve temples that were the subject of building work after the Severan period (Walsh, forthcoming).On an empire-wide level, the temples of Isis and Sarapis, another mystery-cult, also indicate a signifi cant decline in investment in the third century, with not a single iseum or serapeum erected in the Roman world after the Severan period (Wild 1984).
It is important to note that the changing fortunes of temples correlates with a wider decline in the construction and restoration of public buildings in the western provinces during the third century (on Italy see Jouffroy 1986: 321; on Gaul see Bedon, Chevallier, and Pinon 1988; on Noricum and Pannonia see Walsh, forthcoming).To explain why this decline occurred would require a great deal of space so I shall not go into detail, but suffi ce to say one of the major reasons appears to have been a lack of investment by local elites from the post-Severan period onward.Previously, these men had contributed signifi cantly to the urban landscape in order to further their own standing via competition between themselves and with rival cities, while erecting public buildings was also part of their duties when acting as local magistrates (Patterson 2006, 185).However, from the third century forward, such men look to have sought new ways of increasing their social status without carrying the burden of this expenditure (Ward-Perkins 1984: 3-37).Of course, the lack of investment by local elites was not the sole reason for these changing patterns of building activity, with other aspects of the so-called 'Third Century Crisis', such as political turmoil, civil war, economic instability (Greene: 1986: 57-61), andplague (McNeil 1976), also contributing to a lack of resources with which to build and restore civic buildings and temples.
How did mithraea fare in all of this?Rather than exhibiting a similar decline, the construction and restoration of mithraea continued unabated in many areas of the West into the early fourth century.In Italy, excluding those in the city of Rome, nine mithraea exhibit evidence of building activity in the second century, while in the third century sixteen mithraea were the subject of construction and repair (Jouffroy 1986: 334, 337, 349, 357;White 2012: 441-443).In the city of Rome, most of the extant mithraea were either initially constructed, or exhibit some degree of restoration, in the third century (Bjørnebye 2007: 25-51).Of the known mithraea in Gaul, three were built in the second century and four were constructed in the third century (Walters 1974;Gaidon-Bunuel 1991;Luginbühl, Monnier, and Mühlemann 2004;Martens 2004; Wiblé

2004
).In Noricum and Pannonia, in the latter half of the third century, mithraea provide more evidence for construction and restoration than all other temples in these regions combined (Fig. 2).
In Table 1 the sizes of urban mithraea constructed from the end of the Severan period to the turn of the fourth century are provided.The average size of a newly built mithraeum remained relatively small, covering an average area of just c. 65 m², with the London mithraeum the only example larger than 100 m².That mithraea continued to be constructed with relatively small dimensions is quite unusual, for while the offi cial Roman deities had long been worshipped in large well-adorned temples, even the temple precincts of other mystery-cults, such as those of Magna Mater and Isis, had by the late second century reached monumental proportions.For example, the temple precinct of Isis at Savaria in Pannonia covered 2940 m² (Fitz 1998b), just one area of the Iseum complex on the Campus Martius in Rome covered 18200 m² (Wild 1984(Wild : 1750)), and the Campus of Magna Mater at Ostia covered c. 4500 m² (Nielsen 2014: 90).Additionally, unlike the temples of Magna Mater and Isis, those of Mithras do not exhibit any considerable variation in their topography (for a comparative study of mystery cult shrines see Nielsen 2014), with all the mithraea constructed in this period observing the established plan outlined above.Furthermore, just as with previous mithraea, there is no indication natural light was allowed to shine into the inner-shrines of these temples.

Limiting the Size of Mithraic Communities
There is evidence to suggest that the reason newly constructed mithraea continued to remain small in size was because Mithraic congregations took the conscious decision to divide when they reached a certain number of members rather than continuing to expand.To begin with, two Mithraic albums from Virunum in Noricum inform us that a Mithraic congregation, which was active in the town around the turn of the third century, had reached almost 100 members when the fi rst album was produced; but by the time the second album was published a few years later, it appears that the group had split to form two separate congregations of roughly equal size (Piccottini 1994).Further epigraphic evidence of such division is not extant, but in many cases we fi nd mithraea which were erected in towns and settlements that already contained numerous pre-existing mithraea: in Rome, sixteen mithraea have been found (Griffi th 1993), with another sixteen identifi ed at Ostia (White 2012); fi ve have been discovered at both Aquincum (CIMRM 1742(CIMRM -1772;;Póczy 2005: 215) and Poetovio (CIMRM 1487-1618; Vomer-Gojkovič and Kolar 1993); four are known from Nida-Heddernheim (Huld-Zetsche 1986); and three at Carnuntum (CIMRM 1664-1696).It is unlikely that these Mithraic congregations existed entirely independently of each other, particularly in frontier towns that were not especially large, yet they evidently preferred to erect small temples that could facilitate limited numbers rather than construct a mithraeum that could contain them all at one time.Furthermore, in other cases, we fi nd evidence for mithraea that were used by communities that appear to have been much larger than their dimensions would suggest, indicating that the congregations preferred to share the existing mithraeum between themselves rather than expand the temple to hold them all simultaneously.At Tienen, evidence of a cult meal held sometime during the latter half of the third century, which could have fed hundreds of people, was discovered next to a mithraeum.However, the mithraeum covered an area of only c. 90 m², which was nowhere near large enough to accommodate such numbers (Martens 2004: 28-45).While it is possible the feast was held outside the mithraeum, given what we know of the Mithras cult and the evident importance it placed on privacy, it is perhaps more likely that small groups participated in numerous feasts within the mithraeum over the course of several or more days.At Trier, a similar situation may have existed, as a number of lamp moulds with Mithraic imagery were recovered from a pottery factory, suggesting mass production for a Mithraic following that was larger than the small dimensions of the mithraeum would suggest (Gordon 2004: 269-270).If these mithraea were indeed shared by separate congregations as part of a larger community, then it is possible that such arrangements occurred in other locations where only one mithraeum has been discovered.Overall, through a combination of this range of evidence it seems that Mithraic congregations usually sought to limit their numbers, and the size of their shrines appears to be a refl ection of this.This has important implications regarding the levels of commitment generated in Mithraic congregations and it is to this we now turn.

The Benefi ts of Small Congregations and Temples
In order to understand how maintaining a limit on congregation size would result in mithraea continuing to be built and restored while other shrines fell into disrepair we must now take a sociological perspective.To begin, numerous sociological studies of religious movements have demonstrated that commitment is much higher in smaller congregations than in their larger counterparts, as the former have a greater social density (Pinto and Crow 1982;Finke 1994;Zalenski and Zech 1995).Social density, put simply, is a case of maths: in a congregation of 10, nine relationships are available to each member, which would give a combined total of 81 potential relationships between all members; in a congregation of 20 there are 19 relationships available to each member, with 361 possible relationships overall; for adherents in a congregation of 30 there are 29 available relationships available to each member, with 841 potential relationships in total, and so forth.As a result, for an adherent to know his fellows well in a congregation of 30 is more diffi cult than in a congregation of 10.That people in smaller congregations do indeed feel greater affi liation among their fellows than those in larger congregations has been demonstrated in modern surveys, such as the example provided in Table 2. Given that most Mithraic congregations would have fallen into the smallest size-bracket of this survey it is possible that many of the Mithraists could have been particularly close to their fellow initiates in their day-to-day lives as well as in the cult setting.This is given further credence by the fact that, as stated, many of its members were soldiers and customs-offi cials, men who would have spent a lot of time together in small units.As a result, even though the cult of Mithras did not impede its members from joining other cults, exposure to other practices and beliefs that might distract or lessen their commitment to the Mithras cult would be limited.Additionally, the high level of social density would mean that even those of the lowest grade would have a close relationship with the pater, thus all members could be monitored for anything that might be deemed 'deviant behaviour' that could have adverse effect on the solidarity of the congregation.
The limited size of Mithraic congregations would also have served to generate a strong sense of involvement among members during ritual practice.It has been demonstrated that participation in rituals plays a considerable role in fostering commitment as they continually reinforce confi dence in the central beliefs and ideals of the group (Stark and Finke 2000:108).However, as congregations become larger it is often harder to retain the inclusiveness of the ritual practices as more people have to be accommodated for, which can have an adverse effect on the level of commitment.A rational person, if possible, will participate minimally in collective efforts if they feel they can still reap the same rewards as they would if they exerted themselves to the maximum (Olson 1965;Hetchner 1987: 27).Such behaviour is often to be found in religious movements.For example, some people only go to church on Christmas and Easter and do not attend for the rest of the year, but they would still describe themselves as Christian and hope to receive the same rewards (i.e.forgiveness for sins, life after death) as those who attend every week and keep the church going in their absence.Those who seek to gain the benefi ts of cult membership while providing little or no contribution to it are referred to as 'free-rider' members.The larger the congregation, the greater the number of 'free-riders' as there is an increased chance of people failing to contribute, which subsequently will have a negative impact on the overall commitment.In contrast, in congregations where everyone is expected to contribute continuously 'free-riders' are eliminated and thus the average level of commitment remains relatively high (Stark and Finke 2000: 147-150).
Evidence indicates that Mithraic rituals would have been highly inclusive, as each Mithraic grade had certain duties he was expected to fulfi l during these events.For example, in a relief depicting the ritual meal from Dalmatia, the image appears to show the pater and heliodromus, the two highest grades, being served by members of the leo and corax grades, while another initiate holds a torch lighting the scene (CIMRM 1896).The mosaic that decorates the Felicissimus mithraeum in Ostia also depicts various items associated with the grades, such as a fi re-shovel with the leo, a sword with the miles, and a lamp with the nymphus (Clauss 2000: 133-137).That these items were indeed used during Mithraic rituals has been demonstrated via the archaeological record: swords have been found at the Riegel (Cämmerer 1986) and Königshoffen mithraea (CIMRM 1373); a great number of lamps are present in most excavated mithraea (Clauss 2000:120-130); and fi re-shovels were found at the Carrawburgh (Richmond, Gillam and Birley 1951: 20, 87) and Santa Prisca mithraea (Vermaseren and Van Essen 1965: 144).Additionally, an inscription from the Santa Prisca mithraeum in Rome refers to leōnēs offering incense .Given that the small size of Mithraic congregations would have meant that few could occupy the same grade at the same time, each member of the community would have had either a unique role to play during rituals, or at least a role that was shared among only a couple of others.In contrast to Mithraic rituals, the public celebrations of other cults may have involved more people, however, this could have also created greater scope for 'free-riders', which the Mithraic community seemingly sought to avoid.
Furthermore, the architecture of mithraea would have also contributed signifi cantly to the impact of certain rituals on the participants, particularly those experiencing them for the fi rst time.One major aspect of Mithraic architecture was the prevention of natural light from entering the inner-shrines, which meant light effects could be used to great effect during Mithraic rituals.Indeed, such effects appear to have been commonplace in mithraea.At Carrawburgh on Hadrian's Wall, an altar depicting Sol was hollowed out, allowing a lamp to be placed inside, thus illuminating the relief (CIMRM 847).A similar effect appears to have been produced with an altar found in the Koenigshoffen mithraeum in Germany (CIMRM 1367), and a relief from Bilijanovac in Macedonia has various apertures to allow light to shine through from behind (CIMRM 2202).Given the small dimensions of mithraea the sudden appearance of Mithras or Sol radiating light in the darkness would have been all the more impressive, for it would have lit up much of the inner-shrine -like the sun appearing in the middle of the night.In the Mithraic initiate's heightened state, having been blindfolded, stripped, and their hands bound, this would have seemed a mystical experience.Such instances, even with no actual involvement of the divine, still serve to reaffi rm people's belief of their religious explanations (Howell 1997).Furthermore, given that the majority of people who experience 'miracles' are normal, sober people -said 'miracles' appear more plausible when relayed to others.As a result, these events enhance the belief of the group as well as that of the individual who experienced it (Stark and Bainbridge 1997).
Finally, all religious bodies exist to some degree in tension with their sociocultural environment, with tension referring to 'the degree of distinctiveness, separation, and antagonism between a religious group and the outside world' (Stark and Finke 2000: 143).Thus, the focus on small, private temples may have created a level of tension between the Mithras cult and wider Roman society, a situation that would have helped strengthen the commitment of the Mithraic adherents.At the high-end of this scale, cult doctrine can affect members' everyday lives in all manner of ways, such as clothing, diet or who they associate with, thus making them overtly different to 'normal' society.Therefore, to be part of a cult at the higher end of tension spectrum is more expensive, i.e. 'the material, social, and psychic costs of belonging to a religious group' (Stark and Finke 2000: 144).Roman history contains numerous examples where groups were attacked or banned because they met in secret and their practices remained hidden behind closed doors.One notable example of such paranoia can be seen in the occasional persecution of Christian communities, who, like the Mithraic initiates, would meet in private to carry out their ritual practices, while for similar reasons, many voluntary associations also found themselves under suspicion from the government and were often the subject of restrictions or forced to disband (Cotter 1996).Thus, it is likely that Mithraic congregations may have come under suspicion from their neighbours or government offi cials.However, that they appear to have largely continued without interference could indicate the cult had obtained a state of relative equilibrium in its tension with society.This can perhaps be explained by the context of these third century mithraea.Mithraea erected in Rome and Ostia in the third century were to be found in more prominent locations than their predecessors, either in, or close to, large public buildings (Coarelli 1979;White 2012).Those erected elsewhere follow a similar pattern, with the mithraea at Trier (Ghetta 2008: 82-96) and Lentia (Karnitsch 1956) installed in temple precincts, and the Tienen mithraeum found in an industrial area surrounded by workshops (Martens 2004: 26-28).Thus while small groups of Mithraists still met in private during the third century, their meeting place and the identity of the Mithraists themselves would have been more readily apparent.Mithraism may have been a mystery-cult, but in the third century it was a mystery-cult in plain view.
The evidence outlined here demonstrates the cult of Mithras created an environment that would inspire a signifi cant level of commitment from its followers.But how did this manifest itself in the continued construction and repair of temples?The most obvious way would be via monetary donations from congregation members.However, fi nancial support is not the only thing that cults gain from their adherents when they are highly committed; they can also call on their members to donate other things, such as time and energy.In the modern world, the Mormon Church follows a similar pattern of group dynamics to that outlined here for the cult of Mithras.Mormons intentionally limit the size of their congregations, usually to around 300 members, while the group has also remained in signifi cant tension with society due its strict regulations on membership.As a result, this has resulted in high levels of commitment among Mormon congregations (Stark and Finke 2000: 155-157, 212).In correlation with this, it has been calculated that on average 400-600 hours of voluntary service are undertaken by Mormons per week in support of their local Ward.The majority of these tasks are not 'religious' in nature, but often mundane responsibilities such as janitorial and clerical duties (Stark and Finke 2000: 150).This is not an isolated example, with contributions in such a manner being a common trait in religious groups that consist of small, high-tension congregations (Iannaccone, Olson, and Stark 1995).Given their high levels of commitment we can expect Mithraists were much the same and that they were unlikely to have supported the maintenance of mithraea through monetary donations alone, but perhaps also undertaken the construction and restoration themselves by providing manpower, expertise, and materials.Given that mithraea were neither particularly large nor monumental, and all consisted of the same basic plan, it would not have not been beyond the means of a highly committed group of limited numbers to undertake such work.Indeed, in another modern parallel, the Kingdom Halls of Jehovah's Witnesses are built almost entirely by work forces made up of volunteers from their congregations and usually take just a few days to construct (Stark and Iannaccone 1997: 148).As a result, when access to resources became more diffi cult and building activity declined in the third century, mithraea continued to be erected and repaired as the highly committed Mithraists were willing to contribute to the upkeep and construction of mithraea via whatever methods were available to them.

Conclusion
Mithraic congregations were likely to have consisted of a small number of adherents who would have known each other very well, almost certainly to the extent they would have been close in the secular world too, thus minimising any external infl uences that might be detrimental to their commitment.Limiting numbers also meant all members had duties to fulfi l during ritual practices, with little room for 'free-riders' or deviant behaviour.The small dimensions of mithraea also meant that the impact of ritual practices could be considerable, with light effects creating sudden bursts of light in the darkness.That the Mithraists met in private to conduct their rituals would have also put them in tension with the norms of Roman society, further serving to reinforce the commitment of the congregation.In contrast, while the monumentality of their temples and the excitement of their festivals may have impressed onlookers, such attributes did not necessarily generate signifi cant commitment among the followers of other cults.Thus during the third century, when the resources became scarce and the appearance of new temples became increasingly rare, due to their signifi cant commitment Mithraic congregations were likely to have continued contributing to the construction and repair of mithraea in whatever way they could, whether that be through fi nancial support, the provision of materials, or man power.When the Tetrarchs rebuilt Gorsium, they did not reconstruct the temples as such practice was no longer the norm and they evidently felt there was little reason to change this.Yet when they arrived at Carnuntum in A.D. 307, the situation regarding the mithraeum would have been different, as a highly committed group was still using the temple for their ritual practices.As a result, the restoration of this mithraeum, as well as the patronage of other Mithraic communities by senators, governors, and military commanders in this period, should be seen as a politicallymotivated move by which the elites sought to channel the commitment of the Mithraic adherents into support for their own rule.
School of European Culture and Languages, University of Kent

Figure 1 :
Figure 1: The standard plan of a Mithraeum.

Table 2 :
Questions to congregations in Northern California 1963: 'Of your fi ve closest friends, how many are members of your congregation?' and 'All in all, how well do you think you fi t with the group of people who make up your congregation?' (afterStark and Finke 2000, Table 10).