Under the Judgement of the Living God: The Early Christian Funerary Imprecations of Phrygian Eumeneia

Abstract

Since the late nineteenth century, the Phrygian funerary imprecation, known as the Eumeneian formula, has been considered one of the clearest indicators of Christian religious identity on inscriptions from Roman Asia Minor. After a brief précis of the early scholarly history of interpretation of this formula and the historiographical assumptions which underpinned its identification as Christian, this article attempts to reframe how we understand the specific context out of which these inscriptions emerge—that of the wider socio-religious context of Roman Asia Minor—and to examine the degree of continuity which these inscriptions have in terms of religious sentiment with similar pagan examples from wider Anatolia. The central contention of this article is that the Eumeneian formula inscriptions, quite apart from what they can tell us about the socio-political status of early Christians and their relationships with their wider civic environment, are also an important index for understanding early Christian popular religion in the pre-Constantinian period and how ordinary Christians expressed their religious identity in a potentially hostile environment.

1. Introduction

The small city of Eumeneia, situated in south-central Phrygia, arguably made little contribution to the wider history of the Roman world. It was the scene of no great battle, no famous treaties were signed there, it was the site of no particularly memorable monuments or architectural wonders, and it certainly inspired no literary figures of any note. In fact, to many scholars of the ancient Mediterranean, Phrygian Eumeneia may even be so close to anonymity as to be confused with either the Delian city or the festival that bore the same name. However, in histories of early Christianity and its geographic spread, Eumeneia always warrants at least a footnote, sometimes even a whole sentence, for its one claim to fame. That is, by the mid-third century C.E., Eumeneia provides the most extensive, and publicly displayed epigraphic evidence for a thriving early Christian community, which, during this same period, seems to have achieved a marked degree of civic integration, if not predominance (see esp. McKechnie 2019, pp. 210–31; Mitchell 1993, pp. 40–41; Ramsay 1897, pp. 484–568).1

Since the late nineteenth century, scholars have been nearly unanimous on this point. With specific mention of Eumeneia, Adolf von Harnack (1905, p. 358) noted the “very wide extension of Christianity throughout Phrygia”, whilst William Frend (1984, p. 445) was on firm ground when he asserted that “Phrygia must have been one of the most Christianized provinces of the empire by the end of the third century”. Robin Lane Fox (1986, p. 295) wistfully mused, “If Christians in other cities had only put up as many inscriptions as these confident figures in Eumeneia, what curious diversity might we not discover?” Stephen Mitchell (1993, p. 40) expressed similar sentiments when he noted that “Christianity flourished in the cities of southern Phrygia during the third century”, further suggesting that Eumeneia was its “main centre”. Finally, more recently, Graeme Clarke (2005, p. 614) went as far as to suggest—plausibly, but hypothetically—that the Christian population of Phrygia stood between 20 and 30%, adding that “in some instances this went much higher, most notably in the upper Tembris valley and in Eumeneia and its surrounding district”.

The primary data for these claims about Eumeneia are not composed, however, of unambiguously identified Christian inscriptions—like those of the Upper Tembris Valley, where both those memorialized and those setting up the dedications boldly confessed the nomen Christianum—nor by the literary and epigraphic evidence for the Montanist movement, which thrived in Phrygia from the late second century, but by a collection of close to one hundred inscriptions bearing variations on a funerary imprecation warning a potential tomb violator that “he will reckon with God” (ἔσται αὐτῷ πρὸς τὸν θεόν).2 This formula, dubbed by the Scottish archaeologist Sir William Mitchell Ramsay (1851–1939) with the toponymic the “Eumeneian formula”, occurred with a series of variants, a number of which are (arguably) more indicative of its Christian content and worth briefly surveying here (Ramsay 1897, vol. 2, p. 496).3

Drawing on the original collection of examples published by Ramsay in the 1890s, one can see that a significant number use the variation ἔσται αὐτῷ πρὸς τὸν ζῶντα θεόν. “he will reckon with the Living God”, including one inscription belonging to a bishop (ed. Ramsay 1897, vol. 2, no. 362), which reads in full:

Δαμᾶς Διοτείμου κ[α]τεσκεύασεν τὸ ἡρῷον τῷ μήτρωνι Μητροδώρῳ ἐπισκόπῳ καὶ τῷ πατρί μου Διοτείμῳ καὶ ἑαυτ[ῷ]· εἴ τις δὲ ἐπιχειρήσει θεῖνε ἕτερόν τινα, θήσει ἰς τὸ ταμεῖον προστείμου (δην.) φʹ εἰ καταφρονήσει τούτου, ἔστε αὐτῷ πρὸς τὸν ζῶντα θεό[ν].

Damas, son of Dioteimos erected the shrine for [my] mother, for the bishop Metrodoros and for my father Dioteimos and for himself. If someone would lay hands to bury some other, he will give 500 denarii into the treasury. If he would despise this [warning], he will reckon with the living God.

[Trans. Doherty]

Another utilizes an early variant of the Chi-Rho Christogram to substitute the name “Jesus Christ” in place of the usual invocation of God (ibid., no. 371). Yet another concludes with the variant ἔσται αὐτῷ πρὸς τὸ μέγα ὄνομα τοῦ θεοῦ “he will reckon with the Great name of God”, which is seen by some as a circumlocution for the Tetragrammaton (ibid., no. 369):

[τὸ ἡρῷον καὶ] τ̣ὸν ἐπ’ αὐτοῦ βωμὸν κατεσκεύασεν Aὐρ. Ζω̣τικὸς βʹ τοῦ Παπ[ί]ο̣υ Εὐμενεὺς ἑαυτῷ καὶ τῇ γυναικὶ αὐτοῦ Aὐρηλίᾳ Ἀπφίῳ καὶ τῷ ἀδελφῷ αὐτοῦ Ἀ[μ]μια̣[ν]ῷ κα[ὶ εἰ] τιν[ὶ] [ἀλλ]ῷ α̣[ὐτ]ὸ̣ς̣ ζῶν [συγ][χω]ρησει· οὐδενὶ δ[ὲ] [ἀ]λλῷ ἐξὸν ἔσται θεῖναι τινὰ· εἰ δέ τις ἐπιχειρήσει, εἰσοίσει ἰς τὴν Εὐμενέων βουλὴν προστείμου (δην.) ͵αφʹ καὶ ἔσται αὐτῷ πρὸς τὸ μέγα ὄνομα τοῦ θεοῦ.

Aur[elius] Zotikos son of Zotikos son of Papias, Citizen of Eumeneia, erected the shrine and the altar upon it for himself and for his wife Aurelia Apphia and for his brother Ammianos and if some other having agreed while still living (i.1897 arranging for themselves to be buried here too). And none other will it be lawful to bury. But if someone would lay hands upon [this tomb], he will pay 1500 denarii into the public treasury of the council of Eumeneia, and he will reckon with the Great Name of God.

[Trans. Doherty]

A final, sadly fragmentary, inscription specifies that this “reckoning” will be both now and on the day of Judgement (nos. 353–54) (ἔσται αὐτῷ πρὸς τὸν ζῶντα θεὸν καὶ νῦν καὶ ἐν τῇ κρισίμῳ ἡμέρᾳ), an eschatological idea shared by Christians and Jews.4

[—{²ὁ δεῖνα}²—][—ἑαυτῷ—] καὶ τῇ μητέρι Μελτίνῃ καὶ τῷ υἱῷ Γαΐῳ καὶ τῷ ἀδελφῷ μου [Ἀσ]κλᾷ ἑτέρῳ δὲ οὐδενὶ ἐξέσται τεθ{ν}ῆναι {τεθῆναι} χωρὶς τῶν προγ[εγρ]αμένων ὃς δὲ ἂν ἐπιτηδεύσει, ἔσται αὐτῷ πρὸς τὸν ζῶντα θεὸν καὶ νῦν καὶ ἐν τῇ κρισίμῳ ἡμέρᾳ.

[A certain man for himself] and for [his] mother Meltine and for [his] son Gaius and for my brother [As]kla and it is unlawful to inter another, apart from those registered [above], whoever would do this, he will reckon with the living God both now and in the day of Judgment.

[Trans. Doherty]

The cumulative weight of this evidence—surveyed at greater length by Calder (1939), and most recently McKechnie (2019)—has meant that since at least the 1890s this formula has generally been considered a strong indicator of the Christian religious identity of the deceased and, despite its seemingly pithy nature, reveals a great deal about the enculturation of Christianity in third century Phrygia (see McLean 2002, p. 273 for a helpful survey of the relevant literature).

Scholars, then, have long been happy to grant Eumeneia and the wider cultural region of Phrygia this passing distinction, and normally to footnote one of the standard discussions of the Eumeneian formula by Ramsay (1897), his pupil and fellow Scot William Moir Calder (1881–1960) (1939), or the French epigrapher Louis Robert (1904–1985), (Robert 1960), without much further elaboration on its wider religious significance. However, when we stop to consider the matter further, in concert with other factors that can be assembled about third-century Eumeneia and Phrygia more generally, it is worth pausing to give some thought to what the religious implications of these inscriptions are for the study of early Christian popular religion in Phrygia and the ways in which “ordinary Christians” (MacMullen 1990) publicly displayed their religious identity in a potentially hostile socio-religious environment. As most recently reiterated by McKechnie (2019, p. 218), most datable inscriptions utilizing the Eumeneian formula can be safely dated to “a narrow chronological window” of around a generation between the mid-third century through to the early fourth century. This securely places the inscriptions within a period both of significant Christian growth (see Stark 1996, pp. 3–27; Hopkins 1998; MacMullen 1990, p. 156 for different assessments) but also within a period when the most intense persecution of Christians was evident. (see Doherty 2017, focusing there on social and political factors).

In this article, I seek to achieve two modest aims. First, I will briefly recount the early scholarly history of identifying these inscriptions as Christian and some of the historiographical assumptions that underpinned scholars’ claims regarding this issue, and then assess whether these claims still stand up to scrutiny. Second, and more hypothetically, I will attempt to reframe how we understand the specific context of these Christian inscriptions within the wider socio-religious context of Roman Asia Minor, refocusing attention on to the level of cultural continuity displayed in these Christian inscriptions when compared with those of their dedicators’ pagan neighbours. My central contention will be that the Eumeneian formula inscriptions, quite apart from what they can tell us about the socio-political status of early Christians and their relationships with their wider civic environment, are an important index for understanding one aspect of early Christian popular religion, which Ramsay MacMullen (2009) recently dubbed “the second church” (cf. van den Broek 1979). The Eumeneian formula is an important expression of everyday “Christianness”, as the coinage of Rebillard (2012) has it, or perhaps even, to use the problematic evaluative category of Williams (2023), “cultural Christianity”.5

2. Background

Almost as interesting as the inscriptions themselves, the historiography of the Eumeneian formula is worthwhile in its own right and involves intriguing and involved debates between some of the most important historians of Antiquity over the last two hundred years (see Frend 1996, 2003). These debates, some of which I will only signpost to here, were indicative of various wider trends in the historiography of early Christianity and, as such, worth revisiting, if only as a salutary lesson in the importance of understanding past historiographical trends in their historical context.

While the Eumeneian formula had been remarked upon by European travellers and antiquarians for centuries, it was only really in the second half of the nineteenth century that scholars began to pay it the attention it deserved as part of the wider exploration of Asia Minor, which followed on from the imperial ambitions of European powers. Three scholars stand out in importance here: William Mitchell Ramsay, the French Abbé Louis Duchesne (1843–1922), and the Belgian historian of religions Franz Cumont (1868–1947). Ramsay and Duchesne, independently of one another, arrived at the conclusion that these inscriptions were Christian in the 1880s.

The Francophone historian and priest Louis Duchesne (1883, p. 31) first wrote of the Eumeneian formula in an article dealing with the famous Abercius inscription in the relatively conservative Revue des Questions Historiques, noting it as a common example of other Christian inscriptions from the vicinity of the Abercius find, where “l’envahisseur de la sépulture est menacé de la vengeance de Dieu”. Duchesne (1895) elaborated on this just over a decade later when he returned to the topic in the Mélanges de l’École Française de Rome, by then with far more confidence regarding its Christian identification. While Duchesne was something of an amateur when it came to archaeology, his instincts here were sound, though neither the Eumeneian formula nor Phrygia found much of a place in his later monumental Histoire ancienne de l’Église (1906, using trans. Duchesne 1909, vol. 2). Here, his remarks on Phrygia (aside from his discussion of Montanism) comprise one curious, and rather stereotypical, passage in which he wrote:

Behind Asia Proper, many Christian communities existed on the plains of Phrygia. Phrygia was essentially an agricultural country, and inhabited by a simple and gentle folk; their native rites were of fabulous antiquity and had not been very deeply influenced by Hellenism. They involved great religious assemblies, near celebrated sanctuaries, and noisy, exciting ceremonies, presided over by wild and fanatic priests, Galli and Corybantes (priests of Cybele), whose religious frenzies were world famous.

(Duchesne 1909, pp. 190–91; cf. Doherty’s commentary, Doherty 2017, pp. 77–78 on the problematic reconstructions of Phrygian religiosity)

For all his rightly celebrated scholarly erudition, Duchesne’s contribution here was largely indirect, though his name and reputation lent considerable scholarly weight to the arguments of subsequent Francophone commentators, especially Cumont. Despite his service to the critical study of the early church, Duchesne was rewarded with the dubious honour of having his Histoire ancienne de l’Église (though its first volume had received the imprimatur) placed on the Index of Forbidden Books by the Holy Office in 1912, following a vicious and reactionary campaign by sectors of the European Catholic press taking Duchesne to be a sceptical crypto-Modernist (see Frend 2003, pp. 137–43). In reality, Duchesne was more of a historian than theologian and it is probably safe to conclude that, like other scholars at the time, he was largely found guilty by association (for nuanced details, see Hill 2002, pp. 23–31). It was left to the Scotsman Mitchell Ramsay and his colleagues to marshal together a lengthier series of arguments in favour of the Eumeneian formula’s Christian identification.

In something of a historical irony, the Presbyterian Ramsay’s scholarship received a much warmer reception in Rome than Duchesne’s, and he was awarded a gold medal by Pope Leo XIII in 1893 in recognition of his historical work. In the same year as Duchesne published his initial comments, Ramsay and a group of companions were conducting one of what were to become their almost yearly expeditions in Asia Minor, finding themselves spending a considerable amount of time among the ruins of Eumeneia (Işıklı). It was as a result of these journeys and a series of new epigraphic finds that Ramsay (1883, p. 401), writing in the Journal of Hellenic Studies, finally committed to the identification of the Eumeneian formula as Christian. “These examples have decided my opinion”, he noted, “on the point about which I long hesitated—many inscriptions in central Anatolia which end with the curse ἔσται αὐτῷ πρὸς τὸν θεόν, must be reckoned as Christian.” The decisive factor for Ramsay seemed to be what he called the “strange curse” (p. 400) he had uncovered which read ἔσται αὐτῷ πρὸς τὸν ☧, even if the restoration of the key chi-rho (☧) monogram has been disputed (see pp. 433–34; the inscription in question appearing in ibid. no. 371). In his later work, Ramsay was increasingly confident of the Christian identification of the Eumeneian formula, noting in the second volume of his magisterial The Cities and Bishoprics of Phrygia (Ramsay 1897, p. 497), that the Eumeneian formula was “by far the most important” of semi-concealed Christian inscription formulae. Ramsay’s more considered discussion here is particularly interesting, not least for the way in which it sets the tone for later developments.

Two historiographical considerations frame Ramsay’s developed discussion: on the one hand, Ramsay is clear that Christians adopted some of the funerary customs of their pagan forebears, including popular religious sentiments, which he considered problematic,6 and on the other hand, Christians needed to carefully conceal their Christianity in a way that was inoffensive to their pagan neighbours but which did not transgress the boundaries of monotheism. In Ramsay’s phrases (Ramsay 1897, p. 498), the Eumeneian formula was calculated “to avoid at once offence against the common public feeling and outrage to private Christian feeling”. Ramsay’s reconstruction here, written at the end of the nineteenth century, gels with his wider understanding of the nature of persecution at the time. Ramsay, who had written extensively on the question of persecution and debated this with figures like Mommsen, believed in a minimalist interpretation of the extent of persecution and that until the Great Persecution under Diocletian, Christians in Phrygia, as long as they were circumspect, were largely left unmolested (the exception being the Montanists).7 More interesting, for our purposes, were Ramsay’s further comments about the Eumeneian formula (Ramsay 1897, p. 498):

In no respect could it jar on the most susceptible of pagans; and yet it contains an idea, which was rarely expressed by them, while it has been at all times ready to the lips of every person, trained in a Christian society. The pagan so often appealed to their God, but rarely to him as a Judge: they often ask him to punish their enemy, but they rarely ask for a fair treatment according to a reckoning of deserts.

Ramsay’s observation here is underscored by his comment about epigraphic Christians, who were:

…the mass of undistinguished and obscure Christians…this unknown multitude of common persons (who, as a rule, had not the courage and heroism to stand forth prominently as martyrs, or the intellectual power to shine as leaders and teachers), that are revealed to us in the sepulchral inscriptions. They are not represented to us in the Christian literature, except when their errors and backslidings have to be castigated; but if we want to see what Christianity practically was as a working influence in the Roman Empire, these common men are well worthy of some share of the attention that is given usually only to the leaders.

(ibid.)

In these comments, Ramsay touches on what has most often been elided in subsequent discussions, and what I will elaborate on below, the Eumeneian formula is an important example of the popular religion of ordinary Christians as they lived their everyday Christianness in pre-Constantinian Phrygia. Ramsay’s concluding comment (1897, p. 498) bears repeating. “It is a characteristic feature that the pagan form is adopted with the smallest possible change, and the least perceptible modification of its spirit.” While, as I will indicate below, subsequent discussion has focused on whether specific instances of the Eumeneian formula were either Jewish or Christian, it was Ramsay’s comments about cultural continuity that are arguably far more important. Returning to this historiographical overview, by the mid-1890s, another scholar of note lent his scholarly weight to the theory that the Eumeneian formula was a key indicator of Christianity: Franz Cumont.

Writing just over a decade after Duchesne and Ramsay, the Belgian Franz Cumont (1895, p. 252), in a lengthy discussion and catalogue of the Christian inscriptions of Asia Minor, referred to the Eumeneian formula as “le plus remarquable” example of vague inscriptions (“Crypto-Christian”8) which were “suffisement claires pour les initiés, mais qui ne pussent donner l’éveil à leurs ennemis”. (ibid., p. 253) concluded strongly that the formula was “particulière aux chrétiens”. Cumont’s extensive knowledge of mystery religions, and, in particular, their penchant for esoteric language and symbolism, was an important factor in his willingness to apply a maximalist approach to interpreting cryptic evidence, though his reasoning was sound and followed on from that of Ramsay, explaining with reference to the Christian adoption of funerary imprecations, “les chrétiens conservèrent cette coutume, comme tant d’autres, en la modifiant” (p. 252). However, with an eye to his well-documented circumspection when it came to commenting too closely on the relationship of nascent Christianity to the oriental religions, it is unsurprising that Cumont made little further comment on Phrygia outside the cataloguing of texts, other than his highly exoticized portrait of Phrygian religion and its relationship with Montanism in his great Les religions orientales dans le paganisme romain (1906).9 While there were occasional dissenters, notably Danish scholar Wilhelm Schepelern (1929), by the time William Calder wrote what became the definitive contribution to the Christian identification of the Eumeneian formula in 1939, Cumont and Ramsay’s conclusions had already become the standard position—at least amongst Anglo- and Francophone scholars.

In his seminal 1939 contribution to William Buckler’s Festschrift, William Calder gave the most complete account of the reasons for considering that “the great majority of gravestones bearing the formula were Christian” (p. 15). While Calder conceded that “isolated cases” of pagan or Jewish usages could be identified; he mustered a considerable amount of evidence in support of its primarily Christian usage; and his arguments are worth briefly reprising. First, Calder asserted, based on a sizeable epigraphic survey, that “Eumeneian pagans did not appeal to the gods to protect their graves” (p. 18). Secondly, Calder observed that the grammatical construction ἔσται αὐτῷ πρὸς was “extremely rare” in pagan inscriptions, citing the handful of inscriptions then known (the number has not appreciably increased since–see below). Thirdly, Calder invokes an argument based on geographical and chronological specificity, noting (Calder 1939, pp. 20–21):

Much progress has been made with the classification of the pre-Nicene tombstones of Asia Minor into pagan and Christian; but always on condition that the monuments of each self-contained area were treated as an interrelated group, whose members reflected, inside the limits of local and often of parochial fashion, the opposing religious influences operating from decade to decade.

Fourthly, Calder extends the survey beyond Ramsay’s initial set of inscriptions, which in many instances remained tantalizingly vague, to include a series of cases where traditional epigraphic indicators of Christianity10 were present, including three instances where the term κοιμητήριον (ed. Ramsay 1897, vol. 2, nos. 375, 376, 379) and others where a fish symbol appears (e.g., Monumenta Asiae Minoris Antiqua IV [eds. Buckler, Calder and Guthrie], no. 355). Calder’s conclusions (1939, p. 23) were emphatic:

Suddenly a new formula, based on a grammatical construction which is very rare in pagan use anywhere, and entirely unknown in earlier pagan use in Phrygia, makes its appearance in two neighbouring cities, known to contain Christian Churches, about the middle of the third century.

Calder (1939, p. 25f.) further notes (with reference to Cumont and other scholars’ use of the “not very happy term” Crypto-Christian):

The term “crypto-Christian”, when understood to mean not that the Christians concealed themselves under pagan formulae of inoffensive or neutral type, but that they themselves devised formulae of a type which offended neither their own conscience nor the prejudices of their pagan neighbours. All the evidence available shows that ἔσται αὐτῷ πρὸς τὸν θεόν (as editors ought always to print it) was of this type.

Since this time, most scholars have accepted, along with Calder (1955, p. 26), that “this formula, having been in pagan use at an earlier period in Lycia and Pisidia, was adapted by the Christians of the upper Maeander and used by them (as dated tombstones show) at least from A.D. 249 to 273”. Subsequent discussions of the formula, however, have concentrated on a handful of potentially Jewish examples from surrounding regions (especially Apamea and Acmonia) and on challenging one or more aspects of the criterion-based approach adopted by Ramsay and Calder (see, e.g., Bij de Vaate and van Henten 1996; van der Horst 2009; Mitchell 1993; Sheppard 1979; Strubbe 1994; Trebilco 2002, 2004). While these have been by no means unfruitful debates, what all this scholarly discussion has largely overlooked is the Phrygian, and wider Anatolian pagan, religious context which informs these inscriptions.

While almost every scholar has remarked on the adapted nature of the curse formula, they have, in a strange way, avoided discussing what this might mean for the relationship between these texts and their wider context and what this might tell us about the popular religion of early Christians in Phrygia. Calder (1939, p. 23) is representative here when he writes, oddly considering Ramsay’s comments quoted above, that the Eumeneian inscriptions lack “traces, however slight, of pagan religious feeling and practice”. While several examples of a syntactically identical formula used by pagans elsewhere in Asia Minor are always noted, these inscriptions are often summarily dismissed with comments noting it is “extremely rare” (Calder 1939, p. 20) and by implication insignificant.11 Here, an argument from geographic determinism is invoked: if it does not occur in Eumeneia itself, it is not relevant. Conceptually, however, the syntactical construction occurring among pagan inscriptions simply indicates that it is one possible grammatical construction in the epigraphic Greek familiar in wider Asia Minor, and the corollary of this is that the pagan neighbours of the Eumeneian Christians would easily be able to discern its meaning and significance. This observation brings me to the key purpose of this article, situating the Eumeneian formula within its Phrygian, and wider Anatolian, pagan religious context.

3. Discussion

While I am persuaded by the arguments of McKechnie (2009) and others in seeing the Eumeneian formula as primarily a strong indicator of Christian rather than Jewish identity, there is a problem in this identification and one that has been insufficiently discussed in the more recent literature. This might be classified as one of a continuity of religious sentiment or enculturation. Elsewhere I have suggested some possible sociological reasons for this (Doherty 2017), yet what I wish to explore here are the implications of this formula as an example of popular religion, which expresses both the way in which the Eumeneian Christians verbalized the divine/human relationship, and secondly, as reflecting general beliefs (here not confined only to the Eumeneian Christians) about divinity common in Asia Minor in the early Christian era. By popular religion, I draw on the distinction made by Roelof van den Broek (1979, pp. 1–2), when he defines this as the “kind of religiosity which found expression in certain practices which enjoyed a great popularity among the common people in the Church and which, in fact, were mainly nothing else but Christian continuations of ancient pagan usages”. This is particularly important when one comes to funerary practices, for, in Broek’s justifiable view (ibid, p. 16), “it is difficult to find anything that showed a more tenacious life than funeral practices; whoever changes them risks being accused of a lack of piety”.12 This was the type of religion that more recent writers such as Ramsay MacMullen and Nadya Williams, along with Broek, have indicated was practised by the vast majority of “ordinary Christians”, but which was often frowned upon or ignored by the Christian intellectual elites. As MacMullen (2009, p. xii) has written regarding these elites:

They liked to speak of religious practice as it should be. Theirs was a sometimes theoretical, often normative view; and it is well known to us today, indeed it overwhelms us, in thousands of surviving sermons, hundreds of canons published through church councils, and in an always active exchange of correspondence among bishops, filled with opinions about what the world ought to be like. What the bishops didn’t see, however, we ourselves can and ought to see, with an effort. What they saw but were not interested in or chose to ignore, purposefully, we can recognize as actual practice, differing from exalted norms, differing in social strata, differing according to time and region as Christian habits changed—all, reality.

The Eumeneian formula is precisely this kind of popular religion, a point clearly noted by Ramsay but little explored in subsequent scholarship. That such a reading of these inscriptions has often eluded comment is abundantly clear from a survey of the relevant literature. While the different epithets given to the Christian God have received some attention from Paul Trebilco and others, the rest of the formula is virtually ignored. Similarly, while the nature of the formula as being related to the various formulae known under the collective name of grave imprecations has often received passing comment, this is normally only to say that this is common practice in Asia Minor, where the hallowed customs of antiquity safeguarded the inviolability of the tomb as the eternal home of the deceased (see Lattimore 1962). This problem was evident in the historiography as early as the writings of Ramsay (1897, p. 514), who wrote that this formula contained:

…a sentiment which is as much out of keeping with ordinary pagan expression as it is characteristic of Christian feeling: one of the most marked effects that Christianity had on common sentiment is that, among Christian peoples, references to Divine Judgement, justice, fairness, are so frequent.

However, in Asia Minor, this was simply not the case, and Ramsay knew this. This begs the question as to why no more detailed comment of this kind has been forthcoming. The kinds of questions I wish to address, however, are what does this formula indicate about the Christian practices of the Eumeneian Christians? How did this formula function to assist Christian assimilation into the pagan polis? Should this formula be read as a sign of potential “syncretism”? Or should this formula instead be understood as representative of a separation between what might be deemed “popular” religion vis-à-vis the theologically reified religion exemplified by the apologists?13

The Eumeneian formula, in its call for revenge upon or the “reckoning” of a potential grave robber with the Christian God, mirrors a grassroots approach to theodicy (here understood not in the technical sense as it is used in Christian theology but rather simply as the relationship regarding the functions of divinities in the administration of justice), which is as much a by-product of Anatolian indigenous religion,14 and perhaps also beliefs about magic, as it is of a Christian (or Jewish) milieu. That it is linked to the more common curses on those who desecrate graves found throughout Asia Minor (and conveniently collected by Johan Strubbe 1997) shows not only the commonality of these formulae but also how a previously held religious conviction could fairly easily be reinterpreted, assimilated, or “Christianized” with minimal fuss. All scholars agree on this, but few tend to take the religious aspect further than this, instead focusing their attention on the social implications (e.g., questions surrounding concealment of religious identity to avoid persecution). It would be tempting, in the first instance, to see the Eumeneian formula as a form of popular religion that underwent censure, like the type of practices mentioned by Augustine in the Confessions, which were frowned upon precisely because “they bore so close a resemblance to the superstitious rites which the pagans held in honour of their dead” (Confess. VI. 2.). However, that this shared understanding of theodicy is not just an example of a so-called degenerate folk or lay religion is probably best evinced by the fact that the Eumeneian formula was not just adopted by otherwise unexceptional Christians but also by at least one bishop and a number of civic councillors. In corollary to this, it should also be noted that the epigraphic phenomena studied below were used equally by rural farmers as by urban priests and official functionaries: the Eumeneian formula was used alike by butchers (see, e.g., New Documents Illustrating Early Christianity [eds. Horsley and Llewelyn] vol. 1, p. 86) through bishops (Ramsay, vol. 2, n. 362). What I suggest this ostensibly demonstrates is that the religious sentiments underlying the Eumeneian formula pervaded the thinking of a strong cross-section of Christians dwelling in the Anatolian interior. On this point, it is worth moving on to another factor of these inscriptions—their position in their physical social environment.

The Eumeneian formula is, by its very formulaic relation to the long held sepulchral customs of Anatolia, a visible sign of Christian assimilation into the social world of the Greco-Asiatic city. Far from being displayed in an isolated extramural cemetery or concealed catacombs, these monuments, bearing a subtly Christianized variation of the customary grave imprecation formulae, probably stood in a very public necropolis or perhaps even in intramural burial (a custom, while repugnant elsewhere in the ancient Mediterranean, was popular in Roman Asia Minor, particularly in the case of esteemed citizens). Tombstones in Asia Minor were publicly displayed at the roadside and the various entrances to the city. The banality of these monuments within their Anatolian context is further exemplified not only in the use of the grave imprecation formula, the verbal forms, or the classic memorial formula, but also in the kind of monuments themselves. The monuments adorned or marked ἡρῷον, or hero shrines, customary sepulchral monuments throughout Asia Minor, dating back until, at the very latest, the Hellenistic period, and later adapted by Christians for martyr shrines as part of the cult of the saints. Other monuments, perhaps marking the graves of more humble occupants, were in the shape of βωμοί, literally altars, traditionally used for the sacrifices or feasting on behalf of or for the dead, again practices carried over into the cult of the saints. While we might be inclined to see this as a simple example of the Eumeneian Christians following the material sepulchral customs of their neighbours, passages from Tertullian and later from Augustine relating to North African Christians, and the celebration of feasts for the dead should make us pause to think further. Indeed, it is not impossible (or, for that matter, unlikely in the slightest) that elements that lie behind what was to develop into the cult of the saints had their origin in pagan funerary banquets.15

Moving on from these initial considerations, I suggest it is appropriate to look in more detail and in a tentative yet probing way at the possible relation between the Eumeneian formula and a series of other phenomena of popular religion. In the remainder of this article, I will discuss the possible relationship between the Eumeneian formula and four popular religious practices found elsewhere in Roman Asia Minor: (1) defixiones; (2) the “appeals for justice” tablets; (3) the Beichtinschriften (‘Confessional Inscriptions”); and (4) the pagan “Cult of the Angels” inscriptions.

When we note that in Antiquity, the grave was often viewed as having domestic connotations in the afterlife, then the concern with its potential desecration becomes understandable. The economic outlay for a tomb (which probably fell outside the means of the vast majority of individuals) and the potential for grave goods would no doubt tempt an intrepid tomb desecrator in Antiquity as much as it has modern archaeologists. The most common reason for the disturbing of graves in Asia Minor, however, was that a grave was preserved for a family alone, and given the nature of burials, it was often possible for an unregistered corpse to be interred without anyone being the wiser. Given the likely cost of erecting a monument for one’s family, Strubbe (1994), the foremost authority on these imprecations, has suggested that the poor were the most likely desecrators. For my exploratory purposes here, however, I would like to point to another, less wholesome reason (if the idea of grave desecration can ever be called wholesome) found in the fascinating, and often softly spoken of world of ancient magic, in particular the phenomena of defixiones.16

While the term defixiones is fairly modern coinage, it describes the widespread ancient phenomena (even in Phrygia17) of lead curse tablets of various kinds. These tablets often invoked the help of chthonic deities or the restless spirits of those who died by foul means (i.e., murder victims) to punish some kind of rival. The use of these was, of course, not limited to an anti-social group of would-be sorcerers (just as horoscopes are not limited to astrology enthusiasts today). Their popularity was vast and was adopted for all manner of things, from using the gods to woo a beautiful woman to cursing the genitals of a would-be rival in such pursuits, from making one’s horse run faster in the hippodrome to breaking the legs of a rival’s horse. As one example, deposited in a grave in Attica and usually dated to the fourth century BCE aptly demonstrates:

[Side A] I bind Theodora in the presence of the one at Persephone’s side and in the presence of those who are unmarried. May she be unmarried and whenever she is about to chat with Kallias and with Charias—whenever she is about to discuss deeds and words and business…words, whatever he indeed says. I bind Theodora to remain unmarried to Charias and (I bind) Charias to forget Theodora, and (I bind) Charias to forget…Theodora and sex with Theodora.

[Side B] [And just as] this corpse lies useless, [so] may all the words and deeds of Theodora be useless with regard to Charias and to the other people. I bind Theodora before Hermes of the underworld and before the unmarried and before Tethys. (I bind) everything, both (her) words and deeds toward Charias and toward other people, and (her) sex with Charias. And may Charias forget sex. May Charias forget the girl, Theodora, the very one whom he loves.

(Defixionum Tabellae [hereafter DT] [ed. Audollent] no. 68 (Trans. Gager 1992, p. 90))

While the defixiones are a fascinating study in themselves, their importance in the present context is how they were made effective. In order to “bind” a god or angry spirit, it was necessary both to engrave the text and to place the tablet in a particular place. The most appropriate place for this was in graves, particularly those that belonged to people whose deaths had been violent or untimely. The engraving of a text was seen as something of a sympathetic action, like the invocation of a formula or a name. The carving relied on the power invested in language by ancient cultures. The invocations of the Eumeneian formula and other grave imprecation formulas function in a sense very similar to the sympathetic magic found in the defixiones and the related phenomena of the magic papyri. That is, the formula invokes, through the medium of the inscribed word, the deity named, that is, the Christian God. The inscriptions both warn a potential offender but also invoke the deity to action. This intimate and functional relationship with the deity is found repeatedly throughout Asia Minor, for instance in the surviving magical amulets from Ephesus, but most particularly in the context of funerary inscriptions (see esp. Strubbe 1997). However, we need to be cautious about pushing the similarities between defixiones and the Eumeneian formula too far. While the defixiones and related magical phenomena (magic papyri and magical amulets) display to varying degrees a utilitarian posture towards the deities invoked, another related group of lead tablets, defined by Versnel (1991, p. 60) as “appeals for justice” portray a religious sentiment that is closer to that displayed in the Eumeneian formula.

The appeals for justice, or judicial prayers, which display many similar phenomenological elements to both the defixiones and the “Eumeneian formula”, differ from the former in one vital aspect. Whilst the defixiones approach the deity in a manipulative posture, the appeals for justice maintain a deferential approach to the deities in question, appealing to their sense of justice and their role in its administration. They are entreaties to the justice of the gods, rather than ritual bindings. As one example from Asia Minor, usually dated to the second century BCE, reads:

Artemis “dedicates” to Demeter and Kore and all the gods with Demeter, the person who would not return to me the articles of clothing, the cloak and the stole, that I left behind, although I have asked for them back. Let him bring them in person to Demeter even if it is someone else who has my possessions, let him burn, and let him publicly confess his guilt. But may I be free and innocent of any offense against religion…if I drink and eat with him and come under the same roof with him. For I have been wronged, Mistress Demeter.

(DT, no. 2 [trans. Versnel (1991, p. 72)])

The Eumeneian formula, like these appeals for divine justice, does not take a utilitarian approach to the deity; it rather leaves the punishing of an offence under the vague auspices of the Christian God, and this is a key aspect of understanding the cultural continuity that these inscriptions indicate. If the Christians of Eumeneia were so unspecific in their threats, how could they expect that the sanctity of their tombs would be respected? The answer to this question, I believe, lies in the understandings of how the pagan gods (and clearly also the Christian God) were perceived to act in the administration of justice by the Anatolian people at this time.

The presence of an ever-present or watching divine judgement was a common idea in Greco-Asiatic religion (see, e.g., Mitchell 1993). In an area where secular authority was often powerless to prevent crime, it became increasingly common for the meting out of chastisements for sins to fall exclusively into the hands of the gods or lesser deities (see, e.g., Chaniotis 2004). Perhaps the most revealing, and certainly most interesting, example of these beliefs is found in the Beichtinschriften (“Confessional Inscriptions”) of Phrygia and neighbouring Lydia (see esp. Belayche 2004; Chaniotis 2004; Gordon 2004; Rostad 2002; 2020; Vikman 2021 for the growing literature). These inscriptions narrate an expiatory rite undertaken under the guidance of a local priest in which the person responsible for the inscription confesses on stone a hidden sin that the priests have revealed is responsible for their current state of misfortune. The inscriptions generally take narrative form, recording the sins and the expiatory rites required for setting things right with the deity in question. While the collection of these inscriptions is quite large and continues to grow (see, e.g., Petzl 1994), probably the best example, if for no other reason than its comic value, is an inscription from Silandos in Lydia dated to 235/6 CE that relates to the confession of a certain Theodoros, a sacred slave who had a weakness for female flutists and his fellow slaves’ wives. As a result of his misbehaviour Zeus evidently struck Theodoros blind, a condition that would have been much worse, no doubt, if Theodoros had not quickly dedicated the necessary menagerie of wildlife required by the god as propitiation.

THEODOROS: Because I have been brought by the gods to my senses, by Zeus and the Great Men Artemidorou. (I have atoned and set up this inscription).

ZEUS: I have punished Theodoros on his eyes for his offences.

THEODOROS: I had sexual intercourse with Trophime, the slave of Haplokomas, the wife of Eutychis, in the ‘praetorium.’

ZEUS: He takes the first sin away with a sheep, a partridge and a mole.

THEODOROS: While I was a slave of the gods of Nonnos, I had sexual intercourse with the flutist Ariagne.

ZEUS: He takes away with a ‘piglet’, a tuna, (another) fish.

THEODOROS: For my third sin I had sexual intercourse with the flutist Aretousa.

ZEUS: He takes away with a chicken, a sparrow, a pigeon. A kypros of barley and wheat, a prochus of wine, a kypros of clean wheat for the priests, one prochus.

THEODOROS: I asked Zeus’ help.

ZEUS: Look I have blinded him for his sins. But, since he has appeased the gods and has erected the stele, he has taken his sins away. Asked by the council, (I responded that) I am kindly disposed, if he sets up my stele, on the day I have ordered. You may open the prison. I set the convict free after one year and ten months.

(Beichtinschriften Westkleinasien [hereafter BIWK] [ed. Petzl], no. 5, II. 2–26) (trans. Chaniotis 2004, p. 28))

What this inscription reveals, other than that Theodoros was a serial philanderer, is something of the notions about the theodicy of the Anatolian gods. Offences relating to moral or religious crimes, even a bit of secret adulterous hanky-panky in the praetorium, were not beyond the reach of the all-seeing gods. In this understanding of divine omniscience, it is easy to hypothesize how the Eumeneian formula would appear to an outsider, for by crossing this unnamed god, a potential desecrator risked blindness, impotence or worse. What this inscription, and a similar inscription examined below, also demonstrates is that the gods (probably in the above case through the priests) were understood to communicate their displeasure and outline to the offender erecting the Beichtinschriften the offences they had committed. How this divine/human communication took place is particularly interesting when we look at another ‘Confession Inscription’ that can also be placed as epigraphic phenomena contemporary with the Eumeneian formula: the “cult of the Angels” inscriptions (BIWK, no. 3 (Trans. Sheppard 1980, pp. 92–93)):

Great is Men Axiottenos, King in Tarsi! When the scepter had been set up in case anyone stole anything from the bath house, since a cloak was stolen, the god took vengeance on the thief and made him bring the cloak to the god after a time, and he made a proclamation through an angel that the cloak should be sold and his powers written up on a stele. In the year 249.

This inscription narrates how an angel, acting as an intermediary for the Great Man Axiottenos, delivered the instruction to the unnamed thief regarding the necessity that he set up an inscription recording his transgression and subsequent punishment. The inscription, which is dated 249 from the Sullan era (i.e., 164/5 CE), also follows the Anatolian practice of “raising a scepter” (Gordon 2004), a ritual in which the priests would set up a scepter to indicate inter alia the transfer of rights of property to the god, thus making the damaging or stealing of the said property a matter of sacrilege as well as general theft. What is important for my purposes here, however, is that the need for divine propitiation is communicated to the offender through the auspices of an angel.

Angelic messengers (often associated with pagan chthonic deities like Hekate) were a common feature in Anatolian religion, in particular in the areas around the city of Stratonicaea in Caria (see, e.g., Sheppard 1980). These angels functioned in an intermediary sense, acting as messengers for a higher deity or (more importantly for our purposes here) as protectors of the grave. This idea of invoking the protection of intermediary quasi-divine figures in the protection of a sacred space like a grave is, of course, from a phenomenological viewpoint quite common and is witnessed in a variety of cultural contexts across both the ancient and modern worlds. But is it used specifically in the Christian inscriptions, utilizing the Eumeneian formula?

The answer to this question lies in a problematic but very interesting pair of inscriptions concerning a character by the name of Roubes (a name seen by Louis Robert as a Graecized version of the Hebrew name Reuben). The first of these inscriptions reads (from Hellenica [ed. Robert], vols. 11, 12, pp. 429–35):

Property of one who still lives. I, Lycidas, cite God as my witness, that I built the shrine by my own labors, as my brother Amianus was reluctant, and I authorize my sisters Phronime and Maxima to be placed (in it). If anyone inters another, he will have to reckon with God and the Angel of Roubes (ἔστε αὐτῷ πρὸς τὸν θεὸν καὶ τὸν ἄνγελον τὸν Ῥουβῆδος).

(Trans. Sheppard 1979, p. 176)

So who was the “Angel of Roubes”? The answer is, perhaps, found in an extended inscription nearby (Cities and Bishoprics of Phrygia, vol. 2, ed. Ramsay 1897, no. 232), which contains several problematic elements worth mentioning,18 but for our purposes here the final line reads:

Γάϊος π[ρ]αγματικός, ἠδ’ ἀλόχῳ φιλίῃ Τατίῃ τέκεσίν τε ποθητοῖς ὄφρα τὸν ἀΐδιον τοῦτον ἔχωσι δόμον σὺν Ῥούβῃ μεγάλοιο θ(εο)ῦ Χ̣ρ̣(ιστοῦ)19 θεράποντι.

A man—Gaius the lawyer, trained in the arts—built this tomb while he was alive for himself and his dear wife Tatia and their lamented children, that they might have this eternal home together with Roubes, the servant of the Great God Christ.

(Trans. Sheppard 1979, p. 178)

Roubes here, probably to be identified with the same named in the Lycidas inscription, is normally seen as a respected member of the local (Jewish?) community. The options here are either that being dead, Roubes has attained the status of an angel in some kind of euhemeristic sense or that the angel of Roubes is an intermediary guardian angel associated with the deceased Jew Roubes. Such an association might be seen, for instance, in the story in the Book of Acts where Rhoda sees Peter arriving at the house of John Mark when he was believed to be languishing in prison, and when she leaves Peter out in the cold and runs to inform the others present, they tell her that it must be his angel (Acts 12:12–15). Moreover, within the massive corpus of Jewish pseudepigraphal literature, the idea of the dead becoming guardian angels over the living is sometimes referenced (see, e.g., Test. Asher 6:5f.; Jub 35:17; I Enoch 100:5). More interestingly, with regard to the Roubes inscription, is that this inscription, like the confession inscription above, also reveals the involvement of angels in the administration of religious justice in a way that is very similar to how angels are frequently recorded as behaving in the Judaeo-Christian literature. While some of the inscriptions of nearby Stratonicea in Caria evince that worship was specifically given to the angels by pagans, it is important to note that such a belief is not inherent in what is written on the Roubes inscription above.20 What I would suggest here is that the Roubes inscription be understood as invoking the “Angel of Roubes”, much in the same way as pagans in Termessos in nearby Pisidia (Tituali Asiae Minoris [ed. Heberdey], no. 365) could invoke the chthonic deities (ἔσται αὐτῶ πρὸς τοὺς κατυχομένους) as intermediaries who performed a functional role in protecting the rest of the deceased. This inscription contains language by which, while not compromising his monotheism, the Christian Lycidas was able to invoke the supplementary defence against grave desecration provided by the “angel of Roubes”, an angelic intermediary (however we understand this) whose role and efficacy would be readily recognizable to his pagan neighbours.

In the evidence from this and other contemporary inscriptions, we see a Christian inscription that adapts similar religious sentiments from the shared religious context of the Anatolian interior and phrases them in a sense that is both acceptable to Christian claims of monotheism and, at the same time, both innocuous in sentiment and functionally efficacious with regard to local pagans. Like elements of the ‘Confession Inscriptions’ and the judicial prayers, the Eumeneian formula places the transgression of desecrating the grave under the auspices of the deity (here the Christian God), making the matter not only a criminal offence but also a religious one. At the same time, however, it takes a deferential posture toward the deity. The formula does not seek (at least actively), like the defixiones or magic papyri, to use the gods as a labour-saving device.

4. Conclusions

By probing these related Anatolian pagan inscriptions and highlighting a series of conceptual parallels between them and the Eumeneian formula, one might suppose that I consider that the Christians of Eumeneia were in fact highly “syncretistic” and compromised to their pagan surroundings—along the lines alluded to, but cautiously avoided, by Ramsay and others. This is not the case. Instead, I would like to suggest that one of the key reasons for the success of Christianity in Eumeneia, and perhaps also the success of Judaism elsewhere in Asia Minor, in carving spaces for themselves in the civic context, relates precisely to the way in which they were able to appeal to and adapt the widely understood forms and socio-religious notions of its surrounding culture while at the same time maintaining their monotheistic confessions. For Christians, we are dealing with the popular religion of ordinary followers—how they expressed their Christianness in a specific everyday context. Through language and approaches to theodicy and divine/human relationship, which were similar to the Beichtinschriften, defixiones, judicial prayers, and the Anatolian pagan cult of the angels, the Eumeneian Christians advertised to the surrounding pagans that they shared religious sentiments similar to their neighbours. It is possible that their pagan neighbours saw little difference between themselves and the Christians among them. The latter’s language was non-threatening in the sense that it was not aggressively anti-pagan. As Éric Rebillard (2009, p. 75) observed with reference to funerary maledictions more generally, “Christians were acting no differently than pagans or Jews, regardless of their respective religious beliefs”. While the tracing and comparing of religious phenomena shared by early Christians and their pagan and Jewish contemporaries can often fall into “parallelomania” or pose a variety of other theoretical problems (see Smith 1990; cf. Rebillard 2009, pp. 73, 75), I would suggest that, as a methodological starting point, examining these shared features in the Anatolian epigraphic habit, even across blurry religious divides, may be a more fruitful means of drawing further out distinctive aspects of ordinary Christianity in this region. This work also relieves the all-too-predominating text-based approaches to the past. Following this orientation, I believe the Eumeneian formula should be understood as an example of the popular religion of ordinary Phrygian Christians as they sought to adapt the traditional pagan and Jewish customs widely practised in the contemporary socio-cultural environment of third-century Asia Minor.