Hostname: page-component-74656f9884-skzmx Total loading time: 0 Render date: 2026-11-06T01:05:09.533Z Has data issue: false hasContentIssue false
  • English
  • Français

THE RHETORIC OF RELIGIOUS CONFLICT IN ARNOBIUS’ ADVERSVS NATIONES

Published online by Cambridge University Press:  22 April 2019

Konstantine Panegyres
Konstantine Panegyres*
Affiliation:
The University of Melbourne
*
konpanegyres@gmail.com
Rights & Permissions [Opens in a new window]

Extract

In this paper I discuss the ways in which the early Christian writer Arnobius of Sicca used rhetoric to shape religious identity in Aduersus nationes. I raise questions about the reliability of his rhetorical work as a historical source for understanding conflict between Christians and pagans. The paper is intended as an addition to the growing literature in the following current areas of study: (i) the role of local religion and identity in the Roman Empire; (ii) the presence of pagan elements in Christian religious practices; (iii) the question of how to approach rhetorical works as historical evidence.

Information

Type
Research Article
Information
The Classical Quarterly , Volume 69 , Issue 1 , May 2019 , pp. 402 - 416
Copyright
Copyright © The Classical Association 2019 

In this paper I discuss the ways in which the early Christian writer Arnobius of Sicca used rhetoric to shape religious identity in Aduersus nationes. I raise questions about the reliability of his rhetorical work as a historical source for understanding conflict between Christians and pagans. The paper is intended as an addition to the growing literature in the following current areas of study: (i) the role of local religion and identity in the Roman Empire; (ii) the presence of pagan elements in Christian religious practices; (iii) the question of how to approach rhetorical works as historical evidence.

I. INTRODUCTION

I will put forward two complementary arguments: first, that in Aduersus nationes Arnobius sought to create an artificial rhetorical division between Christians and non-Christians not entirely representative of the realities in Africa Proconsularis in the second and third centuries a.d.; and second, that creating the perception that Christianity was not another part of paganism, but a complete rejection of paganism, was a vital rhetorical strategy in Arnobius’ work. Throughout Aduersus nationes Arnobius paints a picture of conflict between Christianity and paganism. However, there is considerable evidence that Christianity was far from being the absolute opposite of paganism, and that similarities between Christians and non-Christians abounded. Why, then, did rhetoricians such as Arnobius seek so forcefully to stress this difference? In this paper, I shall explore three of the rhetorical arguments used by Arnobius to differentiate Christianity from paganism. The attempts by Arnobius to create the appearance of difference between non-Christian cult and Christianity, and whether or not these strategies reflected reality or were merely rhetorical inventions generated for intellectual debate, are the topics of this paper.Footnote 1

I begin with a discussion of the relationship between rhetoric and religion in the time of Arnobius. Then I turn to look at three of Arnobius’ arguments: that Christians and pagans were distinguished as separate racial groups, that Christianity was more intellectual than paganism, and, finally, that Christian cult was rational, pagan cult irrational. In each case, I contrast the arguments of Arnobius with external documentary or literary evidence. This comparison, I shall argue, strongly suggests that Arnobius on a number of topics rhetorically overstates the differences between Christianity and paganism. I finish by considering why this might be so.

II. BACKGROUND: RHETORIC AND RELIGION

The initial problem we must confront is how the idea of Christianity was constructed to seem to be separate from other religious groups, and whether ‘Christians’ considered themselves as being in conflict with ‘non-Christians’ or as belonging to a continuum of ‘non-Christian’ practices. Complexities immediately emerge: who are the ‘Christians’ Arnobius had in mind while writing Aduersus nationes? And who are the ‘non-Christians’? One might think that ‘pagan’, the term used by early Christian authors to describe non-Christians, denotes a unified body of people who believe a certain class of religious beliefs; one might then also think that ‘Christian’ is a similarly unified concept. The evidence presents a more complicated picture.Footnote 2 The reification of the terms ‘pagan’, ‘non-Christian’ and ‘Christian’ has become a central problem in recent scholarship. Many older works on the subject treat the rhetorical binary of ‘pagan and Christian’ as if the terms represent diametrically opposed sets of people; this way of thinking is misleading.Footnote 3 The title of at least one recent work (C.P. Jones's Between Pagan and Christian) can however be seen as a play on that distinction.Footnote 4 In reality, ‘Christians’ might perform so-called ‘pagan’ acts and traditions; while ‘pagans’ might also worship Christian saints, Christ or God.Footnote 5 In the following, I wish to examine the ways in which Arnobius reified notions of ‘pagan’ and ‘Christian’ in his rhetoric, and compare it with evidence taken from local religious realia.

It is well known that Arnobius was not too familiar with the new Christian religion.Footnote 6 He was instead better acquainted with the religion of the non-Christians. The story of Arnobius’ conversion from paganism to Christianity, as recollected by Jerome (Chron. 326–7), shows how Arnobius spent much of his life attacking Christianity in his rhetoric before dreams led him to turn his polemics and propaganda on non-Christians instead.Footnote 7 To place Arnobius in the context of the intellectual climate of his contemporaries, it is first necessary to understand his audience and his rhetorical influences.

Who was Arnobius writing for?Footnote 8 He was probably not writing directly for the local rural people who worked on the vast imperial estates surrounding Sicca Veneria, his home, which is now El Kef in Tunisia;Footnote 9 nor could he have hoped to change greatly the habits of those worshipping in the cult of Saturn or its Christian complement, deus frugifer.Footnote 10 The most probable answer is that he was writing in the hope that he might win over the wealthy landowners who controlled the vast imperial estates, and who had the power to help with the conversion of their workers; non-Christian writers, especially Porphyry, formed the other, far more significant, part of his audience.Footnote 11 Arnobius was no doubt influenced by Tertullian, Eusebius and Cyprian; and one can surely also ascribe the polemical rhetoric of Aduersus nationes to his position as professor of rhetoric in Sicca.Footnote 12 Aduersus nationes was most likely written some time between 296 and 311,Footnote 13 in the years prior to and during the Diocletianic Persecution of 303 to 312/13.Footnote 14 This is important; it reveals that the work was formulated at a crucial period, when rhetorical hostility between religious and philosophical groups was at boiling point, and ideas of religious conflict and difference between paganism and Christianity were becoming crucial to rhetoric.

A number of polemical works were written by both Christian and non-Christian intellectuals in this period,Footnote 15 and more than one scholar has pointed to the possibility that such intellectual debate may have been responsible for building tensions leading to the Diocletianic Persecution.Footnote 16 One important development may for instance have been the movement of the capital and of the imperial court to a Greek city under the Tetrarchy, where pagan intellectuals could more easily exert influence.Footnote 17 Without doubt the most significant of these intellectuals for the study of Arnobius is Porphyry,Footnote 18 a student of Origen,Footnote 19 who wrote a trio of works on pagan soteriology between 270 and 300:Footnote 20 De regressu animae (c. late 290s), Contra Christianos (c.300) and finally De philosophia ex oraculis (c.302). The latter was a work of anti-Christian propaganda disseminated in North Africa prior to the Great Persecution.Footnote 21 As the leading anti-Christian author of his day, Porphyry's ideas had much sway in directing the imperial religious agenda;Footnote 22 moreover, some evidence could suggest that Porphyry was writing such works for oratorical delivery in the presence of the imperial court.Footnote 23 The purpose of these polemical writings was, in a basic sense, to demonstrate the falseness of Christian theology, and to provide an alternative religious philosophy.Footnote 24 Arnobius was familiar with the works recently written by Porphyry; as a new Christian convert, he had formerly used parts of Porphyry's arguments in his own criticisms of Christianity. Simmons has convincingly shown that Books 1 and 2 of Aduersus nationes may be directed against Porphyry's Contra Christianos and Philosophia ex oraculis; indeed, Aduersus nationes seems to be a direct reply to Porphyry's Contra Christianos, though it could also allude to Tertullian's Ad nationes.Footnote 25

In this so-called ‘propaganda war’ of which Arnobius was part, rhetoric and oratory seem to have played an important though underappreciated role.Footnote 26 As one scholar has noted with regard to the age of Tertullian, ‘one cannot over-emphasize the importance of rhetoric to the educated and literate’.Footnote 27 Aspects of Aduersus nationes indicate that the work could be delivered as oratory; for instance, the use ad nauseam of rhetorical questions, often for pages without break; and also the extensive alliterations and phonetic wordplays which mock pagan customs.Footnote 28 Porphyry for his part is important for his contributions to rhetorical theory and issue-theory, for he developed a number of original conclusions building from the work of the second-century rhetoricians Minucianus and Hermogenes; unfortunately, only a few fragments of Porphyry's rhetorical treatises remain.Footnote 29 Whereas Arnobius owes a debt to the rhetoric of Latin apologetic authors of North Africa, such as Tertullian and Cyprian, Porphyry owes a debt to Athenian rhetoricians, perhaps even to the foremost rhetoricians at Athens during his formative years.Footnote 30 In the writings of both authors, the ‘propaganda war’ prior to the Diocletianic Persecution was fought in similar terms. Consider, by way of an example, two passages, the first from Porphyry's Contra Christianos, the second from Aduersus nationes: in the first, Porphyry blames the plague of Rome on the cult worship of Jesus, which seems to have been a recent phenomenon in the city;Footnote 31 in the second, Arnobius argues that such devastations predated the emergence of Christianity, and therefore could equally be attributed to the pagan gods, but in fact ought to be attributed to nature itself rather than to deities.Footnote 32 Such attacks differ not in substance but in intention: one demonstrates the faults of Christianity, the other those of paganism. In the same way, both groups of intellectuals supplied their own similar philosophies of salvation, hoping to gain more adherents than their opponents.Footnote 33 In short, binary polemical rhetorical stances emerged, coming under the banners of ‘paganism’ and ‘Christianity’ respectively, with no apparent room for an intermediary position. According to this rhetoric, Christianity had to be seen to be the opposite of paganism, while paganism had to be seen to be the reverse of Christianity. This rhetorical opposition is reflected in modern scholarship.Footnote 34

One must also understand, however, what rhetoric Arnobius had read, and how his style and tone developed.Footnote 35 He was certainly well versed in classical Roman literature, particularly the theologians and the satirists, but also Virgil and Lucretius.Footnote 36 We shall first deal with the theologians. Arnobius had read Cicero's De natura deorum and De diuinatione in detail,Footnote 37 and recounts how some people had formed such a hatred towards De natura deorum that they suggested that the Roman Senate should order the book to be destroyed for spreading fictions about gods;Footnote 38 for Arnobius, the folly of Cicero's work somehow confirmed the veracity and authenticity of Christianity.Footnote 39 Beyond Cicero, Arnobius’ discussions of minor, forgotten or odd gods seem to reveal his reliance on Roman sources for local religion, such as the indigitamenta, the books of the pontiffs containing the names of various di certi and di minuti and their functions and etymologies; Arnobius explicitly mentions the use of such indigitamenta.Footnote 40 A further influence on Arnobius’ theological writing was Varro, whose works are referred to most often in Aduersus nationes.Footnote 41 Frend long ago noted that Arnobius writes with ‘Varronian skepticism’,Footnote 42 referring to the attitude with which both authors treat religious matters.Footnote 43

Arnobius was also acquainted with classical satire, and the techniques of satire are present throughout Aduersus nationes. He is particularly adept at satirizing obscenity, and the structure of the work, beginning with a defence of Christianity (1 and 2), then a harsh critique of ‘pagan’ gods (3 to 5) and ‘pagan’ rites (5 to 7), means that he does not have to combine the defence of Christianity with the attack on paganism, as Tertullian had done; this makes his attacks on paganism especially forceful and devastating.Footnote 44 He was familiar with Lucilius, and one passage satirizes the insemination of Ocrisia.Footnote 45 But Arnobius was more than familiar with Cicero's rhetoric and the satire of the classical Roman poets: he took it upon himself to combine these influences to attack the very religion of which they were part. Rhetorical polemic and satire are a fitting combination, but an unlikely one for a Christian author. Arnobius devotes the majority of Aduersus nationes to the assault on ‘paganism’, and this is also the strongest aspect of the work; the section on Christianity is by contrast weak and unassured. He clearly knows more about non-Christian deities, and thus is able to satirize everything he knows. This stems from a lack of understanding about what precisely Christianity is, or what it should be; his text reveals a time in which Christianity was a fluid idea. In this struggle for Christian self-definitionFootnote 46 intellectuals such as Arnobius found a rhetorical answer to those asking the nature of their new religion: create the perception of conflict between paganism and Christianity, negate aspects of the old religion, differentiate Christianity from all forms of paganism, and advise Christians who also carry out the old traditions to cease from doing so.

III. THE RHETORICAL STRATEGIES OF ARNOBIUS

It is worth assessing in detail these rhetorical strategies, and comparing them with the realities of Christian and non-Christian interaction in archaeological and epigraphical evidence.Footnote 47 One must remember, when doing so, that in the third century Christianity was a demographic minority; the terms ‘Christian’ and ‘non-Christian’ do not denote a demographic balance, though they give that impression. The rhetoric of Arnobius attempts to give the opposite impression.

A Christian race?

It is in the first place significant that Arnobius talks of ‘we’—by which he means Christians—and ‘you’—by which he presumably means the pagan nationes. This is precisely the same strategy used by Porphyry to attack Christians.Footnote 48 Throughout Aduersus nationes, we find Arnobius describing the Christians as a ‘race’ of people: dicere, postquam esse in mundo Christiana gens coepit, terrarum orbem perisse, multiformibus malis affectum esse genus humanum (1.1.1).Footnote 49 This is misleading, because, as we have seen, Arnobius was himself unsure of what ‘Christianity’ meant, and so could not have known precisely to what sort of person he was referring here other than to a rhetorical binary religious ‘reality’. Yet, the Christians are for him still ‘us’, and this is made especially clear in a number of passages in the first book: ut ante milia annorum decem ab insula quae perhibetur Atlantica Neptuni, sicut Plato demonstrat, magna erumperet uis hominum et innumeras funditus deleret atque extingueret nationes, nos fuimus causa? (1.5.1); ut ex Macedoniae finibus unus exortus adolescens Orientis regna et populos captiuitate ac seruitio subiugaret, nos fecimus atque excitauimus causas? (1.5.5); ut modo Romani uelut aliquod flumen torrens cunctas submergerent atque obruerent nationes, nos uidelicet numina praecipitauimus in furorem? (1.5.6). We need not labour the point with more examples, but it is worth considering how the early chapters of the first book are filled with references to ‘us’ as ‘Christians’; Arnobius is constructing and defending a Christian identity in these passages which he has not attempted to define. Similar is the construction of Roman identity. At the start of Book 4 Arnobius addresses ‘you Romans, lords and leading citizens of the world’ (4.1.1). Once more, this is done without suggesting precisely who these Romans are; some of them, like Arnobius himself, must be ‘Christian’, others must be ‘pagans’, and others again must worship Christianity as part of their ‘paganism’.

In Africa Proconsularis, the evidence suggests that it is extremely difficult to differentiate Christians from pagans, because some who considered themselves ‘Christian’ worshipped pagan deities, and some who considered themselves Roman pagans also worshipped Christian religious figures. It was a religious melting-pot. The main deity in North Africa was Saturn: he was the lord of the universe and the provider of plenty and abundance.Footnote 50 The cult of Saturn worshipped him as the bringer of the first crops of the harvest. The famous Boglio stēlē found in central Tunisia only 48 km south-east of El Kef depicts Saturn enthroned above images of a man working with a yoke of oxen, of workers with sickles cutting wheat, and of grain transportation and the sacrifice of animals for worship.Footnote 51 The stēlē is an affirmation of Saturn as the lord of the harvest; it also represents ‘one of the most significant “pagan” affirmations of the belief in the divine providence’.Footnote 52 Epigraphy supplies further evidence of the cult in use in the late third century: the Cherchel mosaic, with its famous depictions of agrarian production; an inscription from the fundus Aufidianus; and the reaper of Mactar's epigraph.Footnote 53 To Arnobius, Saturn is also the lord of the universe, provider of agrarian fertility and prosperity as deus frugifer (6.10). Saturn controls the winds, lands and waters;Footnote 54 he carries a sickle, accompanied by the sun and the moon;Footnote 55 he is worshipped by incense and offerings on a flaming altar;Footnote 56 another stēlē depicts a sacrificial bull in front of ears of grain;Footnote 57 and he is furthermore found with a lunar crescent by his left arm.Footnote 58 These are local examples of a broader phenomenon of which Arnobius was clearly both witness and judge:Footnote 59 but the close relationship between ancient pagan cult and ‘new’ Christian forms of worship is also evident here.Footnote 60 In Arnobius, Saturn is also the bearer of fruit and produce (1.29), and the carrier of the sickle and the fruit basket (6.12.5). At the same time, evidence shows the monogram of Christ surrounded by the ancient Punic symbol of Tanit, the dolphin, now appropriated as the ΙΧΘΥΣ fish symbol of Christ;Footnote 61 the pine cone, a symbol of Saturn, was being used in the ceremonies of both Christians and non-Christians to represent their respective supreme deities.Footnote 62 Tellingly, Tertullian, responding to those who accuse Christians of worshipping the sun, admits that Christians ‘are not far from your Saturn and days of rest’ (non longe a Saturno et sabbatis uestris sumus) (Ad nat. 1.13.5). As Miriam Taylor observes, the point of Tertullian's line of argument is to show that ‘Christian practices are not so strange or so distant from Roman customs as some Romans seem to assume’.Footnote 63 Elsewhere, Arnobius betrays his knowledge of syncretic cult in Africa Proconsularis, a result, first of the Romanization, then of the Christianization, of older Punic religion;Footnote 64 for instance, he is aware of the cult of Mithras and its relation to the cult of Aion-Khronos, competitors of the cult of Ba'al.Footnote 65

In sum, this evidence from the province Africa Proconsularis offers a very different picture to that presented in Aduersus nationes. It suggests a world of slow integration and assimilation, rather than of conflict, competition and difference; nor was this phenomenon restricted only to North Africa, but is also well attested in, for example, Britain, Italy and Gaul.Footnote 66 It is rather in the works of such authors as Arnobius that we find artificial ideas of religious conflict and competition, which were formulated in order to extract Christianity from the intellectual traditions of pagans. There is no clear ‘we’ and ‘you’ that distinguishes Christians from pagans in the archaeological and epigraphic evidence of the time of Arnobius, between roughly 250 to 320, but rather a grey area remains, where the absolute terms ‘Christian’ and ‘pagan’ become empirically meaningless unless stated explicitly in documentary evidence; indeed, the difference between the ways in which people worshipped and venerated Saturn and the Christian God is sometimes almost imperceptible.Footnote 67 This colours our understanding of the perception of religion Arnobius wishes to give to the reader. Religion is for him a clear binary of ‘pagan’ deities, on the one hand, and a supreme Christian God, on the other; but, as we have just seen, this view was not true of the world in which he lived. Arnobius must have known this, and therefore we must look at his text from two perspectives: his wish to create the perception that Christianity was somehow special and distinct from pagan practices, and the reality of paganism from which he was consciously attempting to extract Christian religious practice.

Christian intellectualism and pagan superstition

One main rhetorical strategy with which Arnobius differentiated Christianity from non-Christian cult was his endeavour to bring about the perception that Christianity was more intellectual than paganism. To Arnobius, the pagan countryside presents a smorgasbord of superstition, and it is literally oozing with bizarre spirits and gods (4.7.19–24). Venus is seen to preside over camps and young men, Perfica over obscenities, Pertunda over the marriage couch: the point is that the ‘pagan’ spirits seem to inhabit everything, and this is absurd for Arnobius. These passages owe much to Arnobius’ detailed knowledge of actual sources for minor figures in pagan ritual, such as the indigitamenta discussed earlier. Elsewhere he begins to play with names to satirize further the absurdity of these spirits, and groups them according to etymological similarities: Peta, deus nemorum Nemestrinus est, Patellana numen est et Patella … Nodutis dicitur deus, qui ad nodos perducit res satas, et quae praeest frugibus terendis, Noduterensis … in tutela sunt Orbonae orbati liberis parentes, in Neniae, quibus extrema sunt tempora … ossa Ossipago ipsa memoratur … dicite, o quaeso ita, ut uobis propitiae faueant Peta Puta Patella … (4.7.2–4.8.2). The vigour and satire of Arnobius’ wordplay in this passage is mesmerizing. Especially notable are the connections Arnobius makes between the names of the spirits and the things they represent, and παρονομασία seems to be his favourite effect (Orbonae orbati; ossa Ossipago).Footnote 68 These excerpts (and there are many like them) are important, because they have a special aim: namely to discredit non-Christian spirits by showing how they are used to represent silly, and in many cases immoral, things. This, however, is not the only reason why Arnobius rejects the non-Christian spirits. They are in his view unnecessary or needless, and it is shameful to abuse deities by assigning them to meaningless things (4.6.1–15):

Lateranus, ut dicitis, deus est focorum et genius adfectusque hoc nomine, quod ex laterculis ab hominibus crudis caminorum istud exaedificetur genus. quid ergo? si testa aut materia fuerint quacumque alia fabricati, foci genios non habebunt, et ab officio tutelae quisquis iste est Lateranus abscedet, quod regni sui possessio non luteis constructa est formis? et quid, quaeso, ut faciat, praesidatum focorum deus iste sortitus est? per humani generis coquinas currit inspiciens et explorans, quibusnam lignorum generibus suis ardor in foculis excitetur, habitudinem fictilibus contribuit uasculis ne flammarum dissiliant ui uicta, curat ut ad sensum palati suis cum iucunditatibus ueniant rerum incorruptarum sapores, et an rite pulmenta condita sint praegustatoris fungitur atque experitur officio. itane istud non foedum, quinimmo, ut uerius dixerim, non contumeliosum, non impium, ad hoc tantum inducere deorum nescio quas fictiones, non quas dignis honoribus prosequare sed quas rebus turpibus et infami praeficias actioni?

These are the spirits of the rural regions. To Arnobius, these are the places in which the old religion thrives most actively, and superstition is an element of the countryside. Arnobius’ dislike for rural regions and his preference for the city form part of a crucial problem for early Christianity: that it had great trouble converting the rural areas, but found the conversion of the cities an easier task.Footnote 69 A parallel for Arnobius’ attack on rural ‘paganism’ is in Tertullian, where the agricultural traditions are denounced for their utter baseness (Tert. Apol. 21). Hence, rural regions were negligible to Arnobius for two reasons: they represented the dominance of what he viewed as old and unrefined paganism, and they were also intellectually backward, far from the supposed refinement of the urban centre. In Aduersus nationes, it is the urban Christianized centre that is moral, while the non-Christian rural regions are immoral in their degradation of deities. This reflects the general pattern of the spread of Christianity, which had more impact in the cities.Footnote 70 In attempting to create a moral distinction between countryside and city, Arnobius has revealed once more his tendency to form binary notions of ‘Christian’ and ‘non-Christian’ to serve a rhetorical function.

The reality was less clear-cut. Sicca was situated in a region rich in agriculture. Much of its wealth was derived from agrarian production. The main crops were wheat and grapes; and viticultural production was at its height in the reign of Probus, when a large number of vine plantations were instituted.Footnote 71 In the fourth century, laws were made which ensured that certain areas of land would not go out of cultivation.Footnote 72 It was noted earlier that Africa Proconsularis was for the most part covered by vast private and imperial estates.Footnote 73 I wish to explore further the exchanges between city and country in the regions surrounding Sicca. It is certainly the case that many towns, cut off from the arterial roads, retained their agricultural character;Footnote 74 the population must have largely comprised those working on the imperial estates. These estates were the property of vastly wealthy aristocratic families, who might live in Italy and not even see their African estates. Columella discusses the dangers of overseas landownership (1.1.20), a passage which most likely refers to North African property;Footnote 75 he also advises that it is more profitable for an owner to limit his landownership to a small, manageable estate, because large estates would suffer owing to their sheer size (1.3.12–13).Footnote 76 In many cases, an aristocratic owner of a vast overseas estate would have to subordinate management to gangs of debt-bondsmen and slaves.Footnote 77 Pliny notes how the spread of vast estates (latifundia) had caused the ruin of agriculture in Italy and Africa (HN 18.35). And yet, North Africa also managed to be one of the major food-suppliers in the empire.Footnote 78 In the first two centuries a.d., the region surrounding Sicca was administratively dependent on Carthage; Sicca was situated in the Siliana valley, and in Augustan times this region had been a centre of colonization and veteran settlement.Footnote 79 Cults were also found in agrarian regions. Stēlai in rural regions attest to the existence of priesthoods sacrificing and holding ritual for the cult of Saturn.Footnote 80 It is certain that cult worship was stronger in rural areas; but it is very difficult to get closer to evidence revealing the spread of Christianity in the countryside, or the manner in which it occurred. The best evidence is that Christianity infiltrated local religions with Christian equivalents for religious spirits. Deities such as Lateranus were replaced by the names of saints. Three agricultural incantations from Italy show that the Christian God performed the role of non-Christian deities by taking up the old ‘pagan’ function.Footnote 81 These changes reveal a Christianity becoming ‘paganized’ just as much as paganism was being ‘Christianized’.

Rational Christianity and irrational paganism

The final rhetorical invention of difference between Christianity and paganism I shall explore is Arnobius’ attitude towards the use of non-Christian food in religious rituals. Here Arnobius’ aim is to prove that Christianity is rational but paganism is irrational or even mad.

The chief objection raised by Arnobius against the use of wine and food in pagan rituals is that there is neither a clear nor a sane reason for doing so, and nobody can provide him with an explanation (7.29.1–6):

merum thuris est socium, quod explanari consimiliter poscimus cur ei superfundatur incenso. nisi enim ratio cur fiat ostendetur nec habebit expositam sui causam, non iam istud errori obiciendum est ludicro sed ut dicatur expressius insaniae dementiae caecitati. ut enim iam saepius dictum est, debet omne quod geritur causam sui habere perspicuam nec caliginis alicuius obscuritate contectam. si ergo fiditis facto, sponte monstrate liquor iste cur detur [id est uinum superfundatur] altaribus. numquid enim numinum corpora sitim sentiunt aridam et eorum necesse est siccitates umore aliquo temperari? numquid nobis ut mortalibus moris est cenis intermiscere potiones, pari etiam more post solidos uictus liborum et pultium hostiarumque caesarum, quo putrefiat facilius ac percoquatur cibus, frequentissimo inrigant accipiuntque se uino?

The use of wine as an offering is furthermore an insult to the gods (7.30.7).Footnote 82 This string of rhetorical questions forms an impressive edifice of questions about the character of ritual which non-Christian authors are to answer. The passage provides clear insight into Arnobius’ rhetorical strategy, which is to delineate what he sees as the ‘madness, folly and blindness’ of ‘paganism’ by forming a long series of questions which undermine many of the old rituals and rites of non-Christians; it also clarifies his philosophical position, which is that, if an explanation (causam) cannot be found for something, then it is to be doubted.

Yet, we find that Arnobius’ argument turns against him. If the use of wine and food as offerings to deities is ‘madness, folly and blindness’ (insaniae dementiae caecitati), then it follows that it will be so no matter what religious group is performing such rituals. We must therefore turn to look at how Christianity incorporated food and drink into its rituals, and find out to what extent this was either a continuation or a repudiation of non-Christian practices. In fact, we find that Justin Martyr had already explicitly referred to the use of wine (or wine mixed with water) during Christian rituals, and how after prayer wine, bread and water would be brought in and distributed to those present in the ceremonies (Apol. 1.65–6). Similarly, Clement of Alexandria also allows wine into Christian ceremonies, though not without cautioning that it should be used sparingly and mixed with water (Paed. 2.2). Would Arnobius have criticized this practice of using wine in ritual if he knew that other Christians were doing it? Or is his criticism really directed at fellow Christians who drink wine, in the attempt to show them that they are only following a foolish old pagan practice? That is difficult to know. Arnobius’ pupil Lactantius also criticizes the use of wine in religious rites (Diu. inst. 6.1.1–2), saying that it does nothing but help people venerate demons (Diu. inst. 2.14.13) and that all such food-offferings are meaningless to the gods (Diu. inst. 7.6.6). Lactantius’ argument is ostensibly directed at pagans, but the implication is that Christians should also avoid all such rites. Yet, Christians clearly did not avoid them. McGowan has shown how early Christian Eucharistic meals consisted of a variety of foods, many of which are no longer used. Cheese, milk, honey, oil and salt were all parts of the ritual of communal Christian meals. Bread and wine were also used, but wine could be replaced by water during ascetic eucharists, and sometimes the cup was altogether omitted.Footnote 83 Christ had eaten bread and wine at the Last Supper, but Christian eating rituals came to include many other foods, and the significance of Eucharistic meals was not to the early Christians limited by reference to the Last Supper.Footnote 84 Communal meals thus involved the consumption of various other types of food as well as bread and wine; they also involved the giving of thanks (εὐχαριστία) or blessing (εὐλογία). The ‘Christian’ Eucharist was also commonly confused with ‘pagan’ ritual meals in the early centuries. Christian gatherings were often compared with pagan cult meetings: Paul urges Christians to avoid ‘pagan’ meals, but because so much business was performed in ‘pagan’ temples and symposia in Graeco-Roman society, it was also inevitable that Christians would take part in ritual ‘pagan’ eating and that pagan feasts would resemble Christian feasts and vice-versa.Footnote 85 This sort of imitation worked both ways. Justin Martyr ascribes to the work of demons the fact that members of the cult of Mithras supposedly imitated the Christian Eucharist by eating bread and drinking water (Apol. 1.66), while Tertullian also makes comparisons between the communal feasts of the Christians and the ‘pagan’ religious collegia, for instance the collegia Saliorum or the collegia belonging to the cults of Dionysus or Serapis (Tert. Adu. Valent. 1.1).Footnote 86 These are further indications that Christian and non-Christian practices in the second and third centuries cannot be easily distinguished, that the two often overlapped and that religious traditions often described as ‘Christian’ or ‘non-Christian’ in fact belonged to both.Footnote 87 Arnobius’ rhetoric creates a binary out of a much more complex religious situation.

IV. CONCLUSION

In Aduersus nationes, Arnobius offered an assault on non-Christian religious practices, and with the power of his rhetoric sought to create the perception that there was great difference between Christians and non-Christians, and that Christianity was a complete negation of ‘paganism’. For clear reasons, Arnobius wished for his new faith, Christianity, to mark a complete break from older religious practices; but increasingly we find that, even while Arnobius was writing, ‘Christians’ were taking up the old ‘pagan’ religious traditions as part of their concept of Christianity. One reason for this may have been that people found it easier to Christianize their old local customs than to invent completely new ones; another is that Christianity may have meant something quite different to the common person than it did to an intellectual such as Arnobius. In any case, it will sometimes be profitable to view notions of religious conflict as the inventions of a ‘rhetoric of difference’ or ‘rhetoric of conflict’ conceived of by early Christian and late pagan authors such as Arnobius as part of their polemics and propaganda. This phenomenon was a result of the intellectual debates of the late third century. The rhetoric of these debates has too often been interpreted uncritically, as if it were representative of the reality of interactions between paganism and Christianity. But perhaps we should not accept all of this rhetorical conflict at face value. After all, Arnobius himself was a pagan who voluntarily became a Christian, and so the details of his own life testify in favour of religious integration rather than conflict.

References

1 I have used the following critical texts of Adu. nat. in the preparation of this paper: Marchesi, C. (ed.), Arnobii Aduersus Nationes Libri VII. CSLP (Turin, 1953 2)Google Scholar; Bonniec, H. Le (ed. transl. comm.), Arnobe. Contre les Gentiles (Contre les Païens). Tome I, Livre I (Paris, 1982)Google Scholar; Champeaux, J. (ed. transl. comm.), Arnobe. Contre les Gentiles (Contre les Païens). Tome III, Livre III (Paris, 2007)Google Scholar; Fragu, B. (ed. transl. comm.), Arnobe. Contre les Gentiles (Contre les Païens). Tome VI, Livres VI–VII (Paris, 2010)Google Scholar. Where the Budé editions were unavailable, I relied upon Marchesi. Translations (where given) are my own, unless otherwise indicated. Cameron, Averil, Christianity and the Rhetoric of Empire: The Development of Christian Discourse (Sather Classical Lectures 55) (California, 1991), 14Google Scholar and passim is essential background to this study, as are the following works: Levieils, X., Contra Christianos: La critique sociale et religieuse du christianisme des origines au concile de Nicée (45–325) (Berlin, 2007)CrossRefGoogle Scholar; Ratti, S., Polémiques entre païens et chrétiens (Paris, 2012)Google Scholar.

2 See Dowden, K., European Paganism: The Realities of Cult from Antiquity to the Middle Ages (London, 2000), 4Google Scholar; Jones, C.P., Between Pagan and Christian (Cambridge, MA, 2014), 1–8CrossRefGoogle Scholar, especially 7. The terms ‘pagan’ or ‘paganism’ remain disputed in contemporary scholarship. The view for a long time has been that paganus was used as a derogatory term for non-Christians by early Christian authors, but—as Cameron, Alan, The Last Pagans of Rome (Oxford, 2011), 14–32Google Scholar demonstrates—this view is inaccurate, as in many cases paganus was used neutrally. A useful explanation is that paganus referred to those not included within the Christian community and was not necessarily a hostile term: Mohrmann, C., ‘Encore une fois: paganus’, VChr 6 (1952), 109–21Google Scholar, esp. 114 and 120. For the term nationes and its relation to gentes and ἔθνη, see Löfstedt, E., Late Latin (Oslo, 1959), 74–5Google Scholar.

3 The problem is about terminology. The need to attribute certain things to paganism or Christianity forced interpretations into a binary opposition; this has not been unchallenged in scholarship. For instance, Dowden (n. 2), 4 suggests that ‘paganism is simply the negative of Christianity’, an idea already refuted by Dagron, G., L'Empire romain d'Orient au IVe siècle et les traditions politiques de l'hellénisme: Le témoignage de Thémistios (Travaux et Mémoires 3) (Paris, 1968), 1Google Scholar; and cf. Fox, R. Lane, Pagans and Christians (London, 1986)Google Scholar for a masterly corrective. Recently these phenomena have been reconceived as integration between pagans and Christians rather than ‘conflict’ or ‘competition’: see Frankfurter, D., ‘Hagiography and the reconstruction of local religion in late antique Egypt: memories, inventions, and landscapes’, CHRC 86 (2006), 13–37Google Scholar; id., ‘Onomastic statistics and the Christianization of Egypt: a response to Depauw and Clarysse’, VChr 68 (2014), 284–9; Pina-Cabral, J. de, ‘The gods of the gentiles are demons: the problem of pagan survivals in European culture’, in Hastrup, K. (ed.), Other Histories (London, 1992), 45–61Google Scholar; Trombley, F.R., Hellenic Religion and Christianisation c.370–529, 2 vols. (Leiden, 1993–4)Google Scholar; Testa, R.L., ‘Le relazioni tra pagani e cristiani: nuove prospettive su un antico tema’, Cristianesimo nella Storia 30 (2009), 255–76Google Scholar.

4 Jones (n. 2), 4.

5 Distinction between ‘pagans’ and ‘Christians’: Rebillard, E., Christians and their Many Identities in Late Antiquity, North Africa, 200–450 c.e. (London, 2012)CrossRefGoogle Scholar. Christianization of ‘pagan’ deities: Simmons, M.B., Arnobius of Sicca: Religious Conflict and Competition in the Age of Diocletian (Oxford, 1995)CrossRefGoogle Scholar, 16, 183–218, 319–34.

6 Micka, E.R., The Problem of Divine Anger in Arnobius and Lactantius (Washington, DC, 1943), 158Google Scholar; Fragu (n. 1), xxxii–xxxvii; Simmons (n. 5), 1–45; Barnes, T.D., ‘Review: monotheists all?’, Phoenix 55 (2001), 142–62CrossRefGoogle Scholar; Duval, Y., ‘Sur la biographie et les manuscrits d'Arnobe de Sicca’, Latomus 45 (1986), 69–99Google Scholar; van der Putten, J.M.P.B., ‘Arnobe croyait-il à l'existence des dieux païens?’, VChr 25 (1971), 52–5Google Scholar argues that Arnobius does not theoretically exclude the existence of pagan gods, but rather indicates how the Christian God is supreme over them; Geffcken, J., Zwei griechische Apologeten (Leipzig and Berlin, 1907), 290Google Scholar suggests that Arnobius’ concept of the gods might have been taken from Plotinus, Enn. 2.9; see also Lactant. Diu. inst. 1.5.26.

7 Jer. Chron. 326–7 is discussed by Simmons (n. 5), 94–130.

8 Cf. Courcelle, P., ‘La polémique antichrétienne au début du IVe siècle. Qui sont les adversaires païens d'Arnobe?’, RHR 147 (1955), 122–3Google Scholar.

9 Kehoe, D.P., The Economics of Agriculture on Roman Imperial Estates in North Africa (Göttingen, 1988), 11CrossRefGoogle Scholar; Shaw, B.D., Rulers, Nomads, and Christians in Roman North Africa (Aldershot, 1995)Google Scholar.

10 Stēlai reveal the existence of the cult worship of Saturn in combination with worship to Saints, Christ and the Christian God: Beschaouch, A., ‘Une stèle consacrée à Saturne le 8 novembre 323’, BACTH 4 (1968), 253–68Google Scholar; Simmons (n. 5), 184–215.

11 Simmons (n. 5), 22–32; id., Universal Salvation in Late Antiquity: Porphyry of Tyre and the Pagan-Christian Debate (Oxford, 2015) also provides useful information on the competition between Porphyry and Arnobius.

12 Simmons (n. 5), 31–2, 190–2; Le Bonniec (n. 1), 8.

13 Simmons (n. 5), 47–93, 216–42, 319 dates it between 302 and 305, but Cornell, T. (ed.), The Fragments of the Roman Historians (Oxford, 2013)Google Scholar, 1.47 puts forward necessary cautions, suggesting a date between 296 and 311; Edwards, M.J., ‘Dating Arnobius: why discount the evidence of Jerome?’, AntTard 12 (2004), 263–71Google Scholar challenges these views with some controversial but thought-provoking arguments, dating it to 327. Simmons has unfairly called Edwards's position an ‘absolutely ridiculous conclusion’ (Simmons, M.B., ‘Review: B. Fragu, Arnobe. Contre les gentiles. Tome VI. Livres VI–VII’, BMCR 2012Google Scholar, http://bmcr.brynmawr.edu/2012/2012-08-36.html). It is significant that the only passage in which the work is explicitly mentioned dates it to the reign of Diocletian, between 284 and 1 May 305; see De uir. ill. 79 (cf. the notice in Chron. 326–7): Arnobius sub Diocletiano principe Siccae apud Africam florentissime rhetoricam docuit scripsitque aduersus gentes quae uulgo exstant uolumina; cf. De uir. ill. 80, Firmianus, qui et Lactantius, Arnobii discipulus, sub Diocletiano principe. It is also significant that between writing Chron. (c.380) and De uir. ill. (c.392–3) Jerome accessed Lactantius’ letters, which probably contained reliable information about Arnobius: Simmons (n. 5), 51; Kelly, J.N.D., Jerome: His Life, Writings and Controversies (London, 1975)Google Scholar, 83 n. 19. The controversy over the date is not yet settled. The most recent discussion is Edwards, M.J., ‘Some theories on the dating of Arnobius’, Ephemerides Theologicae Lovanienses 92 (2016), 671–84Google Scholar.

14 Croix, G.E.M. de Ste., the, ‘Aspects of “Great” Persecution’, HThR 47 (1954), 75–113Google Scholar; Frend, W.H.C., ‘The failure of the persecutions in the Roman empire’, P&P 16 (1959), 10–30Google Scholar; Davies, P.S., ‘The origin and purpose of the Persecution of a.d. 303’, JThS 40 (1989), 66–94Google Scholar.

15 On Arnobius’ polemic and its contexts: Amata, B., Problemi di antropologia arnobiana (BSR 64) (Rome, 1984), 142Google Scholar; Föllinger, S., ‘Aggression und Adaptation: zur Rolle philosophischer Theorien in Arnobius’ apologetischer Argumentation’, in Fuhrer, T. and Erler, M. (edd.), Zur Rezeption der hellenistischen Philosophie in der Spätantike: Akten der 1. Tagung der Karl-und-Gertrud-Abel-Stiftung vom 22.–25. September 1997 in Trier (Stuttgart, 1999), 13–32Google Scholar.

16 Millar, F., The Emperor in the Roman World (31 b.c. – a.d. 337) (Ithaca, NY, 1977), 573Google Scholar; Geffcken, J., The Last Days of Greco-Roman Paganism (Amsterdam, 1978), 56–74Google Scholar; Liebeschuetz, J.H.W.G., Continuity and Change in Roman Religion (Oxford, 1979), 252Google Scholar.

17 Millar (n. 16), 9 argues that the Greek philosophical climate around the court contributed to the outbreak of Diocletian's persecutions; cf. Corcoran, S., The Empire of the Tetrarchs: Imperial Pronouncements and Government, a.d. 284–324 (Oxford, 1996)Google Scholar, passim.

18 Hadot, P., Porphyre et Victorinus (Paris, 1968), 1.86Google Scholar; perhaps also Cornelius Labeo was an influence on Arnobius’ intellectual development: Kroll, W., ‘Die Zeit des Cornelius Labeo’, RhM 71 (1916), 309–57Google Scholar.

19 Euseb. Hist. eccl. 6.19.

20 The question has been raised as to whether Porphyry was a Christian or converted to Christianity at some point: Socrates, Hist. eccl. 3.23 with Schott, J.M., ‘“Living like a Christian, but playing the Greek”: accounts of apostasy and conversion in Porphyry and Eusebius’, JLA 1 (2008), 258–77Google Scholar and Kaabia, R., ‘Arnobe de Sicca du paganisme au christianisme: l’évolution cultuelle d'un lettré romano-africain’, in Ruggeri, P. and Cocco, M.B. (edd.), L'Africa romana: momenti di continuità e rottura, 3 vols. (Rome, 2015), 2.1217–28Google Scholar.

21 Simmons (n. 11), xii, 3–91 (trio of works on pagan soteriology); id. (n. 5), 216–42 (De philosophia ex oraculis), suggesting how Arnobius seems to respond directly to arguments contained in De philosophia ex oraculis in the first book of Aduersus nationes; id., ‘Review: R. Berchman, Porphyry Against the Christians’, JECS 16 (2008), 263–5, at 264 notes: ‘the extant fragments reveal more a work written for pagans than a predominantly anti-Christian book’. If it were not a predominantly anti-Christian book, it is however difficult to find other reasons for which Constantine had  Porphyry's anti-Christian works burned in 325 (Socrates, Hist. eccl. 1.9.30); see Digeser, E.D., ‘Lactantius, Porphyry, and the debate over religious toleration’, JRS 88 (1998), 129–46Google Scholar, at 131–2; id., The Making of a Christian Empire: Lactantius and Rome (Ithaca, NY, 2000), 90–6; cf. Johnson, A.P., Religion and Identity in Porphyry of Tyre: The Limits of Hellenism in Late Antiquity (Cambridge, 2013), 5, 300CrossRefGoogle Scholar.

22 Johnson (n. 21), 24; Smith, A., Porphyry's Place in the Neoplatonic Tradition: A Study in Post-Plotinian Neoplatonism (The Hague, 1974), 21Google Scholar.

23 Riedweg, C., Pythagoras: His Life, Teaching, and Influence (Ithaca, NY, 2005), 155–61Google Scholar.

24 Simmons (n. 5), 23; thus Augustine called Porphyry's works studies in the magical arts (De ciu. D. 10.29).

25 Cf. Beatrice, P.F., ‘Le traité de Porphyre contre les Chrétiens: l’état de la question’, Kernos 4 (1991), 119–38CrossRefGoogle Scholar, at 136; Le Bonniec (n. 1), 57.

26 Frend, W.H.C., ‘Prelude to the Great Persecution: the propaganda war’, JEH 38 (1987), 1–18Google Scholar. Scholarship has tended to maximize the literary aspects of these works while minimizing their oratorical polemical functions; for the fragments of Porphyry's Contra Christianos, consult Becker, M., Porphyrios, Contra Christianos: Neue Sammlung der Fragmente, Testimonien und Dubia mit Einleitung, Übersetzung und Anmerkungen: Texte und Kommentare (Berlin, 2016)Google Scholar with Cameron, Alan, ‘The date of Porphyry's ΚΑΤΑ ΧΡΙΣΤΙΑΝΩΝ’, CQ 17 (1967), 382–4CrossRefGoogle Scholar; Barnes, T.D., ‘Porphyry Against the Christians: date and the attribution of fragments’, JThS 24 (1973), 424–42Google Scholar; id., ‘Scholarship or propaganda? Porphyry Against the Christians and its historical setting’, BICS 39 (1994), 53–65; Berchman, R.M., Porphyry Against the Christians: Studies in Platonism, Neoplatonism, and the Platonic Tradition (Leiden, 2005)Google Scholar. On rhetoric in the world of Arnobius: Heath, M., ‘Rhetoric in mid-antiquity’, in Wiseman, T.P. (ed.), Classics in Progress (Oxford, 2002), 419–39Google Scholar.

27 Dunn, G.D., Tertullian (London, 2004), 25CrossRefGoogle Scholar.

28 de Labriolle, P., Histoire de la littérature latine chrétienne (Paris, 1924), 265Google Scholar is a strong proponent of this. On early Christian rhetoric more generally: Wilder, A.N., Early Christian Rhetoric: The Language of the Gospel (Cambridge, MA, 1964)Google Scholar; Hughes, F.W., Early Christian Rhetoric and 2 Thessalonians (Sheffield, 1989)Google Scholar; Lukas, V., Rhetorik und literarischer ‘Kampf’: Tertullians Streitschrift gegen Marcion als Paradigma der Selbstvergewisserung der Orthodoxie gegenüber der Häresie (Frankfurt am Main, 2008)Google Scholar; Sider, R., Ancient Rhetoric and the Art of Tertullian (Oxford, 1971)Google Scholar; Cameron (n. 1). For a comparative analysis of the rhetoric of Arnobius and Firmicus Maternus, see Semperena, L. and Guadalupe, M., ‘Rhétorique et argumentation dans l'apologétique latine de la période constantinienne’, in Puertas, A.J. Quiroga (ed.), Rhetorical Strategies in Late Antique Literature: Images, Metatexts and Interpretation (Leiden, 2017), 44–72Google Scholar.

29 Heath, M., ‘Porphyry's rhetoric’, CQ 53 (2003), 141–66CrossRefGoogle Scholar.

30 On the culture of Latin apologetic: Edwards, M.J., Religions of the Constantinian Empire (Oxford, 2015), 19–40CrossRefGoogle Scholar.

31 Euseb. Praep. euang. 5.1.10 = A. Harnack, Porphyrios, Gegen die Christen (Berlin, 1916), fr. 80: νυνὶ δὲ θαυμάζουσιν εἰ τοσούτων ἐτῶν κατείληφεν ἡ νόσος τὴν πόλιν, Ἀσκληπιοῦ μὲν ἐπιδημίας καὶ τῶν ἄλλων θεῶν μηκέτι οὔσης; Ἰησοῦ γὰρ τιμωμένου οὐδεμιᾶς δημοσίας τις θεῶν ὠφελείας ᾔσθετο.

32 Adu. nat. 1.3, arguing that pagans blame disasters on the offences of the Christians (iniuriis … offensionibus).

33 Adu. nat. 2.13. On pagan and Christian theories of salvation, see: Chadwick, H., ‘Christian and Roman universalism in the fourth century’, in Wickham, L.R. and Bammel, C.P. (edd.), Christian Faith and Greek Philosophy in Late Antiquity (Leiden, 1993), 26–42Google Scholar; Simmons (n. 5), 264–303; Hirschman, M., ‘Rabbinic universalism in the second and third centuries’, HThR 93 (2002), 101–15Google Scholar; Buell, D.K., ‘Race and universalism in early Christianity’, JECS 10 (2002), 429–68Google Scholar; Simmons, M.B., ‘Via universalis salutis animae liberandae: the Pagan-Christian debate on universalism in the later Roman empire (a.d. 260–325)’, StPatr 40 (2006), 245–61Google Scholar; id., ‘Porphyrian universalism: a tripartite soteriology and Eusebius's response’, HThR 102 (2009), 169–92; id. (n. 11).

34 The notion of conflict is reflected in some contemporary scholarship, influenced by Momigliano, A. (ed.), The Conflict between Paganism and Christianity in the Fourth Century (Oxford, 1963)Google Scholar. In recent work the discourse has been broken down, though it was immediately questioned by P. Brown, ‘Recensione’, The Oxford Magazine (May 16, 1963), 300–1. See Testa, R.L. (ed.), The Strange Death of Pagan Rome (Turnhout, 2013)Google Scholar; Salzman, M., Sághy, A. and Testa, R.L. (edd.), Pagans and Christians in Late Antique Rome: Conflict, Competition, and Coexistence in the Fourth Century (Cambridge, 2016)Google Scholar, passim. John Chrysostom's representation of pagan attacks upon Christian doctrine reveals the difficulty Christians had in differentiating themselves from paganism: ‘How do you accuse us of polytheism when you have three gods?’ (πῶς … ἡμῖν ἐκαλεῖ πολυθεΐαν;): see John Chrysostom, In Joann. Hom. 17.4 (= PG 49.112) with translation and discussion in Cameron (n. 2), 29.

35 Arnobius has been described as an early Voltaire: Labriolle (n. 28), 265. Style and tone: Gabarrou, F., Le Latin d'Arnobe (Paris, 1921)Google Scholar; Hagendahl, H., La prose métrique d'Arnobe (Göterborg, 1936)Google Scholar; K.B. Francke, Die Psychologie und Erkenntnislehre des Arnobius (Diss., Leipzig, 1878); Viciano, A., Retórica, filosofía y gramática en el Aduersus nationes de Arnobio de Sica (Frankfurt and Bern, 1993)Google Scholar. Greek influences: Festugière, A.-J., ‘Arnobiana’, VChr 6 (1952), 208–54Google Scholar; Fragu (n. 1), xxiv–xxv; Plato as supreme philosopher: Adu. nat. 2.36, though being pagan he is uester Plato at 2.13–14. Arnobius’ classicizing or archaic style: van der Putten, J.M.P.B, ‘L'emploi de nec et neque dans Arnobe’, VChr 25 (1971), 51Google Scholar; T. Lorenz, De clausulis Arnobianis (Diss., Breslau, 1910); Mohrmann, C., Études sur le latin des chrétiens III (Paris, 1961), 138Google Scholar.

36 Arnobius’ knowledge of Roman poetry: Fragu (n. 1), xxiv–xxix; Freund, S., Vergil im frühen Christentum: Untersuchungen zu den Vergilzitaten bei Tertullian, Minucius Felix, Novatian, Cyprian und Arnobius (Paderborn, Munich, Vienna and Zurich, 2003), 256–342Google Scholar; Le Bonniec, H., ‘Echos ovidiens dans l’Aduersus nationes d'Arnobe’, in Chevalier, R. (ed.), Colloque présence d'Ovide (Paris, 1982), 145–57Google Scholar.

37 On Cicero, see Arnobius’ adulation at Adu. nat. 3.6; Le Bonniec (n. 1), 47; id., ‘L'exploitation apologétique par Arnobe du De natura deorum de Cicéron’, Caesarodunum 19 (1984), 89–101; but note Edwards (n. 30), 39. Cf. Festugière (n. 35).

38 Arn. Adu. nat. 3.6–7; L. Táran, Collected Papers: 1962–1999 (Leiden, 2001), 463.

39 MacCormack, S., ‘Cicero in Late Antiquity’, in Steel, C.E.W. (ed.), The Cambridge Companion to Cicero (Cambridge, 2013), 251–306CrossRefGoogle Scholar, at 256–7.

40 Adu. nat. 2.73: non doctorum in litteris continetur Apollinis nomen Pompiliana indigitamenta nescire? Cf. Lipka, M., Roman Gods: A Conceptual Approach (Leiden, 2009), 70–1CrossRefGoogle Scholar. On Arnobius and the indigitamenta, see now Champeaux, J., ‘À l’école de Varron: Arnobe et les indigitations’, in Garcea, A., Lhommé, M.-K. and Vallat, D. (edd.), Polyphonia Romana: hommages à Frédérique Biville (Hildesheim, 2013), 743–53Google Scholar.

41 As well as drawing on Varro's style, Arnobius cites Varro's information thus: e.g. Adu. nat. 6.3.8 (ut tradit in Admirandis Varro), 3.39, 3.40, 3.41, 4.3, 5.8 (Varro ille Romanus multiformibus eminens disciplinis et in uetustatis indagatione rimator), 6.3, 6.11, 6.23, 7.1, 7.2. Cf. Le Bonniec (n. 1), 48.

42 Frend (n. 26), 14; Momigliano, A., On Pagans, Jews, and Christians (Middletown, CT, 1987), 61–5Google Scholar, 71.

43 See Champeaux, J., ‘Arnobe lecteur de Varron (Adu. nat. III)’, REAug 4 (1994), 327–52Google Scholar; Lehmann, Y., Varron théologien et philosophe romain (Collection Latomus 237) (Brussels, 1997)Google Scholar.

44 Sogno, C., ‘Persius, Juvenal, and the transformation of satire in Late Antiquity’, in Braund, S. and Osgood, J. (edd.), A Companion to Persius and Juvenal (Oxford, 2012), 363–86CrossRefGoogle Scholar, at 369.

45 Adu. nat. 5.18.5. Cf. Masterson, M., ‘Authoritative obscenity in Iamblichus and Arnobius’, JECS 22 (2014), 373–98Google Scholar.

46 On Christian self-definition, see generally Meyer, B.F., Sanders, E.P. (edd.), Self-Definition in the Greco-Roman World. Vol. 3: Jewish and Christian Self-Definition (Philadelphia, 1983)Google Scholar, passim. A useful brief discussion is Löhr, W., ‘Modelling second-century Christian theology’, in Paget, J.C. and Lieu, J. (edd.), Christianity in the Second Century: Themes and Development (Cambridge, 2017), 151–68CrossRefGoogle Scholar, at 168. For other treatments, see Pocock, J.G.A., Barbarism and Religion. Vol. 1: The Enlightenments of Edward Gibbon (Cambridge, 2004), 65–6Google Scholar: the early Church's struggle for self-definition is discussed in Jurieu's, P. Histoire critique des dogmes et des cults (Paris, 1704)Google Scholar.

47 Cf. McCracken, G.E., Arnobius of Sicca: The Case Against the Pagans (Westminster, MD, 1949), 23Google Scholar; Simmons (n. 5), 7, 113–17; Edwards, M.J., ‘The flowering of Latin apologetic’, in Edwards, M.J., Goodman, M., Price, S. and Rowland, C. (edd.), Apologetics in the Roman Empire: Pagans, Jews, and Christians (Oxford, 1999), 197–222Google Scholar; Testa, R.L., ‘Il terrore delle leggi in difesa dell’insatiabilis honor della Chiesa: la retorica della rappresentazione cristiana dell'Impero’, in Paño, M.V. Escribano and Testa, R.L. (edd.), Política, religión y legislación en el imperio romano. Politica, religione e legislazione nell'impero romano (IV e V secolo d. C.) (Bari, 2014), 117–38Google Scholar; vital background in Price, S., ‘Religious mobility in the Roman empire’, JRS 102 (2012), 1–19Google Scholar.

48 νυνὶ δὲ θαυμάζουσιν εἰ τοσούτων ἐτῶν κατείληφεν ἡ νόσος τὴν πόλιν, Ἀσκληπιοῦ μὲν ἐπιδημίας καὶ τῶν ἄλλων θεῶν μηκέτι οὔσης; Ἰησοῦ γὰρ τιμωμένου οὐδεμιᾶς δημοσίας τις θεῶν ὠφελείας ᾔσθετο, Euseb. Praep. euang. 5.1.10 = Harnack (n. 31), fr. 80. The passage indicates competition between rival cults. On rivalry and imitation between the cults of Asclepius and Jesus: Yeung, M.W., Faith in Jesus and Paul (Tübingen, 2002), 83–95Google Scholar, especially 89; Wells, L., The Greek Language of Healing from Homer to New Testament Times (Berlin and New York, 1998), 13–102Google Scholar, 120–54.

49 van der Putten, J.M.P.B., ‘Arnobiana’, VChr 25 (1971), 40–50Google Scholar, at 42–3, on Arnobius’ use of gens.

50 Simmons (n. 5), 319–34.

51 M. Leglay, Saturn Africain. Histoire (Paris, 1966) (= SAH), 1.97–9, 1.190–1; for the image, see Leglay, M., Saturne Africain. Monuments, 2 vols. 1: Afrique Proconsulaire; 2. Numidie-Maurétanies (Paris, 1966)Google Scholar (= SAM), 1.227; Picard, G.C., Les religions de l'Afrique antique (Paris, 1954), 120Google Scholar.

52 Simmons (n. 5), 193.

53 Lepelley, C., Les cités de l'Afrique romaine au bas-empire, 2 vols. (Paris, 1979–81), 1.34Google Scholar; Peyras, J., ‘Le fundus aufidianus: étude d'un grand domaine romain de la région de Mateur’, AntAfr 9 (1975), 181–222CrossRefGoogle Scholar, at 216; CIL 8.11824 = ILS 7457.

54 SAM 1.230 no. 1 = Merlin, A., Inscriptions Latines de la Tunisie (Paris, 1944)Google Scholar (= ILT), 573.

55 SAM 1.230 no. 1 = ILT 573.

56 SAM 1.236 no. 5; SAM 1.237 no. 7; perhaps the ritual referred to at Adu. nat. 7.29.4.

57 SAM 1.291 no. 1, plate 7, fig. 1.

58 Ferchiou, N., ‘Témoignages du culte de Saturne dans le Jebel Mansour (Tunisie)’, CahTun 26 (1978), 9–25Google Scholar, at 23 fig. 1.

59 Bonniec, H. Le, ‘“Tradition de la culture classique”: Arnobe témoin et juge des cultes païens’, BAGB 2 (1974), 201–22Google Scholar.

60 Cf. Brown, P., The Cult of the Saints: Its Rise and Function in Late Antiquity (Chicago, 1981), 5Google Scholar; Mora, F., Arnobio e i culti di mistero: Analisi storico-religiosa del V libro dell'Aduersus Nationes (Rome, 1994)Google Scholar.

61 e.g. I.Carth. 3.271 (= Ennabli, L., Les Inscriptions funéraires chrétiennes de Carthage. Vol. 3: Carthage intra et extra muros [Paris and Rome, 1991]Google Scholar).

62 See Simmons (n. 5), 193–211.

63 Tertullian is perhaps referring to the sun cult, which emerged in the 270s at Rome: Lane Fox (n. 3), 593. On this statement of Tertullian, see Taylor, M., Anti-Judaism and Early Christian Identity: A Critique of the Scholarly Consensus (Leiden, 1995), 14Google Scholar.

64 Cadotte, A., La Romanisation des Dieux: L'interpretatio romana en Afrique du Nord sous le Haut-Empire (Leiden, 2007)Google Scholar; Benabou, M., La résistance africaine à la romanisation (Paris, 1976)Google Scholar; Leglay (n. 51 [SAH]).

65 Blomart, A., ‘Frugifer: une divinité mithriaque léontocéphale décrite par Arnobe’, RHR 210 (1993), 5–25CrossRefGoogle Scholar.

66 e.g. M. Garcia, Saint Alban and the Cult of Saints in Late Antique Britain (Diss., Leeds, 2010); Ginzburg, C., I benandanti: Stregoneria e culti agrari tra Cinquecento e Seicento (Turin, 2002 2)Google Scholar; Hen, Y., Culture and Religion in Merovingian Gaul, a.d. 481–751 (Leiden, 1995)Google Scholar. The Gosforth Cross (a.d. 930–50) in Cumbria is an outstanding example of the unification of Christian and pagan symbolism: Berg, K., ‘The Gosforth Cross’, JWI 21 (1958), 27–43Google Scholar.

67 It was for theologians such as Arnobius and Porphyry to supply the philosophical difference, if there was little practical difference. Consider the way in which the Christian God gradually assumed Saturn's role as the protector of agriculture in Sicca: Simmons (n. 5), 211–15. It seems that, on undertaking such assimilation, Christian authors accepted the necessity for peaceful integration in practice while strongly rejecting it in rhetoric.

68 Arnobius has perhaps adapted παρονομασία, or the device of etymologizing wordplay, from Varro, who also used it to describe religious matters (Ling. 5.61).

69 See e.g. Dowden (n. 2), 7–8. On Arnobius’ treatment of rural divinities, see now Champeaux, J., ‘Au pays des dieux: la géographie religieuse d'Arnobe dans l’Aduersus nationes’, in Julia, M.-A. (ed.), Nouveaux horizons sur l'espace antique et moderne: actes du symposium ‘Invitation au voyage’ juin 2013, Lycée Henri IV (Bordeaux, 2015), 47–56CrossRefGoogle Scholar.

70 See Curran, J.R., Pagan City and Christian Capital: Rome in the Fourth Century (Oxford, 2000)Google Scholar; Testa, R.L., ‘Concluding remarks: Vrbs Roma between pagans and Christians’, in Salzman, M., Sághy, A. and Testa, R.L. (edd.), Pagans and Christians in Late Antique Rome: Conflict, Competition, and Coexistence in the Fourth Century (Cambridge, 2016), 399–407Google Scholar.

71 Simmons (n. 5), 187.

72 Warmington, B.H., The North African Provinces from Diocletian to the Vandal Conquest (Cambridge, 1954), 66Google Scholar.

73 Kehoe (n. 9), 11.

74 Shaw (n. 9), lxxx.

75 Gsell, S., ‘Esclaves ruraux dans l'Afrique romaine’, in Mélanges Gustave Glotz (Paris, 1932), 1.397–415Google Scholar, at 1.415 n. 1.

76 Kehoe (n. 9), 118.

77 Kehoe (n. 9), 118; Gummerus, H., Der römische Gutsbetrieb (Leipzig, 1906), 84Google Scholar; Neeve, P.W. de, Colonus (Amsterdam, 1984)Google Scholar, 102 n. 173; Martin, R., Recherches sur les agronomes latins et leurs conceptions économiques et sociales (Paris, 1971), 347–8Google Scholar.

78 Kehoe, D., ‘Private and imperial management of Roman estates in North Africa’, Law and History Review 2 (1984), 241–63CrossRefGoogle Scholar.

79 Kehoe (n. 9), 11–12.

80 See Beschaouch (n. 10).

81 Jordan, D., ‘Two Christian prayers from south-eastern Sicily’, GRBS 25 (1984), 297–302Google Scholar; id., ‘Cloud-drivers and damage from hail’, ZPE 133 (2000), 147–8.

82 On Arnobius’ treatment of these topics, cf. also Bonniec, H. Le, ‘Un témoignage d'Arnobe sur la cuisine de sacrifice romain’, REL 63 (1985–7), 183–92Google Scholar.

83 McGowan, A., Ascetic Eucharists: Food and Drink in Early Christian Ritual Meals (Oxford, 1999), 3CrossRefGoogle Scholar.

84 McGowan (n. 83), 12.

85 Alikin, V.A., The Earliest History of the Christian Gathering: Origin, Development and Content of the Christian Gathering in the First to Third Centuries (Leiden, 2010), 35CrossRefGoogle Scholar.

86 On this subject, see Diosono, F., ‘Professiones gentiliciae: the collegia of Rome between paganism and Christianity’, in Salzman, M.R., Testa, R.L. and Sághy, M. (edd.), Pagans and Christians in Late Antique Rome (Cambridge, 2016), 251–72Google Scholar.

87 See Trombley (n. 3), passim.