HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

The Alexandrian Riots of 356 
and George of Cappadocia 

Christopher Haas 

I N LATE JUNE 346, Gregory of Cappadocia died in Alexandria 
after a long illness. As the imperial appointee to the a­
triarchate of Alexandria, Gregory had persecuted the fol­

lowers of Bishop Athanasius and had generously rewarded his 
fellow Arians within the city during the six years of his tenure 
on the throne of St Mark. Gregory was not, however, a 
successful promoter of the Arian cause in Alexandria. Indeed, 
the growth and viability of the Arian community there was 
largely the work of Gregory's countryman, Philagrius, a veteran 
prae/ectus Aegypti, who effectively employed all the powers of 
patronage and coercion in the hands of the imperial admini­
stration.! 

By mid-October 346, Athanasius returned from exile to a 
thunderous welcome at Alexandria (Hist. Aceph. 1.1). Gregory 
Nazianzus (Or. 21.29) says that Athanasius' restoration was 
greeted with "universal cheers, the pouring forth of unguents, 
nightlong festivities, the whole city gleaming with light, and 
both public and private feasting.» This return ushered in the so­
called 'Golden Decade' of Athanasius' episcopate, during which 
he vigorously lobbied for support among eastern bishops, 
launched a missionary initiative to the southern kingdom of 
Axum, and undertook a major program of church building in 
Alexandria.2 But after the assassination of Athanasius' imperial 

1 The principal sources for Gregory's episcopate and Philagrius' prefecture(s) 
are: Festal Index 11-18; Athan. Ep. Ency., ApoL c. Ar. 30, Hist. Ar. 10-21, Fest. 
Ep. 10; Greg. Naz. Or. 21.28; Socrates HE 2.11, 14; Sozomen HE 3.6, 12. 

2 Socrates IlE 2.24; see also E. R. Hardy, Christian Egypt (New York 1952) 
61 f. For recent surveys of Athanasius' career see the important collection of 
articles in C. KANNENGIESSER, ed., Politique et theologie chez Athanase 
d'Alexandrie (Paris 1974: hereafter' Kannengiesser'); M. Tetz, in Theologische 
Realenzyklopadie 4 (Berlin 1978) 333--49; G. C. Stead, in Dizionario Patristico 
e di Antichita Cristiane 1 (Rome 1983) 413-32; A. S. Atiya, in The Coptic 
Encyclopedia 1 (New York 1991: hereafter' CoptEnc') 298-302; T. D. Barnes, 

281 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

282 THE ALEXANDRIAN RIOTS OF 356 

patron Constans in 350, the Arian emperor Constantius felt he 
had a free hand to increase pressure on the recalcitrant Atha­
nasius. Imperial notarii presented the Alexandrians with a series 
of admonitions to abandon Athanasius and embrace the 
theology of the imperial court. 3 One notarius, Diogenes, spent 
four months in late 355 trying in vain to effect Athanasius' 
removal: "since the people and the judges strongly resisted 
Diogenes, [he] returned [to Constantinople] without success."4 
Constantius, however, was determined to impose his will on 
the city. In early January 356, the dux Syrianus and yet another 
notarius, Hilary, arrived in Alexandria accompanied by troops 
gathered from Egypt and Libya. 5 An uneasy stand-off ensued 
until the night of 8 February. Encouraged by Arian elements 
within the city, Syrian us and his troops made an armed attack 
on the Church of St Theonas during a nocturnal vigil. 
Athanasius had been officiating at the service, and only the 
intervention of his close attendants secured his narrow escape 
from the subsequent violence. 6 Within a year Alexandria again 
had an imperial appointee as bishop, George of Cappadocia, an 
Arian "who ruled by force rather than by priestly moderation" 
(Sozom. HE 4.10). 

This chapter in the tumultuous history of Arian/Homoousian 
conflict in Alexandria typifies a scenario played out repeatedly 
in the city's Christian community during the middle decades of 
the fourth century. Yet, on closer examination, the ongoing 
violence of 356 and George's subsequent episcopate manifested 
much broader intercommunal tensions within Late Antique 
Alexandria.? George's turbulent reign marked a watershed in 

Athanasius of Alexandria: Theology and Politics in the Constantin ian Empire 
(forthcoming). A useful bibliographical essay in C. Kannengiesser, "The 
Athanasian Decade 1974-1984," TheoSt 46 (1985) 524-41. 

3 Athan. Hist. Ar. 31; Hist. Aceph. 1.7f; Festal Index 25. 
4 Hist. Aceph. 1.9: populo 7Jero resistente Diogeni 7Jchementer et iudicibus, 

re7Jersus est Diogenes sine effectu. 
5 Athan. Apol. ad Constant. 22ff; Sozomen HE 4.9. 

6 Festal Index 28; Hist. Aceph. 1.1 Of; Athan. Apol. ad Constant. 25, Apol. de 
fuga 24. 

7 The enormous bibliography on Ptolemaic and early Roman Alexandria 
contrasts with the little on the Late Antique period. See, however, M. 
Rodziewicz, Les habitations romaines tardi7Jes d'Alexandrie a La lumiere des 
fouilles polonaises a Kom el-Dikka (=Alexandrie III ~Warsaw 1984]); P. M. 
Fraser, "Alexandria, Christian and Medieval," in CoptEnc 1 (1991) 88-92; H. 
Heinen, "Alexandria in Late Antiquity," in CoptEnc 1 (1991) 95-103; C. J. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 283 

the social and religious history of Alexandria. His unexpected 
campaign against the city's pagan cults galvanized the diffuse 
pagans into a recognizable community that defended its 
religious patrimony with bloodshed. His inept renewal of 
violent persecution against the Athanasians doomed the Arian 
cause and laid the foundations for unswerving popular support 
of the Alexandrian patriarchate in its contests with imperial 
authority over the next three centuries. 8 

Shortly after the followers of Athanasius were expelled from 
the churches in June 356, a large crowd of pagans looted the 
newly-renovated "Great Church" in the Caesarion. Deliberate 
acts of desecration and various sorts of pagan rituals accom­
panied this pillaging. Although Athanasius (the main source) 
describes these events in great detail, modern accounts 
generally ignore them, perhaps because modern commentators 
tend to exhibit a nearly exclusive preoccupation with ecclesias­
tical matters during Athanasius' patriarchate.9 Moreover, mod­
ern scholars often gloss over the year-long interlude between 
Athanasius' exile and the entry of the Arian bishop, George of 
Cappadocia, in February 357, and some even deny the 
existence of this hiatus in episcopal tenancy. 10 Far from being an 

Haas, Late Roman Alexandria (forthcoming). For the wider Egyptian context 
of this period see the useful summary of R. S. Bagnall, "Late Roman Egypt," 
in Dictionary of the Middle Ages 10 (New York 1988) 453-56, and his fuller 
exposition in Egypt in Late Antiquity (Princeton, forthcoming). 

8 General treatments of this conflict in Hardy (supra n.2) 58-66; W. H. C. 
Frend, The Rise of Christianity (Philadelphia 1984) 536-43; C. W. Griggs, 
Early Egyptian Christianity (Leiden 1990) 142-46 (largely derivative of Hardy 
and Frend); fuller analysis in M. Simonetti, La crisi ariana del IV secolo 
(Rome 1975) 226-30, 326-33; E. D. Hunt, "Christians and Christianity in 
Amrnianus Marcellinus," CQ NS. 35 ( 1985) 186-200; J. Matthews, The Roman 
Empire of Ammianus (Baltimore 1989) 441-44; and M. Caltabiano, 
"L'assassinio di Giorgio di Cappadocia," QuadCatStCl7 (1985) 17-57, who 
rightly emphasizes how fifth-century church historians reinterpreted the 
events of 361 as a manifestation of theological conflict, thereby discounting 
social and political factors. Consequently, the near contemporary sources 
acquire an even greater importance in understanding George's episcopate. 

9 A. Martin's otherwise outstanding analysis of Athanasius' episcopate only 
briefly notes these events in a section on "Alexandrie la violente": H istoire 
"Acephale" et Index Syriaque des Lettres Festales d'Athanase d'Alexandrie 
(=Sources Chretiennes 317 [Paris 1985]) 102L 

10 H ist. Aceph. 2.1 f; Festal Index 18f; careful reconstruction of the 
chronology in Martin (supra n.9) 89-97; cf W. Bright, "Georgius of Cap­
padocia," DCB 2 (1880) 637-40. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

284 THE ALEXANDRIAN RIOTS OF 356 

isolated incident, the role of the city's pagans in Alexandrian 
civic strife only increased during the course of George's iII­
fated episcopate. These tensions culminated in the murder of 
the Cappadocian bishop and two imperial officials in Novem­
ber 361. The scale of this public violence suggests the need for 
a careful assessment of its significance for the history of the 
intercommunal relations in Alexandria. 

After the defeat of the usurper Magnentius at Mursa in Sep­
tember 351 (and his later suicide in 353), Constantius directed 
his energies toward eradicating any remaining pockets of dis­
sent that might threaten the unity of the empire.11 Pagans and 
Homoousians were singled out as special objects of his wrath. 
Against the former, Constantius enacted a series of laws in 
353-356 that banned nocturnal sacrifices, closed and demol­
ished a number of temples, and forbade sacrifice. 12 During his 
sojourn in Rome in 357, the emperor also ordered the removal 
of the altar of Victory from the Senate house, the first salvo in a 
decades-long struggle over this important symbol of traditional 
Roman religion. 13 Against the adherents of Nicaea, Constantius 
convoked councils at ArIes (353) and Milan (355), leading to the 
expulsion of Hilary of Poi tiers and Liberius of Rome from their 
episcopal sees. Their exiles resulted largely from vocal support 
for the bishop whom Constantius regarded as the villain behind 
the church's ongoing schism-Athanasius. Serious allegations 
that Athanasius had" subverted all Egypt and Libya» during the 
revolt of Magnentius and Decentius fueled Constantius' 
antipathy for the Alexandrian patriarch (Socrates HE 2.26). Such 
accusations were apparently more than the malicious 
whisperings of Constantius' courtiers: Magnentius' envoys to 
negotiate with Constantius had passed through Alexandria on 
their way to the emperor. 14 

II Zos. 2.51f; Epit. de Caes. 42.4-8; Eutrop.10.12. On Constantius' relations 
with Athanasius see K. M. Girardet, "Constance II, Athanase et l'edit d'Aries 
(353): a propos de la politique religieuse de l'empereur Constance II," and L. 
W. Bernard, "Athanase et les empereurs Constantinet Constance," in Kannen­
giesser 65-91 and 127-43. More generally see R. Klein, Constantius II. und die 
christliche Kirche (Darmstadt 1977); C. Vogler, Constance II et l'administra­
lion imperiale (Strasbourg 1979). 

12 Cod. Theod. 16.10.2-6; Liban. Or. 72.8. 

13 Ambrose Ep. 18.22; Symmachus Relat. 3.7. 
14 Marcellinus, Magnentius' magister militum, and Nunechius, a senator 

and possibly a Praetorian Prefect, sent to Constantinople via Alexandria: Petro 
Pat. fr.16, FHG IV 184-91. Athanasius neglects to mention them and speaks 
only of ecclesiastical envoys: Apol. ad Constant. 6-11. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 285 

Four months after Athanasius' ouster, the new prefect 
Cataphronius arrived in Alexandria (10 June 356) with Heraclius 
as comes (Hist. Aceph. 2.1f). On hand to assist them was Faus­
tinus, an Arian catholicus (Athan. Hist. Ar. 55, 58). These hand­
picked administrators of Constantius immediately instituted a 
new regime in Alexandria. Unlike their predecessor Syrianus, 
who had directed his attention solely to the Christians-on one 
occasion even setting three thousand soldiers upon a large body 
of Christians holding prayer services in the cemeteries-the 
new administrators cast their net much more widely (Athan. 
Apol. ad Constant. 27). The pagans of the city were threatened 
with the destruction of their cult images if they failed to support 
the imperial nominee succeeding the exiled patriarch. De facto 
leaders within the loosely-structured pagan community (pagan 
magistrates and ordinary keepers of temples) were especially 
singled out and forced to swear loyalty to whatever nominee 
the emperor might send. 15 In this same context, Athanasius says 
that certain urban collegia were also coerced into professing 
loyalty, perhaps indicating their character as cult associations or, 
less explicitly, as craft associations dedicated to the service of 
their patron god. 16 

Four days after the new imperial administrators arrived, a 
pagan crowd attacked the followers of Athanasius worshipping 
in the Great Church. Not surprisingly, Athanasius (Hist. Ar. 
55f) ascribes the ensuing riot to the collusion, if not the active 
encouragement, of the prefect, the comes, and the catholicus. 
His account of this riot, however, suggests that, even if the 
rioters began their violent demonstration at the prompting of 
the officials, the riot soon took a course that could not have 
been orchestrated beforehand. The violence began in the agora 
when groups of young pagans ('trov uyopalwv in Athanasius) 
attacked the most prominent church in the vicinity, the so­
called Great Church at the Caesarion. 17 Constantius had handed 

15 Athan. Hist. Ar. 54, 760a: ~ouA.£U'tar; Kat Ollila-tar; £8VtKOUr; V£WKOpOUr; 'troY 
£iowA.wv. 

16 !list. Ar. 55, 760a: 'ttV£r; 'troY £pya<nrov. 
17 Ilist. Ar. 55, 760b. Cj Acts 17:1-9, where a tumult in Thessaloniki occurs 

after "the Jews ... rounded up some bad characters from the marketplace 
('trov ayopaiwv o.vopar; 'ttvar; ltOVTlPOur;), formed a mob, and started a riot in 
the city." In a near contemporary text, Plutarch describes Scipio Africanus 
"rushing into the Forum attended by men who were of low birth (ayf.vvnr;) 
and former slaves, who were frequenters of the Forum (ayopaiour;) and able to 

gather a mob": Pluto Aem. 38.3. Hence, ayopaior; becomes a synonym for 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

286 THE ALEXANDRIAN RIOTS OF 356 

over to the church a portion of this former precinct of the 
imperial cult during the episcopate of the Arian patriarch 
Gregory (339-346).18 Only after Athanasius' return in 346, 
however, was it generally used as a church. 19 The much-exiled 
patriarch even risked imperial displeasure by conducting ser­
vices there before it could be officially consecrated, testimony 
to the growing needs of the Christian community (Athan. 
Apol. ad Constant. 14-17). The memory of the building's 
former status as one of the city's preeminent pagan sanctuaries 
was fresh in the minds of Alexandrian pagans, and it evidently 
rankled to see this sacred precinct employed in the worship of a 
condemned Galilean criminal. It was a natural target of their 
indignation. 

After violently disrupting a prayer service, the crowd of 
young pagans used the furniture and curtains of the church for 
a great bonfire in the small forum adjoining the sanctuary-a 
destructive act with clear religious overtones, as the mob then 
sprinkled incense on the blaze and sang praises to their gods. 
For a sacrifice, they seized the heifer that drew water in the 
gardens of the precinct, and relented only upon the discovery 
that it was female (Athan. Hist. Ar. 56). The significance of the 
episode is clear: the sacrifice of a sacred bull was associated with 
the worship of Serapis, the city's great tutelary deity. 20 Another 

'common', 'rowdy', or 'vulgar'. Earlier references include Philo In Flacc. 64, 
95; Leg. 122. This same same social grouping, "those of the Dramas," appears 
in Acts of Peter, BiblHagGr 1502, 1502a. 

Athanasius singles out the catholicus Faustinus as the principal agent 
inciting 'tcOv uyopatwv (Hisl. Ar. 58): "(The Arians) found that Faustinus, 
who is the catholicus by style, but is an uyopato~ in habits and profligate in 
heart, was ready to play his part with them in these proceedings and to stir up 
the heathen." 

18 For the the early history of the Caesarion see the standard discussions 
with full references in A. Calderini, Dizionario dei nomi geografici e 
topografici dell'Egitto greco-romano I.1 (Milan 1935) 118f; A. Adriani, 
Repertorio d'arte dell' Egitto greco-romano SER. c 1 (Palermo 1966) 215-26. 

19 Epiphanius Haer. 69.2.3. On the transformation of this shrine into a 
church see especially A. Martin, "Les premiers siecles du christianisme a 
Alexandrie: Essai de topographic religieuse (lIre_lye siecles)," REAug 30 
(1984) 211-25 at 217f. 

20 On the Serapis cult sec J. E. Stambaugh, Serapis under the Early 
Ptolemies (Leiden 1972); H. C. Youtie, "The Kline of Serapis," HThR 41 
(1948) 9-30; L Yidman, Isis und Serapis bei den Griechen und Romern 
(Berlin 1970); P. M. Fraser, Ptolemaic Alexandria (Oxford 1972) I 246-76. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 287 

group entered the church shouting religious slogans and waving 
tree branches, reminiscent of the worship of Dionysus, the one 
Hellenic god most closely associated with cults of Serapis and 
Osiris.21 A coterie of pagan women (maenads in Athanasius, 
Rist. Ar. 59, 764d) also took part in these demonstrations. 
Hence, allowing for some hyperbole from the patriarch, further 
confirmation that the popularity of Dionysus con tin ued 
beyond the Ptolemaic period, when he appeared as the divine 
patron of the great religious procession of Ptolemy Philadel­
phus. 22 The alliance of Dionysus and Serapis is not surprising: 
well into the Roman era Dionysus and Dionysiac imagery 
frequently occur in tomb decoration; and in the necropolis at al­
Afushi on Pharos, these motifs stand side by side with images 
culled from the worship of Osiris/Serapis. 23 

Why did this riot break out when it did-only four days after 
the arrival of the new administrators-and why did the violence 
assume such a religious complexion? Two complementary ex­
planations may be offered. Despite the installation of a new 
prefect and military comes, no episcopal replacement accom­
panied the new administrators, even though Athanasius had 
been ousted from public leadership for over four months. 
Their only task was to ensure the loyalty of the civil leadership 
of the city for the soon-expected Arian appointee. Once these 
oaths of loyalty were extracted, however, considerable time 
elapsed before the new Arian patriarch appeared. This gap in 
ecclesiastical leadership had the effect of creating a vacuum in 
Alexandria's precariously-balanced intercommunal relations. 24 

21 Athan. Hist. Ar. 57; Fraser (supra n.20) 205f. 
22 Callixenus ap. A tho 197 c-203B; see also E. E. Rice, The Grand Procession 

of Ptolemy Philadelphus (Oxford 1983). 
23 Adriani (supra n.18) 191-97; A. Bernand, Alexandrie la Grande (Paris 

1966) 206-09. 
24 During their long tenure in the episcopate, most late Roman bishops 

could develop extensive networks of local patronage and considerable 
influence in local affairs. This contrasts sharply with the revolving door of 
imperial appointments. In the five years discussed here (356-361), no fewer 
than seven prefects and three strategoi administered Alexandria. The imperial 
officials based their authority upon local traditions of deference and upon 
various forms of coercion at their disposal. On the Late Antique episcopacy 
see H. Chadwick, "The Role of the Christian Bishop in Ancient Society," in 
Protocol, Colloquy of the Center for Hermeneutical Studies (Berkeley 1979); 
E. Ferguson, Encyclopedia of Early Christianity (New York 1990) 1 SO-54 S.71. 

"Bishop." 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

288 THE ALEXANDRIAN RIOTS OF 356 

With no strong hand to guide the Christians, the pagans gave 
free reign to their animosity towards the growing Christian 
community. Their resort to mass violence introduced nothing 
new to Alexandrian power politics. Shortly after Athanasius' 
expulsion in 339, pagans joined with Arians in violent 
demonstrations centered on the churches of Quirinus and 
Dionysius.25 Alexandrian pagans-from upper-class bouleutai to 
paupers in the agora-had received tangible benefits from 
supporting the Arian party under Gregory and Philagrius. 26 

Like the riots of 339, this was yet another opportunity for the 
pagans to show their allegiance to their imperial patron and to 
attack with impunity their principal rivals. 

At the same time, it would not have taken a very discerning 
pagan to realize that Constantius' religious policy over the last 
five years had shifted for the worse. The tide of imperial 
legislation and pronoucements since the defeat of Magnentius 
was increasingly anti-pagan: not only new laws banning pagan 
cultic practices, but also a freer hand to enforce previous legis­
lation after removal of an armed pagan party in the West. Im­
perial representatives in Alexandria reflected this shift of policy. 
Although Syrianus had employed armed violence solely against 
Athanasians and had openly rewarded upper-class Alexandrians 
with bishoprics (and the attendant exemption from taxation), 
the new administration installed in early June, 356, offered only 
threats to the pagans, who had every reason to fear the im­
position of a new patriarch. 

Thus the rioting at the Caesarion may have taken a form 
familiar to students of early modern history, that of the 'church 
and king riot', i.e., a violent demonstration in support of tra­
ditional political and religious usages perceived as threatened. 27 

25 Athan. Ep. Ency. 3f, ApoL c. Ar. 30, Hist. Ar. 10-13. 
26 Ordination and tax exemption: Athan. Apol. ad Constant. 28; grain dole: 

Ep. Ency. 4, Hist. Ar. 31, 54; gifts of oil and other alms: Hist. Ar. 13,72; 
unspecified "promises": Ep. Ency. 3. 

27 The best discussion of the taxonomy of this phenomenon is E. P. Thomp­
son, "The Moral Economy of the English Crowd," Past and Present 50 (Feb. 
1971) 76-136; see also N. Z. Davis, "The Rites of Violence: Religious Riot in 
Sixteenth-Century France," Past and Present 59 (May 1973) 53-91; M. Mollat 
and P. Wolff, Ongles bleus, Jacques et Ciompi: Les revolutions populaires en 
Europe aux XIV' et XV' siecles (Paris 1970); S. Desan, Reclaiming the Sacred: 
Lay Religion and Popular Politics in Revolutionary France (Ithaca 1990) 
165-216. More generally on the structure of popular violence see E. J. Hobs­
bawn, Primitive Rebels (New York 1959); N. ]. Smelser, Theory of Collective 
Behavior (New York 1963); G. Rude, The Crowd in History (New York 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 289 

Far from revolutionary, these riots are often carried out in the 
name of a traditional political authority and with the tacit back­
ing of local administrators. The pagan violence at the Caesarion 
fits this paradigm exactly, and the rioters even chanted such 
slogans as "Constantius has become a Hellene!" and "The 
Arians recognize our deeds!" while they ransacked the newly­
dedicated church. 28 This aspect of the violence becomes all the 
more comprehensible, for the Caesarion had been the focus of 
ruler-cult in the city for nearly four centuries. Clearly, the 
pagans reacted against the desecration of their revered shrine 
and the overturning of traditional notions of divine rulers hip. 29 

Moreover, they probably regarded their Christian emperor 
with suspicion, since he had not only adjured the role of god­
king but had now sent out representatives to threaten other 
ancestral cults that had long maintained the divine order. A 
Syrian visitor during the 350s describes the depth of Alex­
andrian religious devotion: 

I think that in celebrating [this locale], it owes its particular 
renown to the gods, because there-as I have already said-the 
gods are especially honored by [artistic] representations. 
There, one can see every sort of consecrated shrine and lavishly 
adorned temple; sacristans, priests, attendants, haruspices, 

1963); J. S. McClelland, The Crowd and the Mob from Plato to Canetti 
(London 1989). 

On popular violence in the Classical and Late Antique periods: A. W. 
Lintott, Violence in Republican Rome (Oxford 1968), and Violence, Civil 
Strife, and Revolution in the Classical City (London 1982); P. J. J. Vander­
broeck, Popular Leadership and Collective Behavior in the Late Roman 
Republic (Amsterdam 1987); A. Cameron, Circus Factions (Oxford 1976); E. J. 
Jory, "Early Pantomine Riots," in A. Moffat, ed., Maistor: Classical, Byzan­
tine and Renaissance Studies for Robert Browning (=Byzantina Auslraliensia 
5 [1984]) 57-66; R. Browning, "The Riot of A.D. 387 in Antioch," JRS 42 
(1952) 13-20; T. E. Gregory, Vox Populi (Columbus 1979). 

28 Hist. Ar. 56, 761a: "EAAllV YfYOV£ Kwvcrtuv'tto<;, KO:l o~ 'Apnuvot ta llllWV 
£1t£yvwoav. " 

29 This form of analysis has been pioneered in the earlier period by W. D. 
Barry, Faces of the Crowd: Popular Society and Politics of Roman Alexan­
dria, 30 B.C.-A.D. 215 (diss.University of Michigan, 1988), a ground-breaking 
study suggesting that the Alexandrians did not have an ingrained predilection 
to rioting and violence, but instead were far more deferential in their behavior 
than modern historians commonly depict them. Further explication of this 
view in Barry's" Alexandrians, Orators, and the 'Mob': Dio Chrysostom and 
the World of Roman Alexandria," Historia (forthcoming). 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

290 THE ALEXANDRIAN RIOTS OF 356 

worshippers, and the best diviners all abound; and everything 
is performed according to the proper rites. Thus you will find 
altars constantly ablaze with the fires of sacrifices and heaped 
with incense, as well as garlands and censers filled with 
perfumes emitting a divine fragrance.3o 

This reputation for devotion so impressed the author of the 
Expositio that he was moved to exclaim, "Nowhere, in fact, are 
the mysteries of the gods celebrated as they have been here 
-from ancient times up till the present day."3! No wonder that 
the now-threatened pagans seized the opportunity afforded 
them by the prefect to restate forcefully their support for 
tradi tional usages.32 

In this atmosphere of heightened intercommunal tension, 
Constantius' episcopal nomineee, George of Cappadocia, 
entered the city on 30 Mechir (24 February) 357, and took up 
his post with the backing of imperial troops (Festal Index 19). 
George, one of the most intriguing characters in late Roman 
politics, had begun his career as a contractor of military supplies 
in Constantinople, and by the 340s enjoyed a comfortable living 
in Cappadocia, where he lent volumes from his "very large and 
complete" library to the future emperor Julian. 33 Unlike his 
Arian predecessor Gregory, all sources agree that George 
possessed a forceful personality and was a keen opportunist. 34 
Although Athanasius might call him an idolater with an 

30 Expositio totius mundi et gentium 36.19-27 (J. Rouge, ed. and tr. [Paris 
1966]). Although the anonymous author may be referring to the whole of 
Egypt, it can be argued that he never progressed beyond the confines of 
Alexandria and its immediate hinterland: Rouge 30f. 

31 34.7ff: nusquam enim deorum mysteria sic perficitur quomodo ibi ab 
antiquo et usque modo. 

32 A similar incident occurred some fifty years later in the North African 
town of Calama, when a crowd of pagans celebrating a festival (contrary to a 
recent law of Honorius) assaulted a church with the tacit support of the local 
administration: August. Ep. 91. 

33 Athan. Hist. Ar. 75 ({moo£x:'tllv £V KCilV<J't<lV'tlVO\l1tOA£l 't<l~1<l1(ffiv y£vo­
~£vov). Julian Ep. 106 (411c) Bidez-Cumont; see also A.-J. Festugiere, KJulien a 
Macellum" JRS 47 (1957) 53-58. Transfer from the imperial service to the 
episcopate was not unusual under Constantius, e.g. Eleusius of Cyzicus 
(Sozomen HE 4.20). 

H Amm. Marc. 22.11.4; Athan. De Syn. 37; Epiph. Haer. 76.1.1-8; Greg. Naz. 
Or. 21.16. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 291 

executioner's temperament, George's imperial patron Constan­
tius praised him in a letter to the Alexandrians as "the most 
perfect of beings as a guide for your conduct, both in word and 
deed. "35 

Owing his appointment to Constantius and the Arian bishops 
surrounding the emperor, George immediately sought to 
implement the policies of his imperial patron. Like his ad­
ministrative predecessors who initiated Constantius' new 
regime in the city and oversaw operations against the Homo­
ousians, George aimed to break all hints of local opposition­
whether Christian or pagan. According to Athanasius' hostile 
testimony, George instituted a fierce persecution of the fol­
lowers of the ousted patriarch. Clergy and ascetics were im­
prisoned, and the grain dole was withheld from the lower-class 
dependents of the former bishop. Wealthy supporters of 
Athanasius not only had their property confiscated but, in some 
cases, even their houses were destroyed CAthan. Apo!' de fuga 
6)-a particularly effective method of quelling dissent, as seen 
today in Israeli suppression of the Intifada. Worse followed. In 
May 357, soldiers of the dux came upon a band of Athanasians 
praying in one of the extra-mural cemeteries. Those unable to 
escape were severely beaten and many died. The survivors 
were banished to the Great Oasis CApo!. de fuga 6£). 

Although the brutality of George to the Athanasians simply 
followed the policy of the prefect and his recent! y-installed ad­
ministration, George's carefully orchestrated anti-pagan cam­
paign marked a new stage in the official coercion of religious dis­
sent in Alexandria and highlights the centers of pagan strength in 
the city. This new anti-pagan policy, which George imple­
mented with enthusiasm, eventually produced his downfall. 

George's main targets were the temples of the civic cults. 
Supported by the garrison, he attempted to prohibit pagan 
sacrifices and despoil the temples of their votive offerings and 
decorations. 36 To sap further the financial resources of the pagan 

35 Athan. De Syn. 37, Ad Fpiscopos 7; Constantius ap. Athan. Apol. ad 
Constant. 30. 

36 Julian E p. 60 (3 79b); Sozomen II E 4.30; for the broader context of 
pagan/Christian conflict over temples see L. Habachi, "The Destruction of 
Temples in Egypt," in S. Hanna, cd., Medieval and Middle Eastern Studies in 
I1onor of Aziz Suryal Atiya (Leiden 1972); G. Fowden, "Bishops and Temples 
in the Eastern Roman Empire: A.D. 320-435," jThS N.S. 19 (1978) 53-78; L. 
Kakosy, "Paganism and Christianity in Egypt," in CoptEnc 6 (1991) 1865-71. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

292 THE ALEXANDRIAN RIOTS OF 356 

cults, George suggested to the emperor that the temples should 
no longer be exempt from taxation, specifically mentioning the 
long-standing civic cults that had enjoyed immunity since the 
time of Alexander (Amm. Marc. 22.11.6). George also went out 
of his way to insult the temple of Tyche, at one point asking the 
crowd, "How long shall this sepulchre be permitted to stand?"37 
Against Serapis, the city's great divine protector, George incited 
the dux Artemius, a virulent anti-pagan and trusted "friend and 
companion" of Constantius, to sack the Serapeum. 38 His attack 
provoked a violent reaction from the devotees of the god, re­
sulting in clashes between the garrison and the populace.39 

Ammianus Marcellinus relates that during George's tenure, 
Draconitus, an imperial mint official, tore down an altar-per­
haps dedicated to Juno Moneta. Significantly, as the mint was 
located within the precinct of the Caesarion, this act of sanc­
tioned destruction may have been designed to send a forceful 
message to the same pagans who had rioted there before 
George's arrival. And while an associate of Draconitus, the 
military comes Diodorus, was overseeing the construction of a 
church he arbitrarily cut off the curls of some boys, thinking 
that the long curls "were an aspect of the worship of the gods. "40 

This act of anti-pagan zeal (rows of curls sym bolized devotion to 
Serapis) was yet another aspect of the campaign against the most 
popular civic cults of Alexandria. 41 

Another equally plausible interpretation of Diodorus' actions, 
however, would have Yahweh the god unspecified by Ammi­
anus. Thus these Jewish youths maintained unshorn side locks 
(pe'ot) as a token of respect to the God of their fathers. 42 

37 Amm. Marc. 22.11.7: "quam diu," inquit "sepulcrum hoc stabit?" 

38 Artemius was previously commissioned to bring the relics of Andrew and 
Luke to Constantinople. See Theodoret 11£ 3.14; Passio Artem. 4, 8, 18 (ed. J. 
Bidez, GCS [Leipzig 1913]); on his career: J. Dummer, "F. Artemius dux 
Aegypti," ArchPF 21 (1971) 121-44; PLRE I S.7). "Artemius 2." 

39 Julian E p. 60 (378-80). Artemius also employed force to promote the 
emperor's Arian policies within Alexandria, at one point engaging in a violent 
house-to-house search to discover the hiding place of Athanasius: Festal Index 
32. 

40 Amm. Marc. 22.11.9: id quoque ad deorum cultum existimans pertinare. 
41 J. G. Davies, "Was the Devotion of Septimius Severus to Serapis the Cause 

of the Persecution of 202-3?" jThS N.S. 5 (1954) 73-76: more convincing for its 
study of iconography than for its conclusions regarding a Severan persecution. 

42 Although the wearing of pe'ot is best known in the modern period from 
certain Jewish communities of eastern Europe and the United States, there is 
evidence of the usage from Late Antiquity: Bab. Tal. Mak. 20b, following the 
injunction found in Lev. 19.27. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 293 

Attacks on pagans and non-Arian Christians leave no reason to 
doubt that imperial coercion encompassed Jews, who had 
joined with Arians in the rioting of 339 and reappear in the Arian­
sponsored violence of 373, but who are conspicuously absent in 
the conflicts of the episcopate of George. 43 This Jewish quies­
cence in Alexandria would reflect their tenuous status in the 
mid-350s. A minor Jewish sedition during Constantius' 
struggles with Magnentius occurred in Palestine under a certain 
Patricius and was duly crushed by Gallus in 352.44 Afterwards, 
an imperial enactment was directed against both the 
"sacrilegious gatherings" (sacrilegis coetibus) of the Jews and 
the apostatizing Christians who joined them. 45 Constantius' 
administration also levied various taxes (later rescinded by 
Julian) against the Jews. 46 Officially-sanctioned acts of anti­
Semitism in Alexandria would fit this wider imperial context, 
especially when an administration so closely attuned to the 
imperial will governed Alexandria. 

The close connection between religious factionalism and 
urban topography in late antique Alexandria offers another 
avenue for understanding the turmoil of this period. 47 All of 
George's principal enemies had their strongest support in the 
center of the city. The civic cults that George assailed were all 
located, except for the Serapeum, in and around the agora or 
Mesonpedion. Likewise, the most important churches were 
situated near the Via Canopica, the broad colonnaded boulevard 
bisecting the city; and some of Athanasius' most ardent 
supporters could be found among the sailors and dockworkers 
of the of the city's twin harbors. In addition, from what little is 
known of the fourth-century Jewish community it appears that 

43 Athan. Ep. Eney!' 3f, 7; Theodoret HE 4.18.19, quoting a letter of Bishop 
Peter, Athanasius' successor. 

44 Aur. Viet. Caes. 42.11; Hieron. Chron. S.a. 352; Socrates HE 2.23; Sozomen 
HE 4.7; see also H. Trankle, "Der Caesar Gallus bei Ammian," MusHeh.l 33 
(1976) 162-79 at 167ff; M. Avi-Yonah, The Jews of Palestine: A Political 
History from the Bar Kokhba War to the Arab Conquest (New York 1976) 
176-81. 

45 Cod. Theod. 16.8.7; text and extensive discussion in A. Linder, The Jews in 
Roman Imperial Legislation (Detroit 1987) 151-54. 

46 Julian Ep. 204 (396-98); Linder (supra n.45) 154--60. 
47 A similar analysis is applied to factionalism in Alexandria during the 480s 

in C. Haas, "Patriarch and People: Peter Mongus of Alexandria and Episcopal 
Leadership in the Late Fifth Century," ByzSt N.5. 1 (1991, forthcoming). 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

294 THE ALEXANDRIAN RIOTS OF 356 

Jewish habitation was clustered near the Eastern (or Great) 
Harbor.48 

By contrast, Arianism had for decades found its strongest sup­
port in Alexandria's extra-mural regions. Arius' parish church in 
a suburban district attracted ascetics and shepherds among his 
early followers.49 Indeed, Gregory and Philagrius had recruited a 
mob of shepherds to attack Athanasius' supporters in 339 
(Athan. Rist. Ar. 10). This topographical connection was not lost 
on George, who had spent his early career in Constantinople as 
u1to8£KTT1~ 'taJltaKWV acquiring business acumen and a reputa­
tion for ruthlessness. so George apparently understood that the 
topographical differentiation of Alexandria did not permit a 
favorable 'business climate' in a region dominated by his urban 
opponents, largely concerned with maritime trade and work­
shop manufacture. Consequently, George gained monopolies 
on papyrus production and reed cutting and a special tax on the 
extraction of nitre-economic activities concentrated in the 
suburbs. s1 George's control over the collegium of grave-diggers 
and coffin-bearers, "keeping them in his own hands, not for 
humanity, but for profit," confirms this reliance on the periph­
eral areas (Epiph. Raer. 76.1.5ff). Such a shrewd businessman 
probably also took advantage of Constantius' grant of immunity 
from taxation to grave-diggers (Cod. Theod. 16.2.15.1). In 
addition, George not only profited from these concerns and 
became a patron of the suburban inhabitants, but also kept a 
close eye on the extra-mural districts-regions where dissent 
tended to flourish in late Roman cities. 

After antagonizing Alexandria's two most influential com­
munal groups (and perhaps also the Jews), George came to rely 
increasingly on military force to maintain his position. Yet even 
the spears of the prefect could not secure George in his see. In 

48 Socrates HE 1.35; Athan. Apal. c. Ar. 9, fp. Eney. 5; Haas (supra n.7) ch. 
4,5. 

49 Epiph. II aero 69.3.1 f; on Arius' church of Baucalis, the nearby martyrium 
of St Mark, and the surrounding region of Boucolia, see B. A. Pearson, 
"Earliest Christianity in Egypt: Some Observations," in B. Pearson and J. 
Goehring, edd., The Roots of Egyptian Christianity (Philadelphia 1986) 
132-59. 

50 Greg. Naz. Or. 21.16; Athan. De Syn. 12, Ad Episeopas 7, I1ist. Ar. 75. 
51 Epiph. II aero 76.5; papyrus: Expositio tat. mund. 36.1-9; Strab. 17.1.15; 

BGU IV 1121 (P.Se/eet. 41: lease on a papyrus marsh, 5 B.C.); Plin. HN 13.76; 
nitre and other quarried materials: P.Sakaon 24f (P. Thead. 35f); see also Haas 
(supra n.7) ch. 2. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 295 

late August, 358, the plebs attacked him in the church of St 
Dionysius and he barely escaped with his life. 52 By early 
October he realized how untenable his position had become and 
he left the city, "being driven away by the multitude" (Festal 
Index 30). The Arian patriarch spent the next three years taking 
an active role in the synods and councils that occupied the 
imperial church. 53 As frequently happens, a newly-dominant 
ideology soon shows fissures within a once solid phalanx of 
opposition. So too, ascendant Arianism in the 350s evolved into 
several competing theologies, each of which was opposed to the 
Council of Nicaea. 54 George, very much involved in these 
debates, tended to follow the lead of Acacius of Caesarea, but 
his brutal conduct in Alexandria haunted him even in Arian 
circles, as he had to defend himself at the Council of Ariminum 
(359) against accusations of "plundering and violence"-an echo 
of similar allegations brought against Athanasius at the Synod of 
Tyre (339).55 

Meanwhile in Alexandria, control of the churches passed back 
and forth between the two main parties. Though Athanasius re­
mained securely in hiding, his partisans took possession of the 
churches nine days after George's departure and retained them 
for little more than two months, when the dux Sebastianus, 
reinforced by new troops from Egypt, expelled the Athanasians 
(late December, 358) and restored the churches to George's 
followers. 56 

The following June, the infamous notarius Paul (nicknamed 
"Catena") arrived from court with an edict renewing persecu­
tion of George's opponents (Hist. Aceph. 2.5). Paul conducted 
investigations under a mandate from Constantius to root out 

52 !list. Aceph. 2.3. The anonymous author's simple use of plebs does not 
allow speculation on composition of the crowd and its specific sympathies, 
whether Homoousian or pagan. 

53 Sozomen !lE 4.6.16; Theodoret HE 2.23f. 

54 A balanced, recent treatment in R. P. C. I-lanson, The Search for the 
Christian Doctrine of Cod (Edinburgh 1988) 348-86, 557-637. 

55 Socrates HE 2.39f; Sozomen HE 4.17 (ap7taywv Kat u~pEwv),cf 2.23ff; on 
this aspect of Athanasius' episcopate see T. D. Barnes, "The Career of Athana­
sius" -a depiction to be viewed in light of D. W. I-I. Arnold, "Sir Harold Idris 
Bell and Athanasius: A Reconsideration of London Papyrus 1914," both in E. 
A. Livingstone, ed., Studia Patristica 21 (=Papers presented to the Tenth Inter­
national Conference on Patristic Studies, 1987 [Leuven 1989]) 390-401 and 
377-83. See also Hanson (supra n.54) 239-46. 

56 Hist. Aceph. 2.4; Sozomen HE 4.10. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

296 THE ALEXANDRIAN RIOTS OF 356 

treasonous activity throughout Egypt, after Constantius had 
received stolen petitions directed to an oracle of Bes at the 
hallowed shrine of Abydos (Amm. Marc. 19.12.3-16). Petitions 
to oracles (especially from persons of rank) were deemed 
treasonous, and consequently numerous high-ranking pagans 
were dragged before the notarius' tribunal at both Alexandria 
and Scythopolis in Palestine. Parnasius, a pagan praefectus 
Aegypti at the time of George's expulsion from Alexandria, was 
tried and exiled. 57 Andronicus, a town councillor from Her­
mopolis and former student of Themistius, was likewise hailed 
before Paul. 58 An aged Alexandrian philosopher named Deme­
trius confessed on the rack only to sacrificing at the shrine and 
denied ever consulting the oracle (Amm. Marc. 19.12.12). Al­
though Ammianus implies that numerous persons were tried, 
tortured, and convicted, the harvest from Paul's inquisition was 
meager: Parnasius is the only named individual not eventually 
acquitted. 59 

This hardened imperial policy, however, set the stage in 
Alexandria for ongoing persecutions by Faustinus, the new 
prefect-an experienced hand in violent coercion of Alexan­
drians, for he had stirred up the 'trov ayopatwv in June 356 
before the assault on the Caesarion (Athan. Hist. Ar. 55, 58). His 
new assistant, the dux Artemius, eventually became the prin­
cipal agent of George's policies during the Cappadocian's 
absence. The sources depict Artemius ruthlessly harrassing 
Athanasians and continuing a vigorous campaign against urban 
pagan cults. 60 

57 His prefecture (357-359) followed that of Cataphronius, the initiator of 
Constantius' hard-line policies in 356. It may be conjectured that a pagan had 
been chosen, in part, to mollify the pagan crowds that rioted in the summer of 
356. If so, Parnasius' rapid fall from imperial favor suggests that Constantius 
considered his softened policy to have gone too far-especially when his hand­
picked bishop had been thrown out of the city, and the prefect had failed to 
support the bishop with sufficient force. Indeed, the vindictive George may 
have had some hand in targeting Parnasius for trial. The Hist. Aceph. (2.5) 
goes so far as to claim that Paul's activities in Alexandria were motivated solely 
by a policy to exact vengeance on George's enemies: Paulus ... proposuit 
imperiale preceptum pro Georgio et domuit mullos ob eius vindicta. On 
Parnasius: Amm. Marc. 19.12.10; Liban. Ep. 29, Or. 14.15f. 

58 Amm. Marc. 19.12.11; Liban. Ep. 77. 

59 A recent discussion of the context of these trials in Matthews (supra n.8) 
217f; an exhaustive treatment, rightly emphasizing the religious dimensions of 
the episode, in R. von I-Iaehling, "Ammianus Marcellinus und der Prozess von 
Scythopolis," JAC 21 (1978) 74-101; see also H. Funke, "Majestats- und 
Magieprozesse bei Ammianus Marcellinus," JAC 10 (1967) 145-75. 

60 Festal Index 32; Julian Ep. 60 (379b); Greg. Naz. Or. 21.20. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 297 

Alexandrians saw Artemius as the mere instrument of George, 
and hence also of Constantius. All the more reason, then, that 
George "was regarded with greater aversion than before" when 
he returned to the city accompanied by imperial troops on 26 
November 361 (Sozom. HE 4.10). He immediately "com­
menced a cruel persecution against the pagans" (HE 4.30), who 
"had more cause to hate him than any other body of men, 
especially on account of the insults he offered their images and 
temples; and having, moreover, prohibited them from 
sacrificing, or performing the ancestral rites" (5.7). George felt 
so confident in his newly-restored position that he ignored long­
standing Alexandrian protocol, insulted the city magistrates, 
exiled noted pagans, and imperiously gave orders to both civil 
and military officials. 61 

To George's misfortune, events elsewhere led to the radical 
alteration of his status: at a post station in western Cilicia, 
Constantius died of a fever on 3 November-three weeks 
before George returned to Alexandria.62 The emperor had been 
moving his forces west to face his second-cousin Julian, 
elevated to the purple by his troops the previous year. News of 
Constantius' death finally reached Alexandria at the end of 
November. When the Prefect Gerontius announced the 
transfer of power to the assembled magistrates and people, a 
crowd quickly formed and "attacked George with shouts and 
reproaches as if they would kill him at once" (Sozom. HE 5.7). 
The authorities were able to extricate him from the mob, and he 
was placed in prison, doubtless to await his extradition to 
Constantinople for trial (Hist. Aceph. 2.9). 

Although George had clearly alienated virtually the entire 
urban population, rendering himself" exceedingly obnoxious to 
all classes," the driving force behind the riot apparently came 
from clements within the pagan community (Socrates HE 3.3). 
Other anti-pagan officials in Constantius' local administration 
were likewise under detention, including Dracontius, the 
director of the imperial mint who had overturned an altar in the 
sacred precinct of the Caesarion, and Diodorus, the comes who 

61 Sozomen HE 4.30. The celebrated pagan physician and teacher Zeno was 
exiled from Alexandria during this period, later restored by Julian: Julian Ep. 
58 (426); Liban. Ep. 17l. 

62 Amm. Marc. 21.15.1-6; Consul Canst. s.a. 361; Hieron. Chron. s.a. 361-
testimony to the slowness of communications within the empire. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

298 THE ALEXANDRIAN RIOTS OF 356 

had shorn the locks of Alexandrian boys.63 George's principal 
military supporter Artemius, although unmolested in his post as 
dux, could not render any aid to the Arian bishop. Within a year 
Julian recalled Artemius and executed him at Antioch in his 
presence. 64 

On 24 December, less than a month after the imprisonment of 
George and his companions, a mob broke in, murdered them, 
and dragged their bodies through the streets with ropes tied to 
their feet. Their mutilated corpses were then placed on camels 
and paraded the length of Alexandria to the shore, where their 
bodies were burned and the ashes cast into the sea. 65 Cremation 
prevented veneration of their relics as martyrs to the faith (as 
happened when Artemius' bones were transferred to Constan­
tinople). Moreover, in a land where the the careful preservation 
of corpses had attained the level of a highly-developed science, 
burning their bodies served as a special insult with unmistakably 
religious overtones. The entire process (execution or flogging, 
ceremonial parade andlor dragging of body, cremation) was a 
long Alexandrian tradition, and constituted a carefully structured 
ritual of civic purgation.66 Julian reacted to this outburst in a 
sharply-worded letter to the Alexandrians, deploring their extra-

63 Amm. Marc. 22.11.9; M. Peachin, "Praepositus or Procurator?" Historia 36 
(1987) 248f, argues that Ammianus mistakenly identifies Dracontius' office as 
that of a praepositus, instead of the more widely-attested procurator. 

64 Passio Artem. Anh. 3, 167-75 (ed. Bidez). Ammianus claims that the 
Alexandrian pagans rioted upon hearing that Artemius was dead (22.11.8), but 
the Passio Arlem. states that Artemius was beheaded in Daphne, the luxurious 
suburb of Antioch. Julian, who presided over his trial, did not arrive in 
Antioch until July 362. 

65 The events are described with slight variation in Amm. Marc. 22.11.8-11; 
Sozomen !l E 5.7; Socrates HE 3.2f; Hisl. Aceph. 2.8ff; Chron. Pasch. p.546 
Dindorf. Sozomen and Socrates attribute George's murder to insults that he 
offered to pagan cult objects discovered during the construction of a church, 
but the same story is repeated for the pagan-Christian clashes of 391, where 
the sequence of events fits far better. The announcement of Constantius' death 
provided more than sufficient motivation for an Alexandrian crowd to rise up 
against the hated patriarch, finding legitimation for their actions in the 
acquiescence of the imperial authorities. 

66 E.g. Jewish victims in 38, Christians in 250, Homoousian Christians in 
373, Hypatia in 415, Proterius in 457; a full analysis of this ritual in C. Haas, 
"Alexandria's Via Canopica: Political Expression and Urban Topography 
from Augustus to 'Amr ibn al-'As," in N. Swelim, ed., Alexandrian Studies in 
Memory of Daoud Abdu Daoud (Cairo 1992); comparative material from else­
where in the empire in K. M. Coleman, "Fatal Charades: Roman Executions 
Staged as Mythological Events," JRS 80 (1990) 44-73. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER HAAS 299 

legal violence and calling upon the law-sanctioning authority of 
Serapis, while denouncing George, "the enemy of the gods" ('tep 
8£01<; £X8pep: Ep. 60 [379c]). 

The pagan emperor took no punitive measures against the 
Alexandrians and merely instructed the new prefect Ecdicius to 
send him George's huge library, which the book-loving em­
peror had coveted for many years. 67 Several Christian historians 
attempt to quash the rumor of Athanasian responsibility for the 
murders. Although the sources do not state explicitly that a 
primarily pagan mob murdered George and his companions, 
Julian's letter, in conjunction with pagan violence against George 
at the time of his arrest, permits ascribing the Cappadocian 
bishop's death with some certainty to the pagans. Yet both the 
A~h~nasians and the Jewish community had ample reason to 
reJOIce. 

The ultimate failure of George's episcopate demonstrates that 
administrators sent from Constantinople to the great Egyptian 
port squandered their effectiveness by attempting to conduct 
with equal vigor the emperor's anti-pagan and anti-Homoousian 
policies. In the end neither Alexandrian group was sufficiently 
weakened to ensure long-term conformity to Constantius' 
wishes. Since the emperor alienated both groups, it proved 
impossible to co-opt one party and tip the scales against the 
other in favor of imperial policy. Furthermore, Constantius' 
enmity ensured that the Jewish community sat on the sidelines 
during this tumultuous struggle for hegemony within the city. 

Even though he lacked the crucial backing of the Alexandrian 
Christian community, George was the first Christian leader to 
wage a concerted campaign against paganism in the city. He 
proved to be an extraordinarily potent antagonist with the em­
peror's active support through imperial edicts and the troops to 
enforce them. For contemporaries, George directed imperial 
policy in Alexandria, denouncing various Alexandrians to the 
emperor. In Ammianus' colorful words (22.11.5, tr. Rolfe), 
"George poured into the ready ears of Constantius charges 
against many, alleging that they were rebellious against his 
authority; and forgetful of his calling, which counselled only jus­
tice and mildness, he descended to the informer's deadly prac­
tices." Since he had (in Julian's words) "exasperated Constantius 
against [the Alexandrians]," Constantius gave George a free 

67 Julian Ep. 106 (411),107 (377f). 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

300 THE ALEXANDRIAN RIOTS OF 356 

hand, so much so that military authorities like Artemius and 
Sebastianus were placed under his control. 6H Once his imperial 
patron was dead, George's authority inevitably collapsed and he 
became easy prey for the embittered crowd. 

Athanasius' supporters once again seized the churches and ex­
pelled the Arians, but the bishop remained hidden among the 
ascetic communities for another month or so, until an edict 
from Julian restored him to his see. In late January or early 
February 362, Athanasius entered Alexandria in triumph. 69 

Though he found it politic, owing to Julian's hostility, to with­
draw discretely eight months later, his brief, self-imposed exile 
ceased with the death of the pagan emperor. 70 Athanasius' place 
in Alexandria was secure. 71 George's disastrous episcopate 
spelled an end to the Arian community. When the Arian bishop 
Lucius was installed with military force a decade later, he found 
no viable group of Arians to welcome him.72 Mindful of 
George's example, Lucius stayed close to the protection of the 
soldiers.73 

Activist pagan elements, however, realized that imperial pa­
tronage could also work in their favor, especially with a co­
religionist on the throne. Julian's letters to the Alexandrians 
addressed the pagans as a recognizable community-perhaps 
ascribing to them more communal consciousness than war­
ranted at the time. Nonetheless, the gap in ecclesiastical leader­
ship, followed by George's anti-pagan persecution, galvanized 
the pagans into a distinct community that could stoutly resist 
external threats. The urban pagans demonstrated that they were 
still an influential factor in the dynamics of Alexandrian power 
politics. George's episcopate, however, presaged the future 
course of intercommunal relations in the Late Antique city. It 
remained to be seen whether the scales would tip in favor of the 

68 Julian Ep. 60 (379a)-a concentration of powers that prefigured the 
combined prefecture/patriarchate of the hated Melkite appointee Cyrus 
(630-642): llistory of the Patriarchs 1.14 (ed. Evetts, p.489 [225]). 

69 Festal Index 34; Socrates lIE 3.7; Sozomen lIE 5.6; Greg. Naz. Or. 21.27. 

70 Festal Index 35; llist. Aceph. 3.5f,4.3f. 
71 So secure that the Arian emperor Valens was unable to install a bishop to 

his liking until after Athanasius' death: Hist. Aceph. 5.1-7; Festal Index 37; 
Socrates HE 4.20. 

n Lucius had made an abortive attempt to take up the episcopate of the city 
in September 367: Festal Index 39; Hist. Aceph. 5.11ff. 

73 Socrates II E 4.20f; Theodoret HE 4.18f; Sozomen HE 6.19f. 



HAAS, CHRISTOPHER, The Alexandrian Riots of 356 and George of Cappadocia , Greek, 
Roman and Byzantine Studies, 32:3 (1991:Autumn) p.281 

CHRISTOPHER IIAAS 301 

Christians, jf a virulent anti-pagan, enjoying both the unq ues­
tioned support of his urban congregation and the all-important 
patronage of a distant, yet potent, imperial power occupied the 
throne of St Mark. 74 

V ILLANOV A UNIVERSITY 

January, 1992 

74 Portions of this paper were delivered at the 1991 meetings of the 
Byzantine Studies Conference and the American Historical Association. I am 
indebted to the kind comments of Alan Cameron, John Fine, Bill Barry, 
Mieczyslaw Rodziewicz, Gary Johnson, Marie Gingras, and David Evans.