0% found this document useful (0 votes)
49 views232 pages

Julian The Apostate - Against The Galileans

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
49 views232 pages

Julian The Apostate - Against The Galileans

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 232

JUL ats Against

the Galileans

edited and translated~


“5 ;

R. Joseph Hoffmann
Julian’s Against
the Galileans

Flavius Claudius Julianus (33 1/332—363 CE),


better known to history by the name imposed
by his Christian opponents, Julian “the Apos-
tate,” was a nephew of the first Christian
emperor, Constantine I. Julian is one of the
most fascinating figures of late antiquity.
More information is available about him from
both pagan and Christian sources than about
any other Roman emperor. His reign inspired
both admiration and contempt.
Julian’s ambitious program was to rein-
state the religion of his ancestors and, in the
process, to subdue the growth of the Christ-
ian church, which had achieved legitimacy
under the reign of his uncle. Once in power,
he immediately sought to revive the religion
of classical Rome, to reform-the pagan priest-
hood, to revitalize training in classics and
pagan philosophy, and—as an affront to
Christian prophecy—to rebuild the Jewish
Temple in Jerusalem.
This is the first modern English transla-
tion of the complete corpus of Julian’s Against
-the Galileans and related writings, including
letters and edicts bearing on Julian’s attitude
toward Christians and the Church, together
with the most famous accounts of his short
career written by the Christian historians
Socrates Scholasticus and Sozemon, both
composed within a century of Julian’s death.
It not only puts the work of the philosopher-
. emperor into historical perspective but also
offers important insights into the waning days
of pagan philosophy and the growth of the

(continued on hack flan)


W

——
Julian’s Against
the Galileans
Julian’s Against
the Galileans

edited and translated


by
R. Joseph Hoffmann

@) Prometheus Books
59 John Glenn Drive
Amherst, New York 14228-2197
Published 2004 by Prometheus Books

Julian's Against the Galileans. Copyright © 2004 by R. Joseph Hoffmann. All rights
reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, digital, electronic, mechanical,
photocopying, recording, or otherwise, or conveyed via the Internet or a Web site
without prior written permission of the publisher, except in the case of brief quota-
tions embodied in critical articles and reviews.

Inquiries should be addressed to


Prometheus Books
59 John Glenn Drive
Amherst, New York 14228-2197
VOICE: 716-691-0133, ext. 207
FAX: 716-564-2711
WWW.PROMETHEUSBOOKS.COM

08 07 06 05 04 5 43271

Library of Congress Cataloging-in-Publication Data

Julian, Emperor of Rome;.331-363.


[Contra Galilaeos. English]
Julian's against the Galileans / edited and translated with an introduction by
R. Joseph Hoffmann.
p. cm.
Includes bibliographical references.
ISBN 1-59102-198-7 (hardcover : alk. paper)
1. Christianity—Controversial literature—Early works to 1800.
2. Christianity and other religions—Roman—Early works to 1800. I. Hoffmann,
R. Joseph. II. Title.

BR160.3.J85 2004
273'4—dc22
2004009151
Library of Congress Cataloging-in-Publication Data
To the Ancestors
ELOL SINYOD OV TOVDLOV SOvap.

CONCORDIA UNIVERSITY LIBRARY


: Pe
ih) Y a ay

at 4]
f :
: Ys Wonka

Ris ips 7
eee OS ok
at Dm Lea vA

ee eat
ik Of A aie 1

a |
Tee ene
; a Tape
iy : "

tea ree2
ae
5 8 se te ae
i 7 eR fiipewe
ae
; iene nf isa
an here, aq
eters a ih : A

Sei * aa Rare on ave’ Maia.” ate :


is i ’ Ue . Ps

i, a ; : en ait eres
m ea ae :m 4 ’ :cae 5 ais sible fa :
; f : nae ,

(ea eae ae
7 ie ’
ite r

i vite mi re ao is — 4 : ih

‘ trop ge fie YA) A _

bast! wy iy ie " hin a0 Hy Bey tp


: bad i)

4
oes ia he : =

Re.
Lop! oe J = ye, at Pon, 2 a A nel

ih ; > en s
’ a 1. — =

ve > Fe bed se UG
™ a a

e 2 ' yy t

ohaie

-
CONTENTS

List of Abbreviations

Introduction: Julian the Restorer

A Note on Text and Translation

AGAINST THE GALILEANS

FRAGMENTS OF THE CONTRA GALILAEOS 143

MISCELLANEOUS AND OCCASIONAL WRITINGS


ABOUT CHRISTIANITY AND JUDAISM 147

—_ . Concerning Christian Teachers 147


2. To Maximus 150

3. To Theodorus, On His Appointment as High Priest


and Guardian of the Temples 152

4. To Arsacius, a High Priest of Galatia 159


CONTENTS

5 . To Photinus 156

6 . To Hecebolius 158

7. To the Bostrians 159

8 . On Funerals 161

LETTERS CONCERNING ATHANASIUS


AND THE ALEXANDRIANS 165

Letter 21: Julian Caesar, the Augustus, to the


People of Alexandria 167

Letter 23: To Ecdicius 170

Letter 24: An Edict to the Alexandrians 171

Letter 46: To Ecdicius 171

Letter 47: To the People of Alexandria 172

JULIAN AND THE JEWS 177

To the Community of the Jews 182

LIVES OF JULIAN 185

Socrates Scholasticus, The Education and Parentage of Julian 186


Sozomen, A History of Julian, Called Apostate 192

Bibliography 211
LIST OF ABBREVIATIONS

Abbreviations for ancient works cited frequently in the text. See the bibli-
ography for the full list of ancient authorities and texts.

Abst. On Abstinence (Porphyry)


Adv. Jul. Against Julian (Cyril of Alexandria)
Amm. Res gestae (History) (Ammianus
Marcellinus)

Caes. Lives of the Caesars (Suetonius)

C. Cels. Contra Celsum (Origen)


CG Contra Galilaeos (Julian)

Cod. Theod. Codex (Theodosius)


De princ. De Principiis (Origen)

Dial. Tryph. Dialogue with Trypho (Justin Martyr)

Enn. The Enneads (Plotinus)

Ep. Epistles (Julian)

Ep. ad SPQ Ath. Epistle to the Athenians (Julian)

Ep. ad Them Letter to Themistius (Julian)


10 ABBREVIATIONS

Frag. Fragments (Eunapius)

Greg., Or. Orations

Hist, cel: Ecclesiastical History (Philostorgius)

Libanius, Or. Orationes (Libanius)


Myst. De mysteriis (lamblichus)
Or. IV Hymn to Helios (Julian)

Or. V To [Cybele] the Mother of the Gods (Julian)


Or. VI To the Uneducated Cynics (Julian)
Or. VII Against Heraclius (Julian)
PG Patrologiae Graecae Cursus Completus
(Cyril of Alexandria)
Philos., Hist. Historia Ecclesiastica (Philostorgius)
Socrates, HE Historia Ecclesiastica (Socrates)
Sozomen, HE Historia Ecclesiastica (Sozomen)
Theodoret, Hist. Eccles. Historia Ecclesiastica (Theodoret)
Theog. Theogony (Hesiod)
Zosimus Nova Historia (Zosimus)
INTRODUCTION
JULIAN THE RESTORER

“Time and the gods are at strife; ye dwell in the midst thereof.”
Swinburne, Hymn to Proserpine

I. THE POLITICAL JULIAN

When Flavius Claudius Julianus, better known to history as Julian


the Apostate, died in a poorly planned campaign against Persian
forces in Ctesiphon (Mesopotamia) on June 26, 363, he was thirty-
one years old. According to one of the many biographies that his
brief and eventful reign inspired, he was lanced from behind by an
unknown assailant from his own forces. Realizing his wound was
fatal, he cupped his hands to gather his blood, flung it toward the
sun—which in its aspect as Helios he had once regarded as the pre-
siding god of his rule—and cried, “Galilean—You have conquered.”
The words are almost certainly fabricated for effect by the church
father Theodoret;! Julian was fighting a Persian army, not Christians,
but when word of his death reached Constantinople and Rome the
interpretation of the event was swift and irreversible. Whoever killed
Julian was doing God’s work. The man who was killing Christ's
church had been put to death by the decree of Christus imperator.
According to a legend recounted by the Palestinian church historian

iaiheods) HES:25.5:

11
Ge INTRODUCTION

Hermias Sozomen the omens were in place—the oracle at Bablas


had ceased to utter the prophecies of Apollo, the reconstruction of
the Temple in Jerusalem, ordered by Julian to commence during his
absence, had been disrupted by earthquake and fiery storms spewing
from the foundations, and most portentous of all, a crusading sol-
dier racing to join forces with Julian along the Tigris had received a
vision: a heavenly council of apostles and prophets had been con-
vened and had decided the emperor must die.” For Sozomen, the
occlusion of the sun by a dust storm on the bright June day proved
that his death was not due to simple military miscalculation. The
Persians would not claim responsibility for it; no one from among
the Roman soldiery came forward to confess to it; and no conspiracy
was uncovered. Even the most ardent Christian despisers of Julian
profess ignorance in the matter, though Sozomen calculates that the
assassin was probably a Christian.* Those who loved or admired
him had other ideas: his bodyguard Callistus, who celebrated
Julian’s deeds in a heroic poem known to the historian Socrates
Scholasticus, said Julian had been stabbed by demons,’ and his
friend Libanius, a Syrian sophist, remained confident that whoever
killed Julian “was neither Persian nor Saracen,” but a group of liber-
tine soldiers looking for the right time to put an end to the puritan-
ical regime his religious ideas had brought about.°
Julian’s inglorious end stands in uneasy symmetry with his early
life and exploits as Constantius’s western lieutenant. When he was
nominated Caesar at the age of twenty-two by a vacillating and sus-
picious Constantius, he was pulled away from a life of philosophical
commitment and personal insecurity. Julian was keenly aware of his
cousin's obsessive readiness to purge the family of conspirators,
would-be usurpers, and rivals,° a tendency realized immediately
2s S0ZOMER. HE 5.20,,21) 22oe2.
3. Sozomen, HE 6.2 erroneously or misleadingly attributes the suggestion to
Libanius.
4. Socrates, HE 3.21.
5. Sozomen, HE 6.1.
6. After Constantine's death in 337, Constantine II (Augustus 337-40) and
Constantius IT (Augustus 340-61) had unequivocal title, while an array of unillus-
trious caesars—Magnentius, Decentius, Vetranio, and Nepotian—failed to establish
their claims to power and were regarded as usurpers.
JULIAN THE RESTORER 13

after Constantine's death in 337 when Constantius targeted the fam-


ilies of Constantine’s half-brothers, but spared Julian and his half-
brother Gallus because of their age.? Born in Constantinople,
Julian's father, Julius Constantius, was himself half-brother of the
emperor Constantine through Constantius Chlorus, and his mother,
Basilina, was Julius’s second wife. The half-brothers of Constantine’s
fold were in fact a paradigm for the uneasy relationships that would
later affect the heirs of Constantius; it was only after years of tense
relations that Constantine had decided to legitimate his siblings,
making Julius Constantius, Julian's father, one of his consuls in 335.
Julian was, apparently, especially devoted to his mother,’ who
according to reports by both Ammianus? and Libanius! was a
noblewoman, the daughter of the praetorian prefect Julius Julianus,
who had come to win Constantine’s always unpredictable favor. Per-
haps more relevant for Julian's intellectual development is the fact
that his mother had been tutored by the Scythian eunuch Mardo-
nius, and following Basilina’s death in 339, Julian was sent away to
Nicomedia and to Mardonius’s care and tutelage.
To stay for a moment with the foreground of Julian’s political
biography, the date November 6, 355, was a pivotal one, not only in
Julian’s life but in the life of the Empire. Recounting the events of that
year, Ammianus says that Gaul was in a desperate situation, with
every report out of the region bringing worse news than the one
before it. Constantius was torn between sharing the imperial power,
which could never, his advisers reminded him, be done without risk
to the emperor, and presiding over the destruction of Gaul by hordes
of unopposed barbarians." (Gallus, Julian’s half brother, had been
elevated to the purple, accused of treason, and executed as a conspir-
7. On the purge, see M. DiMaio and D. Arnold, “Per vim, per caedem, per
bellum: A Study of Murder and Ecclesiastical Politics in the Year 337 AD,” Byzantion
62 (1992): 158ff.
8. Orso we may judge from the fact that he founded a city (Basilinoupolis) in
her honor; see Lib. XVIII.187-88. It is usually thought that Basilina was an Arian. It
is entirely possible that she was a pagan, a fact which makes Julian’s being entrusted
to Mardonius—almost certainly a pagan pedagogue—more explicable.
DeAmmMeEe2ro23.
10. Libanius, Or. 18.9.
LeAnna 58725:
14 INTRODUCTION

ator in 354.) Constantius decided on Julian only reluctantly, after the


young scholar had spent seven months under protection and obser-
vation near Milan, when the emperor's wife Eusebia persuaded Con-
stantius that Julian was too much the hermit to plot intrigue or pose
a political threat.!? In practical terms, however, the purges of Con-
stantine’s family had by this time been so effective that Julian was the
only viable candidate to restore order along the Rhine frontier.
Ammianus sets the scene: a bright day in November with the
emperor taking his place next to Julian “on a high platform sur-
rounded by eagles and standards.” In the oration, Constantius praised
the bond of kinship which in ordinary life was expendable to him.
After receiving the acclamation of the assembled troops, Constantius
vested Julian in the purple robe of his ancestors and declared him
caesar, to a din of approval—shields being struck against the knee. As
though to suggest that the display was not altogether to his liking,
Ammianus adds the detail, “Julian stood before [the emperor] dejected
and wearing a slight frown.”!3 Several days later, to seal the bond of
kinship that Constantius had extolled, Julian was married to Constan-
tius’s spinsterish sister Helena, through whose physical inspiration,
apparently, he foundit simple to practice chastity.!4 There followed a
progress from Turin to Vienne where he was to take up command of
the army. Along the route, Julian learned that Cologne (Colonia Agrip-
pina) had been taken by Alamanni forces after a long siege, which he
interpreted as an omen of worse things to come. When he arrived in
Vienne, however, there was an outpouring of affection, including one
voice, an old woman’s, that had a special resonance: “This is the man,”
she cried out, “who will restore the temples of the gods.”!®

12. Julian, Or. III.11 8C and Epistula ad SPQ Atheniarum, 274A-B.


13. Amm., 15.8.10.
14. One of the few references to her by Ammianus (16.10.18), recounting
events of the year 357, asserts that Helena was summoned to Rome by Eusebia, the
wife of Constantius, who administered a drug to her sister-in-law designed to
induce abortion if she conceived. Eusebia is also accused of causing Helena to lose
a male child in Gaul, by bribing the midwife to cut the umbilical cord too short.
These comments must be assessed in the context of Ammianus’s general contempt
for the family of Constantius and his desire to extol Julian’s reputation for philo-
sophical chastity.
15e- Amn S76 ale
JULIAN THE RESTORER i

The fledgling caesar who took command of an army at Durocor-


torum is described in vivid and conflicting ways by biographers.
Julian had been sent to Athens in 354 after Gallus’s execution, largely
for safekeeping, but also for further study with the Neoplatonic
philosopher Priscus.'° There he came to know the future Christian
bishop who would write tirelessly against him, Gregory Naziazen.
Gregory describes a young man already conniving to “restore the
temples of the gods,” but cunning enough to keep his intentions
secret. As Gregory reads the character of his former schoolmate,
nothing good could have come from one so obviously unsuited to a
public role, “a neck seldom steady, a frequent shrugging of shoulders,
an eye always scowling and in motion, together with a frenzied
aspect, a gait irregular and tottering, a nose breathing only contempt
and insult, with ridiculous contortions of countenance expressive of
the same thing; immoderate and very loud laughter, nods, as it were,
of assent and drawings back of the head as if in denial without any
visible cause; speech with hesitancy and interrupted by his breathing,
disorderly and senseless questions, answers no better, all jumbled
together ‘without the least consistency or method.”'’ Gregory sees
Julian as an unpromising enthusiast, an unsystematic philosophical
dilettante—an assessment partly born out by the most oblique sec-
tions of his works against the Christians and the most effusive pas-
sages in his letters to philosophers.'® But to his admirers, Julian was
a promising and regal figure, whose philosophical works “rivaled
those of the Tyrian old man.”!? Ammianus describes him at his
investiture as having eyes at once “delightful and awe-inspiring and a
face to which animation added charm.”*° What for the Christian

16. Ep. V. Julian had met Priscus in Pergamon. His life was written by
Eunapius.
17. Gregory, Or. 5.23.
18. A good example is his approach to the teacher Maximus (Ep. 12, “To Max-
imus, Philosopher”): “I sleep with your letters as though they were healing drugs of
some sort, and I do not cease to read them constantly, as though they were newly
written and had only just come into my hands.”
19. Libanius, Or. 18; Porphyry, with whom Julian was often compared. Cf.
Socrates, HE 3.23.
20. Amm., 15.8.16. The contradictory descriptions have played an important
role in assessments of Julian’s character, reign, religious ideas, and achievements.
16 INTRODUCTION

writer appeared to be a convulsive insecurity in his physical presence


was for his Roman biographer a disarming spontaneity.”
However he may have appeared to contemporaries, his early
skirmishes against the barbarian leaders suggest that Julian had the
makings of a good military strategist and administrator. He man-
aged to secure a peace with the local barbarian leaders who were
assailing Cologne, wintered at Senoae (Sens), and spent the fol-
lowing spring and summer (357) in operations there and in Lug-
dunum (Lyons) and Tres Tabernae (Saverne). Then, near Argen-
toratum (Strasbourg), Julian had his epiphany as a leader against the
German forces. Ammianus takes special delight in describing the
determination of the barbarian armies, who were willing to thrust
themselves headlong at their Roman enemies when it became clear
that their early successes had not resulted in puncturing Julian’s line.
“The enemy, who were willing to squander their lives for victory,
tried repeatedly to find weak spots in the fabric of our line. As they
perished one after another and the confidence of the Romans who
were striking them down increased, fresh hosts took the place of the
slain, till the incessant cries of the dying stupefied them with fear.
Then at last they gave. way under the stress of disaster and put all
their energy into attempts at flight. . . Anyone who was present will
vouch that they had more reason to pray than to flee.”??
By any account the victory was stunning for a young leader, and
after the battle, perhaps in awe of their own accomplishments, the sol-
diers attempted to proclaim Julian augustus. He rejected the title,
indeed even rebuked the troops for their impetuousness “and solemnly
affirmed with an oath that this was an elevation which he neither ex-
pected nor desired.”*? The demurer was, however, also pragmatic, since
Julian was always aware that Constantius regarded success on the bat-
tlefield, even when effected in his name, as a threat to his imperial rule.

21. Ammianus acknowledges that his “truthful telling” of the acts of Julian's
life result in a panegyric: “He was reckoned to be the reincarnation of Titus, the son
of Vespasian, in the glorious outcome of his campaigns very like Trajan, as merciful
as Antoninus, and in his striving after perfection and truth the equal of Marcus
Aurelius, on whom he endeavored to model his own career and character” (16.1.4).
22. Amm., 16.12.48.
2S aeAM TN uOMOn OO
JULIAN THE RESTORER iy

The following years saw less spectacular but steady advances


against the barbarians. In 358 he managed to pacify mutinous
troops who had been put on short rations, subdue Frankish troops
along the Rhine, and repopulate a number of cities that had been
devastated by years of raids by Alamanni forces. By 359 he was in a
position to carry out further strikes against the Alamannic leaders in
the Rhine region and to launch guerilla raids across the river near
Mogontiacum (Mainz), where their armies were encamped. The
effect of these successes, which are attested even by the church father
Hilary of Poitiers, is that Julian managed to endear himself to the
provincial élites (those who had the most to lose from barbarian
incursions) and was seen by them as a savior of the old Roman
order. An inscription found near Beneventum Apulia bears witness
to his popularity:

To FLAviuS CLAUDIUS JULIANUS, MOST NOBLE AND SANCTIFIED CAESAR,


FROM THE CARING TOCIUS MAXIMUS, VIR CLARISSIMUS,
FOR THE CARE OF THE RES PUBLICA FROM BENEVENTUM.~4

Popularity came at a price. The contemporary accounts suggest that


in the engagements the Romans had lost 243 men, the barbarians
more than six thousand—not counting those carried off the field by
their comrades. News of such stunning victories could not be sup-
pressed, even though Julian’s reports to the emperor were deliber-
ately understated and modest: “The Germans have been defeated.”
Constantius’s courtiers made it their task to find fault, to appease the
emperors vanity by belittling Julian’s achievement. The sarcastic
word for him in the court was “Victorinus.” At the same time they
managed to convince Constantius that Julian’s victory was only a
realization of the emperor's good fortune, “ascribing to his lucky star
any success in any corner of the world.”*> Constantius’s habit of
snatching praise for himself seems to have gone far beyond exagger-
24. Corpus inscriptionum latinorum, ed. T. T. Mommsen (Berlin, 1883), ix.1562.
The phrase vir clarissimus suggests Tocius was at the highest level of the social scale
in his community and thus that Julian had succeeded in conveying a positive image
of himself as Caesar among these strata.
25, 7Amm.,21612:68:
18 INTRODUCTION

ation: “When this battle was fought near Strasbourg,” Ammianus


complains, “from which [the emperor] was forty days march away,
his account of the action stated that he had drawn up the plan of
battle, taken his place by the standards, put the barbarians to flight,
[and that he alone] . . . had fought and conquered or inclined a mer-
ciful ear to the entreaties of native kings. Disgraceful to relate, [Con-
stantius] said nothing of the glorious exploits of Julian.”*° Unknown
to him directly, Julian’s position in the court was becoming unten-
able, not for any failures on his part but because, to cite the senti-
ment put into the mouth of one of his detractors, “his successes are
becoming a bore.” While the armies were beginning to compare his
exploits to the wars against the Carthaginians, though with lower
loss of life, his enemies argued that he sought a hero’s death in Ger-
many rather than face a traitor’s death like that decreed for Gallus.
There were further successes in 358, against the Franks and the
Alamanni, while Constantius for his part was barely able to keep
Sapor of Persia from his goal of recovering Armenia and Meso-
potamia through an extravagant game of correspondence. The sym-
metry of the events of 358 does not escape Ammianus, who com-
pares Julian’s uncharacteristically harsh treatment of the treacherous
German kings Suomar and Hortar, prostrate at Julian’s feet for
defying the terms of a treaty, with the ineffectual scribbling of Con-
stantius, reminding the Persian king that “Rome has never emerged
a loser from an entire war.”*’ The successes seemed if anything to
enflame opinion against him: “Endless silly jokes were bandied
about, such as ‘He is more of a goat than a man’—an allusion to his
wearing a beard—and ‘his victories are becoming a bore’ ‘Babbling
mole, ‘ape in purple, ‘Greek dilettante; and other such names were
applied to him.”78
By the winter of 360, Constantius could no longer bear to listen
either to stories of Julian’s successes or to the insincere taunts of the
sycophants who despised the young caesar. The punishment was to
strip Julian of many of his troops and officers, outwardly because

26. Amm., 16.12.68.


27. AMM 7102=9 ch 17.5.8:
28. Amm:, 17sl0.11
JULIAN THE RESTORER 19

Constantius claimed he needed them for the impending confronta-


tion with the armies of King Sapor.”° This time, however, Julian was
threatened with both adulation and mutiny. The Petulantes did not
want to leave their home region of Gaul after receiving orders to
march to the east. At Lutetia Julian was again acclaimed augustus. As
before he refused the title, but this time his denial was tossed aside.
He was raised on a shield, adorned with a neck chain which had for-
merly been the property of the chief of the Petulantes, twisted in the
form of a diadem.*° Ammianus, at least, suggests that Julian was
quick to perceive what would happen if he did not accept the accla-
mation, saying that Julian did so as much out of fear as resignation.
Having accepted the title of augustus, he promised each man a
donative of five gold pieces and a pound of silver, then withdrew
into philosophical seclusion—a characteristic reaction for him—to
reflect on the change that had taken place in his life. He confided to
an associate that on the night before his elevation he had had a
vision of the genius of Rome, who revealed to him that he was des-
tined for greater things than leading armies as Constantius’s
second.*!In later correspondence to the Athenians, however, Julian
says that he was never more than the figurehead of the armies, and
that real power remained with the generals and armies in Gaul, who
watched his every move and could have turned coat at any moment
if his actions failed to satisfy their ambitions.*”
Julian’s own account emphasizes his reluctance to accept the title
of augustus, and records that he did so only after seeking counsel
from Zeus. Whatever reaction he expected from the gods, there was
little doubt what he could expect from Constantius, and so he
addressed the emperor in the hope of finding a peaceful solution
and to explain the outcome of events at Lutetia. Ammianus, the lover

29. Julian, Ep. ad SPQR Ath., 280D, 283B and Libanius, Or. 12.58; cf. Amm.,
20.4.1-5.
30. This event is multiply attested as marking a turning point in Julian’s polit-
ical self-consciousness: Amm., 20.4.10-11; Libanius, Or. 12.57-58; 18.92-93; John
of Antioch, Frag. 177; Eunapius, Frag., 14.4. The date is usually fixed at February or
March of 360.
BieeAmin, 20.55).
32. Julian, Ep. ad SPQ Ath. 277D; 278A-B.
20 INTRODUCTION

of historical symmetry, punctuates his account with stories of Persian


advances against the Roman position in the winter of 360 which
kept Constantius tied up in Constantinople. Poetically, the historian
wants to suggest that Constantius’s star is waning and Julian’s rising,
but Julian himself in his letter to the emperor seems to have
believed, or had come to believe, that he was destined to rule as
augustus. Julian in the same period was in Paris weighing how best
to approach the subject, knowing that backing away from the events
at Lutetia was not possible. In his letter he insists on his “fidelity to
his undertakings” but he also assumes a chastising tone: “Ever since
you sent me into the horrid din of war,” he says, “I have been con-
tent with the delegated authority you have entrusted to me.” But as
if to contradict the voices of the sycophants who had slandered him
mercilessly, Julian asserts that he “never laid stress on [his] own dan-
gers.” If the army has now made its preference known through revo-
lution, “it is because the troops who have worn themselves out
without reward in much hard fighting, have carried out a long-
standing plan.”*3 The letter also berates Constantius for sending the
young men in Gaul into a terrible situation and the raw indifference
of ordering worn-out troops eastward at the end of a hard struggle.
For these and other reasons, Julian implies (and Ammianus makes
explicit) the gods have withdrawn their approval of Constantius: he
is pictured as unfit to rule, while the armies’ choice of Julian make
them instruments of the gods’ will.** Unsurprisingly in the letter
addressed to the Athenians, written to refute any suggestion that he
was a usurper, Julian stresses that he became augustus at the instiga-
tion of the gods. In November 361, recounting his acclamation,
Julian wrote to his philosophical mentor, Maximus of Ephesus,
saying that while he became augustus against his will, the choice was
the gods’, and that after the event he had treated his enemies with
justice. He also relates that he himself led the troops in propitiating

33. Amm., 20.83.


34. Amm., 20.8.5-10. On the authenticity of the epitome of the letter pre-
sented by Ammianus, see Michael DiMaio, “The Antiochene Connection: Zonaras,
Ammianus, Marcellinus and John of Antioch on the Reigns of the Emperors Con-
stantius II and Julian,” Byzantion 50 (1980): 163ff.
JULIAN THE RESTORER 21

the traditional gods because it had been revealed to him that his des-
tiny was to revive their worship and receive benefits for doing so.35
Constantius received Julian’s chiding, accommodating, and
mildly threatening letters from envoys in Cappadocia, where he was
stuck on the Bosphorus awaiting news of Sapor’s advances. After
flying into “a passion unusual even for him,” he dispatched a mes-
senger to Julian rejecting all of the proposals out of hand and telling
the young caesar to stand aside. This message was read out in Gaul,
at Julian’s bidding, to a throng of soldiers and townspeople sum-
moned for the occasion, to the accompaniment of flourishes.
Leonas, the quaestor, made it as far as the passage wherein Constan-
tius condemned the acclamation and ordered Julian to be content
with the subordinate office. At this point a general commotion and
shouts of “Julianus augustus” caused Leonas to hurry back to the
beleaguered Constantius, while Julian directed his army across the
Rhine to rout a Frankish tribe, the Attuarii, “whom he had no trouble
in defeating.”*° During the standoff of 360 Julian devoted his atten-
tion to the security of the Rhine and Constantius continued his oper-
ations against the Persians. Julian celebrated his quinquennalia in
Vienne and obtained for the anniversary a dazzling diadem that con-
trasted sharply with the “cheap crown, befitting the captain of an
athletic meeting, ”*” he had worn previously. Soon after the event, he
received news that his wife, Helena, had died—a minor disturbance
that he saw to by having her remains conveyed to Rome to be buried
in the family estate next to her sister Constantia, the wife of Gallus.
According to his biographers, Julian was torn between wanting a
decisive outcome to the stalemate with Constantius and heeding
certain augurs that the emperor was near death. Ammianus turns the
accusation, later exploited by Christian writers, that Julian was prac-
ticed in the black arts into the more congenial suggestion that the
emperor was always ready to explore all roads to wisdom. Julian was

35. Julian, Epistle VIII.414B; Julian was stationed at Naissa and describes the
adulation to the theurgist-philosopher Maximus: “The gods command me to
restore their worship in its original innocence and have promised me rewards if |
obey them with a good will.”
36. Amm., 20.9.10.
Bi Aimm.;, 2. 1a:
a INTRODUCTION

skilled in the interpretation of dreams, and like his ancestors he set


great stock in the divine power of the augurs and auspices, especially
the reading of entrails and the flight of birds.** Apparently he sub-
scribed to Aristotle’s view that dreams are always “certain and reli-
able” when men are in a deep sleep, and that when signs of future
events turn out to be mistaken, it is not the sign but the interpreta-
tion of it that is at fault—a maxim of Cicero’s.*? In rallying his
troops to fight against Julian, Constantius compared his cousin’s
treachery to that of Gallus. “Julian... has become so madly pre-
sumptuous on the score of some trivial victories over half-armed
Germans as to recruit from his auxiliaries for an ambitious design
against the state a small band of desperadoes who will stick at
nothing. He has trampled underfoot the rule of law, the nursing
mother of the rule of law.”4° The ploy was meant to trivialize Julian's
achievements and to paint him a hubristic interloper, usurper, and
coward. But Constantius had received news that Julian’s troops had
made a lightning-fast march through Italy and Illyricum, had seized
the strategically important path at Succi, and would soon be joined
by auxiliaries for an invasion of Thrace. The armies drew closer to
each other just as Constantius’s resolve was melting away. The
emperor was haunted by dreams. He dreamed of his father’s spirit
“holding out to him a fine child, [which he took] and placed on his
lap, where it shook from his grasp and threw to a distance the orb
which he was carrying in his right hand.”*! In the autumn Constan-
tius rallied his troops once more and moved toward the Antiochene
suburb of Hippocephalus, where he saw “in broad daylight the
headless corpse of a murdered man lying with his feet towards the
west.” By the time he reached Tarsus, he had developed a fever, and
beyond Mobsucrenae (the last outpost in Cilicia) the fever became
so violent that he was often delirious. In a lucid moment, just before

58. Amis Qi.


39. Amm., 21.1.7: This digression in Book 21 portrays Julian as obsessively
concerned with omens and the practice of augury, while outwardly still often pro-
fessing to be a Christian. To conceal his devotion to the old gods he attended
church “on the holy day the Christians celebrate in January and call epiphany.”
40. Amm., 21.13.4-5.
41. Amm., 21.14.1.
JULIAN THE RESTORER eo

his death, it is reported—but disputed—that he named Julian his


legitimate successor.
Constantius had reigned as augustus for twenty-four years. He
was forty-four when he died,*? his wife pregnant with a daughter who
would grow up to marry Gratian. Ammianus, always Julian's par-
tisan, nevertheless gives a candid appraisal of the Arian emperor's
character, saying he prefers to make a “distinction between his good
and bad qualities.” Among the good, he lists the fact that Constan-
tius was born to the role: he behaved like a king and refused to court
popular opinion—even when it might have done him some good.
He was sparing in conferring honors, kept the military in their place,
and, in general, was a good judge of character—or at least of the char-
acter of opportunists. On the bad side, Ammianus finds Constantius
dull-witted, an inferior speaker, pretentious (“when he turned his
mind to versifying he produced nothing worthwhile”). His frugality
in eating and drinking was accompanied by a sexual moderation for
which he was so famous that no one ever accused him of bad
behavior. But Ammianus also finds the Spartan athleticism of Con-
stantius (no emperor was a better archer, or more skilled with the
javelin) a disguise for cruelty—he calls it a barbarity to rival that of
Caligula and Domitian—and suggests that he exhibited all the signs
of paranoia that had characterized the reigns of Nero and Com-
modus. Easily suspicious and driven to extremes of envy by gossip
and rumors, especially those planted by his wives or “the shrill
eunuchs and court officials who applauded his every word,” he was
not above inventing evidence in the prosecution of his enemies—
making a “mountain of mischief out of a molehill of evidence” and
“showing himself the deadly enemy of justice, although his great
object was to be thought just and merciful.”** Constantius is vilified
by Ammianus for two further weaknesses: first, his failure to bring the
greed of the tax collectors under control, which led to crushing exac-
tions in the provinces and earned him what was perhaps an unde-
served reputation for avarice. But the taxing of the provinces paid the
bills for foreign wars, and Constantius prided himself on his military

425 Amm., 2191522;


43. Amm., 21.16.1, 5, 8.
24 INTRODUCTION

prowess despite humiliating defeats over the course of his reign.


Taxes also financed the memorials to his “successes” in civil con-
flicts—triumphal arches in Gaul and Pannonia, and assorted lesser
works and monuments designed to record his deeds for future gener-
ations. The second weakness cited by Ammianus was Constantius’s
understanding of the “simple religion of the Christians” which he
“bedeviled with old wive’'s tales.” No better a theologian than he evi-
dently was a poet, he nonetheless took the Arian cause to himself and
became its champion—“raising complicated issues that caused much
dissension,” and transporting “throngs of bishops hither and thither
to what they call synods.” Ammianus remarks, in language Julian
would have applauded, that Constantius’s attempts to settle theolog-
ical disputes “only resulted in clogging the roads.”*4
Julian, now augustus, was eager to usher in a new imperial golden
age. Entering Constantinople triumphantly on December 11, 361, he
honored the predecessor with the funeral rites appropriate to his sta-
tion,*° then set about immediately to pare down the imperial bureau-
cracy that had grown bloated during Constantius’s time. His biogra-
phers underscore his naiveté and even acts of injustice in the earliest
days of his reign, especially in his choice of interrogators to deal with
the unfinished business (as Julian saw it) of his half brother Gallus’s
execution for treason. Two of the conspirators against Gallus,
Apodemius and Paulus Vinculus, were burnt alive and others were
forced into exile, under sentence of death, until the end of Julian's
reign. In reforming the palace, he is accused even by Ammianus of
showing a “lack of concern for the discovery of the truth quite unbe-
coming in a philosopher,”*® though for his rooting out corruption
and especially his treatment of the eunuchs and the cooks and bar-
bers, the Christian historians of his reign offered a softer assessment.”
44. Amm., 21.16.18, 19.
45. These are variously described: Amm., 22.2.1-5; Sozomen, HE. 5.1.6-8,
Socrates, HE. 3.1; Gregory Naziazen, Or. 5.16-17 and Libanius, Or. 18.120-21.
46. Amm., 22.3.1-12; cf. Libanius Or. 18.130ff.
47. See Socrates, HE. 3.1; Cedrenus 1.532.18ff. Ammianus (22.4.5-6) offers
an amusing episode to underscore the need for reform: On one occasion the
emperor sent for a barber to trim his hair. When the barber presented himself,
splendidly dressed, Julian was astonished and said, “I sent for a barber, not a
treasury official.”
JULIAN THE RESTORER rake

Traditionally emperors were expected to show clemency at the begin-


ning of their reign; Julian, however, was eager to appease the army and
seems to have believed that offers of grace would be seen as a sign of
weakness. Ammianus points to the case of Ursulus, Constantius’s
comes sacrarum largitionum*® as a case in point: When Julian was sent
to Gaul as caesar, Constantius had ordered that he should be given
only his subsistence and deprived of the means of making gifts to his
troops, thereby increasing the chance of mutiny or even assassination.
Ursulus had actually ignored the order and arranged for the new
caesar to receive whatever he asked for from the treasury; this interven-
tion was remembered by many at court who took Ursulus’s execution
as a sign of the new emperor's deficient sense of judgement. To cover
his embarrassment, Julian circulated a counterreport that he had acted
against Ursulus only because the army had demanded his head in
exchange for a remark he had made about the incompetence of the
Gallic forces after the battle of Amida. Yet this early case of blame
dodging and courting the favor of the military and various social élites
would characterize the whole of Julian’s reign and is one of the few
areas severally corroborated by both Christian and pagan biographers.
His early reputation for summary justice, however, is as impressive as
his attempts to reform the imperial bureaucracy.”
The reforms were one way of restoring the prestige of the impe-
tial power, which had suffered considerably from Constantius’s
encouragement of the sycophants. In pursuing the reforms Julian
followed the general pattern of other neo-Flavian emperors in
viewing the provincial vicars as intercessors between himself and the
general population. In a letter sent to Alypius, the vicar of Britain,
for example, he praises the official for his “mildness” and “modera-
tion in the use of force” which characterized the execution of his
48. Custodian of the Sacred Largesse, the fund for apportioning military dona-
tives at the discretion of the emperor.
49. For example: Pendatius was charged with involvement in. the death of
Gallus and threatened with exile, but later acquitted (Amm., 22.3.5); Apodeimus
was executed because of his “eagerness” for the death of Gallus and Silvanus, as was
Paul the Catena, head of the agents in rebus, according to John of Rhodes (Artemii
Passio 21; cf. Libanius, Or. 18.152); and most famous of all, the eunuch Eusebius
was condemned for his complicity in the death of Gallus; cf, Sozomen, HE 5.5.8
and also Amm., 22.3.12.
26 INTRODUCTION

office. Such characteristics, Julian goes on to say, are the qualities of


the good king, and it is commendable that the emperor is being rep-
resented in this way to the provincials.*° By the same token, Julian
courted the army because it had raised him to power and sustained
him, and the senatorial aristocracy because he required its approval
for legitimacy—a dependence that stemmed from incessant sugges-
tions that he had plotted the ruin of Constantius from the beginning
of his rule as caesar. Of special significance was the consulship of
Claudius Mamertinus, whose speech in praise of the reforms, deliv-
ered in Constantinople in January 362, is preserved.” Claudius, with
a transparently self-serving interest, presents Julian as restaurator—
the inaugurator of an age of renewal that could be compared to the
days of Augustus. Had not Augustus brought in a new order by
forming a partnership with the senate, based upon a system of
honors and benefices given by the emperor to the senators in
exchange for their role as intercessor between emperor and the pop-
ulace? As Claudius trumpets it, it was this system that Julian was
restoring; indeed, the consulate of Claudius was itself an illustration
of the new triple bond that existed between emperor, senate, and
people.°? But the perception was not limited to Claudius’s grateful
bombast: the municipal senate in Aceruntia (Apuleia) established a
monument on which Julian is commemorated as the “Repairer of
the World,” and there is additional evidence of the young emperor's
benefactions to the municipal élites—the pattern he established
when he was still caesar and continued after he became emperor.*?
To many wealthy pagans, the idea that a new Augustan age was
dawning was not an absurdity.

50. Julian, Ep. VII.404A.


51. Actio Gratiarum 2.1-2; Gratiarum actio, ed. R. Mynors, in XII Panegyrici
Latini (Oxford, 1964).
52. Claudius Mamertinus went on to hold a number of offices: prefect ofItaly,
Illyricum, and Africa. See note 51.
53. Corpus inscriptionum latinorum, 9.417.
JULIAN THE RESTORER ay

II. JULIAN’S RELIGIOUS REFORMS

Ammianus tells this story as a preface to his description of Julian’s


raids on the Raetian frontier in 361: “While Julian was still at Paris in
the position of caesar he was swinging his shield in various field
exercises when the pins by which it was secured gave way and he was
left with only the handle, which he continued to hold in his strong
grasp. There was a general fright among those present at what
seemed a bad omen, but Julian said: ‘Have no fear; I still have a firm
grip on what I held’”° If the omen is taken to mean what Ammianus
wishes it to mean, Julian used Christianity as a shield to protect him-
self from the trouble that only religious suspicion could provoke.
The questions that greeted him on his entry to Constantinople were
three: Was he Christian or pagan? If a Christian, Arian or Nicene? If
a pagan, a tolerant sort like Tatian or a persecutor like Decius?>°
Julian at the time of his exaltation seems to have harbored no
doubts. He was a pagan, neither of the tolerant nor persecuting
variety, but a philosopher-soldier who took the title pontifex maximus
with earnest intent. Comforted by visions of angels reciting hexam-
eters just prior to Constantius’s death, Julian was certain that his des-
tiny was to restore the religion of his forefathers to Rome. The date
of his “apostasy” from Christianity, if it was that, is unknown;
Ammianus’s record of the year 361 mentions that his pretense was
well established, but that he had given up the faith “some time
before.” More reason, then, that he should have stayed clear of Con-
stantius, who followed the general neo-Flavian habit of immersing

54. Amm., 21.1.14-15.


55. An older view of Julian’s “caution” is represented by Bowersock’s static
notion that Julian came to power resolved to persecute the Christians, but was kept
from doing so by pragmatic considerations—a view unfortunately repeated, but
with amendments, in Rowland Smith's otherwise excellent study of Julian’s reli-
gion, p. 215ff. The theoretical basis for Julian’s treatment of Christians is summa-
rized in the often overlooked letter to Atarbius (not cited by Smith), one of the few
letters written in Julian’s own hand, in which he swears an oath by the gods “that
[1] do not wish the Galileans to be either put to death or unjustly beaten or suffer
any form of injury.” Julian’s consistent philosophy is that as a “disease” Christianity
was to be treated rather than punished.
28 INTRODUCTION

himself in religious quarrels—which he did not fully understand or


have patience with. Few people knew Julian’s true feelings about
Christianity; fewer appreciated the philosophical and political basis
of his devotion to the old gods.
There is some debate concerning how much of the Christian reli-
gion he embraced, or rather how much of the religion he knew.”°
Speculation centers on a brief, and inconclusive, passage in Athana-
sius which gives the name of a certain Basilina as a supporter of the
Arian cause, a later source, which names a Basilina who left her
property to the church when she died, and earlier references to his
mother’s relationship to Eusebius of Nicomedia as evidence that
Julian’s mother was a Christian and saw to it he received “the rudi-
ments of a Christian education,” prior to her untimely death when
Julian was seven years old.*”’ When Eusebius was called to become
bishop of Constantinople, Julian was apparently handed over to his
dead mother’s tutor, the Scythian eunuch Mardonius, to learn the
elements of rhetoric and Greek literature, and it was at this point he
became infatuated with Greek prose, poetry, and philosophy. While
earlier interpreters thought that this ambiguous chapter in Julian’s
biography was to be explained by the fact that Eusebius and Mardo-
nius were teaching the future emperor at the same time,*® it seems
clear that Julian’s education under Mardonius was an unresisted
weaning of the boy from the scant catechism he had learned from
Eusebius, a “withdrawal into the world that Homer had created
[where he] sincerely despised all that lay without its boundaries.”>°
His debts to Mardonius were real and consistently acknowledged,
though it is doubtful that the teacher played the role in his intellec-

56. See G. W. Bowersock, Julian the Apostate (Cambridge, MA, 1978), p. 24;
A. H. M. Jones, The Later Roman Empire, 284-602: A Social, Economic and Administra-
tive Survey (Oxford, 1964), pp. 120-21; and P. Athanassiadi, Julian: An Intellectual
Biography (London, 1992), pp. 24-25.
57. See Barbara Saylor Rodgers and C. E. V. Nixon, eds., In Praise of Later
Roman Emperors: The Panegyrici Latini (Berkeley, 1994), p. 1, s.v., “Basilina,” and see
the brief discussion in Athanassiadi, Julian: An Intellectual Biography, p. 18.
58. Thus Baynes, “The Early Life of Julian the Apostate,” JHS 45 (1925):
251-52 and also J. Bidez, La vie de l'empereur Julien (Paris, 1930), pp. 16-21; and cf.
R. Browning, The Emperor Julian (London, 1975), p. 36.
59. Athanassiadi, 15.
JULIAN THE RESTORER Ae

tual biography that one of Julian’s more impressionistic modern


biographers asserts.°° For all the circumstantial reporting, it can be
doubted that Basilina was a devout Christian, much less a com-
mitted Arian like her cousins.
Julian was introspective in youth and later life; in moments of
crisis and peril his first instinct was to isolate himself from those
closest to him, and he often found consolation or guidance in phi-
losophy or augury. “From my childhood,” he writes, “an extraordi-
nary longing for the rays of the god [Helios] penetrated deep into
my soul; and from my earliest years my mind was so completely
swayed by the light that illumines the heavens that not only did I
desire to gaze intently at the sun, but whenever I walked abroad in
the night season when the firmament was clear and cloudless I
abandoned all else without exception and gave myself up to the
beauties of the heavens; nor did I understand what anyone might
say to me nor heed what I was doing myself.”*! This might lead us
to think that Julian bounced between Mardonius’s formulations of
Greek ideas and the youth’s apprehension of them in frequent
philosophical reveries inspired by the work of Porphyry and Proclus.
But when he was sent for safekeeping to Macellum after the execu-
tion of Gallus, he did come up against (even if he was not notably
influenced by) the teaching of George of Cappadocia. It is George's
particular brand of Christian teaching that Julian reacts to in the
treatise against the “Galileans,” even if this is nowhere admitted,
and even though George is never acknowledged by Julian as one of
his teachers.” The most we learn, from an imaginary dialogue Julian

60. Athanassiadi, 23.


61. Julian, Ep. XI.130D.
62. Ina letter to Ecdicius (Ep. XXIII; cf. XXXVIII) Julian demands that George's
library be sent to him intact, “as one who has ever loved reading.” Sozomen relates
the story of George being attacked by a pagan mob in Alexandria on December 24,
361; cf. Epiphanius, Panarion 76.1 (motives for the murder of George), and below,
pp. 40-41. George's influence, to the extent it was significant, seems to have been
purely negative, and it is difficult to estimate the extent to which Julian may have
regarded the hypocrisy and opportunism of the bishop as representative of ele-
ments he thought characteristic of Christianity in general. See Amm. 22.11., and the
interesting classic discussion in Gibbon, Decline and Fall of the Roman Empire, ed.
J. B. Bury (London, 1909), p. 498.
30 INTRODUCTION

constructs between himself and the murderers of George—the


pagans of Alexandria, perhaps with some help from orthodox Chris-
tians who despised the old bishop—is that the prelate deserved
better than to be torn apart by a mob, though “he might have de-
served worse and more cruel treatment [were it] not at your
hands.”°? This would suggest that if Julian at the age of twelve
endured George's teaching, it was merely that, and that his habit of
seeking spiritual sustenance outside the confines of Christian doc-
trine and its seeming muddle of controversies was well established
by this point.
What survives from the period at Macellum suggests that Julian
was groping for answers but was largely dependent on periods of
“pantheistic exaltation” for deliverance from depression. “Let dark-
ness be buried in oblivion,” he says in describing his state of confu-
sion—the same language he uses in his “Hymn to [Cybele] the
Mother of the Gods,” when he speaks of wandering in the darkness _
and the goddess helping him to purify himself of desperate
thoughts. But, of course, this is not Augustine or Gregory Naziazen
praying for conversion; it is Julian praying for deliverance from the
errors that he thinks*are being thrust upon him from the Christian
side. Ammianus makes the point that from earliest childhood Julian
was inclined to the worship of the pagan gods but was afraid to pro-
fess his beliefs publicly.
This neat assertion is problematical, for the view if Christian
writers is the opposite: Gregory Naziazen recalls the young Julian as
a lector in the church at Nicomedia and one who revered the martyrs
and frequented their shrines—someone who became an enemy of
the church and changed faces beyond all recognition—thus an apos-
tate in the specific sense.°° Socrates Scholasticus and Sozomen agree
that Julian became an enemy of Christianity only in adulthood, and
that as a young man he exhibited loyalty to the faith. To illustrate the
point, Sozomen records a story about the young Gallus and Julian

63.-Ep; XX1.379D;
64. Or. V, and cf. Libanius, Or. VIII.174c.
Co mAM Moo eale
66. Gregory, Or. IV.97.
JULIAN THE RESTORER cel

building a small monument on the tomb of St. Mamas in Cap-


padocia. Gallus’s work progressed satisfactorily, but Julian’s did
not—an omen he recounts (without seeing the fundamental contra-
diction in his argument) to show that even as a boy Julian had no
faith.°’ The Christian testimonies concerning Julian’s early religious
beliefs might be worthless were it not that his apologist Libanius,
whose positive assessment of Julian’s reign the Christian writers
deplored, sees Julian as a convert to Hellenism, a man who was
moved by the gods to slough off superstition and embrace the light
of day.°* But withal, the attempts of pagan and Christian writers to
see his “conversion” in inverted Pauline terms, the evidence that
Julian was ever devoutly Christian (or indeed was baptized)°° is slim;
he seems to have given up any attachment to Christianity and to have
adopted the practices of traditional Roman religion, including
theurgy, as a natural part of his immersion in Neoplatonic thought
prior to being named caesar. In his letter to the Alexandrians written
in 363, he states that he had given up the Christian faith when he was
twenty years old and had been an adherent of the ancestral rites for
a dozen yéars prior to writing the letter.” This corresponds to Ammi-
anus’s “official” view that Julian did not openly profess paganism
until the threat of reprisal was removed in his thirty-first year.
If there was safety in dissimulation in the matter of religious
belief, the death of Constantius changed everything. The Christian
church was the one constituency Julian had no wish to court and no
reason to appease. At the start Julian gave no one to believe that he
intended to discriminate actively against the Christians: he lifted the
ban on the teaching of the Arians and permitted various sectarian
leaders to return to their sees.” But biographers Christian and pagan
saw this early expression of “toleration” as more strategic than altru-
istic. “Julian called the bishops of the sects to him... [and told

67. Gregory, Or. IV.24.26; and cf. Sozomen, HE V.2.12-14.


68. tT ob0dSpov picos Kata Tov BEOdv EneoxYES DMO TOV LAavTEvLGTOV
jepovuevoc, Libanius, Or. XII.34; XIII.11; cf. XVHI.16.
69. Cf. Gregory, Or. IV.52, who thinks that he was, but this is to be able to put
him squarely in the apostate’s camp as a traitor to the Christian mystery.
70. Ep. XLVII.434D.
71. Amm., 22.5.3—4.
oz INTRODUCTION

them] to set aside their differences and to live in peace and har-
mony, knowing that toleration would intensify their divisions and
that he would no longer have to fear a unanimous public
opinion.””? Experience had taught him “that no wild beasts are as
dangerous to man as the Christians are to one another.”’> When he
seeks to explain their intemperance, he not surprisingly points to a
lack of philosophical acumen which they share with the Jews, a
“second rate race .. . also lacking in paideia”: “In my opinion, there
is no reason why their god should not be a mighty god, even though
he does not happen to have wise prophets and interpreters. But the
real reason why they are not wise is that they have not submitted
their souls to be cleansed by the regular course of study [eyKvKAL01g
uoaOnuwao1| nor have they allowed those studies to open their tightly
closed eyes and to clear away the mist that hangs over them.”
Julian’s strategy for dealing with the Christians was threefold:
First, he attempted to isolate Christians from the mainstream of
Roman society by limiting their rights and abrogating certain bene-
fits to which they were entitled under the law. Secondly, he
attempted to circumscribe their influence in a way that does not
seem to have occurred to his predecessors: creating a “pagan church”
to rival the organization of the Christians—one which would
assume some of the charitable functions of Christianity. This reform,
or remodeling, of the temples was seen as part of a total religious
program designed to make Christianity irrelevant in a social sense.
And finally, Julian returned to the custom of conservative pagan
intellectuals like Celsus and Porphyry by actively assailing Christian
belief in philosophical polemic designed to prove the unoriginality
of the faith. In this last effort, Julian wanted to portray Christianity
as'an apostate form of Judaism which did not remain true to the
worthy aspects of a much older religious philosophy. Thus, as a
reformed paganism might show the Christians as less beneficent
than their pagan countrymen, so the renewed literary attack was
designed to point up the inauthentic nature of Christian belief.

72. “Ammy22 5.)


73. Amm., 22.5.2; Julian regarded the Christians as troublemakers (Ep.
XXXVII and cf. Ep. XLI.437D).
JULIAN THE RESTORER 85

With respect to the first part of Julian’s strategy, a law of the Theo-
dosian code specifies that decurions were not permitted to defer or
avoid compulsory military service on the grounds that they were
Christians.’* Ammianus tells us that Julian caused legislation to be
passed removing Christian teachers from their posts,”> a lapse in the
emperor's judgement as he sees it, though Julian’s own rescript on
Christian teachers (Ep. XXXVI) provides some insight into the rea-
sons for the action. His view, apparently, was that Christian teachers
should be sacked not for their belief, but because they were poor
models for their pupils—who would not help but notice the
hypocrisy of a rhetoric or grammar master praising the classical sto-
ries but believing them impious at the same time, because they also
passed along traditional forms of belief and worship.’° Other legal
impediments included a law upheld in a statute dated 405 banning
the Donatist sect in north Africa’’ and a law mentioned by Julian to
the Christians of Bostra (Ep. XLI), adjuring them that if they sacri-
ficed to the traditional gods they would hold on to their citizenship,
but that if they persisted in their beliefs, fomenting the factional
strife to which it led, they would be stripped of their legal status.’8
The rest of Julian’s plan fell more easily into place. It would be
too much to say, based on the two letters which serve as the primary
documentation for this phase of his strategy, that Julian (as
Sozomen wants to urge) admired the organization of the Christian
church and wished to emulate it. But its growth, if not the increase
of its factions and disturbances since the days of his uncle, the first
to legitimate Christian proselytizing, was obviously impressive to
the young emperor. In 362 Julian wrote to the high priest Theodorus
to create him archpriest of the diocese of Asia, with the right to

74. Cod. Theod. 12.1.50.


75. Amm., 22.10.5. The location of this statement in Ammianus is very puz-
zling coming at the end of a discussion of Julian’s tactical indifference to questions
of religion in hearing cases and doling out punishment. On the issue of the passage,
see Thomas Banchich, “Julian’s School Law: Cod. Theod. 13.3.5. and Ep. 42,”
Ancient World 24 (1993): 5-14.
76. Ep. XXXVI, esp. 423A-D.
77 Gods Theed 116.5.37.
78. Ep. XLI.437A-B.
34 INTRODUCTION

appoint all priests to temples in the cities of the region. He takes the
opportunity to address a variety of concerns: first, that the customs
of the forefathers are being forgotten, with lamentable conse-
quences. Second, he demands certain moral qualifications of the
high priest, not unlike those that Christians had long required of
their bishops,”? namely that they lead by example, treating all
people fairly, and that they forego luxury in favor of lives of moder-
ation, “since in our day the concern for divine things has been extin-
guished by the love of vulgarity.”®° Contrasting the apathy of pagans
to the zeal of the Jews in matters of religion, he sees traditional reli-
gion in a state of dissolution symbolized above all in the moral
decline of the priesthood, which then is communicated as lethargy
to the people: Where the Jews are fervent about tradition, the pagans
no longer observe the religious laws and have forgotten the tradi-
tions of the fathers. Late in 362 Julian addressed Arcasius, the arch-
priest of Galatia, in similar language, wondering how it happens
that the sought-after restoration of traditional religious rituals has
not led to a renaissance of pagan belief: “The Greek religion still
does not flourish as I would like,” he begins, “and this is the fault of
everyone who professes it.”*' What is missing, he now thinks, is a
pagan initiative equal to public displays of Christian benevolence—
their kindness to strangers, their care for the graves of the dead, the
(alleged, he says) purity of their lives. Julian’s solution, however,
suggests his desperation; his tactic is to outdo the Christians in dis-
plays of generosity. He orders hostels established in Galatia for the
benefit of strangers and the poor, who are also to have an allocation
of corn and wine. “It is disgraceful,” Julian complains, “that no Jew
ever has to beg and the wretched Galileans take better care of our
poor, as well as their own, than we do.”®2
For Julian the charity and holiness of the Christians is especially
cloying because their “atheism” seems to be showing better than the
faith of the heirs of Homer. But the political interest in creating these

(9. See Lita 3)


80. Ep. XX.452D.
81. Ep. XXII.429A.
82. Ep. XXII.430D.
JULIAN THE RESTORER 30

charities is close to the surface in Julian’s letters on the subject: his


real concern is that Christians were looking more and more to the
church for their protection and security and less to the emperor.
What Christians privately encourage with their benevolence, he sug-
gests, is apostasy from the state, though their practices are not a typ-
ical form of sedition. The view that Christianity was a subtle form of
rebellion against the legitimate rule of the emperor is something
Porphyry had also cited in his books against the Christian sect.83 At
no point does Julian regard Christian charity as genuine philan-
thropy, but he is willing to fight the errors of the Galileans on the
battleground of their own choosing.*4
A key goal of Julian's reforms was to create a series of social insti-
tutions that reinforced his role as supreme patron of citizens and
clients. In the letter to Alypius he outlines his desire to have an
“intercessory hierarchy” through which the goodwill of the clients
and citizens can be maintained through the work of social elites; the
panegyric of Mamertinus, already mentioned, emphasizes the same
program. His model for religious officials was contrived as the
equivalent of the social program: Julian was pontifex maximus; as
such he viewed himself as the lawful mediator between the Empire
and the gods. He regarded himself as having received the gift of
prophecy from the apollonian oracle at Didyma (Miletus), and
increasingly saw himself as able, in inspired states, to communicate
with the gods.®> It infuriated him that religious leaders over whom
he had no practical or moral authority, the Christian bishops, were
usurping this role and seeing to their own succession, while the
supreme pontiff was empowered only to appoint the archpriests of
a diminishing number of pagan dioceses. In a letter to an unnamed
official Julian reminds a negligent public servant, who had let an
assault on a priest go unpunished, that priests must be respected as
much as sacred objects, things set aside for sacred use. He reinforces

83. See Porphyry, Against the Christians, trans. R. J. Hoffmann (New York,
1996), esp. pp. 39-40. '
84. Thus he begins the work against the Galileans as a court case with himself
as presiding magistrate: CG 41E.
85. Letter XVII.451D.
36 INTRODUCTION

the distinction between priests and laity and insists that the former
must be free from the kind of harassment that has, apparently with
no public outcry, been visited on members of the pagan orders.*°
The real issue for Julian is less the security of individual priests (he
acknowledges some moral laxity within their ranks) than the ques-
tion of hierarchy and patronage. A man “who strikes a priest is guilty
of sacrilege,” because he offends at once the priest, the temple, the
emperor, and the gods,*’ and an officer of the peace who does not
punish the sacrilege is guilty of complicity in the crime. Julian sus-
pects, in any case, that the accused man secretly consorts with Chris-
tian bishops and elders over supper, and may even plot with them
to harass the pagans—whose gods, Paul had taught them, are not
gods but things of wood and stone.®®
The model of authority to which Julian appeals had already
been absorbed by the Christian church; bishops since the time of
Ignatius of Antioch had seen legitimacy and authority as extending
from God, through Christ, to the bishops, presbyters,®° and deacons ~
in orderly succession. For Julian, the matter of hierarchy entailed a
question about the religious authority of the emperor. The Christian
church, in appropriating the imperial model, was guilty of theft on
a political level—the same sort of crime Julian accused the Chris-
tians of committing, at a religious and theological level, in ran-
sacking the religious traditions of other nations. But if the latter con-
cerned mainly elements of doctrine and practice borrowed from the
Jews and pagan philosophy, Julian was genuinely concerned that the
theft of Roman political models could be turned against the power
of the emperor himself. The reassertion of the emperor's authority
over the religious institutions of the empire was thus an essential
element of his program.
The third element in his attack on the Christian church was the
use of polemic to discredit the religion that had enjoyed freedom
from official persecution for forty-eight years prior to Julian's acces-

86. Letter XVIII.450D.


87. Letter XVIIL.451B.
88. Paul, 1 Cor. 8.4.
89. Ignatius, Ep. to the Ephesians, 3.6.
JULIAN THE RESTORER of

sion. It is perhaps fair to say that of the three sorts of attack, broadside
was both the preferred way of dealing with religious annoyances and
the least effective. The literary attacks on the Christian church can be
traced to the time of Nero; in extensive form, to the time of the slan-
ders and accusations mentioned by Tertullian in the Apology.
Celsus, Fronto, Galen, and Porphyry had written tirades against Chris-
tianity; Proclus, Plotinus, and Marcus Aurelius were philosophically
contemptuous of it as a religion of slaves. The succession of oppo-
nents was impressive; but so also were the defenses offered by the
Christian fathers, who made the art of apologetics a branch of early
Christian theology. It was in the heat of discussion that important
doctrines like the nature of the soul, resurrection, the defense of
monotheism, and Christian ethical praxis were hammered out.
Julian was thus undertaking to do nothing new when he decided
to put pen to paper “in unfriendly Antioch”™ during the long winter
nights of 362-363 against the mischief of the Galileans, and his urg-
ings were the more hopeless because they were not new. It is often
suggested that he was indebted to Porphyry’s books against the
Christians for his arguments, but there are few direct echoes of Por-
phyry’s criticisms, at least in what survives of the treatise, and the
influence of the great biographer of Plotinus seems to be more gen-
eral than specific.?” Julian was an avid learner but an unsystematic
scholar, and the work as a whole lacks the critical acumen of a Por-
phyry and the originality of Celsus’s second-century attack. Like
Celsus, he uses Judaism, or rather, the construct of Jewish antiquity,
as a point of departure for his own assault, but only in the interest
of establishing the familiar argument that Judaism is thousands of
years older than Christianity, and thus worthy of respect for its
antiquity if not for its strange customs and beliefs. Judaism was to
90. Tertullian, Apology, Il.
91. See Athanassiadi, “The Priest-King and the Philosopher-Priest,” Julian: An
Intellectual Biography, pp. 161-91.
92. As G. Ricciotti has observed, Julian, unlike Porphyry, could “find nothing
good at all about Christianity, which he sees only as a bundle of historical absurd-
ities and moral aberrations.” Julian the Apostate (Rockford, 1999), p. 233. This is
clear from the first moment of the treatise Against the Galileans, where he sets out
to “prove that the beliefs of the Christians are a fabrication (oxevopia),” a human
invention “wickedly put together.”
38 INTRODUCTION

be preferred, not admired.° But it was seen by Julian as the proper


interpreter of its original doctrines—not the upstart religion that
had been on the scene for less than four centuries, and legitimately
so for only a fraction of that time. To the extent that Judaism was at
odds with Christian belief—on the question of prophecy, the role of
sacrifice, or the identity of the messiah—then the more ancient faith
was obviously to be preferred.°* The doctrine of the resurrection of
the body was—for Julian as for all Neoplatonists—an abomination
because it taught a grotesque idea of immortality and denied the
syzygy between the human (rational) and divine (Intelligible) that
had become characteristic of philosophical thought in various and
conflicting formats—Neoplatonic, Epicurean, and Stoic.”°
What Julian seems to have added to the mere polemical thrust,
nevertheless, was something his predecessors could not have con-
templated: the offer to rebuild the ruined temple in Jerusalem.®°
Julian was savvy enough to know that the exegesis of this event dif-
fered in the two communities. For Christians it represented the ful- —
fillment of a prophecy?’ which abrogated Jewish religious authority

93. The suggestion that Julian was a “Philo-semite” or a “proto-Zionist” has


been largely discredited; but see M. Avi-Yonah, The Jews of Palestine (Oxford, 1976),
pp. 185-207; FE Blanchetiére, “Julien: philhelléne, philosémite, antichrétien,”
Journal of Jewish Studies 33 (1980): 61-68.
94. CG 253A-E; 191E.
95. See Breckinridge, “Julian and Athanasius: Two eee to Creation and
Salvation” Theology 76 (1973): 74-76.
96. Amm., 23.1.7; Julian, Ep. XXV.398A-E. The “Letter to the Community of the
Jews,” whose authenticity is doubted, may nevertheless represent Julian’s intentions
accurately (cf. Ep. XLI.369A-398A; and Sozomen, HE 5.22 [Soc., HE 3.20], since a
portion of a letter quoted in Lydus, de Mensibus, 4 [Bidez, I.2:197] reports Julian
saying, “I will use all my zeal to make the temple of the most high God rise again.”
The project is also mentioned in the letter “To a Priest” (XIX.295C). The reference to
el elyon or theos hypsistos may have been suggested to Julian by the rabbis and encom-
passes both the name given to God in the Hebrew Bible (Gen. 14.18ff) as well as the
Phoenician theogony known to Philo of Byblos and mentioned by Eusebius (Praep.
Evang. 1.10). The language suggests as well that Julian may have regarded the restora-
tion of the Temple as significant to his syncretizing religious program.
97. Mark 13.2 and parallels, especially Luke 21.5-6. It is not clear that Julian
was influenced by the famous “documentary” hypothesis of Porphyry concerning the
Book of Daniel, but more with the Christian theory of fulfillment, which saw the
destruction of the temple as proof of God's election of a new people to replace the
JULIAN THE RESTORER 39

and gave Jewish law and custom, as well as the Hebrew scriptures, a
merely theoretical value. For the Christians the destruction of the
Temple invalidated Judaism and validated the prophecies of the
Gospel, an event of such importance that the earliest gospel writer
actually permits himself a glowering anachronism to suggest its
physical destruction was augured at the moment of the death of
Jesus (Mark 15.38). For the Jews, it was a political catastrophe that
signaled the end of their religious hegemony over Roman-ruled
Palestine and the loss of their religious center, and the stunning dis-
confirmation of the sort of messianic and eschatological hopes that
saw Jerusalem and the Temple as the locus for the new age. Chris-
tians of the fourth century regarded Jerusalem as a city “of Greeks,
foreigners, and idolaters,”°® its association with Judaism a thing of
the past. Origen had commented in his treatise against Celsus that
the loss of Jerusalem was “eternal,” a part of God’s judgement on the
ejected Jews, “who never before have been cut off from their ritual
and worship.””? For both faiths Titus’s devastation of the holy site
was a defining moment that formed the irreparable interpretative
breach between the two. Julian saw that political capital could be
gained by siding with the Jews as a benefactor, in effect making them
yet another of his elites and thus enlisting them in his campaign to
destroy Christianity.
In 362 he appointed Alypius, formerly the governor of Britain,
to begin the work of reconstruction, with the intention of creating a
monument to his generosity and religious largesse.!°? Enormous

Jews. Cf. Ep. Barn. 4.6-7; the more graphic verdict is Justin’s: “The custom of circum-
cising the flesh, handed down from Abraham, was given to you as a distinguishing
mark, to set you off from other nations and from us Christians. The purpose of this
was that you and only you might suffer the afflictions that are now justly yours; that
only your land be desolate, and your cities ruined by fire, that the fruits of your land
be eaten by strangers before your very eyes; that not one of you be permitted to enter
your city of Jerusalem. Your circumcision of the flesh is the only mark by which you
can certainly be distinguished from other men. .. . As I stated before it was by reason
of your sins and the sins of your fathers that, among other precepts, God imposed
upon you the observance of the sabbath as a mark.” Dial Tryph. 16.1.
98. Eusebius, Commentary on Psalms 86.2-4 (PG 23.1044c).
99. Origen, C. Cels. 4.22.
100. Amm., 23.1.2.
40 INTRODUCTION

sums (immodicis) were allotted for the project, augmented by contri-


butions from the Jewish patriarch, voluntary offerings, costly gar-
ments, and donations of jewels from Jewish women. The work is
said to have begun with clearing the temple site of centuries-old
debris from the valley of the Tyropoeon, which divided the ancient
city into two parts. The basilica built near Calvary during Constan-
tine’s time, close to the traditional location of the tomb of Jesus, was
not disturbed during the excavations. If the references in Ammi-
anus’s account do not represent an interpolation by Christian edi-
tors, the omens seem to have been bad from the beginning: tremors
and outbursts of “fireballs” made it impossible for the workers to
advance the job of reconstruction, and the project, always a thing
of fits and starts, was abandoned as Julian prepared to cross the
Euphrates en route to his final campaign against the Persians.

One of the persistent themes of Julian’s antagonism toward the


Christian church was his belief that they were troublemakers,
causing political discord in any community where they formed the
majority: “they turn everything upside down.”'? The most famous
instance of this is his reaction to a popular uprising in Alexandria,
ending in the violent murder of the Arian bishop George of Cap-
padocia when he threatened to demolish the temple to the city’s
tutelary deity. George is called “a human snake” by Ammianus,
apparently a reference to a poisonous reputation earned by serving

101. Ricciotti, Julian, p. 225, notes that toward the close of 362 earthquakes
along the Palestinian coast and parts of Syria leveled Gaza, Eleutheropolis (Beit-
Jibrin) and Nicopolis to the southwest of Jerusalem, and that Jerusalem itself was
affected by the tremors.
102. The Christian sources include descriptions by Gregory, Or. 5.3; Socrates,
HE 3.20; Sozomen, HE 5.22.2ff; John of Rhodes, Artemii passio 58; Philos., Hist. 7.9;
Theophilus, AM 5855; Theod. HE 3.20.4ff; and see the discussion by Robert
Panella, “The Emperor Julian and the God of the Jews,” Koinonia 23 (1999): 15-31.
Further, the PhD dissertation by Jeffrey Bross, “Apostate, Philo-Semite, or Syncretic
Neoplatonist: Julian’s Intentions for Rebuilding the Jewish Temple” (University of
California, Santa Barbara, 1992).
103. Ep. XXXVII.376C.
JULIAN THE RESTORER Al

as a part-time chef d’espionage for Constantius when it served his


interests. The bishop miscalculated his popularity on the day he
posed a memorable question to the Christian posse that surrounded
him, within earshot of the pagan citizens: “How long shall this sep-
ulcher stand?” He was first trampled and then torn apart by an angry
mob, the viciousness of which seems to have startled even Julian.
In his letter to the Alexandrians, Julian’s scolding tone toward the
perpetrators leaves no room for doubt that while the method for
removing George deserved to be condemned, his fate was not
entirely unwelcome.’ In a second case, Julian writes to Hecebolius
condemning the Arian Christians for attacking the small minority of
Valentinian-Gnostic Christians in Edessa and disturbing the peace of
the city.'°° His response shows Julian at once sarcastic and severe:
“Since [the Galilean’s] most worthy law tells them to sell what they
have and give it to the poor, so that they can attain more easily the
kingdom of heaven, I have decided to assist them, and order that all
the treasuries of all their churches be confiscated throughout Edessa
and distributed to the soldiers; and that such property as they own
be assigned to me and converted to public use. This is done so that
poverty will teach them a lesson in civility, but also to ensure they
inherit the heavenly kingdom for which they yearn.” And in a
third case (never proved), the Christians were accused of setting fire
to the temple of Apollo at Daphne in October of 362, a structure
which Julian had sought to aggrandize by erecting a colonnade
around the original structure, which had been built during the reign
of Antiochus Epiphanes (216-163 BCE).'°8 The persistent sense that
Christians were reckless, unphilosophical, and disloyal became
more intense as Julian approached the Persian campaign, and his

104. Amm., 22.11.1-11. Ammianus, however, makes the point that the popu-
lation of Alexandria was given to frequent outbreaks of violence.
105. Julian, Ep. 21.380A. Julian’s primary concern, however, is that the
manner of George's death was a desecration of the city reflecting the violent dispo-
sition of its people. Ammianus comments on this characteristic of Alexandrian life,
22.11.4.
106. Julian, Ep. XL.
107. Julian, Ep. XL.424D.
108. Amm., 22.13.1-2; the view of Christian writers was that it was destroyed
by divine intervention.
42 INTRODUCTION

earlier dissimulation, if it can be called that, gives way to an unam-


biguous animosity. Their religious doctrines set them apart from the
Greek ideal, and their religious rites—which he regarded as both
morbid and uninspired—were further proof of their inferiority.’
The basic themes of his polemical masterpiece were fixed by the
winter of 362.

III. THE FINAL CAMPAIGN

Julian’s various strategies to bring the Christian movement under


control while advancing the cause of the ancestral faith weakened as
he became increasingly involved in planning for the Persian cam-
paign. No one is flattering about his preparations. According to
Ammianus, he paid too little attention to those who advised caution
because “he interpreted caution as delay.” He was “passionately
eager to avenge the past,” and avoided human counsel out of suspi- ~
cion. Instead, he resorted, as he had done before battles in the past,
to augury and sacrifice. On one occasion he sacrificed a hundred
bulls and whole flocks of other animals, and he developed an insa-
tiable need for white birds. Rituals proliferated as needed precau-
tions to win the protection of the gods. His obsessive religious exer-
cises seem to have made no impression on the troops, who are
described as lost in an almost constant round of drinking and
carousing, especially the Petulantes and Celts, “whose indiscipline
at this time passed all bounds.”"° Impatient with traditional modes
of inquiry, Julian brought in anyone with a reputation for divina-
tion, however shady, and then decided that the time was right to
reopen the Castalian springs that Hadrian had closed off after a
ritual of reconsecration demanded by the god. The legend of the
springs was that Hadrian had closed the prophetic shrine because it

109. In the Misopogon, 344A, he alleges that the Christian churches were
nothing but “loathsome mausolea” built over the bones of their martyrs, a sentiment
repeated in the treatise against the Christians (CG 335B, C) when he argues that the
Galileans have done nothing “but fill the world with tombs and sepulchers.”
110. Amm., 22.12.5-7.
JULIAN THE RESTORER 43

was there he had learned that he was to be emperor, and he feared


that other visitors might receive the same message.

Julian immersed himself in a flurry of fund-raising activities to sup-


port the campaign against the Persians, with the goal of bolstering his
personal prestige and solving once and for all the problem his cousin
and predecessor had been unable to put to rest during his reign. As he
moved through Asia Minor to Antioch he tried to ingratiate himself to
the various communities with gifts and benefactions; but when an
earthquake in the region was followed by a severe drought, and Julian
refused to divert funds for the campaign in order to offer relief to vic-
tims of the disaster, the Antiochian senate became agitated and
declined to offer him any additional material support. Ammianus
speaks of Julian's “longing” for war, and of his reputation in Antioch
steadily declining. He was caricaturized in verse as a monkey, or a
dwarf who tried to make himself larger than life by puffing out his
chest and squaring his shoulders. The caricature included a standard
image of Julian as a goateed imbecile staring down the giants of the
east, “taking big strides like a brother of Otus and Ephialtes.”"* The
profusion of his sacrifices, including one memorable and frantic
search for an Apis bull (thought to be the harbinger of good harvests
and success against enemies), earned him the title “axe-man” rather
than priest, and tales were spread about the ostentation he displayed
by carrying the sacred objects personally rather than leaving it to the
members of the lower priesthood. In this respect he was not unlike his
uncle Constantine, who took the office of priest very seriously and was
given to equivalent outbursts of superstition. Julian kept his temper,
then composed a scathing satirical self-portrait, the Misopogon
(“beard-hater”), enumerating the defects of Antioch and her leading
citizens,"3 this while preparations for the war went on.
Li Amm., 22; 12.10.
112. Mythological giants. Amm., 22.14.4ff.
113. See Friedhelm L. Miiller, “Die beiden Satiren des Kaisers Julianus Apos-
tata” (Symposion oder Caesares und Antiochikos oder Misopogon). Griechisch und
deutsch mit Einleitung, Anmerkungen und Index. Palingenesia 66 (Stuttgart, 1998).
Ad INTRODUCTION

“I have committed [ | terrible sins,” he ironizes, “for though |


was coming to a free city which cannot tolerate unkempt hair, I
entered it unshaven and with a long beard, like men who are at a
loss for a barber. One would have thought it was some Smicrines he
saw, or some Thrasyleon, some ill-tempered old man or crazy sol-
dier, when by beautifying myself I might have appeared as a
blooming boy and transformed myself into a youth, if not in years,
at any rate in manners and effeminacy of features.”
Julian had come to Antioch in the autumn of 361 and stayed on
until March 362. Rich and important as the city was commercially,
Julian thought her glory depended on two things, the famous shrine
of Apollo and its school of rhetoric. Both of these had been neg-
lected by the citizens during the reign of Constantius. A Christian
church had been built in Apollo’s grove in the suburb of Daphne,
and Antioch’s most distinguished rhetorician, Libanius, was practi-
cally unknown in his home town. As a commentator has observed,
“‘Julian’s behavior at Antioch and his failure to ingratiate himself ~
with the citizens illustrates one of the causes of the failure of his
pagan restoration. His mistake was that he did not attempt to make
paganism popular, -whereas Christianity had always been demo-
cratic. He is always reminding the common people that the true
knowledge of the gods is reserved for philosophers; [yet] even the
old conservative pagans did not share his zeal for philosophy.
Antioch, moreover, was a frivolous city. The Emperor Hadrian three
centuries earlier had been much offended by the levity of her citi-
zens, and the homilies of... Chrysostom exhibit the same picture
as Julian’s satire. His austere personality and mode of life repelled
the Syrian populace and the corrupt officials of Antioch.” "4
The Sassanid Persians and before them the Parthians had been
natural competitors for hegemony in Mesopotamia since the days of
the late Republic. Yet it is often pointed out that Julian had no
reason, other than to settle an old score, to renew hostilities in the
east. There is also the possibility that he needed to win back the
respect and allegiance of his armies, since relations between Julian

114. W. C. Wright, Julian, vol. 2, Orations 6-8. Letters to Themistius, To the Senate
and People of Athens, To a Priest. The Caesars. Misopogon (Cambridge, MA, 1923), p. 89.
JULIAN THE RESTORER A5

and his officers had reached a low point. Victory over the Persians
would bring Julian the respect he craved and glory to the army, and
fighting a traditional enemy, rather than the Germanic intruders to
the west, would bring a special sweetness to any victory.
In early March 363 Julian set out on his final campaign. Ammi-
anus is careful to mention that he began his journey under an aus-
piciously sunny sky. When he entered Heliopolis a colonnade sud-
denly collapsed, with its beams and timbers crushing fifty soldiers
and wounding many others who were encamped around it. Julian
accelerated his pace towards Mesopotamia, intending to take
Assyria" by surprise—his trademark strategy of striking quickly and
where least expected. He crossed the Euphrates on a flotilla of boats
and entered the town of Batnae, where he encountered another
omen: a hayrick loaded with fodder collapsed under the weight of
groomsmen, burying fifty of them under its weight. Entering the
market town of Carrhae, the scene of a memorable disaster for the
Roman army during the time of Crassus, Julian performed sacrifices
in accordance with the lunar customs of the region, and for a small
space, between March 19 and March 29, the emperor's mood began
to clear; the signs had improved. After an especially restful night,
when he ordered his horse “Babylonius” to be brought to him, the
animal rolled on the ground in a fit of colic and dumped its jewel-
encrusted harness. Julian announced triumphantly, “You see:
Babylon has fallen, stripped of her finery.”"° After performing a sac-
rifice to the Magna Mater to confirm the omen, he is reported to
have been in good spirits and full of confidence that his armies
would triumph. Even a desperate letter from his friend Sallustius,
the prefect of Gaul, urging him to defer the campaign made no
impression on him."”
The omens in April turned dark again, and on April 7, toward
sunset, a small cloud “suddenly spread out over a darkening sky”
and with it flashes of lightning and thunder. One lightning bolt

115. For Ammianus, the region of Assyria encompasses Babylonia and Assyria.
IGs Amm~) 233.6.
117. On this, see A. D. Nock, ed. and trans., Sallustius, De diis et mundo (Cam-
bridge, 1926).
46 INTRODUCTION

struck and killed a groomsman by the name of Jovian and two of his
horses (the irony of his name was not missed by later commenta-
tors), and the interpreters decreed the lightning bolt had been “advi-
sory”—a warning from the gods to desist. But in this case the
Etruscan soothsayers were ignored by the emperor who consulted
the philosophers who, contemptuous of the diviners, conveniently
turned the omen on its head: it signaled, they said, the gods’ high
regard for Julian and the propriety of his battle plan.
And there were successes. Julian’s forces took the fortress of Anatha
and received the surrender of several local princes along the way. He
continued surprise attacks throughout Assyria, ravaged the countryside
between the rivers, and sensibly avoided the well-guarded fortresses of
Thilutha and Achaiachala."® The cities further to the south, Diacira and
Ozogardana, were sacked and burned by the Roman army, as was the
better-defended city of Pirisabora after a brief siege.
With the battle for Pirisabora the Romans got their first taste of
a sustained Persian resistance, and as they moved southward local
inhabitants became increasingly aggressive, flooding and otherwise
impeding their route. Just before reaching Ctesiphon, the Romans
came first to a deserted Jewish setthement and then to Maioza-
malcha, a sizeable Persian enclave with strong fortifications. Julian
must have imagined that in taking this city he would have driven the
Persian army to panic, since he had begun to feel the enterprise was
under the protection of a god."
The battle of Maiozamalcha was Julian’s last major victory in the
Persian campaign. Many Persian troops had escaped by stealth and
retreated to Ctesiphon, where the Persian army was entrenched and
waiting. As the Persian resistance grew stronger, increasingly using
guerrilla tactics against the Romans, and as Roman supplies dwin-
dled, Julian began for the first time to consider the possibility of
defeat. A sizeable number of troops was lost; Julian himself was
almost killed a few miles from Ctesiphon, and as they approached
the city the enormity of their miscalculation must have been clear.
They had relied on the cumulative effect of previous battles to

118. Amm., 24.2.3-4; Libanius, Or. 18.227-28; Zosimus, 3.151-52.


119. Amm., 24.4.1-31; Libanius, Or. 18.223-27; Zosimus, 3.20.2-7.
JULIAN THE RESTORER 47

damage Persian morale, but after preliminary skirmishes with Per-


sian advance troops it became evident that the battle would be a
long one—and one for which the troops were not equipped. Ammi-
anus, who witnessed the events, is reflective about the deceptiveness
of early Roman successes: “No star relieved the darkness of the
night, which we spent as one does in moments of doubt and diffi-
culty; no one dared to sit down or close his eyes for fear.”!° Julian
ignored the caution of his generals that the army was unprepared for
a protracted battle— “enraged by the possibility of defeat,” Libanius
says. But by this point the omens were unequivocally evil:

At dead night Julian after a short period of restlessness and troubled


sleep had roused himself, as was his habit, and was writing in his
tent, after the example of Julius Caesar. He was a lost in the profound
thoughts of some philosopher when he saw in the gloom... the
shape of the genius of the Roman people, which [had] appeared to
him in Gaul when he rose to the dignity of Augustus. Now it was
departing in sadness through the curtains of his tent with its head and
horn of plenty veiled. For a moment Julian remained in a state of
stupor; ‘then he rose above all fear and committed the future to the
will of heaven. The night was far spent and he was fully awake, so he
left his bed, which was on the ground, and betook himself to prayer,
using the ritual appropriate to avert evil. Then he thought he saw a
blazing light like a falling star, which clove its way through part of the
air and vanished. He was horror struck by the thought that the star of
Mars had appeared to him in this manifestly threatening form.’

Haunted by the certainty of failure, Julian chose retreat and


ordered his fleet burned, as he had decided to march through the
province of Assyria and allow his troops to live off the land as they
moved westward.!22 The Persians had the upper hand. Outflanking

120; ;Ani 2 5.151,


121. Amm., 25.2.10-20.
122. The failure of Julian is attested by a number of pagan as well as by Christian
writers. Libanius emphasizes that the boats would have been useless in the circum-
stances (Or. 18.262-63), and see also Ephraem Syrus, Hymns, 2-3; Gregory Naziazen,
Or. 5.12; Theod., HE 3.25.1. Further on the boatburning, see Michale diMaio,
“Infaustis ductoribus praeviis: The Antiochene Connection,” Byzantion 51 (1981): 502.
48 INTRODUCTION

Julian’s forces, they harassed the troops and burned crops along the
way, effectively starving them out before they could replenish their
supplies. The men suffered from fatigue and burning sun, scores
died of exhaustion. Ammianus, ever reluctant to call the emperor's
motion a retreat, suggests that the Persians had decided against
pitched infantry battles and preferred ambush to direct encounter,
keeping watch over the progress of the Roman troops from the sur-
rounding hills. In one such ambush, Julian, who had moved for-
ward to reconnoiter, was dismayed to learn from a scout that the
army had been attacked from the rear and rushed to join the belea-
guered troops without his breastplate and armed only with his
shield. In fact, the Roman forces were under attack on all sides, and
Julian went about frantically in an attempt to shore up the defenses,
“flying from one danger spot to another.”!”? In this reckless and con-
fusing moment a cavalry spear struck Julian solidly, piercing his ribs
and lodging in his liver. There are various reports about what came
next: according to some, the blow was enough to knock him from
his horse. In another, he tried in vain to extract the lance from his
body. Encomium has it that he insisted on being returned to his
horse so that he could continue the fight, “after the manner of the
great general Epaminondas.” Ammianus’s version of Julian’s
deathbed oration, one worthy of Socrates, is basically an exonera-
tion of his policies and a rebuttal to the charges of rashness and
immoderation that greeted news of his failure in Mesopotamia: “I
have enjoyed imperial power. This came to me as a gift from the
gods to whom I am kin, and I have kept it, to the best of my belief,
free from stain, showing moderation in the conduct of civil matters
and waging war, whether offensive or defensive, only after mature
deliberation.”'** He died on June 26, 363.

123. Amm., 25.29-16,


124. Amm., 25.3.18-25.
JULIAN THE RESTORER 49

Ill. THE RELIGION OF JULIAN

Julian tells us in his letter to the Alexandrians that he shared the


faith of the Galileans until he was twenty, but had not practiced it
for a dozen years.'”° However many interpretations this remark may
invite, it is clear that Julian’s agenda to restore the ancestral religion
of his pagan predecessors was formulated specifically to underscore
his rejection of Christian doctrine and practice. His attack on the
“creed of the fishermen” and the “carpenter's son,”!2° was also his
way of purging himself of a teaching which he persistently refers to
as the “disease” of the Galileans—the name he imposed on the
Christians to emphasize the geographical and social obscurity of
their origins. !7
Despite the barely repressed desire of certain Christian commen-
tators to see Julian as a new Decius—or a would-be Nero but for his
untimely death—there was in fact a more sinister aspect to Julian’s
program than earlier apologists had confronted in doing battle with
impatient emperors and pagan critics. It is the simple fact that the
persecutors before Diocletian not only had not been Christians, but
also had known very little about the religious enemy at the door.
The juridical way in which Christians were handled from the time of
Trajan (98-117) suggests nothing so much as disinterest in the
Christian church at an intellectual level, and the processes and spo-
radic persecution directed against them were targeted against associ-
ations, not specific ideas or practices. The latter were, of course, the
fodder for the attacks of variably informed critics such as Celsus and
his intellectual successors, who seem to have relied on rumor and
report, especially calumnies circulated by Jewish teachers and lapsed
proselytes, for their information. Julian, by contrast, less than two
generations after the legitimation of Christianity in the Empire,
found the religion philosophically defective, doctrinally unoriginal,
125. Ep. XLVII.434D.
126. Ep. XLV (To Photinus, preserved only in a Latin version of Facundus Her-
mianensis, dating from around 564), where he refers to “the whole mistaken folly
of the base and ignorant creed-making fishermen.”
127. Gregory, Or., 4.76; but even St. Jerome was troubled by the geographical
beginnings of the Church: Ep. 129.4.
50 INTRODUCTION

and liturgically impoverished. If Julian had not been reputed to be


the most philosophical emperor since Marcus Aurelius, or if his
information about the faith had been second-hand, like that
embedded in Celsus’s treatise, then his apostasy might have been
seen as unimportant, a caesura in the succession of faithful (more or
less) heirs of Constantine. But Julian’s devotion to the vocation of
philosophy was famous, acknowledged (if belittled) even by the
staunchest of his Christian opponents and attested by his tireless
correspondence with teachers, former schoolmates, dabblers, and
hangers-on, even in the midst of war.'*® As newly appointed caesar,
in the heat of the Gallic campaigns, he wrote to Eumenius and Phar-
ianus, two Athenian students, “Let your whole effort be to under-
stand the teachings of Aristotle and Plato;.let this be your only voca-
tion, the foundation, the base, the house, the roof.”'*? He combined
this philosophical commitment, moreover, with intense personal
piety, devotion to the cults, an interest in the rituals and theologies
of towns and cities throughout the empire, a genuine concern for
the physical integrity of shrines and holy places, a desire to reinvig-
orate local ancestral traditions he thought had been squandered
through impiety—disloyalty—and lassitude, and an almost naive
religious optimism that his loyalty to a cause imposed on him by the
gods would be rewarded by the gods.!°° Above all, he did not think
or write in disgruntlement as a pagan threatened by a new and
unapproved form of superstition (though he is keen to see Chris-
tianity as something new) but as a former member of a religious
association about which he is knowledgeable, whose holy books he
can quote, often from memory, and whose theologians, still battling
among themselves as “homoousians” and “Arians” he can refute
intelligently. These factors combined to make Julian a new kind of
enemy for the Church—an apostate—and this accounts for the
energy spent on his vilification following his death in 363.
His unpopularity among Christians seemed increasingly to tor-

128. G. W. Bowersock, Julian the Apostate (1978), p. 88, suggests that Julian’s cor-
respondence was the result of an inability to relate to his countrymen at a social level.
129. Ep. Ill.441D.
130. Cf. Ep. XXXVII.376C and especially Ep. VIII415D.
JULIAN THE RESTORER as

ment Julian just prior to the Persian campaign, since, perhaps inno-
cently, he expected their gratitude for following what he regarded as
a fair and even course concerning their right to practice their rites in
public.'*' He boasts that he has exercised leniency when he might
have been severe, and even Sozomen catalogues Julian’s “persecu-
tion” as a program of artful maneuvering rather than an assault on
the fabric of the church and its teachers.
!3?The most that can be said
of the disabilities suffered by Christians during his brief reign—
restrictions on teachers, revocation of military commissions, and
(doubtfully) threats to strip recalcitrant proselytizers of citizen-
ship—must all be assessed against the perception of many church
leaders that Julian was always guided by appearances: to be rash
would be seen to be unphilosophical—imprudent and intemper-
ate—and his confidence that it was the Christians who were truly
unphilosophical would finally be shown up in the kind of chaotic
displays and social unrest that accompanied their doctrinal disagree-
ments. Sozomen notes that however intent Julian was in promoting
the pagan cause, he was “equally intent in refusing to employ vio-
lent means [against the Christians] which might prove embarrassing
or seem tyrannical.”!7? This accounts for the sense felt by many
wriiters that the true measure of Julian’s opposition to Christianity
was unknown at the end of his reign, and that his death intervened
between a set of tactics that had not succeeded in bringing about the
restoration of “the Hellenic faith,”!°* and more direct measures that
would have suffocated the Church if he had returned victorious
from the Persian front. Theodoret, Gregory Naziazen, and John
Chrysostom all register certainty that Julian would have unleashed a
violent persecution if he had returned a victor.'*°

131. See the letter to Hecebolius (Ep. XL.424C) and to the Bostrians (Ep.
XLI.436A), where he expresses unhappiness with the response of the Galileans to
his leniency.
132. Sozomen, HE 5.16-17. But Sozomen also makes the point that by his
actions Julian encouraged even if he did not instigate persecution.
133. Sozomen, HE 5.17.
134. Ep XXII.429C.
135. Theod., HE 3.16; Greg., Or. 5.9; John Chrysostom, Contra Julianum, 22.
The most famous defense of the thesis that Julian was a persecutor in disguise,
biding his time and weighing his means before a fully fledged campaign against
52 INTRODUCTION

JULIAN’S NEOPLATONISM

All of this speculation leaves unanswered the fundamental question:


What elements of Julian’s religious and moral faith led the normally
pragmatic emperor to take the side of a dying religious cause? During
the period he was caesar, and an object of Constantius’s suspicion,
Julian is characterized by observers, both Christian and pagan, as a
soul in turmoil. There are traces of enthusiasm and excess in his let-
ters, especially those to philosophers and former teachers.3° Certain
letters are little more than panegyric, and others have the superficial
quality of stoic extemporations on the difficulty of suppressing
human love. “Where will I find a loving friend like you,” he writes on
learning of the departure of his friend Sallustius, “who will sustain
me, counsel me with wisdom, correct me with love, support me in
the Christians, is Bowersock’s Julian the Apostate (London, 1978), which sees him as
an implacably intolerant persecutor. A more nuanced view is that of Smith (Julian’s
Gods, esp. pp. 214-16), which sees a change in attitude in Julian’s approach to the
Church, partly as a response to provocation. On this reckoning, it is argued that
once he became confident that he could act with impunity against the Christians,
Julian decided on “the eradication of Christianity as a social force in the Empire.”
Smith cites in favor of his assessment the fact that Julian punished Palestinian Con-
stantia for its Christian associations (it was merged with pagan Gaza) and when the
Caesareans destroyed the last temple in their city “there was not only a fine but civic
demotion and higher taxes to boot.” The information comes from Sozomen, 5.3.
While there should be little doubt about Julian’s mindset against Christianity, the
question of his intentions remains open. Like his predecessors he regarded Chris-
tians as troublemakers at a social level; hence most of the penalties assessed were
social penalties designed to make the profession of Christianity an expensive
choice. An example is the confiscation of the Church’s wealth in Edessa, which he
Says is generous in view of what they might have deserved by presuming on his phil-
anthropia: “sword and exile and fire” (Ep 40.425A). A balanced assessment of the
evidence would suggest that Julian’s policy hinged on challenges to an official
policy of repressing the growth of the Church and advancing the growth of the tem-
ples; to the extent he was frustrated in this effort, his policies toward Christianity
varied, but did not become violent.
136. Libanius comments on this trait, as does Ammianus (27.7.3-4), who
deplores that when Julian greeted his former teacher, Maximus of Ephesus, his
enthusiasm was “unbecoming to an emperor.”
JULIAN THE RESTORER oe

doing noble deeds humbly, speak freely to me taking the sting from
his words.”!*’ But this apparent effusiveness is a philosophical pose.
As a Neoplatonist, though of a decidedly eclectic variety, Julian seems
to have felt a constant pressure for purification of the emotions that
impeded the soul's ascent to the One, and he gives literary vent to this
struggle in his epistles to intellectuals, priests, former schoolmates,
and teachers. For Julian the use of language in this way is a form of
exercise, marking his attempt to cultivate the cathartic or purifying
virtues of his philosophical caste. “Write, and do not cease writing to
me continually,” he implores his “philosophical” mentor, Maximus.
“No, rather come to me by heaven’s intercession, and know that
while you are away from me I cannot be called alive—except insofar
as I am able to read what you have written.”!38
In the system of Plotinus, the ultimate goal of ethical asceticism
was godlikeness, or “assimilation to God,”!*? and Julian relied more
than averagely on the assistance of professional philosophers,
teachers, and charlatans as guides on his journey. At least a part of
the enthusiasm incipient in his letters to teachers is familiar from
the “love-language” of Plotinus’s system, one which freely employed
words like intercourse (Ouoimotc), rapture, ecstasy, and nakedness to
convey the accessory levels of “undressing,” “entering,” and
“uniting.” The concept of the lover journeying to his beloved was
often construed!*° in language which was frankly sexual, and the
master-pupil relationship which Julian cultivated with teachers such
as Maximus and most explicitly, perhaps, in relation to the sophist
philosopher Priscus' frequently and deliberately exploits the Neo-
platonic idiom.

137. Or. VIII.243c. The language is analogous to Augustine’s lament over a


friend’s loss: Confessions, 4.8.13-9.14, also steeped in the Neoplatonic idiom.
138. Ep. XII 383B; and cf. Or. VII.235A; Letter to Themistius 264D. Julian's effu-
sive admiration for Maximus is also recorded by Ammianus (27.7.3), who says that
on hearing word of the teacher's arrival, Julian suspended a meeting of the Senate.
Maximus was with Julian at Constantinople in 362 and at his deathbed on the Per-
sian front. According to Ammianus (29.1), he was executed at Ephesus under
Valens in 371, having been implicated in a conspiracy against the emperor.
139. Plotinus, Enn. 5.3.7: dere navta.
140. Enn. 6.9.11.
141. Ep. V.425D.
54 INTRODUCTION

Yet there was an unavoidable contradiction in the Neoplatonic


system for Julian—at least in the unextrapolated and drier forms
developed by Plotinus and Porphyry. Carried to its logical extreme,
the system of Plotinus was not for the masses but for a philosophical
élite. Julian may have had no misgivings about belonging to this
caste, but its religious aspect “was so inward and so abstract that it
rejected on principle anything so material as temples and formal wor-
ship, the very things which were the chief foundation of ordinary
pagan religion.”!4? Plotinus himself had been contemptuous of reli-
gious externals and rituals (“It is for the tokens to come to me, not for
me to go to them,” he is reported to have said about religious cere-
mony)!43 and therefore contemptuous equally of the accoutrements
of pagan, Jewish, and Christian liturgies.'*4 These were for the weak-
minded, the hylic (earthbound) men who were not able to move
beyond the world of appearances to the intelligible (divine) world.
Plotinus or perhaps his teacher, Ammonius Saccas, had already found
a clientele among some Christian sects, especially the Egyptian and
Syrian Gnostic schools, but where the system attained the status of a
theosophical creed, as in the Gnostic cults, it tended to quarantine
itself from the worship and life of ordinary believers. It was exclu-
sivist, hierarchical, and élitist rather than, as some have claimed about
the Egyptian Gnostic communities, proleptically democratic.!*°
Within the system, God (10 €v) is the source of a dizzying profu-
sion of beings or emanations (the mpdvodoc) constituting the
entirety of the existing universe, while the emanations, through a
process of “reconversion” (€niotpodn), move toward the One by fits
and starts. For the Platonic demiurge or creator, Plotinus inserts the
concept of Intellect (votdc) and for the Platonic notion of a world-
soul he offers yoy or soul-in-itself. The intelligible world thus con-
sists of Mind (intellect) and Soul as emanations of the One; while
through a further outpouring of emanations comes the material

142. Ricciotti, Julian the Apostate, p. 31.


143. Porphyry, Vit. Plot. 10.
144. The ambivalence of Julian toward Mithraism must also be regarded in
this context.
145. See, for example, the work of Elaine Pagels, The Gnostic Gospels (New
York, 1978).
JULIAN THE RESTORER ae

world and material beings, including man—a mixed creature com-


prising the hylic or material element as well as a spiritual element
that “directs” him by degrees back to the true source of being, often
imaged as the “fatherland” (matpic).!4° Most popular understand-
ings of Plotinus’s philosophy, unlike the starkly metaphysical
Gnostic renderings, chose to emphasize the practical means of rec-
onciling the two “motions” of emanation from the One and return
or epistrophe, by envisioning levels or grades of attainment, roughly
equivalent to the grades of perfection known in religious circles
through the mystery cults, especially Mithraism. Julian, like many of
his soldier-comrades, would have seen no inconsistency in being
both a Neoplatonist and an adherent of the Mithraic and Cybeleian
mysteries, except the high symbolic content and private pursuit of
the latter contrasted with the public display of political, social, and
contemplative virtues that Neoplatonism demanded.!*’
146. Julian either misses or is uninterested in the Christian adaptation of certain
Neoplatonic themes, especially those in early writers like Paul (e.g., 1 Cor. 15.35-49).
147. Cf. Or. IV, the Hymn to Helios, and further on the subject, David Ulaney,
“Mithras and the Hypercosmic Sun,” in Studies in Mithraism, ed. J. R. Hinnells
(Rome, 1994), pp. 257-64, and the popular treatment of Julian’s Helios-worship, M.
Clark, “The Emperor Julian and Neoplatonism,” Sunrise (October/November 1996;
December 1996/January 1997). Starting with Plato’s notion that the universe itself
“came into being as a living creature possessing soul and intelligence,” Julian enun-
ciates the doctrine (Or. IV, the Hymn to King Helios) of the hierarchical nature of the
universe and all its parts, wherein the One Supreme Cause sends forth from itself
gods or powers that rule over lesser and lesser degrees of living beings, until every-
thing is included in the cosmic embrace. The sun and moon and the heavenly bodies
“are only the likenesses of the invisible gods.” The same oration contains Julian’s
allegorical description of the constitution of the universe, its substance, origin,
powers, and energies which are the sun’s gift to its domain, including the mysterious
“fifth substance, aether” (after Aristotle) which binds the whole together. Following
Plotinus he describes a chain of being, emanating from the “uncompounded
Cause”: Helios, the god behind the visible Sun, is lord of the intellectual worlds, not
only “the common father of all mankind,” who “continually revivifies [the substance
of things generated] by giving it movement and flooding it with life,” but also “the
mind of the universe” bestowing through Athena “the blessings of wisdom and intel-
ligence and the creative arts.” Helios gives to the “divided souls” (mankind) the fac-
ulty of judgment, and bestows on all nature the generative power. Julian emphasizes
that Helios brings about the various activities of his solar realm, not directly to the
beings, but through the means of countless other gods, angels, daemons, heroes, and
others. Referring to the Phoenicians, he cites their teaching that “the rays of light
everywhere diffused are the undefiled incarnation [embodiment] of pure mind.”
56 INTRODUCTION

Plotinus had emphasized the significance of “aloneness” (ut


solus cum solo uniaris) as a perquisite of the soul's journey, beginning
with the acquisition of wisdom (prudence, justice, fortitude, and
temperance—political virtue) with respect to one’s relations with
others, but ending with a self-mastery which required abnegation
and self-denial of a more austere nature. While Julian is known to
have demonstrated levels of mastery in regard to the political
virtues, he is also criticized for falling short in each category: impru-
dent in the Persian campaign and in relation to his paganizing
agenda; unjust in his attempts to curry favor with the army and the
social élites when first elevated as augustus; excessive in his final
hours; intemperate in his craving for augurs, omens, priests, sooth-
sayers, and sacrifices. His widely reported philosophical withdrawals
were partly meditative, partly, as Ammianus reports, laced with expe-
riences of dreams and visions such as one might expect of any reli-
gious zealot of the era.!48
But the strands of Julian’s attachment to Neoplatonism must be
understood in terms of a philosophical tradition that winds from the
early disputes over the “true” interpretation of Plato’s philosophy in
the work of Plotinus.and Porphyry to the more radical reformula-
tions of Iamblichus of Chalcis and the Syrian schools, which devel-
oped an increasingly radical view of the separation between the One
() mavtn Appytoc) and the sensible world, configured as links in a tor-
tuous chain of intermediaries. The artificial complexity of
Iamblichus’s thought resembled the Syrian gnostic schools, which
doubtless took ideas from and gave notions back to the impulsive
mystics who propagated Iamblichus’s doctrine. Central to this
teaching is the subdivision of the sensible world into an upper level,
the home of gods, heroes, angels, and demons, and a lower, material

148. Amm., 25.2.10-21. In the Hymn to King Helios, Julian recalls that these
periods of withdrawal began in childhood: “A vehement love for the splendors of
this god took possession of me from my youth; in consequence of which, while I
was a boy, my rational part was ravished with astonishment as often as I surveyed
his ethereal light; nor was I alone desirous of steadfastly beholding his diurnal
splendors, but likewise at night, when the heavens were clear and serene, I was
accustomed to walk abroad, and, neglecting every other concern, to gaze on the
beauty of the celestial regions with rapturous delight.”
JULIAN THE RESTORER od

level inhabited by mankind. Uniquely, man mediates between these


worlds, since he alone among sensible creatures has intimations of
the gods as well as a vague aspiration for the supreme good. Yet in his
natural state, as in classical Platonic thought, he is bounded and
inhibited by the “prison house of creation.” The theosophical impli-
cations of the system take us beyond Plotinus to a world where “the
divine element is echeloned, and scattered throughout the whole
range of being,”'*? and where all phenomena must be scrutinized as
broken fragments or images possessing a reflection, however dim, of
the divine. The unusual, the quaint, the horrific, the surreal, and the
ambiguous might contain information channeled from the realms of
all-seeing power: a calf born with two heads—or in Julian’s instance,
a collapsed hayrick or a dyspeptic horse—might reveal as much
about the outcome of events as the predictions of soothsayers or the
contemplations of an enlightened conscience. There is something to
be said for the claim that this phase of Neoplatonic teaching, because
of its rejection of reason as a way of knowing the intelligible Good,
represents the end of Neoplatonism as a philosophy and its transfor-
mation into a form of theurgy and magic. Iamblichus himself seems
to have avoided vulgar displays of magic, though he is presented as a
magician on several occasions by the biographer Eunapius.'°° In fact
the term “theurgy,” which is much debated in relation to Julian’s own
religious conspectus, is greatly misunderstood:!” it does not relate to
a subclass of magicians or hierophants, but simply to men and
women from a variety of religious and philosophical castes'°? who
claimed to be god-evokers. Iamblichus himself had proclaimed that
149. Ricciotti, Julian the Apostate, p. 33.
150. Eunapius presents lamblichus as being challenged by his earliest followers
to do ever greater feats of magic. His demurer on one occasion (“It is irreverent to
the gods to give you this demonstration, but for your sakes it shall be done.”) has
significant cognates in the reticence of Jesus, according to the gospel tradition, to per-
form miracles: cf. John 4.48; 2.4; Mark 1.40-43; 6.5, etc. Eunapius, Lives of the
Philosophers, 459, in W. C. Wright (trans.), Philostratus and Eunapius: The Lives of the
Sophists (London, 1922), pp. 368-73. See also R. J. Penella, Greek Philosophers and
Sophists in the Fourth Century A.D.: Studies in Eunapius of Sardis (Leeds, 1990).
151. See Rowland Smith, Julian's Gods, pp. 91-113.
152. The Chaldean and Iamblichan “theurgies” discussed by Smith, pp.
91-113, provides a good theoretical framework for understanding Julian's milieu, if
not specific aspects of his own practice.
58 INTRODUCTION

“it is not by thinking that men are linked to the gods, but . . . by the
efficacy of unspeakable acts performed in the appropriate manner,
acts which are beyond all comprehension, and by the potency of the
unutterable symbols which are comprehended only by the gods.”!”°
These symbols or “tokens,”—herbs, salves, signs, and displays of
necromancy—“accomplish their proper work without intellectual
effort by their own virtue.” In this sense the tokens of theurgy are effi-
cacious, a moment of revelation made available to those trained in
interpretation. Other ancient forms of divination and prophecy
envisaged a relationship with the divine through the divine being’s,
or the gods’, immediate self-disclosure. Without challenging—
indeed, taking for granted—the legitimacy of the Hebrew and Apol-
lonian modes of revelation, 9eovpyia, as it was understood by the
later Neoplatonists, was a bringing of divinely hidden truths to light
by a sort of detective work: 8eaywyia. The difference, while signifi-
cant, is essentially one of emphasis; in the theurgic system of
Iamblichus and his disciples much less integrity can be accorded the
divine being than in the ancient prophetic systems, where the divine
will is known through revelation or Adyoc.!** Thus in his discourse on
theurgy, Iamblichus discusses the marks of “true” theurgy in relation
to physical and mental states involving possession, dreaming, and
rapture, and certain visible tokens of genuine (as opposed to fraudu-
lent or deceptive) divination, such as healing powers, levitation,
orgiastic purification, the invocation of spirits, and visions of
specters.'°? Similar theological calculuses were developed by Iam-
blichus and his school for the study of numbers and harmonics.!5°
153. Iamblichus, Myst. 2.11.
154 Without stretching the analogy, which the church fathers almost unani-
mously rejected prior to the theosophical reworkings of the Pseudo-Dionysius,
Nicene Christian theology offered a reductionist soteriology, emphasizing the
savior as “consubstantial” and sui generis (unigenitum). Gnostic Christianity, on the
other hand, conformed to the Neoplatonic model.
155. lamblichus, Theurgia III.7 (Theurgia, or The Egyptian Mysteries, trans. A.
Wilder, London, 1911).
156. The “break” in the Neoplatonic tradition is sometimes thought to consist
in the dispute between Iamblichus and Porphyry over the role of religious symbols
and the place of liturgy and ritual in the attainment of philosophical truth, as the
following exchange, composed from Porphyry’s Letter to Anebo (Epistula ad
Anebonem) and lIamblichus’s De mysteriis, a refutation of Porphyry’s critique
JULIAN THE RESTORER 59

Essentially, however, theurgy in the Neoplatonic context, whether


primarily mystical (mystagogic), or thaumaturgic-liturgical, was a
form of soteriology: the goal of theurgy is not merely the under-
standing of the cosmic order but clarifying the relationship between
individual destiny and the noetic (the origin of all: } advm CLPPNTOG
apxn) through special rituals and techniques. In its practical aspect, it
seeks to answer the Platonic question, How is one saved from the
world of appearances and brought into the light of the One; recipro-
cally, how can flashes of the noetic illuminate everyday existence, so
that the future becomes decipherable rather than mysterious and
(addressed ostensibly to an Egyptian priest) suggests. (Page numbers refer to respec-
tive sections in the two epistles):
Porphyry’s Questions: (Section 3) Since the gods are unlimited, undivided,
and uncircumscribed in power, how does it happen that religion and ritual treat
them as allotted to different spheres of influence?
lamblichus’s Replies: The differences do not lie in the gods, but in the recipi-
ents and their varying capacity to receive different types of divine power (30-33: cf.
n3)3()9)
Porphyry’s Questions: (Section 4) Why are shocking and obscene events and
language used in the Mysteries?
Iamblichus’s Replies: (a) Because this is the correct way of worshiping the gener-
ative forces (38-39), (b) because it “inoculates” those elements in matter seeking
order, showing them the better by demonstrating the worse (39). (c) Repressing emo-
tions makes them more destructive, moderate release is cathartic (39-40). (I, 11)
Porphyry’s Questions: (Sections 4 and 5) Why are so many things at sacred rites
performed as though they were directed to beings who are swayed by emotions?
Iamblichus’s Replies: Prayer and invocation do not mean that the gods are per-
suaded to come down to us, but rather that by their means we adhere to and
become like them (42, 46-49). The “anger” of the gods is not their turning away
and desertion of us by them. The situation is rather that we render ourselves inca-
pable of receiving their beneficence. And the same argument holds true for expia-
tory sacrifices (43-44). (I, 12-15; VIII, 8)
Porphyry’s Questions: (Section 6) How is it that some (planetary) gods are
givers of good things, but others of evil?
lamblichus’s Replies: These things arise from a misunderstanding of astrology,
which talks of benefics and malefics. All the gods are good, but material conditions
may distort that which emanated from the divine in a state of harmony and lead to
conflict (53-57). (1-18)
Porphyry’s Questions: (Section 8) What is the difference between gods and
daemons?
Iamblichus’s Replies: The governance of the gods is all-embracing and unre-
stricted. That of the daemons is limited in time and place; daemons do not com-
pletely transcend that which they rule (63-64). (1, 20)
60 INTRODUCTION

unknown: “You must look toward the light and toward the rays of the
Father, whence your soul was sent to you clad in much nous . . .“; “Do
not look upon nature; the name is enmeshed with fate”; “Let the
immortal depth of the soul be opened up; spread out your eyes
strongly and upwards.”!°’
Following his initiation into the “entrance to philosophy” by
Maximus, Julian seems to have been attracted to the rites of the so-
called Chaldean Oracles, the “canonical” text of Neoplatonic
theurgy, or so we may judge from a letter written to Priscus from
Gaul in 358, asking the philosopher to bring with him a copy of
Iamblichus’s commentary on the Oracles. In fact, Julian says little
about theurgy or the Oracles as such in his writings, but there is evi-
dence to suggest that he remained attached to theurgic praxis
throughout his career as a military commander and emperor.!°®
According to Ammianus, the practices, now intertwined with tradi-
tional augury, intensified throughout his period in Illyricum, just
prior to the show-down with Constantius in 361, when Julian
seemed to undergo periods of depression if confronted with
ambiguous portents or forced to deal with charlatans.!°?
Julian’s letters addressed to Iamblichus are without doubt for-
geries,'©° but the emperor's inclination to Iamblichan doctrine cannot
be seriously doubted. He describes Iamblichus as “a truly godlike man,
the third after Pythagoras and Plato,” and one “who has reached the
highest wisdom that man can attain . . . so that no one can be anything
more perfect than he.”'*! The most important names associated with
Iamblichus’s school are those of the Pergamese teacher Edesius!®? and
157. Chaldean Oracles, 115, 102, and 112, and Smith’s discussion, Julian’s
Gods, p. 101. On the Oracles generally, H. Lewy, Chaldean Oracles and Theurgy: Mys-
ticism, Magic and Platonism in the Later Roman Empire (Paris, 1978).
158. Some of the evidence for this is arguable (cf. Smith, p. 91), but Julian
does petition for perfection in theurgy in his Hymn to Cybele, Mother of the Gods.
159; Amo mir 2 2.2.12 28.
160. Eps. LXXIV-LXXIX. Or—in view of their innocuous content—letters
written by another admirer and wrongly ascribed to Julian by virtue of the latter’s
professed admiration for the philosopher's doctrine: see Wright, III, xlix.
161. Hymn to Helios, Or. IV.157C, D.
162. According to Eunapius, Edesius for either political or intellectual reasons
refused to accept Julian’s lavish gifts at Pergamon with the retort, “You have heard
me teach many times but are still ignorant of my soul.”
JULIAN THE RESTORER 61

his pupils Chrysanthius, Eusebius of Myndus, Priscus, and Maximus—


all of whom had associations with Julian, as did Eunapius, the biogra-
pher of Iamblichus, and the philosopher and mysteriously influential
thetor Eustathius, a hierophant and sage who impressed the Persian
king Sapor with his oratory, if not with his magic.!°? Of these, Maximus
occupies the preeminent position, especially as an arbiter and judge in
philosophical discussions: “ottog éoa.”!°4 On his arrival in Pergamon,
the aged Edesius’s reticence toward Julian (or perhaps toward his
potential as a philosopher) left him in the care of Chrysanthius and
Eusebius, Priscus having gone to Greece and Maximus having been
called to Ephesus. Eunapius makes the point that Julian was torn
between Chrysanthius’s irrationalism (his devotion to the occult was
famous), and Eusebius’s quiet pursuit of dialectic “when Maximus was
not present to correct him.” The picture of Maximus, Julian’s “twin
soul,” that emerges is not flattering: on the one hand a sycophant and
a power-broker, on the other hand someone whose eloquence and per-
sonal magnetism made him impossible to ignore. Eunapius speaks of
his “grand nature,” but at the same time as one who despised rational
proofs, dialectical reasoning, and was completely devoted to magic,
“even to the point of playing the madman.” Julian’s devotion to the old
teacher may have been encouraged by a tale told him by Eusebius
involving a scene at the temple of Hecate,'® who had a special role as
prophetic intercessor in the Neoplatonic theurgy, especially that repre-
sented in the cosmogony of the Chaldean Oracles.'°° In the tale told by
Eusebius, Maximus convinced a number of his friends that he could
demonstrate to them how far he surpassed ordinary mortals; he then
burned a single grain of incense before the statue of Hecate in their
presence, whereupon the goddess began to smile, then laugh out loud,
until at last “the torches she held burst into flame.” The element of fire
was especially significant as “proof” of Maximus’s semi-divine stature,
because of its symbolic centrality in Neoplatonic, especially the

163. Amm., 17.5; cf. Julian, Eps. XLIV and XLV.


164. “He has said it.” Ep. 16.452B.
165. Libanius observes (13.11) that a “certain spark of the mantic art” caused
Julian to lose his hatred of the ancestral gods and become docile to the oracles.
166. See the excellent discussion of this subject in Rowland Smith, Julian's
Gods, pp. 99-104.
62 INTRODUCTION

Chaldean, theurgy. “The fire . . . which has occupied a vessel above the
heavens . . . is infinitely in motion, a boundless eternity. It is not within
the grasp of the blessed gods unless the mighty father should plan his
purposes so that he himself might be looked upon.”!°’ On hearing the
story, whatever Eusebius may have wished its effect would be, Julian
kissed Chrysanthius on the forehead and set out for Ephesus in pursuit
of Maximus.
According to his adversary and former schoolmate, Gregory
Naziazen, Julian was initiated into the mysteries at Ephesus by Max-
imus and became an adept in theurgy under his tutelage.'°® As one
modern commentator has observed, the “conversion” of Julian in
350-351 owes a great deal to his “human exemplar” and the distinc-
tive color of his Neoplatonism is probably best sought in Maximus’s
teaching.!°? Julian does not seem to have been converted by the cool
truths of Neoplatonic doctrine but by the thaumaturgic displays that
almost unexceptionally impressed all of Maximus’s disciples—even
those not able to duplicate his feats. This is not to suggest, as an
older school of interpretation claimed,'” that Julian’s conversion
was totally without a philosophical motive or interest,!” but that it
was as much a devotional need as intellectual commitment that
sealed his loyalty to the pagan cause and the ancestral rites, and
inevitably shaped his attitude toward Christianity thereafter. Gre-
gory tells the story of the initiation in a detail that calls attention to
its basis in hearsay and Christian legend, though the essential out-
line of the event is plausible. Accompanied by a hierophant, Julian

168. Gregory, Or.V.34; Eunapius, 475.


169. Smith, Julian's Gods, p. 186.
170. See Arthur Darby Nock, Conversion (Oxford, 1933), pp. 14, 185, 218-19.
171. The Epistle to Themistius is the primary source for Julian’s understanding
of politics. His primary model is the optimus princeps of Plato’s Laws, especially
where Plato discusses the “more divine nature” of the king (ovoems Seic8a1 Sa1-
Lovimtépac). While admiring the examples of Solon, Pittacus, and Lycyrgus, Julian
remarks that he “has been assigned by God the same position formerly occupied by
Heracles and Dionysus, who being at once philosophers and kings purified almost
all the earth of the ills infesting them” (253c-254B). Julian’s refusal to play an
active role in politics, preferring the “Socratic” to the “Alexandrian” style, is based
on the conviction that Oe@pia exceeds npdtetc. See further on this period the dis-
cussion of Athanassiadi, pp. 56-58.
JULIAN THE RESTORER 63

entered the mithraeum in fear, “greeted by strange sounds, revolting


exhalations, fiery apparitions, and prodigies.” Comforted by the
words of his guide (“We loathe but no longer fear [the demons]"),
Julian emerged from the abyss reconciled to his future and “from
that day on was possessed.”!72
The basic accuracy of Gregory’s final comment on Julian's pagan
“baptism” is borne out in the events of his subsequent life as caesar
and emperor, as well as by the testimonies of Libanius, Ammianus,
and Julian himself. In his Hymn to King Helios (Oration IV), com-
posed over three days, and only six months before his death, Julian
rehearses the episodes that brought him to the worship of Helios
and the proofs he feels secured and justified his faith in the god's
sovereignty. “From my youth I have longed for the rays of the god to
embrace me; from early days my mind was enthralled by light ethe-
real (thy didvoiav €€totaunv) such that I not only wanted to gaze at
the sun without stopping but I wanted as much to walk alone at
night, beneath the clear and luminous sky, giving my whole soul to
the beauty of Heaven, having no regard for anyone nor anything that
was said to me.”!”? Claiming to be ignorant of the science of
astrology, Julian records that even as a child, and lacking the
“proper” tools of interpretation—the Platonic view of the divinity of
the cosmos—he sensed the natural affinity of the sun and enlighten-
ment, and revered the sun as the medium that dispelled the clouds
of ignorance, now associated in his mind with Christian doctrine.
The association between “darkness” and Christianity would
become more pronounced in Julian’s mature work, especially in an
oration delivered against the teaching of the cynic Heraclius (Ora-
tion VII). In it Julian recalls a spiritual crisis that he seems to have
undergone close to the time of his initiation into theurgy. What pro-
voked the crisis is anyone’s guess—a delayed reaction to the
slaughter of his relatives during his childhood may have been
responsible!”*—but the language of the oration seems also to sug-

172. Greg., Or. IV.55-56; Theod., HE 3.1.


173. Or IV.130B-131A.
174. “Having been so tormented by the death of my kinsmen as to want to
hurl myself into the bowels of Tartarus” (Or. VII.229D).
64 INTRODUCTION

gest that he had come to believe that the religion of Constantine’s


sons, and especially that exemplified by Constantius himself, was
not passively offensive but actively repugnant. He prays to Zeus
“that this wicked zeal for impiety should not overtake all men,” and
urges Helios “to take care especially of your child Julian and cure
him of this sickness, for you see how he is infected with smoke and
filth and darkness, and there is a risk that the fire you sowed in him
will be extinguished.”!”° Notably, the sickness and disease to which
he refers make Christianity and error indistinguishable, and Julian
thetorically favors the epithet “disease” as a way of describing the
Christian faith in his letters. The devotion to Helios finds its closest
approximation in the pre-Christian devotion of Constantine himself
to the Sol Invictus, but Julian’s mystical conviction—that he was in
direct relationship with the god, and that the god not only guided
but admonished, chastised, and befriended him in time of need—is
distinctive. Thus Julian records that Helios himself recognizes the
divine particle—the splinter of light—within him, and that the god
preserves him from danger.!”°
The sense of divine favor is consistent in Julian’s orations and
letters, if somewhat fluctuating from moment to moment. In his
letter to the Athenians, he-claims that he sought divine guidance
“with outstretched hands” when nominated to be Caesar by Con-
stantius, and saw the goddess Athena atop the Acropolis, “who sent
guardian angels to my side from Helios and Selene.”!”’ In a similar
episode, he sought divine counsel about whether to send an
embassy to the Empress Eusebia in Milan, and was warned by the
gods that if he did so he would be killed.!”8 These visions are men-
tioned also in reports from Ammianus, who says that the emperor
was. in the habit of telling his closest friends about his dreams and
apparitions, and that the night before he was proclaimed Augustus
he was visited by the genius publicus—the guardian spirit of the

175. Or. VII.228B, 229B, D.


176. Or. VII.229D. Julian comments that he was saved from his wretchedness
through the intervention of Helios and Athena.
177. Ep ad SPQ Ath. 275B.
178. Ep ad SPQ Ath. 275C, D.
JULIAN THE RESTORER 65

state—who upbraided him for his recalcitrance and threatened to


abandon him forever if he declined the dignity.!”9
Julian’s composite Neoplatonism, his reputation as a mystic of
sorts, a visionary given to philosophical rhapsodies and the practice
of theurgy, yields an uncertain portrait of his religious faith. If he was
a Christian until his twentieth year, what sort of Christian was he?!®°
If his acknowledgement of the principality of Helios predated his
“philosophical” conversion under Maximus, as he seems to suggest
in the concluding section of Against Heraclius, does this nascent
paganism nullify the outward attachment to Christianity during his
youth? Libanius suggests that Julian's religious hypocrisy was so well
developed, “that if Aesop wanted to tell a new fable, it would not
about the ass that hid under the skin of a lion, but of the lion that
hid under the skin of the ass.” The lion is Julian; the ass, the ancient
symbol of opprobrium for the ignorant Christian.!®' At another level
of interpretation, however, Julian's religious “hypocrisy” conforms to
the eclecticism of his age. One sees repeatedly in his letters and ora-
tions the syncretism that characterized the religious melting pot of
late antiquity: the supreme God or ruling power can be addressed as
Helios, or as Zeus, invoked solely, or in combination with Athena. In
the Hymn to the Mother of the Gods, addressed to the Great Mother

1792 Amim, 20.5-10.


180. The many attempts to answer this question can perhaps be put aside in
favor of the following: Julian’s mother, Basilina, was, if a Christian, an Arian; his
first guardian, Eusebius of Nicomedia, was a defender of the Arian theology; Con-
stantius himself, who controlled the fate of both Julian and Gallus, professed the
heresy; and both George of Cappadocia, whose library Julian inherited, and Aetius,
who had once tried to convert Julian to Christianity on learning of his pagan incli-
nations, were Arians. In his writings Julian does not distinguish between Arians and
orthodox Christians, but does seem to know of the doctrinal disputes dividing the
two contingents. The extent to which the confusions of the post-Nicene period in
Christian theology may have contributed to his abandoning Christianity cannot be
assessed; however, Ammianus “represents” Julian’s attitude in the following: “[He
knew from experience] that there are no wild beasts so hostile to mankind as are
most Christians in their hatred for one another” (Amm., 22.5.4).
181. Libanius, 18.19; on the ass image, in addition to the famous graffito from
the second (?) century, cf. Tertullian, Apol. 1.16 (“Nam et somniastis caput asininum
esse Deum nostrum” etc.); Ad nationes 1.11, 14; Minuc. Felix, Octav. 9. Tertullian traces
this absurdity to Cornelius Tacitus, who uses it against the Jews (Hist. V.4).
66 INTRODUCTION

Cybele, the focus, if there is one in this meandering tract, is on the


redemption of Attis by castration as an allegory of the soul’s
epistrophe to the One; yet it is possible that the hymn itself is
impregnated with allusions to the Mithraic zodiac and cultic
astrology.!82 The double and triple aspects of gods (trimorphism)
made it possible for Julian to mix and match deities according to
need, or the severity of a crisis: one such moment was the call from
Constantius for Julian to return to Constantinople when he had
been in Athens for less than three months. His prayerful approach is
to Athena, regnant on the Acropolis, though underlying his appeal is
the notion that Athena acts as intercessor between himself and
Helios, an aspect often associated with Cybele.'®? It should be
stressed, however, that this does not represent a “confusion” of
names in Julian’s theology,!®* but is rather an essential feature of the
Iamblichan system, reflecting the provisionality of all revelations of
the divine nature. In the same way, Julian can speak of the God of
the Jews, without irony, as the Most High God,'®° who under any
other name is a revelation of the One. In the Hymn to Helios, he
intones his conviction that the many names of the “intellectual”
(intermediate) gods are always to be associated with the source of
their power, and hence are modes of knowing the supreme god: “We
say that [Helios’s] royal establishment among the intellectual gods,
from his middle order between the planets; for if we perceived these,
or as many other properties, belonging to any other of the apparent
gods, we should not ascribe the principality among the gods to the
sun. But if he has nothing in common with the rest, except that
beneficent power which he imparts to all, we ought to rely on the tes-
timony of the Cyprian priests, who raised common altars to Jupiter
and the Sun; or, indeed, prior to these, we should confide in Apollo,
who is the attendant of this god; for thus he speaks: Jupiter, Pluto,
Serapis, and the Sun, are one. And thus we should consider that there
is a common, or rather one and the same principality, among the

182. Smith, Julian’s Gods, p. 161.


183. Ep. SPQ Athen. 275A.
184. Thus against Ricciotti, Julian the Apostate, p. 64.
185. Ep. LL.398A.
JULIAN THE RESTORER 6/7

intellectual gods.”!®° When Julian addresses Maximus shortly after


the death of Constantius, he invokes Zeus, Helios, and Athena as a
trinity of powers who have revealed to him that his master would
come to no harm.'*’ Similarly, he invokes Apollo as the “coregent of
Helios”'*® whose rule binds Rome, Constantinople, and Athens
together in a community (the oikoumene) of shared values.
In the oration Against Heraclius,!* Julian alludes to his conse-
cration as “soldier” in the Mithraic mystery. It is sometimes sup-
posed that this initiation took place during his governorship of
Gaul.'”° In another passage Julian is reminded by Hermes, the for-
midable symbol of mystagogic gnosis, of his responsibilities to the
cult. The depth of his involvement in the cult of Mithras is widely
debated. On one reckoning, Julian can be seen as a good miles
Mithrae [soldier of Mithras], whose devotion to the cult was
intensely personal and entailed keeping its secrets.!?! The argument
for this view is chiefly from silence, since in his most extended reli-
gious tractate, the Hymn to Helios, Julian does not mention the
Mithraic rituals and makes no allusion to its doctrine. On another
reckoning, the silence is more telling; Julian does not mention
Mithras, except fancifully, because he was more interested in pro-
moting the well-established gods of the Empire as a part of his polit-
ical agenda to restore the ancestral cults of Rome. Thus Zeus, Apollo,
Cybele, Athena, and the tutelary gods of cities are mentioned fre-
quently, and to a certain extent coalesce, as is typical within Neopla-
tonic theurgy. Perhaps because of its origins, its special association
with the guardian caste, and Julian’s sense that “philosophical”
rulership demanded virtues not altogether patent to Mithraism,'”*

186. Or. IV; T. Taylor's translation (1793), pp. 49-50.


187. Ep. VIII.415A, B.
188. Ep. XI.152D.
189. Or. VII.234A.
190. Thus Athanassiadi, “A Contribution to Mithraic Theology,” p. 38.
191. Rightly challenged by Smith, Julian's Gods, p. 162, as being “impossible
to verify or refute.” Cf. Cumont, Les mystéres de Mithra (Paris, 1913), p. 146.
192. Smith (p. 170) calls attention to Julian’s emphasis on philanthropia and
civilitas, the cultivation of civic virtue and justice. By contrast, Athanassiadi (p. 41)
stresses the moral code of Mithraism would have been appealing to Julian. Or.
VII.233D-234A presents Helios as prescribing four commandments for Julian:
68 INTRODUCTION

he may have seen contradictions between promotion of the latter


and his commitment to Iamblichan doctrine.
By the same token it is possible to exaggerate the incompatibility
of the two systems. The Pergamese and Anatolian schools of Neopla-
tonism represented by Julian’s teachers were not themselves closely
identified with the traditional gods of Roman religion, so the “for-
eignness” of the Mithraic system was not in itself a decisive mark
against it for Julian, whose ability to syncretize gods within his the-
ological calculus is attested by all available literary and epigraphic
evidence. In fact, it is Julian’s religious inclusivism that distinguishes
his philosophical opposition to Christianity from the largely intel-
lectual opposition of Celsus. While Celsus speaks with the voice of
a conservative second-century intellectual, opposed to new and
unapproved cults—especially those associated with the lower
classes—Julian speaks as a religious syncretist and takes the arrows
for his quiver from a variety of religious and philosophical tradi-
tions, domestic and foreign. For all this, it is questionable whether
Julian was initiated into the Mithraic mysteries as early as 351/93 or
that it was Mithraism, as opposed to Neoplatonic theurgy, which
provided that “flash of light” that lifted him from his period of
depression at the age of twenty. The most that can be said is that
there is no good reason to doubt that Julian found some elements of
Mithraic doctrine compatible with Neoplatonic theurgy, and in the
typical pattern of his day regarded Mithras as a mode for under-
standing the revelation of Helios. In the biographical portions of the
Oration against the cynic Heraclius, often thought to reflect Julian's
attachment to Mithraism, a tearful prince receives instruction from
Helios that “initiation” does not provide security against the
demands and commitments of the real world.'°4 This very advice
worship of the gods, loyalty to friends, philanthropic love of his subjects, and
impassibility in judgment and service. The extent to which these are commonplaces
having no particular role to play in Mithraism or are importations into the system
based on Julian’s natural tendency to syncretize Mithraic and Neoplatonic thought
would have significant bearing on deciding the nature of his adherence to the cult.
193. Athanassiadi, p. 39, generally supports Bidez’ emphasis on the centrality
of Mithraism to Julian's religious ideology, but with qualifications regarding the
Paris “coup d'état” (1992, pp. x-xv).
194. OrVII.231C, D.
JULIAN THE RESTORER 69

should perhaps be seen as Julian’s manifesto concerning the cult,


whose basic rituals were mirrored in the ladder of ascent Julian had
already come to know from Iamblichan doctrine,!®> especially with
respect to “solar piety”—the belief that the cosmos is shot through
with divinity and thus to be venerated in its own right for the reve-
latory function it serves as mediating between the One and its
hypostases. If Mithraism did appeal to Julian, it would have been in
connection with the corollaries he saw between it and his broader
philosophical commitment to Neoplatonism: the concept of contin-
uous progress through a ladder of ascent; the movement of the soul
from the confinements of passion to purification; the sloughing off
of greed, private ambition, sensual desire (yitov), and sloth.!°° The
political virtues to be cultivated in the system of Plotinus were the
positive corollaries of the Mithraic levels of divestiture—the shed-
ding of the garments of the flesh through the seven levels of the
cosmos. Porphyry himself speaks of the apogenesis of the soul, lib-
erated from the cycle of birth and rebirth and fully “saved,” !°’ and
it would have been no great leap for Julian to have seen Mithraic
ascent as the cultic equivalent of what was a cardinal doctrine of
Neoplatonic theurgy.
For all this, it is doubtful that Julian’s profound silence about
Mithraism throughout an eleven-year period can be interpreted as
secrecy or obedience. Other than conflational tendencies in the
Hymn to King Helios and the oration against the cynic Heraclius,
there is little to suggest deep commitment to the cult at any time
during his life.!°* It must also be admitted that Julian never whole-
heartedly embraced the darker aspects of the mysteries. He is
acknowledged to have been repulsed by the “stygian symbolism” of
his initiation,!°? and his tendency to associate Christianity with
195. See R. Turcan, Mithras Platonicus: Recherches sur I'héllenisation philosophique
de Mithra (Leiden, 1975).
196. Cumont, Les mystéres de Mithra, p. 146.
197. Abst. 4.16.
198. See the useful review of Rowland Smith’s speculation by Thomas
Banchich, “Julian’s Gods: Religion and Philosophy in the Thought and Action of
Julian the Apostate,” Bryn Mawr Classical Review (1997): 238-39.
199. Gregory, Or. 4.55-56; recognizing Gregory's use of invective to describe
the scene, it is worth quoting the following portion: “He had descended into one
70 INTRODUCTION

graves and sepulchers would seem to suggest a general aversion to


the use of darkness and death as religious topoi: His intellectual pref-
erence to “keep separate the things of Helios... and the bright
Olympians” from the things of the night, associated with ignorance
and death, must not only have influenced his view of Christianity*”
but determined his attitude toward Mithraism as well.
Julian’s devotion to the great mother Cybele cannot be doubted.
It can be traced to his time in Bithynia and several trips to Phrygia,
where he encountered “the marvelous woman [called] Aréte.”
According to Gregory Naziazen, Julian’s real “apostasy” dates from
this period of discipleship when “through impious blood he washed
away the waters of his baptism.”*” If the pattern outlined in Julian’s
correspondence and Gregory's diatribe can be given credit, then it
was Aréte, who may have been a pupil of Iamblichus herself, who
explained the meanings of the Phrygian mysteries to Julian and
arranged for his initiation. It would be ten years later, in March 362,
that Julian would visit Phrygia again, this time as emperor, where he
composed the exegetical Hymn to the Mother of the Gods. Word of
his strange activities must have reached Gallus, and may have
reached Constantius. At any rate, Julian, apparently to quiet reports,
became a lector in the Christian church in Nicomedia, “reading
scripture with great feeling and causing everyone to admire his
of those sanctuaries [the Mithraic cave] inaccessible to the multitude, and feared by
all (as would that he had feared the way leading unto hell before proceeding to such
extremities), in company with the man that was as bad as many sanctuaries put
together, the wise in such things, or sophist more rightly to be called; for this is a
kind of divination amongst them to confer with darkness, as it were, and the sub-
terranean demons concerning future events: whether that they delight more in
darkness, because they are darkness, and makers of the darkness of wickedness, or
that they shun the contact of pious persons above ground, because through such
they lose their power. But when, as my fine fellow proceeded in the rites, the
frightful things assailed him, unearthly noises, as they say, and unpleasant odours,
and fiery apparitions, and other fables and nonsense of the sort, being terror-struck
at the novelty (for he was yet a novice in these matters), he flies for help to the
Cross, his old remedy, and makes the sign thereof against his terrors, and makes an
ally of Him whom he persecuted. And what follows is yet more horrible.”
200. This separation is clearest in the language of Ep. LVI, the edict on
funerals.
201. Ep VI.259C, D.
202; ‘Greg OriV.52:
JULIAN THE RESTORER pa:

Christian devotion.”*° On hearing of his behavior, a distressed


Gallus then commissioned the Arian theologian Aetius to go to
Bithynia and question Julian about his true religious feelings. What-
ever the outcome of these interviews, the two men seem to have
struck up a friendship—probably because Aetius was a passable
philosopher, and arguably because his theology, that of the radical
Arians (Anomoeans), taught that the substance of the son is unlike
that of the Father, thus making Jesus decisively inferior to the
supreme God. Julian was similarly forgiving of the theology of the
Photinians.* Philostorgius relates that Aetius was careful to tell
both Julian and Gallus what they wanted to hear.*°° Julian therefore
continued his life of religious dissimulation in Nicomedia?°°
without interruption, developing a cadre of friends and philosoph-
ical dilettantes around him.
If the nature of Julian’s attachment to Mithraism, especially as it
is made available in the Hymn to King Helios, is fraught with ambi-
guity,*” the theology of the Hymn to Cybele is far more clear. There
are similarities in the theology of the hymns: Like Helios, the Magna
Mater is‘ polymorphic: she is Athena, Rhea, Ceres, and Dio. Like
Helios she permeates the cosmos as “the source of the intelligible
gods,”2° while Attis becomes the mediator between the Great
Mother and the material world. In her generative aspect, she “pro-
duces” emanations in the same way that the Sun produces its light
in the form of solar rays—an image which was common not only to
Neoplatonic philosophy and Gnostic thought, but even to the

203. Socrates, HE 3.1.20.


204. CG 262C; on the theology of Photinus, condemned by both the Nicene
and Arian parties, cf. Sozomen, HE 4.6.
205. Philos. 3.27 and Julian, Ep. 15.
206. Lib. 18.19.
207. In her introduction to the 1992 paperback edition ofJulian: An Intellec-
tual Biography, first published as Julian and Hellenism in 1981, Polymia Athanassiadi
confessed that she had “overemphasized Julian’s Mithraism ... and distorted the
balance of Julian’s religious belief, especially in his imperial years by making him
lean heavily on the cult of Mithra” (p. xiv). By the same token she is undoubtedly
correct in her assessment that Julian’s “was a personality fraught with more contra-
dictions, tensions and inconsistencies than the average man” (p. viii).
208. Or. V.166A, B.
he INTRODUCTION

Nicene fathers who appropriated the image to symbolize the proces-


sion of the son from the Eather, o> é« owtdc. Julian’s contribution
to the interpretation of the Cybele myth—the castration of Attis—is
his application of Neoplatonic exegesis to the story of Attis’s deser-
tion of Cybele (as noetic principle or principality) in favor of sen-
suous passion, the love of a nymph. Attis’s treachery represents the
soul's confusion and entrapment in the material world in contrast to
the divine Cybele’s apathetic or passionless love. His castration
interrupts the process of diremption—the potentially endless rup-
ture of the perfection of the divine being—and sets in motion the
process of redemption: the reversion of the soul from error to truth,
from many to the One. Because it was unthinkable to Julian that a
god could die, he rejected the received version of the myth which
centered on the death of Attis. Like Porphyry and Celsus before him,
Julian will make this point again in his diatribe against the Galilean
belief that Jesus, a god, could die and rise again from the dead.”
Julian is at pains to link the origins of the Phrygian cult to
Rome—“the friend of divinity’—a measure of respect notably
absent in his tepid references to Mithras. Cybele’s history was clearly
linked to Rome’s history, whereas for all the links between Helios
and the Mithraic religion, Mithras himself, at least in Julian’s
theogony, remained curiously Persian rather than Roman, astral
rather than cosmic, and literal rather than allegorical. As a modern
commentator has observed, Julian felt that it was not through the
bloody mysteries of Mithras but through the intellectual patronage
of Cybele that “the Empire would be cleansed of the stain of
atheism.”?!° Her worship provided the most immediate paradigm, a
central myth, for the Iamblichan theurgy he sought to cultivate.”
She was the patron deity of his political program to propagate the
teaching of his ancestors and pagan worship throughout the empire.
209. In what is either an odd or a gratuitous denial, Julian claims he had not
read a work by Porphyry on the same myth, almost certainly On the Cave of the
Nymphs (Or. V.161C), and asks that any similarity be charged off to coincidence.
How he might have anticipated the similarities argues strongly in favor of the uno-
riginality of his interpretation.
210. Smith, Julian’s Gods, p. 178. Cf. Or. V.180B.
211. Smith, Julian's Gods, p. 162. Smith catalogs the influence of the Chaldean
Oracles on Julian's exegesis of the Cybele myth.
JULIAN THE RESTORER is

It is widely thought that the reference to “the stain of atheism”


in the Hymn to Cybele”? refers to Christianity, though the context
of the reference is broad: despite the fact that atheism was an epithet
often applied to Christianity by pagan intellectuals, the term had a
far less restricted meaning. Julian’s own understanding of atheism
embraces all forms of impiety, especially the kind of negligence that
had led to the moral dissolution of traditional cults and rituals. If
Christianity is the irritant and the Church the seducer of people
away from pagan tradition, Julian is clear that those who abandon
the gods do so at a risk, and paraphrases Homer to document his
displeasure: ov yap pot Béuicg Eoti KouCéuEv oovvd’ GnonéuTEv
avdpa tov Oc Ke VEotow anéxOntar waKdpecovv.”3 In the letter to
Arsacius, written in July 362 from Antioch, Julian complains that the
lassitude of his subjects has angered the Mother of the Gods, and
that they must become her suppliants to win his favor. Christianity
is always in Julian's purview as that form of “disease” or “ignorance”
that most impels to atheism, but atheism itself in Julian’s view is the
rejection of the “Hellenic faith,” whether through acceptance of
Christianity or failure to return gifts to the gods,7"4

IV. THE GALILEAN PROBLEM


AND THE CONTRA GALILAEOS

The issues that encircle Julian’s hatred of the Galileans have already
been surveyed briefly: To what extent is Julian to be accounted a
“convert” (or apostate) from Christianity? Are his later recollections
or “glimpses” of the Hellenic faith in his childhood and adoles-
cence, in the words of modern reviewer, nothing more than “the
retrojections of a would-be born-again pagan”?”!? Was his apostasy
provoked by “exemplars” such as Eusebius of Nicomedia, Constan-

212. Or. V.180B.


213. Ep. XXII.432A; paraphrasing Odyssey 10.73: “It is not lawful for me to love
or pity men who have made themselves enemies of the immortal gods.”
214. Ep. XXII.430D.
215. T. Banchich, Review of R. Smith, Julian’s Gods (London, 1995), BMCR
(1997): 22.
74 INTRODUCTION

tius, Eusebia, George, and (even) Gallus, on the Christian side, in


contrast to the example of pedagogues and friends such as Mardo-
nius, Maximus, Priscus, Eustathius, and Aréte? Put in that way, the
question leaves little doubt that the Christian religion for Julian rep-
resented his dark night of the soul and was associated by him with
superstition and the rejection of the Hellenic worldview.”!° The
bishops’ unproductive struggling for a narrow orthodoxy based on
texts he found unoriginal and philologically poor stood in sharp
contrast to the beauties of Homer and the wisdom of Plato: “[My
pedagogue Mardonius said to me] ‘Do you long for horse races?
There is one very cleverly described in Homer. Take up the book and
study it! ... Moreover, in Homer there are so many more plants
more pleasant to read about than those we:see; even so near the altar
of Apollo on Delos did I once see a date palm burgeoning. And
there is the wooded isle of Calypso, and the caves of Circe, and the
Garden of Alcinous. Know that you will never see anything more
delightful than these’””!” In contrast to his memories of life as Mar-
donius’s pupil, his allusions—and there are few of them—to a
Christian childhood are brusque and painful: “Let that darkness be
forgotten,” he says inthe hymn to King Helios and in almost iden-
tical words in his Letter to the Alexandrians.”!> “When I groped all
alone in darkness,” he says in the Hymn to the Mother of the Gods,
referring to the days before his initiation into the Metroac mystery.°
Julian’s references to his sense of isolation—loneliness, blindness—
in childhood must refer, on one level, to the family purges being car-
ried out by his cousin when Julian was only six years old. At the

216, There is enough evidence in Julian’s writings cumulatively to suggest that


he was conscious of having been a Christian in his youth, though this formulation
would be slightly different from Athanassiadi’s view that Julian was never “con-
sciously a Christian” (Of Julian and Hellenism, pp. 24-27 and Julian: An Intellectual
Biography, p. xii). Bowersock’s efforts to show that Julian’s attachment to Hellenism
can be explained in the light of an exclusively Christian upbringing are, in my view,
unconvincing; cf. G. W. Bowersock, Hellenism in Late Antiquity (Cambridge, 1990),
pp. 6, 26, and does not reflect an adequate grasp of the religious and theological
context in which Julian’s religious ideas were shaped.
217. Misopogon, 351C-352A.
218. Helios, Or. IV.131A; Ep. XLVII.434D.
2195 Or, VA74e;
JULIAN THE RESTORER DS

same time, these dynastic atrocities were the work of Christians; they
were performed under the nodding gaze of self-serving bishops like
George of Cappadocia, a “poisonous” man whose arrogance out-
stripped even that of the orthodox Athanasius whom he would later
come to despise.*”° Julian seems to have felt that his escape into the
beauties of Hellenic thought were also an escape from the darkness
of his Christian family.
Julian came to believe, not in a flash but as a cumulative matter,
that Christianity was a cancer on the polis, and that he had been
entrusted by the gods to remove it. Before his departure for Persia he
decided to ventilate his animosity for the Christian faith in a long
diatribe against the followers of “the carpenter's son.” The project is
either outlined, or at least alluded to, in a letter written soon after
Julian arrived in Antioch in July 362, addressed to Photinus, bishop
of Sirmium.*”! In it, Photinus is congratulated for his view that “a
god can by no means be brought forth from the womb,” in contrast
to the muscular defense of the divinity of Jesus put forward by the
Nicene theologian Diodorus of Tarsus, who was in Antioch during
Julian’s time in the city, “whose false myth styles [the Nazarene]
eternal.” “If only the gods and goddesses and the Muses and Fortuna
herself will lend me aid, I shall show that [Diodorus] is a weakling,
a perverter of the law and order of the pagan mysteries of the god of
the underworld, and that this upstart Gaililean god of his, the one
he calls ‘eternal’ by employing a lie as a myth, has been stripped of
any divinity Diodorus ascribes to him by virtue of his humiliating
death and burial.” Julian’s plan, then, was to demonstrate to the
“base and ignorant creed-making fishermen” the folly of their ways.
Libanius is even more direct: the purpose of the work, he said, “was
to refute those writings which make a god and a son of God of the
man from Palestine.”??
220. Cf. Ep. XLVIL.
221. Ep. LV, preserved in Latin fragments by Facundus Hermianensis (5462);
cf. Neumann’s partial reconstruction, Contra Christianos, p. 5. According to
Sozomen 4.6, Photinus was tried, deposed, and banished by Constantius for his
heretical views on the divinity of Christ, which seem to have run afoul of both the
Nicene and Arian parties equally.
222. Ep. 55.
223. Libanius, Or. 18.178.
a6 INTRODUCTION

A modern biographer of Julian has suggested that in judging the


merit of Julian’s Contra Galilaeos, “we are in the position of a sailor
who has to describe a boat sunk at some distance from the shore,
with only the mast and a few other pieces of the rigging rising above
the surface.”224 Byzantine Christianity had no interest in archiving
the remains of the Church’s most vicious opponents, and with the
fifth century ban on anti-Christian teaching*”’ the remaining copies
of Julian’s book, along with the polemical writings of Celsus and
Porphyry, were destroyed or began to disappear. Julian’s work is
known to us, like the fragments of his anti-Christian predecessors,
from the refutations of his opponents, mainly the Against Julian of
Cyril of Alexandria (written around 440),??° and a few quotations in
the writings of Theodore of Mopsuestia, Aretas, and Jerome. Cyril's
refutation ran to twenty books, of which ten have survived, together
with “scraps” in the anthologies of later church writers. Cyril's sur-
viving chapters deal almost exclusively with the first of Julian’s three |
books against the Christians, with a very few quotations from book
two being preserved in the anthologies. Not only has nothing sur-
vived of Julian’s third book, but Cyril says that he has passed over
“those passages injurious to Christ and repugnant to Christians.”
Cyril regarded the work of Julian especially threatening to Christian
belief “and greatly to be feared” because “he alone was worthy of
holy baptism and was trained in the holy scriptures.” That Julian
was baptized can be doubted; that he was passably versed in Chris-
tian and Hebrew scripture is beyond question. Indeed, Cyril’s pur-
pose in dredging up Julian’s arguments against the Church nearly
eighty years after the emperor's death was to show that despite his
impressive command of the biblical writings and his rhetorical skill,
“he did not understand what [the texts] meant”: “[The pagans even
in our day] reproach Christianity up and down, arraying Julian’s
writings against us, saying that they are beyond compare and saying
that none of our [Christian teachers] has the skill to refute him.”227
224. Ricciotti, Julian the Apostate, p. 231.
225. Presumably under Theodosius II in 448 CE. The ban was aimed chiefly
at the works of Porphyry.
226. Pro Christiana religione adversus Iulianum Imperatorem (PG 76).
227. Adv. Jul. 76.809C.
JULIAN THE RESTORER By

Cyril's comment is evidence that even in the fifth century Julian was
seen as a champion of the pagan cause, and feared as a potential
threat to weak-minded Christians who might be swayed by the
power of his rhetoric. The same fear accounts for the heavy-handed
way in which Cyril edited out those passages “injurious to Christ
and repugnant to Christians.”
What can be claimed for the remnant is that it represents a
bowdlerized selection of passages from the first of Julian’s three
books, enough to give us the thrust of his argument but not enough
to permit us to form a judgment of the whole. If, as suspected, the
subject matter of books two and three formed a more direct assault
on Christian doctrine than the matter of book one, it is no wonder
that the existence of Cyril’s refutation itself was thought to be dan-
gerous and that the refutation needed to be suppressed.
Despite suggestions that Julian was more attuned to the anti-Chris-
tian polemic of Porphyry than to that of Celsus, the general pattern fol-
lowed by Julian reflects the argumentative approach of the earlier critic.
The thesis is that Christianity has no connection with Judaism, a reli-
gion not td be admired for its practices but recognized for its antiquity.
On the one hand the God of the Hebrews can be apprehended, at
least, as a national god,?*® even though his partiality and jealousy
make him less than the important gods of the Hellenic faith. Julian
elsewhere pays what may be ironic tribute to the “most high God” of
the Jews,?”? or he may simply, in this case, be citing the famous passage
from Genesis 14.19-20—Melchizedek’s blessing of Abraham in the
name of el Elyon. However we are to understand Julian’s vacillating
understanding of the God of the Hebrews, he finds it a crime that the
Christians have abandoned this deity and proclaimed a new revela-
tion. Both the Jews and Christians err in thinking that God reveals
himself in paltry ways—writing, prophecy, and magical tricks—when
he can be known in the depths of the soul and through the order of
nature. With Celsus, he finds it offensive that a god thought to be
supreme would limit knowledge of himself to an insignificant tribe
and race—the Hebrews first, the much inferior Christians thereafter.

228. CG 106E.
229. Ep LI.398A.
78 INTRODUCTION

The direct purpose of the books against the Christians was to


demonstrate that the Christian scriptures are a fable and to prove
that there is no truth in the doctrine “that that fellow from Palestine
is a god or a son of a god.”*° In short, the Christians had fallen into
the error of worshiping a man as a god—a lapse scarcely to be
defended even by reference to their own scripture, where only the
evangelist—John—dared to call Jesus a god, and even he “did not do
so clearly and distinctly.”?7! By contrast, Paul, Matthew, Luke, and
Mark say nothing about the divinity of Jesus, though Paul is “guilty”
of being the first to entice the Greeks away from their ancestral faith.
Julian was aware that the Christian teachers of his day were fond
of arguing their case on the basis of prophecy. Whereas Porphyry
would claim that these prophecies were fulfillments of particular
historical conditions and predictions applicable to their own day
and region, Julian appeared to believe that the prophecies them-
selves were fables. He ascribes the belief that Jesus was a god to the
followers of Jesus and not to Jesus himself; but in so doing, he does
not find Jesus better than “the company of abject and ignorant apos-
tles” who followed him. He finds no basis in scripture for teaching
the divinity of Jesus, since “Moses taught that there is only one God.
The belief that the Word was the first born-son of God is one of the
ideas falsely constructed by you Christians later.”*°?

The main points of the surviving portions of Against the Galileans can
be summarized simply.
First, the God worshiped by the Jews is a tribal God, a fact Julian
thinks is supported by a careful reading of their scripture. “The God of
the Hebrews did not create the whole universe and has no sovereignty
over the whole; his rule is limited and we must think of him as one
god among the many.”*3? Julian supports this point with the familiar
Neoplatonic axiom that the God of All is not the “property” of a par-
ticular race, since to restrict his acts to a chosen people would also be
an admission of his limitations. The human race possesses its knowl-
230. Libanius, Or. 18.178.
231. CG 2138; cf. 327A.
232. G6 290G:
233, CG 100C;
JULIAN THE RESTORER 19

edge of God by nature—by a natural “yearning” for the divine—and


not by teaching.*** In this absurd notion of “particular revelation,”
Julian finds the Christians and Jews equally culpable, since the
Hebrew idea was simply taken over by Paul (“of all magicians and
deceivers the worst”) and imposed on Christianity. Echoing Celsus,225
Julian asks the Christians, in his role as prosecuting attorney, why God
would limit his attention to Judaea rather than extending it to Greece
and Rome: “[Why] to us no prophet, no anointing with oil, no
teacher, no messenger to announce his love for man... . If he is god
of all and creator of all, why did he neglect us?”?3°
Second, Julian hopes to profit from comparison, not simply
from pointing out absurdities in Christian and Jewish teaching. He
does this by contrasting the creation accounts of Genesis and the
Platonic myth recorded in the Timaeus. It is not that the Greek myth
is superior to the Hebrew myth, Julian maintains, but that the Jews
and Christians fail to recognize the myth as an allegory, whereas the
“Hellenes” are not content with the ridiculous picture of God and
the universe that emerges if the accounts are taken literally. The story
of the creation of man and woman in the Garden is especially per-
plexing to Julian, who finds it ridiculous that a god worthy of the
name would create man and woman without the power of moral
discrimination. Like certain Gnostics, Julian holds that the serpent
must be regarded as the benefactor of the human race, since through
him a knowledge of good and evil came into the world;”*’ yet Chris-
tian teachers regard this knowledge as the source of evil. Julian sees
such a god as morally irresponsible. In this section of the work can
be found Julian’s fundamental objection to Christian doctrine: it
exalts obedience and suppresses the exercise of reason, making igno-
rance virtue and mindless submission the only source of happiness.
For Julian the Genesis myth is “blasphemous” on three counts: first,
in teaching that “the Woman who was created to be a helpmeet
would be the cause of the fall,” it makes ridiculous the idea of an all-
knowing god. Second, it is a blasphemy against the Platonic

235 eGaGels: 47:


236. CG 106D.
237, CG. 93D, 94A.
80 INTRODUCTION

teaching that knowledge is the “principle” through which men


become godlike by making knowledge a form of rebellion. Worst of
all, it teaches that God is by nature envious and grudging, in
teaching that the punishment for desiring immortality (the tree of
life) is for the soul to be cut off from immortality and to become
mortal. In every one of these ways Julian saw not only contradiction,
but opposition between the Hebrew and Platonic cosmogony. The
Hebrew God was petulant, fickle, jealous, grudging, and spiteful—
qualities Julian says that we would abhor in a man, “but, in the
Christian account, they are regarded as divine attributes.*?®
In the third place, Julian argues against the completeness of the
Christian myth. The doctrine of the Neoplatonists saw the universe
as a plenum, full of spiritual beings of every grade and variety. The
Christians too, Julian says, believe in a “spirit,” but they cannot make
up their minds whether this being is “ungenerated” or “generated.”
He refers of course to the great theological debates of the post-Nicene
period concerning the procession of the Holy Spirit. But he fails to
find anything in the writings of Moses that would help them out of
their dilemma, since the spirit Moses refers to does little more than
“move upon the face of the waters.” This points to another weakness
in the Christian teaching about creation, since Moses teaches “that
God is the creator of nothing incorporeal,” in other words, is not the
creator of spiritual beings but essentially the fashioner or organizer
of preexisting material substances.**° By contrast, Plato’s description
is full of references to the creation of invisible beings and the noetic
principles that proceed from God—the intermediate grades between
the unseen or intellectual world and the created order.
Fourth, Julian finds the teaching of the Christians incoherent.
This is an old charge, dating from before the time of Celsus, but by
Julian’s day it had become clear that the more of Greek philosophy
the Christian writers imbibed the more inconsistent their theolog-
ical positions became. He finds it ludicrous that a man, born in

238. CG 155C. Julian’s claim that in the Greek myths God “is never shown to
be angry, resentful, wroth, oath-taking, fickle or turned from his purpose” (160D)
is probably meant to be interpreted strictly in relation to the Timaeus.
239: GG 99G;
JULIAN THE RESTORER 81

Palestine to a mortal woman and a carpenter, should be regarded as


the preexisting logos of God, a teaching attested by only one gospel,
and even there in a completely unconvincing way.2?° But even if it
could be maintained that Jesus was a god or the son of a god, “what
about other saviors—those worshiped by the Hellenes, including
Asclepius, Dionysus, and Heracles?” In some cases the incoherence
does not extend to soteriology but simply to etiology: the Christians
seem to take literally the wholly fabulous story of the tower of Babel
as an explanation of the variety of languages in the world;?“' but
they refuse to credit Homer's story of the Alodac (Odyssey 11.316),
who tried to scale heaven by setting three mountains one upon the
other. Why should one myth be preferred to another, when both tell
the same tale??* It is clear, Julian argues, that the Christians have
abandoned the Hellenic faith and adopted certain teachings of the
Jews.**? But here too they are vulnerable to the charge of incoher-
ence, for while they twist and turn the prophecies of the Hebrews to
suit their own ends, they reject some of the “praiseworthy” customs
of the Jews such as the offering of sacrifice—a tradition Julian osten-
sibly wished to restore to the Jews through the rebuilding of the
Temple, while in the process invalidating the Christian prophecy
ascribed to Jesus, “No stone [of the temple] will be left standing on
another” (Mark 13.2). It is a double apostasy of the Christians, as
Julian sees it, first to reject the myths of the Greeks and then to rebel
against the religion of the Jews.
Julian cites a number of passages to point up the inconsistency
of Christian interpretation of Hebrew prophecy. Moses taught that
all humankind should worship one God; the Christians have made
Jesus into another god.*44 The Jews possess the prophecy that “a

240. CG 213B.
2417 CG 134D-
242. The myth of the giants is preserved in the story of the nephilim in Gen.
6.1-4, which has been conjoined to the story of the flood and its aftermath
(6.5-10.32). The story of the tower “resumes” the narrative of human challenges to
God by providing an aetiology for learning to use bricks in place of stone for the
construction of temples.
243.. CG 235D; 207D.
244, CG 253C.
82 INTRODUCTION

young woman shall conceive and bear a child” (Isaiah 7.14), but
there is nothing unusual about this. The Christian translation of the
passage, “A virgin shall conceive and bear a son,” had been used
since Matthew’s day to underscore the extraordinary event of Jesus’
birth. But applying his own brand of literal exegesis to the passage,
Julian argues that it says nothing about a “god” being born of a
virgin or the “firstborn of creation” being born of a young woman.**
So too, the prophecy of Deuteronomy 18.15-16, “The Lord God
shall raise up a prophet like me,” ascribed to Moses”*° and applied
to Jesus does not suggest that the prophet will be divine. Then there
are the references to Jesus’ links to the House of David which ceased
to exist in the time of Zedekiah: how, Julian wonders, can this be rel-
evant to Jesus when “|[their books teach] that he was not born of
Joseph but of the Holy Spirit?”?*” Jesus did not come from Judaea,
Julian reminds his opponents, and the genealogies of the gospels
disagree and do not establish a coherent family pedigree for him.7*®
A final area of concern for Julian is the fact that Christians, who
make so much of obedience to the will of God, disobey the tradition
that brought them into existence. Christians boast of abolishing the
law or being set free.from the requirements of the law. Does this
make them good Greeks or bad Jews??*? He stresses that the law-of
Moses was supposed to last for all time, and cites Deuteronomy 4.2
(also cited by Jesus in Matthew 5.17-20) to show that nothing could
be added or subtracted from the law.?°° Yet the Christians neglected
the practice of circumcision, and defend their negligence by invoking
Paul's view that it was a circumcision of the heart and not the flesh
that was given to Abraham as a covenant. Julian has little use for this
strategy; as he sees it, this is a simple case of Paul playing the char-
latan to win converts from among the Greeks. It is also proof that
Jesus himself was relatively closer to the Jewish position on the ques-

245. CG 262D; in this section Julian also shows some awareness of arguments
over the status of Mary.
246. Acts 7.37.
247. CG 253C, 261C.
248. CG 261C.
249. CG 305D.
250. CG 320A.
JULIAN THE RESTORER 83

tion, since he had said himself, “Whoever relaxes one of these com-
mandments and teaches others to do so shall be called least in the
kingdom of heaven.”?>! A related issue is the Christian intolerance of
sacrifice. Here Julian will argue that the Jews continue to sacrifice in
the privacy of their homes and in symbolic ways, by eating conse-
crated meats, praying, and “giving the shoulder [of the lamb] to the
priests as the first fruits.” Julian is aware of the technical impediment
to the older sacrificial practices, namely the destruction of the
Temple,*°* but the Christians do not sacrifice because they are in
active rebellion against the Jewish law and tradition; they have
“invented a new sacrifice which does not need Jerusalem.”*°? Their
practice has no claim to validity because it is not grounded in antiq-
uity, whereas the Greeks, at least, “have temples, sanctuaries, altars,
purifications and precepts,” making Judaism and Greek religion
closer to each other than either is to Christianity. Julian’s argument
was intended to be painfully clear: religion is defined by adherence
to ancient tradition. The true Hellenic faith demands such adherence,
as does the law of Moses. Consequently, the true Greek will offer sac-
rifice and ‘worship the gods; the Jew will obey the law of Moses. The
Galileans repudiate both and thus have no claim to legitimacy.

While the effect of Julian’s work is difficult to assess, the emperor


himself probably assumed it would have a devastating impact on the
Christians. There is some reason to think, judging from contempo-
rary references, that its immediate influence was on those who sup-
ported Julian's political and religious ideas and programs. Libanius
knew the work, and the heretic Photinus had probably read it. With
the accession of Julian’s Christian successors and the general malaise
that greeted the pagan cause thereafter, Christian intellectuals devel-
oped a greater interest in the treatise and seem to have used the work
as a compendium of accusations against the Church. Theodore of
251. Matt. 5.10.
P52) CG S06B:
253. CG 306B. Julian does not seem to associate the Eucharistic celebration, if
that is the intended reference here, to Jewish Passover ritual.
BA INTRODUCTION

Mopsuestia and Philip Sideta composed refutations, both lost, as did


Cyril, who decided on a response following renewed pagan hostility
against Christians in Alexandria in the early fifth century. It comes as
some surprise that neither Basil of Caesarea nor Julian’s schoolmate,
Gregory Naziazen, mention the work in their invectives against
Julian, nor is it mentioned by western theologians.
It is easy to lose sight of the fact that Julian’s literary attack on
Christianity was only one part of his larger program to restore the
cults and breathe new life into a dying, or rather suffocating, tradi-
tion. His personal example in this regard was energetic and impres-
sive: he is celebrated by Libanius as someone who has never neg-
lected his religious duty, but has taken the gods as allies in the cause:
“Hermes, Pan, Demeter, Ares, Calliope, Apollo, the Zeus of the
Mountain and the Zeus of the City.”7°4 With so many friends and
allies among the gods, Julian scarcely needed the Palestinian car-
penter to complete the side. His idea of restoration was as compre-
hensive as the creed of the fishermen was exclusive.*°° As the Chris-
tian church had become the locus for a shrinking vision of divine
revelation and the divine presence, Julian’s world was fully inhab-
ited: “[It] was full of-ever-present helpers, manifest gods, and they
lived in the cities’ temples as much as in the caverns of the theurgists.
They had been neglected. ...Temples had closed and oracles had
failed; to restore them was his duty to the Empire and the gods.”?°°
For Julian, this was a tragedy in which the Christians were complicit:
their refusal to worship the ancestral gods was clear; their tendency
to cause social unrest—to profit from religious confusion—was also
clear. If the Galileans could not be corrected through reasonable
argument or the acceptance of “true paideia” perhaps they would
succumb to ridicule; thus the strident and prosecutorial tone of his
writing. For all that, it is not clear that the Contra Galilaeos shows
Julian at his best. Like the Christian apologists or the invectives of
his more famous predecessors, Celsus and Porphyry, Julian is deter-
mined to win his case without subtlety. Unlike Porphyry, he has

254. Libanius, Or 12.87, 91; cf. 15.79.


255. See Smith’s excellent summary, Julian's Gods, pp. 215-17.
256. Ibid., pp. 217-18.
JULIAN THE RESTORER 85

nothing at all good to say about Jesus; unlike Celsus, he is unwilling


to grant his opponents even a notional foothold in Judaism. He
seems to have been sensible of the fact that the battleground on
which he fought was a different one and that the methods he used
against the church needed to be rhetorically unsparing. The result is
a hodgepodge of accusations, specious arguments, sarcasm, unar-
gued propositions, adventitious allusion, and special pleading—all
quite normal in the religious debates of the third and fourth century,
and not only in pagan-Christian interchange, but in Christian
attacks on the heresies as well. Tertullian’s contempt for Marcion
and Irenaeus’s hatred of the Gnostics are rhetorical cousins of
Julian’s narrative assault on Christian teaching. What “heresy” was
for the Church's apologists, Christianity was for Julian. To tolerate it
would be the death of faith, the leveling of Olympus, the plunging
of the halls of heavenly light into the stygian gloom.”°”

257. Cf. Eunapius, frag. 28.


ee
hn

See
A NOTE ON TEXT
AND TRANSLATION

he earliest independent reference to the Contra Galilaeos is Liba-


nius‘s comment (Or. 18.178) that in his books “Julian has
made the doctrine of the Christians look ridiculous.” The Council of
Ephesus meeting in 431 does not mention Julian’s work but con-
demns Porphyry’s to be burned, and an order of Theodosius II dated
448 similarly condemns the works of Porphyry without alluding to
Julian. An edict of Justinian in 559 envisages a more comprehensive
destruction of anti-Christian books, but here again only Porphyry is
named. The invective of Gregory Naziazen already discussed? may
have relied on a knowledge of the Contra Galilaeos, but does not name
or quote the treatise. Cyril's refutation, written perhaps between 429
and 441, was provoked by a resurgence of pagan activity in Alexandria
and by a worry that the earlier antagonists of Julian had been, by his
reckoning, heretics themselves.*°? It cannot be known how widely cir-
culated the treatise was, but judging from the measures taken by
Julian’s successor Jovian beginning in 364 to reestablish religious tol-
eration, there is no reason to think that the work was equally well
known in all parts of the empire. It cannot have been as widely
known as Porphyry’s work, and its prominence in Alexandria should
be explained in the context of local political and religious tensions.

258. See note 219 above.


259. He seems to have Theodore of Mopsuestia and Philip Sideta in view.

87
88 A NOTE ON TEXT AND TRANSLATION

The letters of Julian were collected and published before the end
of the fourth century; Eunapius (346-414) knew them and used
them in writing his history.2°° Among pagan collectors, Libanius,
Aristophanes of Corinth, and Zosimus (450-501) had key roles to
play in the preservation of the epistles and in deciding which were
safe for keeping. It is likely that the church historians Socrates,
writing in Constantinople ca. 440, and his contemporary Sozomen,
who quotes from ten letters and mentions fourteen that have not
survived, possessed a “mixed” collection of Julian’s writings. We
would doubtless have a fuller vision of Julian’s view of Christianity
if certain letters had not been suppressed by their owners, as being
too dangerous for circulation in the religious environment of the
fifth century. Others doubtless were destroyed by Christians who
would have regarded them as disrespectful to Christianity. By the
same logic, certain letters (e.g., Ep. XLI) would have been preserved
as showing Julian in a bad light (Ep. LXXXI, a forgery) or as an imi-
tator of Christian practice (e.g., Ep. XX). Because some letters were
mutilated, others expanded, others preserved in handbooks as
models of the epistolary style, the question of authenticity has been
a constant feature in Julian scholarship.
The modern collection of Julian’s letters begins with the 1696
Leipzig edition of Spanheim (Latin translation) which also included
the Contra Iulianum of Cyril. The canon of letters expanded from an
original inventory of 48 in 1499 (the Venetian corpus) to 63 by
1630 (the Paris corpus of Martinius, Rigalt, and Petau). The more
recent editions of Julian’s complete works are those of E C. Hertlein
(Leipzig, 1875-1876) and J. Bidez, Rochefort, and Lacombrade
(Paris, 1924-64). The partial reconstruction of the Contra Galilaeos
by K. J. Neumann (Contra Christianos, Leipzig, 1880)?* which sup-
260. See R. Blockley in Fragmentary Classicizing Historians of the Later Roman
Empire I and II (Liverpool, 1981-83).
261. Juliani imp. Librorum contra christianos quae supersunt. Neumann's textual
commentary is included in the Latin prolegomena. Amendments are provided by
Gollwitzer, Observationes criticae (Erlangen, 1886) and by Asmus, Julians Galiléer-
schrift (Freiburg, 1904). See also the occasional corrections provided in W. C.
Wright, The Works of the Emperor Julian, 3 vols. (London, 1913-23), and E. Masarac-
chia, Giuliano imperatore Contra Galilaeos (Rome, 1991). Neumann’s choice of title is
inexplicable, considering Julian’s consistent use of the more disparaging term.
A NOTE ON TEXT AND TRANSLATION 89

ports the present translation relied on Spanheim’s pagination. Tex-


tual numeration in this translation of the Contra Galilaeos refers to
that edition. References to the letters and orations of Julian appear
according to Hertlein’s standard pagination in his edition of the
complete works. In cases where the Greek is uncertain or where a
hypothetical reconstruction is warranted to complete the sense of a
statement, insertion brackets < > have been used.
In the critical notes I have followed the standard practice, unless
otherwise dictated by context, of referring to the works of Julian by
citing only the source (thus, Or.VII.161 or Ep. XXII.356B). Refer-
ences follow the Loeb edition numeration, for ease of reference,
with cross reference, in the case of the letters, to Hertlein. The edi-
tion of the letters edited by Bidez and Cumont (Epistulae, leges, poe-
matia, fragmenta varia, Paris, 1922) is the standard for questions of
textual accuracy.

R. Joseph Hoffmann
Bae

i
JULIAN’S AGAINST
THE GALILEANS

39A-42E:?°

he time has come for me to say for the benefit of all how I dis-
covered beyond any doubt that the stories of the Galileans*
are the inventions?® of deceivers and tricksters. For these men
seduce people into thinking that <their> gruesome story is the truth
by appealing to the part of the soul that loves what is simple and
childish.?
I propose therefore to deal with what they consider their pri-
mary teachings. And I should say at the start that if my readers wish
to refute me, the way to do so is to proceed as though this were a
case at law. Moreover, as it is their views that are on trial, their coun-
tercharges (and whatever other trivialities they may wish to raise)
must wait until they have defended themselves. It will be best all
262. Marginal numbers in the English text refer indicatively to the pagination
of Spanheim’s 1696 edition of Cyril of Alexandria’s polemic against Julian, Pro
Christiana religione, as rearranged by Neumann in his 1880 reconstruction of
Julian's treatise.
263. TaAdtaaiot: Julian’s, like Epictetus’s, designation for the Christians is
designed to stress the insignificance of the founder. See Gregory Naziazen, Or. IV.
76 (115).
264. Here oxevapia, the equivalent of katackevaoua, implying malicious
design.
265. See Paul, 1 Cor. 13.10-12.

91
O@ JULIAN’S AGAINST THE GALILEANS

round if they hold their fire until they have answered for themselves:
So, for the present, let us leave these countercharges”®® to one side.
43A-52C
The best way forward is this: First, to consider briefly how we
came to have a conception of God. Then, to compare what Greeks
and Hebrews have to say about the divine being; and finally, to ask
those “who are neither Jews nor Greek,”*°’ namely the Galileans,
why they chose the belief of the Jews above our own, only to
abandon Jewish belief to follow that strange path of theirs.
<These Galileans> have accepted not a single admirable or
important belief from those that we Greeks hold; <nor any> from
those imparted by Moses to the Hebrews.?°° They have instead taken
on the mold that has grown up around these nations like powers of
evil—denial of the gods?® from Jewish recklessness;*”’ <and from
us> laziness and superstition as a consequence of our way of life.?7
This, they say, should be considered the most excellent way of
revering the gods.?”2

266. While it is not clear what Julian may mean by the term avtikatnyopetv,
he alludes to the general pattern of Christian polemic against pagan practices.
267. Ovte “EAAnvos ovte Tovdatoc. 1 Cor. 24. In the Ad Nationes Tertullian
rehearses a number of derogatory names for the Christians originating in pagan cir-
cles, including the designation “third race”: “If you attach any meaning to these
names, tell us what are the first and second races so that we can know something
of this ‘third’” (Ad Nat. 1.8). Julian’s polemic harkens back to the charge common
in Diocletian’s time: that Christians were officially stateless persons and thus lacked
an €@vos.
268. On the ambiguity of the origins of Christian beliefs, see Celsus, C.Cels. V.33.
269. a8éortn¢: see Tertullian, Apology 10.
270. Julian’s complaint is repeated in the rescriptus on Christian teachers,
written from Antioch in 362. On the tradition that Julian assisted the Jews materi-
ally in the rebuilding of the Temple in recognition of Yahweh's stature as a tutleary
God: Amm., 23.1.2.
271. In addition to Julian’s own defense of his asceticism in the Misopogon,
Libanius in the Epitaphios Iuliano comments on austerity measures introduced in the
imperial court, which included the expulsion of barbers, cupbearers, cooks, and
jugglers. So, too, Amm., 22.4.2-5; 23. Even Christian historians such as Socrates
(HE III.53) were approving of Julian’s correction of “dissolute morals” while
deploring his policy toward the Christians. See further the discussion in P. Athanas-
siadi, Julian: An Intellectual Biography (London, 1992), p. 97 and note 40.
272. Se0c€eBera.
JULIAN’S AGAINST THE GALILEANS 93

It is not by teaching?” but by nature that humanity possesses its


knowledge of God, as can be shown by the common yearning for
the divine that exists in everyone, everywhere—individuals, commu-
nities, and nations.*’* Without having it taught to us, all of us have
come to believe in some sort of divinity, even though it is difficult
for all to know what divinity truly is and far from easy for those who
do know to explain it to the rest.
Besides this notion of divinity which all people seem to have in
common, there is another: all of us are naturally dependent on the
heavens and the gods that appear there, so much so that even if
someone imagines a god beyond these he invariably assigns him a
place in the heavens. In so doing he is not separating him from the
earth but elevating him to “kingship of the All’?” in the heavens as
being the position of honor; and he thinks of him as seeing from
<heaven> all that happens in the world.
What is the use of calling Hebrews and Greeks as witnesses to
this? Is there anyone who does not stretch his hand toward the
heavens when he prays? Whether he professes one or many gods, if
he has any idea of the divine, he looks to heaven. This is natural, this
feeling that men have, because in the heavenly sphere they see sta-
bility and order, unchanged and undiminished. <They see> harmo-
nious movement unaffected by discord—the regularity of the illumi-
nations of the moon or of the rising and setting of the sun, com-
pletely regular and in accordance with definite seasons of the
years.?”° And so naturally they considered that heaven was a god,
indeed, the throne of a god.
A being of the sort they conceive is <likewise> free from change,
decay, alteration; not subject to increase by having something added

273. &6aKxtdc: He does not mean to imply Christian revelation but the inter-
pretations of Christian teachers.
274. Julian’s acceptance of this trademark Neoplatonic idea is attested by Fir-
micus Maternus, De errore VII-VIII. Julian argues the essential unity of Hellenistic
religion on the basis of the dispersal of the divine logoi and innate apprehension
of the divine. In Ep. XLVII.434C, D, Julian argues that the true logos—the living,
animate, and beneficent image of the intelligible father—is Helios. Cf. Or. IV.133C.
275. Cf. Plotinus, VIII Ennead 17. A faint echo can be found in Eph. 4.5-8.
276. II Ennead 3.5.
94 JULIAN’S AGAINST THE GALILEANS

to it, or to decrease by having something taken away; beyond every


impurity, every possibility of stain.?”” .
This all-moving being is eternal, as we sommpeehena it: traveling
in circuit around the majestic creator, perhaps motivated by a sub-
lime and rarefied soul existing within <it>, as our bodies are ani-
mated by the soul existing within us; or perhaps receiving its motion
from God himself as it moves in unstopping spiral through its
unending course.?’5
44A: Of course, the Greeks concocted their stories about the
gods, those incredible and terrible fables. They held, for instance, that
Kronos swallowed his children and then vomited them out.*”? They
told tales of illicit couplings, Zeus bedding his own mother, having a
child by her, marrying—rather, not marrying her, but lying with her
and then handing her over to someone else. There is a story about
how Dionysus was torn in two and his body reassembled again.?®°
And this is the sort of thing we get from the myths of the Greeks.
75A: Compare with this the Jewish teaching:
A garden was planted by God, and Adam was made, and after-
ward, for the sake of Adam, woman was created. God said, It is not
good for a man to be alone; let us then make a helper resembling
him.?*' And yet, she was no help at all. She deceived him and caused
the fall from a life of pleasure in the garden.?°?
75B: This is complete fable. Is it likely that God would not know
that a creature designed to be a helper would be a curse rather than
a blessing to the one who accepted her? And then—what language

277. V Ennead 8.11.


278. Plotinus had maintained that the existence of souls in diversity was
educed from the primal unity of the soul: IV Ennead 9.3-5.
279. Hesiod Theogony, 453ff.; Apollodorus, .4ff.
280. Dionysus, in a poem by Nonnos, is identified with Osiris. In the ancient
tradition known to Euripides (Alcestis), it is Pentheus who is torn to pieces for
spying on the votaries of the god.
281. Gen. 2.18ff.
282. Because Julian wishes to stress a resemblance between Greek myth and
Hebrew fable, he sees the story of the fall as a repetition of the Pandora-Prometheus
story (Hesiod Theog. 535-70). He sees Eve (“all mother”) as equivalent in deceit to
Pandora (“all-gifts”), though he stops short of saying that Eve was sent deliberately
by the gods to foil the progress of the human race.
JULIAN’S AGAINST THE GALILEANS 95

would the serpent have used when speaking to Eve? 86A: Was it the
language spoken by human beings?
Furthermore, how do fables of this sort differ from the myths
made up by the Greeks? Is it not extremely odd, for example, that
God refused to the beings he created the power to tell the difference
between good and evil? Can anyone imagine anything more absurd
than such a being, one unable to tell what is acceptable from what
is wicked? It should have been evident that <left to himself> man
would not avoid the latter, |mean evil things, nor pursue the former,
I mean good things. The heart of the matter is that God refused to
let man taste of wisdom,?*° even though nothing could be more
important to mankind. .
89B Even the fool recognizes that wisdom includes the power to
tell the difference between the good and what is less good, and so
considered it emerges that the serpent <was really acting as> bene-
factor of the human race.?** Moreover: their God is to be called
malignant,?®° for when he perceived that man had obtained a bit of
the wisdom he once had been denied, and to prevent him tasting
the tree of life, God casts him out of the garden saying something
like, “Aha! Adam has become just like us, knowing good from bad;
forbid him to stretch out his hand and take from the tree of life, eat
from it, and thus live forever.”?°°
94A: Hence, if it is not the case that every one of these tales pos-
sesses some secret meaning, as I think must be the case, then they
are full of blasphemy against God. First, in suggesting <God's> igno-
rance, that the one made to be a helper would be the cause of
calamity; second, to withhold from mankind the very knowledge
which might have made his mind complete; then to be jealous lest
man eat from the tree of life and turn from mortal to immortal: this
would be petulant envy of the worst kind.
96C:

283. Gen. 2.16-17.


284. This platonizing interpretation of the serpent's activity is also a feature of
certain gnostic exegesis; cf., for example, the Testimony of Truth, IX.3147: 14-30;
48.4-13; 50.3-5.
285 BaoKatoc or jealous.
28GeGenns.22.
96 JULIAN’S AGAINST THE GALILEANS

I turn now to consider the true opinions of the Jews and the
ideas handed on to us <Greeks> from the beginning by our fathers.
<Whereas> our account?8’ specifies an immediate creator of this
universe, Moses says nothing at all about gods who may be superior
to this creator. 96D Nor has he anything to say about the nature of
angels. Even though it is often said, in various ways, that they serve
God, it is not clear from what he says whether <angels> are begotten
or unbegotten, or begotten?®* through one god and made to serve
another, or something still different. <Moses> does tell us clearly
how heaven and earth and their contents were ordered: he says, for
instance, that God commanded things such as light and the firma-
ment to be; and again, he says God fashioned certain things—the
heavens, the earth, sun and moon—and that things which already
existed but had been buried such as dry land was then separated
from <encircling> water.
Beyond this <Moses> says nothing about the begetting or
making of a spirit, merely that “the spirit of God moved above the
face of the waters,” but nothing at all about the nature of this spirit,
whether generated or ungenerated.
49A Compare if you will the statements of Plato,7°° namely what
he says about the creator at the time of creation of the world,?°° and
see how it fits the picture of creation painted by Moses. In this way
we can judge whose conception is nobler: 49B Plato, who did jus-
tice to the eternal images, or Moses, of whom it is written that God
spoke to him face to face.?”!
“In the beginning God created the heavens and the earth. And
the earth was unseen and without a form; and over the face of the
abyss, darkness was. And over the face of the abyss God's spirit
287. There is an amount missing from Cyril's discussion at this point, no
doubt a reference to the Platonic cosmogony in the Timaeus.
288. eite 5€ yeyovotes cite &yévnto. Julian has in view the silence of Hebrew
scriptures on the question of whether the spirit of God is begotten or unbegotten.
His general view was that Christian theology was lost in unproductive theological
controversies which were insupportable in scripture.
289. See Julian, Ep. XIX and Or. IV. The doctrine quoted is a paraphrase of
Timaeus 41, A, B.
290. KOOHOYEVEta.
291.. Num. 12:8; Exod: 33.11.
JULIAN’S AGAINST THE GALILEANS a

moved. And God said, let there be light, and there was light. And
God saw the light was good; and God divided the light from dark-
ness. And God called the light Day and darkness he called Night.
And the Evening and morning were the first day.
“And God said, let there be a vault in the middle of the waters.
49C<> And God called the vault heaven. And God said, let the
waters under heaven be gathered into one place, and let dry land
appear; and it happened. And God said let the earth bring grass for
feed and fruit tress bearing fruit. And God said Let there be lights in
the vault of heaven to illuminate the earth. So God set them in the
vault of heaven to govern the day and the night.”?9?
In no place does Moses say that the abyss was created by God,
nor the darkness, nor the waters. 49E And yet it would seem that
after ordering the light to be, and after it was, <Moses> ought to say
something of the night and also of the abyss or the waters. But even
though he often refers to them, <Moses> says nothing to avoid the
impression that they already existed. Nor does he describe the birth
or fashioning of the angels or how they were brought into being: he
treats only the created things in heaven and on earth. As Moses tells
the tale, God is creator of nothing without a body; he merely organ-
izes and shapes the stuff that already exists—since the words “And
the earth was unseen and without form”??? must mean that <God>
thought of wet and dry stuff as original matter, and <this means>
that God is simply the shaper of this matter.
57B-C But Plato must be heard concerning the universe:
“Now the whole of heaven or the universe—or whatever other
name it may be appropriate to call it (for so should we name it)—
did it exist eternally? did it have no beginning? <Did it have> no
point of origin?—or does it indeed have a beginning? It had a begin-
ning. It is seen, manipulated, has a body, and like all objects known
to the senses and made perceptible to us through sensation, being
those which had a beginning <the universe too> has come into

292. Gen. 1.1-17; abbreviated.


293. adpatog Kai dxatacKkevatoc. Julian’s point is that the Hebrew myth
incorporates an older mythology, in which preéxistent matter is shaped by the
demiurge. God is the molder of material things that come into being through him.
98 JULIAN’S AGAINST THE GALILEANS

being, and thus as we can see, had a beginning.’* Logic dictates that
this universe came into being, just as a living creature does: ensouled
and truly full of intelligence through the foreknowledge of God."
57D-E We only need to compare <these accounts> point for
point: What kind of speech does the god of Moses make? And what
does the god of Plato say?
58A: “And God said, let us make man in our image and likeness.
And let <them> have governance over the fish in the sea and the birds in
the air, over cattle, and everything on earth and every thing that crawls
upon the earth. So God created man. In the image of God he created
him. Male and female he created them. And he said, Be fertile, multiply,
and stock the earth and subdue it. 58B Take command over the fish of
the sea, the birds in the air, over the cattle, and over the earth.*?°
Listen now to the words which Plato attributes to the demiurge:
“Gods of gods—What things of which I am the fashioner and
father will endure unchanged and indestructible forever, so long as it
is my will. Observe, everything that has been established can be ©
uprooted. But to will to uproot what is soundly planted and harmo-
nious: that is the work of an evil being. And so, since you have been
brought into being, you are not altogether indissoluble nor immortal.
58C But you have found through my will something stronger to estab-
lish you than when you first came to be; accordingly, you will not, by
any means, be uprooted or undergo death. Listen now to the word
that I say to you: Three sorts of mortal beings are yet unborn: and the
heaven is incomplete unless they are given birth, as <heaven> would
not <then> have within itself all kinds of living things. If these, how-
ever, come into being to receive a share of the life that is in my power
to bestow, they should become equal to gods. And so, that they may
be mortal, and that the All may truly be All, busy yourself with the
fashioning of living things according to your nature. Imitate the power
that I demonstrated in fashioning you. 58D And the part of them
suited to receive the same name as the immortal beings—that which
is called divine and beckons to those who follow you in the ways of
justice—this power which I have sown and begun, I will deliver to

294. Plato, Timaeus 30A.


295. Gen. 1.26-28 (LXX).
JULIAN’S AGAINST THE GALILEANS 99

you. As for the others, in fashioning the blend of mortal and im-
mortal, create living beings and give them birth, support and increase,
receiving them back again when they perish.”2%°
65A-B But before considering whether this <narrative> is purely
a fable, learn something of what it means.
Plato calls those things which are visible <by the name of> gods:
sun, moon, the stars, and the heavens—but <he regards them>
merely as images of the invisible gods. The sun which we see with
our eyes is a likeness of the intellectual principle, the invisible sun;
and so the moon we see with our eyes, and the stars: these are like-
nesses of the intelligible. Clearly Plato knows of intelligible and
unseen gods who are immanent within and exist alongside the cre-
ator, and proceeded or originated from the creator himself.
Accordingly the creator in Plato’s narrative speaks of “gods” when
he means invisible beings, and when he means the visible gods he
speaks, apparently, <about that which comes> “of-gods.” And the cre-
ator of both is the one who made heavens and earth and the sea and
stars and begot archetypes of these within the intelligible world.
Notiee the excellent saying which follows this:
“There are <yet> three kinds of mortal things,” namely, human
beings, animals, and plants, and to each of these he gives its own
particular definition. 65D “Now if each one of these,” he continues,”
should come to exist through me, it would be, through necessity,
immortal.” And it is certain that in the case of the intelligible gods
and the visible world, the cause of their immortality is solely to be
sought in the fact that they were brought into being by the creator.
So, when he says, “The part of these which is immortal must be
given to them by the creator,” he refers to the reasoning soul. 65E
“For the rest, blend immortal with mortal.” It is clear, therefore, that
the creating gods 2°” got from their father the creative power <which
gives us> all living, mortal things on the earth. For if there were no
difference between heavens, man, and animals, by God, right down
to the little colonies of crawling things and fish that swim in the sea,
then there would surely have had to be one single creator for all of

296. Plato, Timaeus 41A-C.


297. oi Snwovpyikoi Geot.
100 JULIAN’S AGAINST THE GALILEANS

them. But <as it is>, there is a great divide between immortals and
mortals. And this cannot become either greater through adding to it
nor less by taking something away; nor can it be combined with
what is mortal and hence subject to fate; 66A and so it follows that
the group of gods who were responsible for mortals is different from
the group of gods who were the cause of immortal beings.
Now, as Moses seems to have failed to give a thorough rendering
of the creator of the universe, let us compare further the opinion of
the Hebrews and the views of our <true> fathers concerning the two
nations. 99E Moses says that the shaper of the universe selected the
Hebrew people, indeed <says> that only to this people did he show
care and concern, and he takes responsibility only for it. Where other
nations are concerned, and by what sort of gods they are ruled, he
says nothing—that is, unless we take the sun and moon?’ as their
allocation. But more of that a bit later: 1OOA Here I want only to sug-
gest that Moses himself and the prophets after him, and the Nazarene
Jesus, and that magician and deceiver, surpassing all others of every
place and time, Paul <all these> say that he is the God of Israel and
Judaea alone and that the Jews are his chosen people.
Listen to their very words, first of all to those of Moses:
You will say to Pharaoh: Israel is my son, my firstborn. And I
have said to you: Let my people go in order that they might serve me.
But you refused to let them go.*??
100B And further on:
And they <said> to him: The God of the Hebrews has sum-
moned us. Therefore we will go three days’ journey into the desert
to sacrifice to the Lord our God. 3°°
He soon speaks again in a similar vein:
The Lord God of the Hebrews sent me to you to say, Let my
people go, in order to serve me in the wilderness.32™
298. Deut. 4.19.
299 NeExOd AD 2:
300. Exod. 4.23.
301. Exod. 5.3. Missing from Cyril’s text are the words of the prophets and
Jesus promised by Julian. While Paul's religious exclusivism is undermined by spe-
cific passages (e.g., Rom, 11.1-11), Julian may have been less taken with the struc-
ture of Paul's argument than with specific passages showing Paul's debts to Judaism
as in Rom. 11.2. Similarly, Jesus’ words in Matt. 6.31 or 8.4.
JULIAN’S AGAINST THE GALILEANS 101

106A, B Moreover, that God from the start cared only for the Jews
and chose them for his own has been clearly stated not only by
Moses and by Jesus but also by Paul, albeit it is an odd thing coming
from him. He changes his ideas about God according to his situa-
tion—just as a polypus changes its colors to match the rocks.3°2 So
that here we find him saying the Jews alone are God’s, and here in an
effort to get the Greeks to join sides with him, saying “Don’t think he
is the God of the Jews alone, but the God of the gentiles also.”3%
106C Thus we are justified in asking Paul why God, if <he is>the
God of the gentiles and not only of the Jews, <would send> the
sacred gift of prophecy to the Jews in a bounty, as well as Moses and
the chrism [consecrated oil] and the prophets and the Law—along
with their puzzling and banal myths.
For you hear them crying, “Man eats the food of angels.”3%
At last God sends Jesus to them as well: but to us <Greeks> he
sends no prophet, no chrism, no teacher, nor any messenger to
announce 106D <God’s> philanthropy which, though it come late,
would someday extend even to us. Even so, he stood by for innu-
merable, or if you'd rather, thousands of years, <during which time>
mankind in its ignorance served idols, to use your expression; and
from where the sun rises to where he sets, and north to south in its
compass <God saw> only the tiny tribe which had settled some two
thousands years previously in a corner of Palestine.
Yet, if he is the God of all alike, the shaper of everything, why did
he overlook us? 100C Is it not preferable to think that the God of the
302. Theognis; cf. Misopogon 349D.
303. Rom. 3.29 and Gal. 3.28.
304. Psalm 78.25: [aKove1g yap avt@v Powvtwv], dptov ayyéAwv eEpayev
&vOpenoc: Literally in Hebrew, “bread of the mighty ones.” Julian perhaps has in
view the whole myth envisaged in Psalm 78, whereby God takes pity on the tribes
in the desert by opening the doors of heaven (a phrase found only in 78.23) and
providing manna (cf. Exod. 16.1-36; Num. 11.1-35). It is called “bread of angels"
only here in the Old Testament. Early Christian exegetes doubted on the basis of
Mark 12.25 whether there was eating and drinking among the angels; cf. G. F.
Moore, Judaism (Cambridge, MA, 1927), 1:405. The Christian fathers referred to the
eucharist as the angelic bread or bread of angels in conjunction with John 6.50, 58;
see John Chrysostom, Commentarius in sanctum Ioannem apostolum et evangelistam,
Hom. 46. It is not clear whether Julian means to imply that the Christians echo the
Jews with this cry, or refers simply to Jewish claims on behalf of their status.
102 JULIAN’S AGAINST THE GALILEANS

Hebrews is not maker of the whole cosmos with power over it all,
but only, as I have suggested, a god of the limits, whose dominion
is bounded on all sides 106D-E—a god who should be thought of
as one of a club of other gods?2% Or are we to pay special attention
to you because you or one of your kind thus pictured the God of the
universe, even though your picture is only the vaguest hint of him?
Is this not pure partiality? God is a jealous God,>°° you say—but <of
what has he> to be jealous, so much that he settles the score for the
sins of the fathers with their children??”
115D But regard now what we teach in opposition to your doc-
trine. Our authorities maintain that the fashioner <of the universe>
is both the common father and lord of all that exists, while the gods
of nations and the gods who protect cities have been delegated spe-
cific responsibilities by him. Each has been given a role to play in
strict accordance with his character. 115E. So it is that in the father
all things are complete and the all is unified, while in the distribu-
tion of gods one trait or another <tends to be> dominant: so Ares
rules contentious nations; Athena those who are wise as much as
warlike; Hermes those that are more cunning than daring; and, to be
brief, each nation ruled by a god exhibits the character of its own
god. If common experience is not adequate witness to the truth of
our teachings, we will allow that ours are the fictitious traditions and
the wrong doctrines and, if so, 116A that we really ought to be per-
suaded and embrace the teachings you hold to be true. But if the evi-
dence is on our side, and if history teaches through experience that
our account is the true one, and does not support your accounts,
why continue this endless game of make-believe?
Here, tell me why it is that the Celts and Germans are violent
while Greeks and Romans for the most part are humane <in their
305. GAAoIg voeioBa Sevic: be conceived as one among other gods.
306. OEdc¢ Cnrwre.
307. Exod. 20.5. The principle of vengeance is particularly alien to Julian's pla-
tonic conception of the divine artificer and is strictly speaking “incomprehensible”
within his system: see Plato, Timaeus 27, 28, and esp. 29b: The creator makes the
world of generation because he is good, “and the good can never have jealousy.”
The implicit point of Julian’s play on this theme is that jealousy requires a rival and
the use of the term in reference to God is tantamount to an admission that the
Hebrew God shares power with other gods.
JULIAN’S AGAINST THE GALILEANS 103

laws> and political life, even if courageous in waging war. Or <why


is it that> the Egyptians are clever at crafts while the Syrians are
womanish, hot-tempered, quick to learn, and quick to flee from
danger? 116B If you maintain that there is no good reason for these
differences among peoples, or that <though the differences exist>
they occur by dint of chance, how then will you maintain that the
universe is governed by a foreknowing power?2%%
If, however, anyone should say that reasons exist for these differ-
ences, then let that person explain them to me and teach me in the
name of the creator himself. 131B As for the laws of men, these obvi-
ously have been created by men in conformity with their natural
inclinations.*” 131C I mean to say that civil and moral laws were
established by those in whom civilized instincts were dominant
above other instincts, <while> those of a different inclination fash-
ioned vicious and inhuman laws. And the shapers of laws have
scarcely been successful in shaping the characters of men or
improving by law what is <inherent in> their essential disposition.
So it was that the Scythians refused to accept the religious rantings
of the possessed Anacharsis.*"° Nor will you find that Gauls and
Iberians have any knack for geometry or philosophy or pursuits of
<an intellectual> kind, fields in which the Roman hegemony has
long been established. 131D Likewise, those whose gifts are for dis-
cussion and argument refrain from studying anything else: so deci-
sive is nature’s power over us, it appears.
And where do these differences of character and of laws originate?*"
134D Moses for his part gives a preposterous explanation for differ-

308. Tpovoia. Julian repeats the conventional wisdom concerning the char-
acter of nations; cf. Herod. 4.76; Lucian, Anacharsis, and Julian, Misopogon 359B.
309. ovolc OiKEIvoUG.
310. Julian refers to a story told by Herodotus (4.48) designed to show antag-
onism between cultures, especially Scythian dislike of Greek ways. Anacharsis
encounters the cult of Rhea-Cybele at Cyzicus, makes a vow to the Mother of Gods,
and attempts to introduce the cult into Scythia on his safe return there. He is shot
with an arrow and killed by King Saulinus when the king discovers the illicit night
procession and the raising of sacred images. Julian tells the story to reinforce his
point concerning cultural diversity, and perhaps also to suggest the novelty of Chris-
tian worship within the context of Greek and Roman religious observances.
311. See Herodotus, 4.76.
104 JULIAN’S AGAINST THE GALILEANS

ences in <customs and> languages. For he explained that the sons of


men convened with a plan to build a great city with a tower at its center,
but that God decided he would <foil their plan by> confusing their lan-
guages. So that no one can accuse me of fabricating this to slander him,
let me cite specifically the following from the book of Moses:
“And they said, let us go and build a city with a tower with its
pinnacle touching high heaven. And let us make a reputation for
ourselves before we are scattered over the face of the earth. And the
Lord came down to see the city and the tower which the sons of men
had built. 135E And the Lord said, See: this people is one and they
all have a single language; and this they have done, such that now
nothing that they propose to do will be impossible for them. Let us
then go down and in that place confuse their language so that no
man should be able to understand his neighbor's speech. So the
Lord God scattered them over the face of the earth and they ceased
building the city and the tower.”>!?
And you <Galileans> insist that we ought to believe this account
while you refuse to believe Homer's story of the Aloadae, where <it
is recounted> that they planned to build three mountains, one on
top of the other “so that the heavens might be scaled.”*!> From my
point of view, one tale is like the other. But say, if you credit the
<story of Moses>, why in the name of the gods do you reject
Homer's myth? I guess that to people as ignorant as you it goes
without saying that even if men the world over had but one lan-
guage they still would be unable to build a tower to heaven, even if
they turned the whole surface of the earth to bricks! 134C Because
a tower of that sort would need innumerable bricks, each brick as
large as the earth itself, even if they are to succeed reaching <merely>
the orbit of the moon. Imagine, for a moment, that everyone came
together speaking one language, one tongue, and that these <very>

312. Gen. 11.4-8.


313. Odyssey 11.316. This section of the epic includes the story of illicit unions
between the Olympians Zeus and Poseidon and the mortal women Leda and
Iphimedeia, the latter of whom bore the giants Ephilates and Otus who threaten to
make war on the gods by stacking mountains to reach the heavens. They are killed by
Apollo. Julian attempts a connection between the Greek myth and the legend of the
nephilim in Gen. 6: 1-4 and the legend of the beginnings of technology in Gen. 11:1-9.
JULIAN’S AGAINST THE GALILEANS 105

people decided to dig stones and turn earth to brick; and even
though they stretch it finer than thread, when would <such a struc-
ture> reach up into high heaven?
Or do those of you who hold this fiction to be the truth believe
also that God trembled before the threatened violence of mankind?
Is it for this reason he had to ‘come down’ to confuse their lan-
guages? 135D And I ask: do you still dare to boast of your knowl-
edge of God??!4
137E I shall return to the question of God’s confusing the lan-
guages. Moses explained the reason as follows—namely, God was
afraid of <men> having one language or a common way of thinking,
lest they find some way to build a 138A road leading up to heaven,
and there do something perverse against <God>. Precisely how God
caused this to happen is not recounted: <we are told only> that God
came down from heaven—because apparently the deed could not be
done while on high, without <God> coming down to earth. But nei-
ther Moses nor anyone else*!> has given insight into the origin of dif-
ferences between characters and customs—this despite the fact that
political constitutions among nations differ, as do their traditions,
in ways far greater than than differences among languages.
138B What Greek would tell us that a man should marry his sister
or his daughter, for example? But this is thought to be a virtuous <prac-
tice> among the Persians. Why, however, go through the list of charac-
teristics—or recall the German love of freedom and lack of self-restraint;
the docility and the mildness of the Syrians, Persians, Parthians—and,
to be brief, all barbarians east and south, and all peoples who are happy
with the rule of despots of some sort. If, then, these greater and more
important differences <of national character> emerged in the absence of
a greater and more godlike providence,*® why should we worry about—
and why worship—<a god> who cares nothing for us?
138C Indeed, is it right that the one who cares nothing for the
way we live, our customs, lives, good government, political structure,
.
waoi Ged?
314
315. ovte GAdoc: That is, no biblical writer or Christian teacher.
316. mpdvova: Julian seems to mean that important differences can be assigned
only to a foreknown purpose: the God of the Jews behaves reactively and out of
0dvoc [jealousy], hence without foreknowledge of consequences.
106 JULIAN’S AGAINST THE GALILEANS

should demand veneration from our hands? It is not. See to what


absurdity your teaching leads! For among the blessings that we see
in the life of a man those which pertain to the soul are primary, and
those which concern the body are secondary.
Thus <I conclude>: if <the God of the Jews and Christians> gave us
nothing in the way of spiritual blessings?!’ nor any thought as to our
material welfare, 138D and if, in addition, he sent no teachers to us,
no lawgivers to us as he did to the Hebrews—Moses and the prophets
after him—for what, precisely, are we supposed to thank him?*'8
But think, has not God given us gods as well: kind protectors
never acknowledged by you, but not the least inferior to the <God>
revered by the Hebrews as first among the Jews of Judaea <who
allege>, as Moses taught, that they and those after them are the only
ones he cared for. For even if the God worshiped by the Hebrews is
none other than the demiurge, yet our beliefs about him are better?!
than theirs, even as he has given us greater gifts than he has given
them, both spiritual and material. Of these I will speak further, but ~
reiterate here that he sent many lawgivers who were not Moses’ infe-
rior; and many were far better.
143A
As I have suggested, if it is not the case that the differences in
laws and peoples are the work of the god holding precedence over a
separate nation and thereunder his angel, demon, hero, or order of
beings working for the whole pantheon and obeying them in all
respects, you must show me what other power brought such differ-
ences about. It is clearly not proof of your case to say simply, “Well,
God spoke and it happened as he said.” For the commands of God
must harmonize with the nature of the things he has created.32° To
put it more precisely: Was it chance that God established that fire
should rise up and earth should sink down? To complete the ordi-
nance, wasn't it necessary that <fire> be light and <earth> heavy?
And is this not also true of other sorts of things: <if differences
in language and political constitutions exist, must not these be
317. aya0O> yoxLKdc.
318. KaAd@c Evyapioteiy.
319. In the sense of being more virtuous.
320. See Oration IV.140A; Plato, Laws 713D.
JULIAN’S AGAINST THE GALILEANS 107

attributed to the preexisting conditions of the nature of the people


who create them?>374
143C
And so too with the things of the gods; for just as mankind is
ordained to die and perish, so also his works are by nature destined
to perish—changeable, subject to every permutation. But as God is
eternal it follows that his ordinances are unchanging. This means
that they comport with the nature of things and constitute the
nature of things. Nature cannot be at variance with nature, that is
with the will of God. How could it depart from harmony???
So I conclude that if <God> ordained the confusion of languages
and the lack of harmony among us, and consequent differences in our
political institutions, then it was not that he did this by a particular or
special ordinance, as if to make lack of agreement a feature of our
natures.*?> Indeed different natures must have been a part of every-
thing that stood to be distinguished, nation from nation. Just look at
the physical difference between Germans and Scythians from Libyans
and Ethiopians. Is this due simply to God willing it, without <the pre-
existing cause> of climate and geography having a joint influence, as
it were, with the gods in the determination of skin color??”4

321. Based on Cyril's paraphrase; Julian seems to compress this point, which
is worked out fully in Oration IV.
322. Julian here expresses one of Porphyry’s criticisms of the Christian faith,
particularly the emphasis on revelation as “interruption” or discord of the created
order: see Hoffmann, Porphyry’s Against the Christians, p. 35.
323. Wright's gloss of this passage is puzzling: “<Julian means> if there were
to be differences of speech and political constitution they must have been adapted
to pre-existing differences of natures in human beings” (III, p. 357, n. 2). Julian’s
point is more acute: Once having established difference of character and language
as part of the nature of humanity, with the result that there should be differences of
political order, there was no further need for God to differentiate by further, partic-
ular revelations (e.g., as to Abraham or Moses). Like an earlier generation of pagans,
Julian would have understood “superiority” of these political ordinances in a prag-
matic way: Rome's difference from the Hebrews and Christians could be demon-
strated by its superior political status in the world. The argument reaches back at
least to the time of Celsus, who enunciates it in a less philosophical style.
324. This is a good example of Julian’s “cooperative teleology” which depends
on a principle of noncontradiction between the nature of things and the role of
God in maintaining harmony. In Julian’s view, things move toward differentiation,
or evolve, partly because they are disposed to do so through natural process, partly
108 JULIAN’S AGAINST THE GALILEANS

146A-B
And yet Moses deliberately drew a veil*?° around.this kind of
reasoning when he fails to attribute the confusion of languages to
God as the sole cause.22° For he says?” that God did not descend <to
earth> alone but that others descended with him, although he does
not say who these were. Obviously those who descended with God
were godlike; and indeed it was not God who acted alone to disturb
the languages but others with him—evidently the same beings who,
acting with God, served to create the divisions in the human race,
each with its distinguishing marks.
148B
You may wonder why I have gone on for so long when I wanted
merely to say a few words. It is this: If, as-you say, the direct creator
of the world is none other than the God proclaimed by Moses, then
<I say> we hold a truer belief in him. We consider him to be the
ruler of all things in a universal sense: but <we believe as well> that
there exist the gods of nations who are like him—just as governors,
each having power over his province, are like a king, but subordinate
to him. And we hold that <God> is not the fractious rival of these
lesser rulers.3?® But if Moses gives his allegiance to a lesser god, and
then makes leadership of the world contrast with his power,>°
because of preexisting factors (climate, geography, etc.) which positively encourage
change and difference.
325. EnEKGALTTE TO TOLODTOV E1dmc¢ . . . Cf. Exod. 27.31: Moses was instructed
to make a veil to draw around the Ark containing the law. Julian here interprets the
passage to refer to the obscurity of the Hebrew account of the origin of language.
326. The phrase t@ 8€@ Ov seems to require this construction as dealing with
God as cause of the world’s languages.
S27 Gengilyn
328. A slightly different view of the familiar pagan argument that Christian
exclusivity in having no gods but the God of Moses is a form of intolerance that
diminishes true religious feeling, patriotic fervor, and proper veneration of the gods.
329. The point is somewhat oblique, but Julian probably has in mind the
Christian teaching (see Eph. 2.1-5) that the kingdom of this world has a lord other
than the Creator and that these archons, led by the “prince of the powers of the air,”
may have been the agents described in Gen. 11.7. The antipathy between the cre-
ator-god and the ruler of the age (aeon) was a fundamental starting point in Chris-
tian exegesis of the fall; see Origen, De princ. 3.2.1. The point is also made by Por-
phyry, apud Macarius Magnes. Julian’s larger, and politically nuanced, point is that
the world would reflect no harmony if it were caught between the rival purposes of
JULIAN’S AGAINST THE GALILEANS 109

surely it would be better to believe as we do: for we recognize the


God of All?*° though with due regard also for the God of Moses.32!
It is far better, I think, not to substitute the worship of one who has
been given <only> his small portion for <the worship> due the Cre-
ator of all.
152B
What a surprise is this law of Moses—I mean the famous deca-
logue. “Thou shalt not steal,” “Thou shalt not kill,” “Thou shalt not
bear false witness.” Let me write down each of these command-
ments which he says were written personally by God:
“Tam the Lord God who brought you out of the Land of Egypt,”
and then the second,
“Have no other gods over me.”
‘Make for yourselves no graven image,” followed by this explana-
tion:
“For I the Lord God am a jealous God, visiting the sins of the
fathers upon children to the third generation.”
“Take not the Name of the Lord your God in vain.”
“Remember the sabbath day.”
“Honor your father and your mother.”
“Commit no adultery.”
“You shall not commit murder.”
“You shall not steal.”
“You shall not bear false witness.”
“You shall not covet anything that is your neighbor
Ig 1332

I ask you in the name of all the gods, with the exception of “You
shall worship no other gods” and “Remember the sabbath day,” is
there any nation that does not think it ought to keep the other com-
mandments? So true is this that there is a price to pay for anyone

a “national” God, such as he imagines the God of Moses to be, and those seeking
to extend their dominion at his expense.
330. Lit., the God with hegemony over all.
331. This is Julian's clearest statement of the belief that the God of Moses is a
deity worthy of honor as a national god, but whose revelation as recorded in the
books of the Christians and Jews puts him on a par with the mythoi of other
national gods—i.e., as the creator of the Hebrew nation.
332. Conflating passages from Exod. 20.2-3, 4, 13-17; Deut. 4.24.
110 JULIAN’S AGAINST THE GALILEANS

who decides to break <any of the commandments>—in some Cases


more severe penalties than those ordained by Moses, but often more
humane.??3
Now as to the commandment, “You shall not worship other
gods,” <Moses> attaches a monstrous 155D untruth about God: “I
am a jealous God,” he is made to say, or later, “Our God is a con-
suming fire.”3°4 Now if a man is jealous or covetous you consider
him immoral, but if it is God—you call the attribute divine? But
how can this obviously false contradiction be at home with reason?
For if <God> is indeed jealous, it must follow that all other gods
who are worshiped receive honor to spite him, and all people who
worship these other gods defy the will of God. Well, then, how is it
that he is not able to restrain the nations if his jealousy demands
that other gods, besides himself, should not be worshiped? 155E
Perhaps he was not able to do it, or perhaps at the beginning he did
not wish to deny these gods their due reverence. As to the first sug-
gestion: it is impious to say that he was unable to do it. As to the
second, it is exactly what we ourselves say. Spare us such nonsense
and avoid incurring the charge of blasphemy!
159E Further, if it is God’s will that “no other god” should be
worshiped, why is it that you worship this so-called son of his,
whom he has never yet recognized or acknowledged as his own? I
shall prove this to you, while you, for mysterious reasons, insist on
giving God a false son.3%°
160D
Nowhere <in the Greek books> is God shown to be angry,
spiteful, petulant, oath-bound, or fickle, leaning now to one side and
now to the other, and suddenly front face again, as Moses tells us hap-
pened in the case of Phineas. Any of you who have read the book
called Numbers know what I mean.**° For when Phineas had seized

333. dirAavOpwnotépac, or of benefit to the human race.


334. Deut. 4.24.
335. Cyril suggests that Julian here launches into a discussion of Christian
abuse and misunderstanding of Greek accounts of the gods.
336. In Num. 25.10-13, Phineas, son of Eleazar, is said to have turned back
God's consuming anger (expressed as fire and earthquake in 26.10f.) as an agent of
his jealousy and wrath.
JULIAN’S AGAINST THE GALILEANS bee

and slain by his own hand the man consecrated to Baal—Peor,


together with the woman who persuaded him, dealing her a shameful
and excruciating wound to the stomach, the God of Moses is made to
say, “Phineas, son of Eleazar, the son of Aaron the high priest, hath
turned my wrath away from the children of Israel, for he was jealous
with my jealousy among them, and in my jealousy I devoured not the
children of Israel.”*3’ Can you imagine a more trivial reason for the
inane anger of God than that concocted by the writer of this passage?
What could be more irrational than this—even if there had been ten
or fifteen, or for argument’s sake even a hundred (no one would sug-
gest a thousand, | think), but let them say as many as that had defied
one of the laws laid down by God, would that then make it right—if
these thousand had transgressed, that 600,000 should be annihilated
by God???8 In my view I think it might be better in every case to save
a bad man along with a thousand virtuous ones than destroy a thou-
sand because of the one! Indeed, if the wrath of a single hero, or one
insignificant demon, is hard for countries and cities to withstand, who
could be expected to endure the anger of such a powerful God,
whether directed against demons, angels, or men?
168B
It is useful to compare <God’s> behavior with the grace of
Lycurgus, or the tolerance of Solon, or even the mildness and benev-
olence of the Romans toward those who have done some wrong.
And mark, too, how much superior our teachings are to these others,
for the philosophers advise us to imitate**? the gods as best we can
and tell us that this imitation is achieved through the contemplation
of ideas; and this sort of study has nothing to do with passion—in
fact, as I must say at the risk of saying the obvious, it is grounded in
freedom from passion.**° Only to the degree that we have been suc-
cessful in the duty of contemplating these realities,**! and learn
freedom from passion, do we become like God.

3372 Numi) 25:11.


338. According to the census concluded after Korah’s rebellion, the total
number of the people of Israel threatened by God was 600 and 1,730 (Num. 26.51).
339. utpetobat.
340. anaeia.
341. Sewpiav.
112 JULIAN’S AGAINST THE GALILEANS

But what sense does this “imitation of God” make to the


Hebrew people? Anger, wrath, hateful jealousy—the God who says,
“Phineas has turned away my wrath from the children of Israel,
because he was jealous with my jealousy among them.” In other
words, God found someone as petulant as himself and aggrieved as
much as he was, and so was made to keep his petulance at bay. This
sort of utterance and other, similar words are seen frequently in rela-
tion to the God of Moses in the sacred writings.
176A-B
It is also to be marked that God does not in later times show
concern only for the Hebrews, and that while caring for all nations
he gave the Hebrews no special privileges or gifts, while giving us
benefits far surpassing theirs. Consider the example of the Egyptians,
who manage to count a few wise men among their ranks, claiming
proudly the successors of Hermes—that Hermes who visited Egypt
in his third epiphany.*4? The Chaldeans and Assyrians take pride in
the successors of Oannes and Belos,**? while the Greeks boast of the
many successors of Cheiron.
Out of it all the Greeks developed a reputation for the sacred
mysteries and theology—much in the same way, as you know, as the
Hebrews, <who> claim to be uniquely gifted in this area.**4
178A
But has God commissioned you to start a <new> science or
philosophical movement? Well, then—what may it be? For the
charting of celestial bodies was advanced by Greeks, who built on
the theories first made by the barbarians in Babylon. So too the
study of geometry, which originated among the Egyptians during the
course of surveying the land and has since become of immense
importance <to all nations>. Arithmetic was begun by Phoenician
342. The “third great Hermes,” identified with Thoth, who had special status
among the Neoplatonists, and hence is important to Julian as a theological cipher.
343. On the fish-god Oannes, see Berosus, History, who claims he taught the
Babylonians the art of civilization “and thus has some analogy with the serpent of
Genesis” (Wright, III, p. 366). Brog is an attempt to give a Greek equivalent for
Assyrian bil, the Baal of the Hebrew Bible.
344. In PG 176C Cyril suggests that Julian goes on to deride the accomplish-
ments of David and Samson as being petty chiefs rather than heroes in the Greek
sense.
JULIAN’S AGAINST THE GALILEANS 143

merchants, and over time, among the Greeks, achieved the rank of a
science. And these three sciences the Greeks blended in the form of
music to become one, connecting astronomy and geometry through
arithmetic and finding within the principle of harmony. They estab-
lished rules for their music soon after, when they learned that har-
mony is perceived through hearing a nearly perfect agreement <in
the intervals>.>4°
184B
Must I list one by one the names and contributions of each
man? As for example, the individual men: Plato, Socrates, Aristeides,
Cimon, Thales, Lycurgus, Agesilaus, Archidamus; or maybe I should
speak of classes <of men>—philosophers, generals, artificers, law-
givers. For it can be shown that the foulest, the most brutish of these
generals conducted themselves with greater justice and civility
toward offenders than Moses did toward the innocent.
190C
And to what jurisdiction**® shall I report you? Should it be to
Perseus, or Aecus, or <perhaps> Minos of Crete, who rid the sea of
pirates and drove the barbarians to the edge of Syria and to Sicily,
extending his frontiers in two directions, and ruling not only over
the islands but over the coastal peoples? For this <Minos> parceled
out the care of mankind to his brother, Rhadamanthus—indeed,
establishing through what he had received from Zeus laws for the
whole earth, leaving Rhadamanthus the role of judge.**
After <Our Mother Rome’s> foundation she was beset by many
wars, and she won them all, prevailing over <her enemies>, ever
increasing in greatness in proportion to the dangers she has faced
and her need for security: so it was, then, that Zeus sent the renowned

345. The purpose of this digression, a lecture on the “unoriginality” of Chris-


tian teaching and story compared to the Greeks and antecedent civilizations, had
become a favorite topos for pagan critics at least by the time of Celsus, who expends
considerable energy attempting to show that Christianity has contributed nothing
new to philosophical discussion; see Celsus (Hoffmann 1987), pp. 55-58.
346. Baodcia—in keeping with the fancy that the Christians are being tried as
in a court, but have, properly speaking, no national law on their side.
347. This introduced <apud Cyril> a discussion of the legend of Minos, Dard-
anus, Aeneas, and the founding of Rome.
114 JULIAN’S AGAINST THE GALILEANS

philosopher Numa.#“8 This is the selfsame Numa, the excellent and


righteous man who lived a hermit in the groves, communing with
the gods in the purity of heart we know as contemplation.
193D-194B
It was Numa, chiefly, who fixed the law of the temple worship.
And from this came the blessing of divine possession, and also the
power which emanated from the Sibyl on the one hand, and from
others who, in those days, spoke oracles in their native tongues.
<And these gifts> were expressly given to the City by Zeus.
Further, <we must remember the omens of> the shield that fell
from the sky**? and the head which manifested itself on the hill, and
from which the throne of mighty Zeus himself received its name.*°°
Are such things to be considered the highest gifts, or merely as sec-
ondary benefits? But, do you, O deluded men, despite there being
still today with us the very artifact which dropped from heaven
above, sent by mighty Zeus or father Ares as sure proof, not promise,
that he will always hold that shield before this city, do you cease to
adore it and reverence it—You who <instead> adore the wood of the
cross, trace its figure on your foreheads, draw it on your house-fronts?
194D '
Who would not be right to despise the most intelligent man
among <the Christians>, or feel pity for the most stupid? For these,
by following you, have descended to such a pit of ruin that they have
let go of the ever-living gods <of our city> in order to embrace the
lifeless corpse of the Jew. <. . .> For I say nothing about the Mysteries
of the Mother of the Gods,?* and I admire Marius.*?

348. Successor of Romulus, a favorite of the gods. It is not entirely clear why
Julian chooses the name “philosopher” for Numa, except to tie him to a succession
of philosopher kings of which he sees himselfa part. Cf. Julian, Or. IV.155A.
349, Livy 1.20 (as the “shields called celestial”) and Aeneid 8.664.
350. Caput/Capitoline (Livy 1.55).
351. The cult of Cybele, the “Magna Mater,” was brought to Rome in about
205. Belief in eternal life was a part of the cult from its early days, and Julian seems
to have suggested that the Christians borrowed the doctrine from the mysteries
~ associated with the cult. Julian’s devotion to her cult is expressed in Oration V.
352. The logic is somewhat contorted: Cyril's refutation mentions that Gaius
Marius was known for his superstitious nature for his attempts to appeal to the pop-
ulares; but he was also considered a novus homo, and an object of veneration.
JULIAN’S AGAINST THE GALILEANS ELS

198B
Now the spirit that comes to men from the gods is attested, but
it is given rarely and to few, and it is not at all easy for everyone to
share in it all at once or every time <it is present>.3°3
We are not surprised, then, that the prophetic spirit has ceased
to move among the Hebrews, and is no longer known to the Egyp-
tians. So too, we see that the ancient oracles of Greece have fallen
silent*>* over the span of time. But as if to provide us with a means
of enquiry sufficient to our needs, our gracious lord and father Zeus
heeded our prayer for communion with the gods and gave to us a
sacred art.>°°
200A
Recall the most outstanding gifts of Helios and Zeus. I naturally
saved these for last. And need I say that <these gifts> are not <ones>
which only we who love Rome acknowledge but also ones which we
share with our Greek cousins. I mean, how Zeus fathered Asclepios
from himself, placing him among the known gods, and then,
through the life-giving power of Helios, revealed him to the <sons
of> earth:*°° After making his visitation to the earth, Asclepios
appeared in the shape of a man, alone, at Epidaurus. After this, he
manifested himself in diverse ways, stretching out his saving right
hand to the whole earth.*°’ He came to Pergamon, to Ionia, to Tar-
entum, and thereafter to Rome. He also traveled to Cos, and then to
Aegae. Thereafter he <was> manifested everywhere, both on land

353. The phenomenon of ecstatic utterance, which Julian knows both from
Christian writings (cf. 1 Cor. 12) as well as from prior expressions in the Cybele
mysterion and in the orphic and Delphic traditions, is the subject of this digression.
Julian does not wish to deny the ecstatic, but to limit it to “true prophecy.”
lamblichus concerned himself with the same procedure in his study of Egyptian
praxis; and cf. 1 Cor. 2.4, 14.6-10.
354. Wright, III, 373, thinks that Julian here refers to unsuccessful efforts to
restore the oracle at Delphi.
355. Divination.
356. Oration IV.144B: Asclepios is called cwmp by Julian, who in the same
oration contrasts him with Jesus.
357. The phrase éni ndoav dpete thy YAV Thy GuTIpLov EavtOU deEiav is an
attempt to relate the Asclepeian soteriology to those circulating in Christian writ-
ings: cf. Phil. 2.5-11 and especially Hebrews 3, 4.
116 JULIAN’S AGAINST THE GALILEANS

and at sea. And though he does not take private patients, still he
raises up those who are sinful and heals the sick.*”*
201E
Yet what great gift of this kind do the Hebrews say was granted
to them by God?—The Hebrews who have persuaded you to desert
to them? If you had only listened to what they taught, it may be you
would have done better for yourselves—although worse than when
you still counted yourselves among us—still, your <philosophical>
position would have been tolerable, perhaps sustainable. At least
you would be worshipping one God and not many; and not wor-
shipping a man, or shall I say, many miserable men.°*? Yes, and even
though you would be following a law that is vicious and hard and
contains much of a savage and brutal kind—unlike our humane and
gentle laws (and in other ways too would yet be inferior to us), still,
<were you Hebrews> you would be more holy and more pure than
your worship now makes you.
191D
Ah, but what has happened? Like leeches you have sucked the
worst from <the Jews> and left the good alone. Yet the name of this
Jesus who first persuaded the wretched and miserable among you*®
has scarcely been known for three hundred years. And during ‘his
lifetime, he accomplished nothing worth mention—that is, unless
one should think that healing a cripple and a few blind men, or

358. It was a convention of anti-Christian polemicists that the Christian boast


that Jesus was the divine son of a merciful God was disproved by the narrow geo-
graphical matrix of his miracles: see Celsus (Hoffmann 1987), pp. 68-69: “What
god comes only to those who already look for his reappearance and is not even rec-
ognized by them?”
359. The apostles and martyrs. The growth of shrines and churches was of par-
ticular concern to Julian. Constantius had handed over church property in Edessa
to the Arians, a situation Julian seeks to remedy (see Ep XL.424.C, D); and Gregory
(Or.3.86D) and Sozomen (HE 5.5.) allege that Julian had begun a program of spo-
liation of churches and martyrs’ shrines. The same worry underlies his edict on
funerals (Ep. LVI).
360. Celsus (Hoffmann 1987, pp. 72-73) argues the same case on the basis of
the social status of the Christians, and other pagan writers exploited Paul’s descrip-
tion of the church at Corinth to document the inferiority of the Christians as a class
(1 Cor. 6.9-11).
JULIAN’S AGAINST THE GALILEANS tly

driving the demons from possessed men in wayside villages like


Bethsaida and Bethany count as mighty works!
205E
And when the subject is purity of life, you cannot be sure that he
spoke of the matter at all.3
You yourselves behave like the Jews who vent their rage and
petulance by razing altars and destroying temples. And you
slaughter us as well: not only those of us who remain true to the
teachings of our ancestors, but even your own—or those you style
“heretics” because they do not bewail a corpse in the manner your
<teachers> demand.*°
206B
These are your own deeds: Nowhere did Jesus or Paul pass on
rules for such actions. And why? Because never could they have imag-
ined <that their followers> would have such power as they have now.
They were happy with seducing naive women and slaves*®? and
through them <their> mistresses, and men like Cornelius and
Sergius.*°* (Should you show me that either of these men is recalled
by any of the acknowledged historians of that time, <as> these events
are said to have happened during the reign of Tiberius or Claudius, I
would be happy to concede that I have misrepresented you.)
361. This is a central premise of Julian’s critique. He seems to have in mind the
disciples’ breaking of the ritual and ceremonial laws of Judaism (cf. Mark 7.1-5) as
reflected in the Gospels and the relative paucity of ethical teaching on Jesus’ part
that does not derive from Jewish law.
362. In an edict to the citizens of Bostra cited by the church historian Sozo-
men (5.15), Julian recalls the slaughter of heretics at Samosata, Cyzicus, Paphlag-
onia, Bithynia, and Galatia mandated by his predecessor, Constantius. The
“heretics” in question were those who did not belong to the Arian sect. The edict
itself is a call to religious tolerance, “not to become embroiled in the feuds of the
clergy,” the causes of which the emperor lays at the doorstep of the bishop Titus and
(according to Julian) his disingenuous calls for calm and restraint on the part of the
Christian community. Julian had a special dislike of Christian burial practice, which
flouted the pagan practice of reserving burial until after dark in honor of the lords
of the underworld; see Ep. LVI.
363. Cf. note 352.
364. The centurion and proconsul, mentioned respectively in Acts 10 and Acts
13.6-12. This allusion suggests Julian's acquaintance with the full body of the
Lucan corpus, as well as his skepticism toward the historical plausibility of the
episodes.
118 JULIAN’S AGAINST THE GALILEANS

209D
I do not know how I came to speak about these things. In any
case—coming back to the point from which I departed when I asked
you, “Why were you so ungrateful as to desert the gods for the Jews,”
Was it that Rome has been allotted its sovereignty by the gods, while
the Jews have been their own masters for such a short while, and
ever after aliens and slaves? Look at Abraham! Was he not a stranger
in a strange land? Jacob? The same, first in Syria, then in Palestine,
finally as an old man in Egypt.
Or Moses, who is said to have led <the Hebrew nation> from
slavery “with arm outstretched.” But after their rummaging about
Palestine, did they not change fortune more frequently than, it is
said, a chameleon changes its colors—first servants of the Judges,
next slaves of foreign nations? And then they began to be governed
by kings. Let me defer for a while asking about the manner of this
governance, bearing in mind what <their> writings tell us3°° that God
did not wish them to have kings, but <consented> only after being
pressed by them and after warning them that they would be governed
badly in such a way. Nonetheless, they had their little space and
farmed their own ground for just over three hundred years.
213A
After that they were a subject people: first to the Assyrians, then
to the Medes, later to the Persians, finally to ourselves. Jesus, who
was proclaimed by you, was nothing other than one of Caesar's sub-
jects. If you will not believe it, I shall prove it further on—but let me
state it here for the record: for you yourselves teach that together
with his father and mother his name was enrolled during the gover-
norship of Cyrenius.*°
213B
But when he was merely a man*°° what good did he do for his

365. Exod. 6.6.


366. 1 Sam. 8.
367. Luke 2.2.
368. Wright's translation, using Neumann's insertion—“when he became
man” (p. 381) for yevouevoc &vOpanog—seems to presuppose the credal formula-
tion of 325, but it is not clear that Julian intends this allusion. Julian’s sarcasm
refers to the time before Jesus was elevated to divinity by his followers.
JULIAN’S AGAINST THE GALILEANS 119

countrymen? No, say the Galileans; <the Jews> did not listen to
Jesus. Indeed? Then how did it happen that this “hardhearted and
stiff-necked people”*®? obeyed Moses. Yet Jesus—who commanded
spirits, walked on the water, drove away the demons and, as you
declare yourselves, “made the heavens and the earth”?””—though
none of his disciples dared to say such a thing concerning him,
except John, and he does not say so directly or unequivocally. But
for the sake of argument let us suppose that it was said: would it not
then stand to reason that this Jesus might have done better for his
friends and countrymen? Might he not, at least, have helped them?
218A
I shall return to this subject further on, when I take up the ques-
tion of the miracles <of Jesus> and the composition of the gospels.
For the present, however, answer me this: Which is the better thing,
to be free without restriction and masters over the wide earth and
sea for two thousand years, or to be for the same time the duty-
bound slaves of others? I think no man is so lacking in self-esteem
that he would willingly choose the latter. It would be as if someone
chose defeat over victory in war: Is anyone as stupid as that? And
speaking of <warfare>, show me in truth a single commander
among the Hebrews like Alexander or Caesar? There is no such man
among you. But by the gods, I cringe to think how I have insulted
the valorous by asking such a question! I mention them because
they are the best-known examples, and generals who are less well
known would not be recognized by the masses of our people—yet
the least among these deserves more admiration that all the generals
ever produced by the Jews.>”
221
Now as to the constitution of a city, the mode of law, the gover-
nance of cities, the superiority of laws, progress in learning, the
369. Ezek. 3.7. The phrase was also current in Roman anti-Jewish polemical
writing, on which see John Gager, Origins of Anti-Semitism: Attitudes Towards Judaism
in Pagan and Christian Antiquity (Oxford, 1985), pp. 15-57.
370. John 1.3.
371. Porphyry seems also to have used heroism as a criterion for the truth of
the Christian gospel. In a controversial passage, Augustine claims that Porphyry
listed Jesus among the heroes; see CG VIII.12; X.21; esp. XIX.22-23. If Julian knew
this tradition he does not support it.
_

120 JULIAN’S AGAINST THE GALILEANS

encouragement of liberal arts—weren’t these things barbarous and


wretched among the Hebrews? That dreadful man Eusebius says,
preposterously enough, that poems in the heroic meter can be
found among the Jews—saying further that the study of logic was
advanced among the Hebrews, since the word they use for logic can
be heard among the Greeks!?7*
But has any medical art appeared among the Hebrews, such as
that the Greeks know from Hippocrates, or the schools that came
after him? Can we say that their wisest man Solomon should be
compared to a Phocylides, a Theognis or Isocrates among the
Greeks??? Surely if you could compare the exhortations of Isocrates
with the proverbs of Solomon you would concede, no doubt, that
the heir of Theodorus is wiser far than the very wisest king <of the
Hebrews>. “Ah,” but they answer, “Solomon was an expert in the
secret wisdom of God!” If true, is it not true as well that this
Solomon, at his wives’ bidding, served our gods also, which even
they teach??’4 What virtue is this! What treasured wisdom! The man
could not even control his passions; the murmurings of a woman
led him from wisdom’s path. Please call no man wise who is
deluded by a woman, or if you still think him wise, then do not
claim he was deluded. Say rather that he trusted his own judgement
and followed his own heart, and the teaching that he received from
a God who had revealed that he should serve other gods.
224E Petulance and envy should not bother the most virtuous
men, for the things are distant form both angels and gods. But you
<Christians> hanker after the intermediate and partial powers
(which we would not be wrong in calling demonic), and in these

372. Eusebius, Praeparatio Evangelica 11.5.5: that Moses and David wrote in
Hexameter.
373. Julian's choice of sages is interesting: Theognis, presumably for repre-
senting the tradition of aristocratic wisdom; Isocrates for his eloquence and reputa-
tion as the moral prophet of the Hellenistic world; Phocylides for the gnomic char-
acter of his philosophical verse.
374. 1 Kings 11.4 refers to Solomon’s apostasy from the God of the Hebrews
at the bidding of his wives; 1 Kings 3.1 refers to an alliance between Solomon and
the king of Egypt. The point is not relevant to his argument against the Christians
but is rather an accusation leveled at the Jews.
JULIAN’S AGAINST THE GALILEANS 121

powers are pride and vanity, while in our gods there is nothing of
the kind.3”°
229C And if you can be happy with reading your own books,
why nibble at the learning of the Greeks? It would seem better to
keep men away from <philosophy> than from the eating of sacrifi-
cial meat.*’° Speaking of this question, Paul says the one who eats is
not harmed, but—O most brilliant and arrogant people!—the con-
science of the brother who sees but does not eat is offended. Our
own learning has caused the best and most virtuous men among
you to quit impiety. So too, almost every one who has retained the
virtue he was born with has quickly parted with your vanities. It is
clear, then, that it would be a finer thing to keep men from learning
than from sacrificial meats. Yet I suppose you know better than I
what effects your writings, as distinct from ours, will have on one’s
intelligence. For in studying yours no man would even achieve ordi-
nary goodness, let alone virtue, whereas from ours a man might
become better than before, even if he had born with no natural apti-
tude for excellence. A man who has such aptitude and has added to
it the benefit of our writing—that man is a gift of the gods to
mankind: <such a man> can light the fire of knowledge, can write a
constitution, rout his country’s foes in battle, travel bravely to the
ends of the earth and back again, like the heroes of old.”

375. This passage is something of a puzzle. It is not likely that Julian is talking
about the intercession of the saints, as Wright (III, p. 385 n. 1) suggests. As he seems
to have 1 Corinthians in view he may be referring to Paul’s reference to wisdom and
the rulers of the world (1 Cor. 2.6), as being derived from the tradition of
solomonic wisdom. In Paul's view it is the mvevya that supplies the “hidden”
wisdom of God.
376. 1 Cor. 8.7-13. Paul makes the eating of meat taken from animals sacri-
ficed to idols a matter of conscience; but he begins the chapter with an aphorism
which apparently irritates Julian: “ yv@org ovoroi 7 S€ Gyan oikodopet.” Julian
finds Paul's idea that “all of us possess knowledge” a direct attack on the Neopla-
tonic view that knowledge is achieved through the exercise of the intellect. It is
important to distinguish Julian’s view on this point from the equally anti-intellec-
tual teachings of Iamblichus concerning the means to acquire divine wisdom,
which is very distant from his argument here.
377. The Old Testament heroes are attacked next, according to Cyril. Julian
also polemicizes against the Old Testament being written in a tongue unsuitable for
learning and heroic tales.
122 JULIAN’S AGAINST THE GALILEANS

229E-230A
Why not put my words to a test: Choose some from.among your
children. Train them in your scriptures. When they have arrived at
manhood, if they have qualities any better than a slave's, then I am
only grumbling out of a bad temper. But you are foolish and arro-
gant, so that you think those scriptures of yours are divinely inspired
in spite of no man ever having become better or braver or wiser than
he was before by reading them. And what of the writings whose
proof is that, reading them, men do acquire wisdom and courage
and justice? These you attribute to Satan and those who do Satan's
bidding.*”6
235B-235C
<We say> “Asclepius heals our bodies, The Muses with the help
of Asclepius and Apollo and Hermes, god of eloquence, train our
souls. In war, Ares as well as Enyo fight for us. Hephaestos distrib-
utes and watches over our crafts. Athena, the virgin goddess, with the
help of Zeus, presides over them all.” Now consider whether we are
not superior to you in every single one of these things—in art,
wisdom, intelligence—not regarding whether we are speaking of the
useful arts or the imitative arts whose end is beauty: Statuary,
painting, household management, healing from Asclepius (with his
oracles found everywhere on earth and apportioned to us at any
time). With God my witness, I know that when I have been ill, Ascle-
pius has cured me by proffering remedies.*” And so if those of us
who are not apostates from the truth do better than you do even in
soul and body and worldly affairs, what reason can you give for for-
saking our teachings and adopting new ones?
238A-B
And why is it that the Galileans do not accept the traditions of
378. See, for example, Origen’s discussion of the influence of demons on
poetic recitation in De princ. 3.3.3. On the tradition in general, see Denise Kimber
Buell, Making Christians: Clement of Alexandria and the Rhetoric of Legitimacy
(Princeton, NJ, 1999).
379. Julian’s ploy at this point is a little oblique, in wanting to suggest that
there are plenty of examples of miraculous intervention in the mysteries, and that
even in the art of healing Asclepios surpasses the Christian God. The natural com-
parison would be between the healings of Jesus and those of the pagan healing god,
but Julian does not make it explicit.
JULIAN’S AGAINST THE GALILEANS 123

the Hebrews, or obey even the Law that God has given them? By
rights you should have rejected our teaching more than theirs, the
religion of your ancestors, the sayings of their prophets; but if any
one should take the trouble to examine your religion, he will find it
a wicked combination of Jewish recklessness and Greek vulgarity.
You have taken the worst rather than the best, the inferior rather
than the sublime from both sides and woven from it a thieve’s cloak.
The Jews have exact laws concerning ritual, innumerable sacred
things, observances which require the offices of a priest. And while
<Moses> prohibited them from serving any gods except the one
“whose portion is Jacob and Israel an allotment of his inheri-
tance,”*°° and while he said as well, “You shall not hate the gods”3*!
—but over the generations it has been thought by countless genera-
tions that the neglect of worship is a prescription for blasphemy,
coming from the people’s arrogance and shame. And this is the only
thing you have received from the Hebrews; in all other respects you
have nothing in common with the Jews. From the superstition of the
Jews you have developed your attack on the gods whom we revere.
But our veneration for every higher power, which typifies our reli-
gious worship—this you have set aside along with the traditions of
our forefathers. Yes, you have accepted the habit of “eating all
things” —even the green herb.?®?
238E
To be truthful, you have outstripped us in vulgarity—a thing that
I suppose happens as a matter of course in all nations. You think you
must adapt your ways to those of the lowest among you, the shop-
keepers, tax collectors, the dancers and whores.**? This is nothing
new, since the Galileans of our own day are like the first to receive
the teaching from Paul, who says himself that those of the earliest
days were people of the vilest sort in one of his letters.**4
BS580. Daita2sodinn
381. Exod. 22.28 (LXX).
382. Oration VI.192D; the reference is to Christian use of Gen. 9.3 against
Jewish dietary observances. Julian finds the Christian view an example of conven-
ient interpretation.
383. Letter XXXVI, the “rescript” on Christian teachers contains the same
charge, that Christians behave like salesmen.
384. 1 Cor. 6.9=11.
124 JULIAN’S AGAINST THE GALILEANS

For I think that Paul would not have written personally con-
cerning these matters if he had not been certain that these men com-
mitted these disgraceful deeds—written in language (even though
coupled with good words along the way) that would make the writer
blush. And even if the good words disguised the bad deeds, even if the
good words were deserved, it does not disguise the evil—and if unde-
served, if only flattery, then <Paul> should dig a hole in the ground
and hide for shame for such wanton and slavish pandering. 245C
Now these are the very words Paul addressed to those who
received his teaching, his precise words, to the men themselves: “Do
not be deceived, for neither idolaters nor adulterers nor homosex-
uals, nor sexual libertines nor thieves nor the greedy nor drunkards
nor extortionists shall inherit the kingdom of God. And of this you
are not in ignorance, brothers, because you were these things; but
you washed yourselves and you were sanctified in the name of
Jesus.”38° Do you not see that he admits the men he addresses were
these things, and then he says they were “washed” and they were
“sanctified,” as though water itself had acquired the power to
cleanse and purify not the body only, but even the soul! But baptism
does not take the sores away from the leper, or the scabs and boils,
the wens and disfigurations, or gout or dysentery or dropsy, or a
whitlow—in fact, <water> takes away no disorder of the body, how-
ever great or small: so shall it then do away with adultery, theft, and
all of the sins of the soul? <Oh you Christians, who run from your
sins like slaves from their true master, thinking that if you are
retrieved, well, at least your state will be no worse than before.>3°°
253A-B
Now: as the Galileans would have us think that while they are not
Jews, they are still in some sense “Israelites” because they hold to the
prophets and revere Moses and his successors in Judaea above all
others, let us see whether they heed these prophets. Let us, indeed,
begin with the teaching of Moses himself, who, they claim, was the
first to foretell the birth of Jesus who was yet to be. This Moses says

385. 1 Cor. 6.9-11 again, here explicit: Julian's reference point for describing
the social and moral inferiority of the Christians.
386. Based on Cyril's paraphrase of an analogy made in Julian’s work.
JULIAN’S AGAINST THE GALILEANS L253

not once or twice or even three times, but repeatedly, that men
should honor the one God—that is, the one men call “The Highest.”
He says on no occasion that they ought to honor another god. Yet he
speaks about angels and lords and acknowledges <the existence> of
many gods, and from them he selects a “highest” and denies rank to
a second, either like or unlike him, such as you <Galileans> have
invented. If from your own books you happen to possess a single
word from Moses on this subject, please do submit it; for the words
<you so often cite>, “The Lord your God shall raise up from among
your brethren a prophet for you, like unto me and unto him shall
you hearken,”**’ were certainly not said of the son of Mary.
Even if I could concede to meet you halfway, that these words
were said of <Jesus>, Moses says that the prophets will be like him,
that is, not a god, and born of men, not of a god. The words, “The
scepter shall not depart from Judah, nor a leader from his loins,”?°*
were certainly not spoken of the son of Mary; they apply to the royal
house of David which ended with the reign of Zedekiah the king.
The scripture can be understood in two ways, where it says, “Until
there comes what is reserved for him,” which you have mistakenly
interpreted to say, “Until he comes for whom it is reserved.”°°
Clearly not a single saying among these has anything to do with
Jesus. He was not even from Judah. For do you not yourselves teach
that he was not born of Joseph but through the Holy Spirit? In your
genealogies, you trace Joseph back to Judah; but you do not even
manage to create a credible line, for Matthew and Luke refute each
other's genealogies with contradictions. But as I will deal with an
examination of the truthfulness of these things in my second book,
we shall leave it until then.*?° Even granting that he was “a scepter

S87.) Deut 6-13:


388. Gen. 49.10, an ambiguous verse used by Jews and Christians to refer to a
messianic kingdom, since the second part of the verse includes the words “nor the
tuler’s staff from between his feet until he comes, to whom it belongs.” Julian's
explanation, historically, is as good as any other.
389. In the LXX version.
390. Cyril's reply to the second book is lost. It presumably dealt in more detail
with contradictions in the scriptural record, if this passage of the treatise previews
some of his argumentation.
Nat JULIAN’S AGAINST THE GALILEANS

from Judah,” then he is not “God born of God,” as you are fond
of saying, nor is it true that “By him all things were made, and
without him nothing was made.”*°? Ah, but you say, the Book of
Numbers says, “Out of Jacob shall arise a star, a man out of
Israel,”393 but this relates only to David and his lineage, for David
was the son of Jesse.
What you take from these writings, prove to me that these are at
variance with my interpretation of them. Indeed, you say, Moses
believed in one god, as can be proved when he speaks of the God of
Israel in the Book called Deuteronomy, “So that you may know the
Lord your God is one God and there in no gods beside him.”** And
he says again, “Hear O Israel, the Lord God is one Lord,”*”? and still
again, “See: Iam and there is no god beside.me.”*?° This is what Moses
says when he proclaims that there is only one God. But of course the
Galileans will say, “But we don’t say that there are two gods or three
gods.” But I will prove that this is precisely what you do say.
262G
I call first <the evangelist> John to bear testimony,??’ when he
says, “In the beginning was the Word and the Word was with God
and the Word was God.” This “word” is said to be with God, is it
not? It makes no difference—save that I may answer Photinus <Sir-
mius>—whether this was the one who was born of Mary or
someone else; that quibble I leave to you. But it is sufficient to point
out that <John> says “with God” and “in the beginning,” and how,
pray, does this agree with the teaching of Moses?
“Never mind,” say the Galileans, “for it conforms to the teaching
of Isaiah. For Isaiah says, “Behold the virgin shall conceive and bear
a son.”*°§ Even if this were stated of a god—and it is not so stated—
391. Julian here being the first pagan writer to show knowledge of the
homoousion formula Nicene Creed.
392. John 1.3. Julian’s broad knowledge of the New Testament canon and Chris-
tian use of the Old Testament is evident in his allusions throughout this section.
393. Num. 24.17.
394. Deut. 4.35.
395, Detty4 39)
SI GmDentys21393
397. See 1 John 5.8.
398. Isa. 7.14.
JULIAN’S AGAINST THE GALILEANS ber

a married woman who has been with her husband before her preg-
nancy could hardly be called a virgin. And let us accept that it is
indeed said about Mary: does Isaiah say that a god will be born to a
virgin? You never refrain from calling Mary the mother of God, even
though Isaiah says nowhere that the one born of the virgin is the
only begotten son of God??? or the firstborn of all creation.4°° And
as for the saying of John that “all things were made by him and
without him nothing that was made was made,” where shall I find
this among the sayings of the prophets? Yet, in quick succession, so
take heed, listen to what these same prophets have to say: “O Lord
our God, make us your own, for we own no other beside you.”4°
And from the king Hezekiah, who is reported to have said, “O Lord
God, God of Israel, you sit above the Cherubim; you are God: you
alone.”*°3 Is there room here for a second god? If you truly believe
that the Word is God born of God, of the same substance as God,
why assert that a virgin is the “mother” of God. How could she—a
human being—bear this god; and since God declares in plain words,
“I am He and there is no one who can deliver beside me,” Do you
presume to call this son of hers a “savior” ?4™
290B-290C Now you may acknowledge that Moses calls the
angels gods, as when he himself says “The sons of God saw that the
daughters of men were fair, and they took wives for themselves as
they pleased.”*°° Still he writes, “And after a while when the sons of
God came into the daughters of men, they bore children to them,
and these became the giants of old, the renowned men.”*" It is evi-
dent that he means angels by these words, which are his own and no
other, because it was not men but giants who were born from this

399. Alluding to John 1.18.


400. Col. 1.15.
401. John 1.3.
402. Isa. 26.13, altered by Julian.
403. Isa. 37.16.
404. This rather tedious recitation is designed primarily to show the Christian
opponents Julian’s virtuosity as an expounder of biblical texts; thus the punning
use of the verb oatew, here meaning both to save and to give birth (as in the case
of Mary).
405. Gen. 6.2.
406. Gen. 6.4.
128 JULIAN’S AGAINST THE GALILEANS

union. For if mortals rather than superior beings with a higher


nature and more powerful stature had been the fathers, he would
not have named the offspring giants. He claims instead that the race
of giants arose from the mingling of mortal and immortal.*” So too,
when Moses discusses the sons of God and says they are not men
but angels, would it not be expected that [at this same moment] he
would have disclosed (had he known of it) the information that
Jesus Christ is the Word, or as you call him the sole-begotten Son of
God?4°8 290E Is it not precisely because Moses was not thinking <of
the Word> that he says specifically, “Israel is my firstborn, my first-
bormSon?/4?
But why does Moses not say this about Jesus? After all, he taught
that there was one God; that he had many sons; that these were then
divided into nations. As to the Word of God being the firstborn Son
of God, or even God, or any of these other fabrications you have con-
cocted, Moses knew nothing and taught nothing about them at all.
2OLA
So, you have listened to Moses himself and to other prophets as
well. And you know that in other places Moses frequently says some-
thing of this sort: “You shall fear the Lord your God; him only shall you
serve.”4"° How then do the gospels teach <in defiance of this> that Jesus
commanded his followers to “go into all the world teaching and bap-
tizing men, in the name of the Father, the Son, and the Holy Spirit?”

407. €« yop Ovntod Kai GPavatov priEews.


408. Julian does not challenge the normative use of prophecy, only the Chris-
tian interpretation of the Mosaic texts which were held to be prophetic. Like Por-
phyry, he prefers the literal meaning of texts (see the following sentence) when a
clear motive for allegorical or prophetic meaning is not present. The overzealous
use of prophetic prooftexts was a constant feature of anti-Christian polemic from at
least the time of Celsus, the combative or propagandistic use of such texts having
arisen in the early missionary preaching in times of persecution.
409. Exod. 4.22.
410. Deut. 6.13, as also in Matt. 4.10, Jesus’ rebuke of Satan in the desert.
411. Matt. 28.19. Julian seems to view this as an example of unguarded
tritheism, but does not develop the point in relation to Mosaic religion, which he
has argued also had room for a plural concept of deity even if it had been subordi-
nated to monotheistic practices in the course of Israel’s history. Nor does he refer to
the growing philosophical defenses of Nicene (homoousian) orthodoxy in the
eastern Empire.
JULIAN’S AGAINST THE GALILEANS as,

These writings speak only about serving Jesus and say nothing about
serving God.*!?
Note too, if you please, how Moses speaks about the gods who
fend off evil when he writes, “The man shall bring two goats from
among the goats for a sin offering and a ram for a burnt offering. Then
shall Aaron bring his bullock for himself and shall make atonement for
himself and his household. And he shall take the two goats and present
them to the Lord before the door of the tabernacle of the law. And
Aaron shall cast lots for the two goats—one lot for the Lord and the
other for the scapegoat. And this goat he shall let loose in the wilder-
ness. In this way you shall send forth the goat that is the scapegoat.”
And of the second goat, Moses writes, “He then shall kill the goat that
is meant for a sin offering for the people in front of the Lord, and
gather his blood in a jar, for the blood to be sprinkled on the step of
the altar <mercy seat> , making atonement for this holy place on
account of the filth of the children of Israel and because of their tres-
passes and sin.”*!3 So it is clear that Moses knew the techniques of sac-
tifice, and did not think them impure, as <the Christians seem to,> as
you will witness when you hear his own words: “But the soul that con-
sumes the flesh of the sacrifice of a peace offering meant for the Lord,
that soul incurs uncleanness and shall be cut away from the people.”4"4
412. The meaning of this passage is unclear: Julian seems to want to say that
the Christians have extended the worship of God to other “beings” through the
baptismal injunction in Matt. 28.19, and so have violated their own monotheistic
foundations, which are based on the exclusive worship of the God of Moses and
Abraham. Cyril’s summary suggests that the Greeks and the Jews have many cus-
toms in common notwithstanding the Greek worship of plural deities and their tol-
erance of soothsaying. He notes also the prevalence of circumcision (rejected by the
Christians) among nations throughout the world and the fact that even Moses on
occasion sacrificed to heathen gods.
413. Lev. 16.15. Julian’s point is to suggest that Moses had learned the rites of
sacrifice and found them effective in dealing with the God of Israel.
414. Lev. 7.20. It is somewhat remarkable that Julian reaches into this most
arcane of Old Testament purifactory texts for arguments against Christian teaching,
especially as his fascination with its description often causes him to lose sight of his
central arguments against Christian practice, or to bury his grievance in reflections
on the Septuagint text. The key point is that Christians have abolished the use of
animal sacrifice while keeping much of the symbolism of the priestly books: Jesus
is thus the lamb of God slain before the foundation of the world, an atoning sacri-
fice for sin. Christian symbolism, however, was remarkably fluid, as Jesus could also
130 JULIAN’S AGAINST THE GALILEANS

Moses himself cautions you with respect to the eating of sacrificed


flesh.4!5
305D
It's best to remind you of what I have already written, as it bears
on the present discussion also: Why is it that after you have deserted
us <Greeks> you will not accept the law of Jews nor the teachings of
Moses? Of course, one of your sharper wits will say, “But the Jews do
not sacrifice either.”*!® But this point I can convince him is a dull
one, since I can also prove that <you Galileans> fail to observe any
of the customs which are observed by the Jews; moreover, that in our
time the Jews continue to sacrifice—in their own homes. What they
eat is always consecrated, and they pray before the sacrifice as well,
giving the right shoulder of the lamb to the priests as first-fruits. But
of course being deprived of their Temple, their sanctuary, they are
prevented from offering these same first-fruits to God in sacrifice.*!”
But why do you not sacrifice? You have invented a whole new
way of sacrificing that does not need Jerusalem. I suppose it is use-
less to ask such a question of you, since it hearkens to my earlier
point—concerning my wish to show that Jews and non-Jews agree in
many things, except-that the Jews believe in only one God. This
alone do we find unique to them and strange to us, since we have
be regarded as the scapegoat (released from death through resurrection, as man was
released from sin); the atoning sacrifice might be regarded as a peace offering and
thus not to be “eaten” by the people, who would thereby incur the guilt of having
usurped a sacrifice intended for God alone. Julian’s imperfect understanding of the
emerging theology of the Christian eucharist is evident throughout this section,
though his arguments reveal a great deal both about pagan perceptions of this the-
ology in late antiquity as well as the relative liberty with which the Old Testament
texts had been liturgically appropriated.
415. Cf. Paul, 1 Cor. 8.1-12.
416. Julian does not take notice of the historical similarity of early Palestinian
Christianity and early synagogue Judaism, which emerge, following the end of the
temple cult, equally as postsacrificial systems.
417. Sozomen (5.22) observes that Julian tried to persuade leading men
among the Jews to resume the sacrificial cult, which they were willing to do if the
Temple were restored. It is clear from his line of argumentation, which stretches the
meaning of sacrifice to include ordinary meals, that Julian does not regard the
Christian eucharist as bearing any resemblance to the vestiges of sacrifice in the
domestic life of the Jews. For him, it symbolizes chiefly the rejection of sacrifice in
the traditional sense.
JULIAN’S AGAINST THE GALILEANS 13d

much else in common with them—the temples, sanctuaries, altars,


purification, even certain laws. In all these things, differences
scarcely exist, except in small matters. <But it is not clear whether the
Christians worship one God, or whether they worship many gods,
because they have departed from the teaching of both the Greeks
and the Jews in this affair.>*!8
314C
Your diet, for example: Clearly you are not as particular as the
Jews, since for some reason you say we are obliged to eat everything,
even as the green herb; trusting in Peter who according to the
Galileans admonished them, “Do not make unclean what God has
made clean.”*!? And what proof do you offer that God has changed
his mind, holding certain things forbidden in the past, but now <for
the sake of Christian appetites> making them pure?
314D
Moses for his part, imparting the law concerning four-footed crea-
tures, says that those with parted hooves or cloven hooves that chew
the cud are pure, and others which do not fit the description are
impure.*7° Ah, but now, after the visions of Peter,*”! the pig begins to
chew the cud, so let us by all means <ignore Moses> and obey Peter!
Is it not miraculous that following Peter's revelation the pig has
acquired this skill? Yet if <Peter> lied when he disclosed that he had
seen this wonderful vision, this revelation—to use your own words—
in the tanner’s house, why should we rush to believe him in other
things? When Moses commanded you to avoid the flesh of swine, did
he not also forbid you to eat things that have wings, things that swim
in the seas, declaring these things unlawful and unclean according to
God's law? Or is it that these rules are too difficult for you to keep?
319D
Why bother to probe these doctrines of the Galileans further when

418. Not quoted from Julian but paraphrased from him by Cyril.
4V9. Acts 10:15.
420. Lev. 11.3, the point being that even if the Jews, like the Christians, no
longer observe the traditional forms of the sacrificial cult, they still respect the dietary
prescriptions related to it, whereas Christians, “who have no need for Jerusalem”
(the Temple), have forsaken the dietary laws pertaining to the cult as well.
421. Acts 10.9-16.
ic JULIAN’S AGAINST THE GALILEANS

it is far more simple to examine how they work in practice? Now the
Galileans will say that God had established an earlier law, but then
decided on a new law.!2? They will say that the first law came about
because of particular circumstances <in the life of the Jews> and was
limited as to time and place, and so gave way after a while to this new
law.’?3 But the falsity of <their position> I can demonstrate not from
ten but ten thousand different passages taken directly from the books
of Moses. Moses says, for example, that the law is for all time, as you
can <read> in the Exodus when he writes, “This day shall be a memo-
tial for you, and you shall keep it throughout your generations; you
shall keep this feast at my command forever; the first day you shall
remove all of the leaven out of your houses.”** There are many verses
which say precisely the same thing <about the law>, but because of
their great number I will not put them in evidence here to show that
the law of Moses was meant to last for all time. And what will you
<Galileans> answer? That there is some word of Moses to equate with
Paul’s rash suggestion that “Christ is the end of the Law.”47°
320B
Where does God offer the Hebrews a second law, different from
the law he had <originally> set down? Nowhere—not even as much
as a revised version of the-original law.**° Listen to what Moses has
422. devtEpos voLoc.
423. Julian seems to have in mind passages such as Jesus’ reformulation of the
law in Mark 7.17-23 and Matt. 5.1-7.28.
424. Exod. 12.14-15: This Passover ordinance would seem especially flimsy
given the opening flourish with which Julian introduces his evidence. According to
Cyril, Julian produced other texts in support of his view that the law was not to be
abrogated, but he does not reproduce them. Strangely, Julian does not quote Jesus’
words to the effect that the law is established until the time of Judgment (Matt.
5.17-20), which might have been set against Paul’s more Hellenistic understanding of
the “provisional” or limited force of the law reflected in the passage from Rom. 10.4f.
425. Rom. 10.4, the complete verse running “Christ is the end of the Law for
them that believe,” being probably addressed to gentile Christian proselytes rather
than Jewish Christians, but Julian takes the verse in an absolute sense.
426. Wright, III, p. 410 n. 3, cites Julian’s comment in Oration V.170: “The
gods, not being ignorant of their future intentions, do not have to correct their
errors.” It was a standard complaint among pagan critics that the Christian God
seemed to lack foreknowledge and that the drama of redemption was based on the
need to rectify a divine error. The criticism was also reflected in certain Christian
heresies, especially the solution of the Marcionites.
JULIAN’S AGAINST THE GALILEANS 133

said: “Add not a word nor take a single word away from what I have
commanded: keep the commandments of the Lord your God which
I command you this day.”*?” And he says, “Cursed be every man who
does not keep them all.”428 But you think it a trivial matter to add
and subtract words from the law. You seem, indeed, to think it
courageous and high-minded to destroy the law completely, if it
suits you, and to create a revised truth that men will find attractive.
327A
But you <Galileans> are so caught up in your error that you have
betrayed even the teachings passed on you by the apostles. These
<writings> too have been changed for the worse and made heretical
by the scribes who came later.*”? At all odds it was not Paul, nor
Matthew, nor Luke, nor Mark who dared call Jesus God. No, it was
the estimable John who did so, I imagine, because multitudes in the
towns of Greece and Italy had already been struck down by this dis-
ease,*°° or perhaps because he had heard that the tombs of Peter and
Paul were being worshiped, albeit in secret. Knowing where things
stood, it was John who first presumed to call Jesus God. It was John
who, giving short shrift to John the Baptist,**! returns to proclaim
<Jesus> the Word of God: “The Word was made flesh,” he says, “and
dwelled among us.”*°?
Nor does John refer to “Jesus” or “Christ,” preferring to call him the

427. Deut. 4.2.


428. Deut. 27.26. Julian expands his criticism, according to Cyril, by citing
Acts 15.29-30, which forbids the eating of meat sacrificed to idols (cf. Paul’s admo-
nition in 1 Cor. 8.7ff., which seems to offer the opposite advice). Julian would have
been unaware of the “Hellenist” crisis of the early church, which accounts for the
different strata of advice concerning the law and dietary rules.
429. Julian is not offering an early theory of synoptic revision, but throughout
this section he argues that the earlier gospel writers were cautious in their evalua-
tion of the person of Jesus and that only later did Jesus come to be regarded as a
god. He associates this development particularly with the Fourth Gospel rather than
with Paul.
430. véooc: Julian’s standard term for conversion to Christianity; cf. Or.
VII.229D, Ep. LVIII.401C (to Libanius).
431. John 1.6-8.
432. John 1.14. Julian takes it as self-evident, especially within the framework
of his own Neoplatonic thought, that the reference to the incarnation of the divine
Logos is an assertion of the divine status of Jesus.
134 JULIAN’S AGAINST THE GALILEANS

Word or God’? but then stealthily, quietly seducing our senses—saying,


for example, that John the Baptist gave testimony to Jesus Christ, and that
we must therefore regard <Jesus alone> as the true Word and God.
333B
I do not, of course, deny that John says these things about Jesus
Christ, though certain heretics have said that Jesus Christ is distinct
from the Word proclaimed by John.*3* But this is not so: The very one
whom John proclaims as God the Word is the same who was recog-
nized by John the Baptist as Jesus Christ. See how <John> cunningly,
almost imperceptibly advances the action of his little stage play and in
the last act, full of impiety, comes on stage to unveil this final bit of
deceit: “No man has seen God at any time. But the only begotten son
of God, the one who is in the bosom of the Father, he has revealed
him.”49> Well then: Am I to think that the only begotten son who
reclines in the Father's bosom also is the God who is the Word made
flesh? Because if it is he and no other, then you have certainly seen God,
since after all, <John> says next, “And he dwelled among you and you
beheld his glory.” Which makes scant sense of the verse, “No man has
seen God at any time.” <If John’s words ring true> then you have seen
God—and if not the Father, then at least God the son, the Word.*°°
But if the only begotten son is one thing*?’ and God the Word
something else, as I have heard it said some of the members of your
433. John 1.1, 18.
434. The Modalists may be in Julian’s view, or the Arians, both of whom
preached varieties of a separationist or subordinationist doctrine that kept the
Logos unaffected by the limitations of Jesus’ humanity.
435. John 1.18.
436. Julian refuses to allow the author of the Fourth Gospel any room for nar-
rative or theological expansion, insisting that he has caught him in a contradiction
brought on by his awareness that Jesus was not the Word. The order of the verses
does evoke a non sequitur; but the Evangelist’s point is that while no one has seen
God, the Word incarnate reveals him. In Letter XLVII.434 to the Alexandrian Chris-
tians in response to their petition for the return of Athanasius from exile, Julian
writes, “You dare not to worship [the gods of your fathers] and prefer instead that
one you have never seen nor your fathers ever saw, namely Jesus, ought to be
revered as God the Word.” He enjoins them to return to the worship of Helios, “the
intelligible father’s living image.”
437. 0 povoyeviig vidc, Etepog 5€ 6 Beds Adyos: Wright introduces the word
“person” into the translation, which because of its associations with Nicene the-
ology I have rendered simply as “thing” or “matter of concern.”
JULIAN’S AGAINST THE GALILEANS 135

sect,*** then it seems that not even John was foolish enough to
declare that <Jesus was God>.
335B
But the beginning of this wicked teaching was John’s doing, and
how can one begin to revile as they deserve reviling the many errors that
you heap one on another, just as you heap the dead bodies of those
newly martyred atop the corpses of those who died in ages past?*2?
You have filled the whole world with tombs and sepulchres, yet
where in your holy books does it tell you to prostrate yourself at the
tombs and pay honors to the dead?**° But you have so far departed
from the truth in this that you will not heed even the words of Jesus
of Nazareth. Listen—what does he say about the gravesites? “Woe to
you scribes and Pharisees, you hypocrites! For you are like whited
sepulchres. On the outside the tomb appears beautiful, but within it
is filled with bones of dead men and all impurity.”**' So then, if even
438. It is not clear what Julian means by the term “sects” or “heresies.” Within
the Church, the term had come increasingly to mean an illegitimate Christian
minority opposed to the teaching of an orthodox bishop; for his part, Julian does
not seem to think the internal doctrinal differences of the Galileans are worth pur-
suing except as they point up philosophical inconsistencies in doctrine.
439. Julian was exasperated by the benevolence of the Christians to the poor
and their record of contributing to the welfare of the sick. In a letter to Arsacius of
Galatia (Ep. XXII. 430D) he praises the Christian maintenance of graves, but, as
here, he sees it within the “detestable” context of Christian veneration of the relics
of the dead, which he finds an affront to pagan religion; cf. Eunapius, Lives, 424.
440. Julian’s revulsion at Christian reverence for the tombs of martyrs is
reflected in his frequent use of the word novnpdcg or “depraved” to describe their
activities. In recounting an expedition made ca. 354 from Nicomedia to New Ilios,
Julian recalls a conversation with a Christian by the name of Pegasius, who showed
him shrines to Hector and Achilles. Perplexed that even Christians in Ilios revered
pagan heroes and refrained from violating their shrines, Julian is informed by Pega-
sius that it was natural for the people of the region to worship the brave among
their own citizens “just as we also worship the martyrs.” In this same letter Julian
finds examples of such syncretism useful in promoting the restoration of Greek reli-
gion, observing that Pegasius was able to move about the shrines “without making
the sign <of the cross> on [his] vile forehead or hissing himself. . . . For the two
things are the meat of their theology, | mean hissing at demons and signing a cross
on their foreheads” (Ep. XIX).
441, Matt. 23.27, though Jesus is speaking of hypocrisy and not of the vener-
ation of tombs, which was strictly forbidden in Jewish law. Julian also seems to have
cited Matt. 8.21 (“Let the dead bury their dead”) to prove that Jesus commanded his
followers to avoid graves.
136 JULIAN’S AGAINST THE GALILEANS

Jesus declared that the tombs are full of pins iia how can you
say that God can be worshiped there?
339E
With so much to condemn you, why do you persist in prostrating
yourself among the tombs? If you wish to know, I will happily tell
you—or rather not I but the prophet Isaiah <when he says>, “They live
among the tombs and in the caves for the sake of dream visions.”***
Here you will see that this practice of sorcery was established long ago
by the Jews, who are said to dwell among the tombs in order to receive
dream visions. And it is certain that after the master’s death your own
apostles did this very thing, that is, <sleeping among the tombs>; and
that they then passed on the practice to the ones who were first to
practice your faith. These no doubt were able to conjure and work
miracles with greater deftness than you do nowadays, displaying their
abominable rites to the initiates in the secret places appointed for
them so that the blasphemy might be perpetuated.**7
442. Isa. 65.4 refers to the Hellenic custom of sitting among the graves in
order to evoke an oracle or vision. Julian assumes the practice was initiated by apos-
tates among the Hebrews, then illicitly carried forward into Christianity in defiance
of Jewish custom.
443. The charge of witchcraft and sorcery was a polemical convention in Jewish
anti-Christian propaganda but did not as such feature independently in earlier
pagan polemic. The received description of Jesus as a magician and his followers as
apprentices seems to have had an association with tomb ritual, which is reflected in
the Lazarus tradition (John 11.1-44), in the Markan story of the rich young man,
and a vestige in Mark 14.51 (the young man who flees, leaving his burial shroud
behind), and in the so-called—and highly controversial—stratum of tradition
known as “Secret” Mark; see M. S. Smith, Jesus the Magician (San Francisco, 1978),
and his “Clement of Alexandria and Secret Mark: The Score at the End of the First
Decade,” Harvard Theological Review 75 (1982): 449-61. A countervailing tradition is
the story of the Gadarene demoniac who is said to “live among the tombs.” The pres-
ence of Jesus and his followers in a location that would have been off-limits to obser-
vant Jews would suggest a certain estrangement from Jewish tradition, as even Por-
phyry recognized, and it is sometimes suggested that Jesus’ withdrawal to “lonely
places” and even the Gethsemane account are redacted stories of cemetery rituals.
Despite a flight of interpretive fancy in this verse (it is not at all clear that Christians
would have utilized cemeteries for the purpose of receiving visions and revelations
from the dead, though likely that eucharists were celebrated secretly at gravesites, cf.
Eunapius, Lives, 424), the tomb tradition is a prominent religious element, begin-
ning with the narrative of the burial of Jesus; the visitation and discovery of the
empty tomb; and more to Julian’s point, the extension of visitation and veneration
to the graves of Christian martyrs in Rome, Antioch, and the East.
JULIAN’S AGAINST THE GALILEANS LY

343C
So you who perform the rites which God has always hated, as we
know from Moses and the prophets, you nevertheless refuse to sac-
rifice animals at the altar. “Naturally,” say the Galileans, “because
<God does not rain down> fire to consume our sacrifices, as in the
time of Moses or in the days of Elijah the Tishbite, but only after a
long span.”*** But I can give you proof in short order that even
Moses taught that it was necessary to build the fire for sacrifices per-
formed before him, as you may tell from story of the patriarch
Abraham.**°
And these cases can be multiplied, as when scripture says that
the sons of Adam offered first-fruits to God, “And the Lord was
pleased with Abel and his offerings, and so Cain was angry and
looked downcast. And God said to Cain, Why are you angry and
sullen? For is it not true that even if you offer what is proper but do
not divide the shares properly you have sinned?”**° And would you
wish me to tell you what these offerings were? “After a long time it
happened that Cain brought an offering to the Lord taken from the
fruits of the earth. Abel also brought an offering from the fatted
firstlings of his flock.”*4”
You see! Say the Galileans, it was not the sacrifice but the divi-
sion of the offering that God condemned, saying to Cain, “If you
offer what is proper but do not divide the shares properly you have
sinned.” This is what one of your own well-read bishops has told
me, though he deludes himself and deceives other men as well in
thinking this way.*** When I put to him this question: In what way
444. Lev. 9.24; 1 Kings 18.38.
445. Gen. 15.7-17. It is not clear why Julian chooses to belabor this point,
especially as Christian refusal to offer animal sacrifice was tied to a particular under-
standing of the sacrificial death of Jesus embodied symbolically in their eucharistic
celebrations, rather than to the absence of miraculous intervention.
446. Julian’s use of the Septuagint of Gen. 4.4-7. It is not clear what specific rel-
evance this quotation would have in Julian’s case against Christian rejection of sacrifice
unless the “bishop” he refers to in the following section had cited the passage to show
that God himself rejected sacrifice, using Cain as a prototype of its inadequacy.
447. Gen. 4.3-4.
448. Aetius, “old friend and correspondent” (Ep. XV), may be the partner in
this discussion, since Julian had particular delight in conversing with the bishop
and refuting him in debate.
138 JULIAN’S AGAINST THE GALILEANS aai
74
fy

was Cain’s division of the offering wrong?, he was silent and could
not wriggle free of my snare with so much as a half-hearted reply. So,
when I saw that I had him, I said, “God did reject the <offering>, as
you say, and with justice; but the piety of the two men was equal.
Both knew that they must offer gifts and sacrifice to God. But when
it came to apportioning the offering, one man received praise while
the other was blamed—but why? What's the reason? Because of all
that exists on the earth, some things possess soul while others are
soulless. And <you would agree> that things with souls are more
valuable than things without souls, at least in the eyes of God who
is the author of life. For that which has soul has some share in the
life of God, having a life in some sense resembling his. And so, God
approved of the man who offered living sacrifice.”**
351A
To turn now to another matter, I ask why you refuse the practice
of circumcision. I hear your answer when you say, “Paul teaches that
circumcision of the heart rather than the flesh was imputed to
Abraham because of his faith.4°° Paul was not speaking of the flesh
and we should heed what he says, as well as what is said by Peter.”
But have you not heard that God is said to have given circumcision
to Abraham as a sign of the covenant when it is written “This is the
covenant you will keep between me and yourselves and your off-
spring generation after generation: You shall cut away the flesh of
the foreskin as a token of the covenant between us, and between me
and your offspring.”*°' Christ moreover instructs us that he did not
come to abolish this law,*°* and teaches explicitly that it is proper to
observe this law. And if he not only teaches the observance of the
law but actually condemns those who violate even the least com-
mandment, what can you say in your own defense—you who have

449. One can only imagine the reasons for the Christian bishop’s silence on
hearing this imaginative exegesis from his emperor.
450. Rom. 4.11, 2.22 are in view.
451. Julian’s rendering of Septuagint Gen. 17.10-11. Paul's argument in
Romans 4 is that Abraham himself was not circumcised “according to the flesh,”
and thus might be seen as a prototype of those who have faith but do not belong
to the covenant people by birth or rituals associated with the ethnos.
452. According to Cyril, Julian paraphrases Matt. 5.17-19.
JULIAN’S AGAINST THE GALILEANS 19

violated every one of them?*°> You cannot have it both ways, for
either Jesus was speaking falsehood or you have failed dismally to
keep the commandment: “The circumcision,” Moses says, “shall be
of your flesh.”454
But the Galileans do not listen; they say instead, “Indeed, we cir-
cumcize our hearts.” And well you should, for as we know there is
in your little company not one evildoer and not a single sinner—so
well circumcized are your hearts! And the Galileans say, “But we
cannot keep the rule concerning the feast of unleavened bread, the
Passover. For <we believe> Christ was sacrificed for our sake once
and for all.”*°° Indeed, and did he then command you himself not
to eat unleavened bread? With the gods as my witnesses I count
myself among those who avoid the festivals of the Jews. But I ven-
erate without hesitation the God of Abraham, Isaac, and Jacob, <for
they were> members of a sacred race, the Chaldeans, learned in the
arts of divination, who became acquainted with the rite of circum-
cision during the time of their wandering among the Egyptians.**°
And the Jews worship a God who has always been gracious toward
me, as hé was <always gracious> to Abraham and those who, like
Abraham, worshiped him. He is a great and powerful God, to be
sure, but he is no God of yours.
356C

453. The logic of the argument is somewhat attenuated by Julian’s tendency to


rely on the apposition of the texts he is using to support his case: that as an obser-
vant Jew Jesus taught observance of the law of circumcision which Paul, in the
interest of his mission to the Greeks, reinterpreted and then set aside, in violation
both of Jewish teaching and practice and the example and teaching of Jesus.
454. Gen. 17.13 in Julian’s rendering, “n nepitopy gota nEpi Thy OdpKa cov.”
455. A reference to the early Christian interpretation of Passover, in particular
its meaning for gentile Christians not accustomed to observing Passover (Heb.
10.10). This continues Julian’s routine of juxtaposing the teaching of Jesus with the
theology of Paul, or in this case the epistle to the Hebrews.
456. This is typical of Julian’s recognition of the Jews’ zeal for tradition,
whereas “[the Greeks] are in such a state of apathy as concerns religious matters that
we have forgotten the traditions of our forefathers” (Letter XX.453D-454B). It is
interesting that Julian here seems to complain that the Galileans reflect more of
Hellenistic indifference to the forms and rules of religion. The Jewish God is wor-
shiped “under other names” by the Greeks, but not by the Christians who have
abrogated his laws (454B).
140 JULIAN’S AGAINST THE GALILEANS

For you have nothing in common with Abraham, who built altars
to <God> and worshiped him with sacrifices on those altars with
burnt offerings. Like the Greeks, Abraham was accustomed to offer
sacrifice daily, and he shared with us Greeks the custom of telling the
future from shooting stars. And for significant things he learned to
augur from the flight of birds, hiring a servant in his house who was
expert in the reading of signs.4*” But someone among you may doubt
this, so here are the precise words spoken by Moses to prove the point:
“After this the word of the Lord came to Abraham in a night vision
saying, Do not be afraid, Abraham. I am your shield, and your reward
will be very great. And Abraham said, Lord, my God, what can you
give me? Since I am childless, and the son of the slave-woman Masek
is to be my heir. And at once the word of the Lord came to him saying,
This man is not to be your heir. The one who will be born from you
is to be your heir. And he brought him out and said, Consider the sky
and read the stars of the heavens: can you count them all? And he said,
this is how numerous your offspring shall be. And Abraham believed
the Lord and righteousness was attributed to him.”*°°
Can you explain to me why he was shown the stars by the one
who brought him out,.whether God or angel? For surely while he was
yet inside the house he knew how vast the array of visible stars were
shining above in the night sky? No, I think <he was brought outside>
because <God> wished to show him the shooting stars as a visible sign
of the promise he had made to Abraham, and to demonstrate that the
fulfillment and sanction of all things can be seen in the heavens.*°?
457. Gen. 24.14, and an interpretation built on Gen. 24.2-43. Maimonides
also seems to have seen Gen. 24.14 as an act of divination.
458. Julian has earlier examined Paul’s use of this passage (Gen. 15.1-6, lib-
erally paraphrased by Julian) in connection with the rite of circumcision; here he
uses it to show that Abraham's believing in God (éxiotevoev ABpady 16 Oe)
involved a reading of the stars or an act of augury. It is remarkable that Julian does
not choose to mention the most famous Christian example of divination in the
Gospel, the star of Bethlehem (Matt. 1.7-10) in his discussion of Hellenic customs.
459. Julian is probably guilty, and perhaps conscious, of an overimaginative
piece of exegesis read back into the patriarchal narratives, stemming from his belief
that the patriarchs themselves were Chaldeans and thus especially skilled in the
interpretation of the skies. The idea that the stars served as a “visible pledge” might
suggest that Julian is thinking of another biblical text (Gen. 9.13) where a sign is
given to Noah in the heavens representing the promises of God.
JULIAN’S AGAINST THE GALILEANS 141

358C So that no one will think my interpretation is rash, I


would ask you to observe what follows in this passage, for what fol-
lows next is this: “And he said to him, I am the Lord your God that
brought you out of the land of the Chaldees to give you this land as
an inheritance. And he said, Lord God, how shall I know that I shall
inherit it? And he said to him, Bring me a heifer three years old, a
she goat of three years, a ram of three years, a turtle dove, and a
pigeon. And he did as God commanded and divided them down the
middle and laid each piece one against the other; but the birds he
did not divide, but the birds came down on the divided carcasses
and Abraham sat down among them.”4°
So you can see how even the revelation of God or an angel was
supported by this use of reading the signs of the birds, how a
prophecy was brought to fulfillment in a complete manner. By the
offering of sacrifices, and not, as in <the case of the Galileans> an
arbitrary fashion.** Further, he says that it was the flocking together

460. Gen. 15.11; lit., “drove them away.” Julian according to Ammianus was
especially interested in the use of birds in augury; cf. Amm., 22.12.1, who also com-
ments on the emperor's “drenching the altars with the blood of bulls and other ani-
mals” before the Persian campaign.
461. This is yet another instance of Julian’s hypogeal style of argumentation:
According to his logic in this passage: The Hebrew patriarchs were Chaldeans and
skilled astrologers; Abraham being a Chaldean looked to the stars for knowledge of
divine things and received signs from God; these were then confirmed for him
through the use of sacrifice; sacrifice is therefore a confirmation of the relationship
between our knowledge of the divine and the revelations one is able to attain
through observation (observation being understood to include dreaming and
ecstatic states, as well as seeing). The Galileans have piecemeal taken on some of
these religious customs, but so modified them that they have lost the sort of coher-
ence which Julian thinks both the Hellenes and the Jews have preserved. Here too,
however, there are noticeable silences with respect to his familiarity with earlier
Christian tradition as reflected in the gospels: the Christian apocalypses (cf. Mark
13.24-27, pars.; Rev. 4-12, etc.) based on Hellenistic Jewish prototypes (1 Enoch,
4 Esdras, etc.) and Jesus’ commendation of divination as a way of foretelling the
coming of the heavenly son of man (Dan. 7.1-27), and Paul resorts to the same
method in Rom. 1.18-20. It is reasonable to suppose that Julian’s failure to men-
tion early Christian persistence in the Hellenistic Jewish custom of divination and
augury stems from his awareness that Christians had by and large given up the prac-
tice (as had the Jews) following the age of Messianic enthusiasm, ca. 70 CE. Here
too, however, the receding of apocalyptic divination among the Jews is not called
into question.
142 JULIAN’S AGAINST THE GALILEANS

of the birds that proved the revelation was true. So Abraham


believed the promise, saying as well that a promise that cannot be
shown to be true is foolishness and stupidity.*% For need I tell you
that one cannot discern truth from mere words: there must be some
sign that the words are true, some proof that everyone can see which
supports the prophecy as it concerns the future. There is, I imagine,
an excuse for your recklessness in this matter: the fact that you are
not permitted to offer sacrifices outside Jerusalem. Yet Elijah offered
his sacrifice on Mount Carmel and not in the Holy City. . . .

462. The implication is that Hebrew prophecy is supported by signs


of
approval, whereas Christians simply assert it to be true that their religion fulfills
the
Old Testament prophecies.
FRAGMENTS OF THE
CONTRA GALILAEOS*

FRAGMENT III:

These things have happened before, and still are happening. How
can they be signs of the end of the world?*

FRAGMENT IV

Moses after fasting forty days received the Law. Elijah was permitted
to see God face to face after fasting for the same length of time. And
what did Jesus receive after fasting for just the same period?*®

463. From Neumann’s reconstruction, Juliani imperator librorum contra chris-


tianos quae supersunt (Leipzig, 1880) (with German translation). The numbering of
fragments here corresponds to Neumann rather than to Wright's revision.
464. From Julian’s Book II (Cyril, Book XII); derived from Theodore of Mop-
suestia who wrote a refutation of Julian around 378. On the critique of Christian
apocalyptic hopes as an early fixture of anti-Christian polemic, cf. Hoffmann,
Celsus, pp. 6-12.
465. From the same source as fragment iii.

143
144 JULIAN’S AGAINST THE GALILEANS

FRAGMENT VI:

Jesus was in the desert when he was [said to have been] led to the
pinnacle of the Temple: how could this happen?*°°

FRAGMENT VII

Furthermore, Jesus—a “god”—requires the comfort of an angel as he


prays, using language that would be humiliating even for a beggar
who bemoans his adversity. And who had told this Luke*®’ the story
of the angel if, indeed, it ever happened, for no one was there to see
the angel as he prayed, since the men had fallen asleep. So it is that
when Jesus comes to them from his prayer he finds them “asleep for
grief,” and he says, “Why are you asleep; arise and pray with me, and
so forth, until, “While he was speaking, a multitude and Judas
arrived,” and this is why John omits this tale about an angel, for he
did not see it either.

FRAGMENT IX

Hear now this sage political advice: “Sell whatever you have and give
it to the poor. Provide for yourself the purse that does not wear
thin.”*°8 Can anyone imagine a finer piece of common sense than
this? For when all men obey your command, there will be no one left
to buy anything; and if this laudable teaching were carried out, what
city, what nation, what family would survive? When everything has

466. The criticism echoes the famous comment of Origen in De principiis 4.3
concerning the transportation of Jesus in Matt. 4.8: “The devil is said to have taken
Jesus up into a high mountain in order to show him all the kingdoms ofthe world
and the glory of them. How could this possibly have happened literally, either that
the devil should have led Jesus up to a high mountain or that with his human eyes
he should have seen all the kingdoms of the world—as if they were lying close to
the foot of a single mountain.”
467. Luke 22.42-47.
468. Luke 12.33.
FRAGMENTS OF THE CONTRA GALILAEOS 145

been given away, neither household nor family has aught to bargain
or to buy. And it goes without saying that in a city where everything
was given away [to the poor] the traders would become beggars.*®

FRAGMENT X:

How can you say, “The Word of God takes away sin,”*”° when he
causes many to commit the sin of killing their fathers and even their
children?*” So we are left with the choice, either to accept this super-
stition or to hold fast the traditions and the ancient beliefs which
our fathers held from the beginning. Is it not the same with Moses,
who was supposed to take away sin but fell prey to the charge that
he caused sin to increase?*”?

FRAGMENT XV

What was written of Israel the writer Matthew transferred to


Christ.*7? He does this to ridicule the naivete of those among the
gentiles who believed.

469. Cf. Ep. XL (to Hecebolius), where Julian quotes the same passage in con-
fiscating the property of the church at Edessa so that “their poverty may teach them
virtue.”
470. John 1.29, conflated with John 1.1.
471. Although many have seen this as a reference to Matt. 10.21 (the dissolu-
tion of families when confronted with the Gospel), Julian may have in view more
recent divisions in household during the time of the last persecutions under Dio-
cletian and Decius.
472. Lev. 16; cf. with Paul’s argument in Rom. 5.20, which Julian seems to
have in view in this passage. Julian seems to think that both Moses and Jesus failed
in their purpose by creating additional opportunities for wrongdoing, Moses by
imposing a law that proved too strict for the Jews, Jesus by offering unconditional
forgiveness without emphasizing moral responsibility.
473. Ad Christum transtulit. Preserved by Jerome in his Latin Commentary on
Hosea 3.11. The passages referred to are Hos. 11.1 and Matt. 2.15, the prophecy used
by Matthew to support the tradition of the Holy Family’s wandering in Egypt prior
to going to Nazareth.
MISCELLANEOUS AND
OCCASIONAL WRITINGS
ABOUT CHRISTIANITY
AND JUDAISM

INTRODUCTION

he following is a selection of Julian’s occasional writings con-


cerning Christianity and Judaism. Because of the sporadic
nature of these compositions, they do not offer a consistent line of
argumentation, but reflect rather the emperor's developing view of
the dangers posed by the Galileans at various junctures in his reign.
Broadly speaking, the epistles fall into three categories: those
written to confidants and teachers, such as those to Maximus (Ep.
XII, and perhaps LIX) and Theodorus (Ep. XVI and XX); those
written to deal with a particular social or religious crisis, such as
those to the Alexandrians (Ep. XLVII and XLVIII) , the Bostrians (Ep.
XLI), and the “people of Edessa” (Ep. XL); and those written to
exhibit or to withdraw philanthropy, such as the letter to the Jews
(Ep. LI) and the edict on Christian teaching (Ep. XXXVI).

1. CONCERNING CHRISTIAN TEACHERS


(AD 362, WRITTEN AT ANTIOCH)*”4
The line of argumentation in this letter is straightforward: Julian accuses
Christian teachers of rhetoric of hypocrisy for “thinking one thing” as
474, Ep. XXXVI.422A-424A (Hertlein, 36).

147
148 JULIAN’S AGAINST THE GALILEANS

believers and teaching their students to respect the works of writers like
Homer who were “consecrated to the gods” and the muses. The more
damaging accusation is that those who profit from hypocrisy appear to be
greedy and thus to violate the gospel’s endorsement of poverty as the key
to salvation.

A sound education, in my view, is not a hard-won command of


well-proportioned sentences and glibness of tongue, but depends on
a sound mind—a mind attuned to the good rather than evil, the mind
that comprehends the difference between what is noble and what is
base. A man who thinks in one way but teaches his pupils to think
another is a failure as a teacher because he fails to be an honest man.
Now if the contradiction between what a man holds to be true
and what he teaches for truth is a matter of no great significance, even
though this should be wrong, still it is, I think, in some sense a minor
matter. But in larger matters this is not the case: What if a man has
philosophical convictions but teaches the opposite of them—what ~
sort of behavior, but the conduct of a complete liar, should this be?
A man who praises what he knows is worthless and promotes what
he privately condemns is a hypocrite, a cheater, a seducer, a peddler
of worthless goods. Any man who desires to teach anything must first
of all be a man of virtue. Above all this means that a man <who
teaches> cannot keep hidden in his soul certain ideas which clash
with the ideas he openly discusses, for I believe that a teacher has a
special responsibility <for the souls> of the young.
And it is for this reason that a man who teaches the young the
art of speaking should be above all a man of real virtue, for it is
<through this art> that the young come to understand and discuss
the works of the ancients, the rhetoricians and grammarians, above
all the sophists. For it will be recalled that these philosophers
claimed to teach the apt use of language as well as other things,
especially moral and political philosophy. I salute all who bring
such high ambition with them to their teaching, of course, leaving
aside for the moment any discussion of their method. Yet I cannot
leave aside the fact that <there are those> who teach their students
many things that do not conform to their inmost beliefs.
MISCELLANEOUS AND OCCASIONAL WRITINGS 149

How can this be? Did not the gods bestow all learning on the
likes of Homer, Hesiod, Demosthenes, Herodotus, Thucydides,
Isocrates, Lysias? And did not these same men think themselves
inspired by Hermes or one of the Muses? How terrible that the men
who have the duty to expound these writers revile the gods whom
these writers loved!
Now while I find this position risible, I am not suggesting that the
teachers should change their views before being allowed to teach;
rather I offer them a choice: either not to teach what they do not find
worthy of belief, or, if they still wish to teach, require them to con-
vince their charges that none of the writers whose works they study—
Homer, Hesiod, and the rest—should be counted guilty of impiety or
Nnaivete or error with respect to the gods, as they have said.
After all, since it is by the work of these writers that they are able
to make a living and receive pay, is it not the most crass admission of
greed that they would sell their souls for a few drachmas? There were
many reasons why, until the present time, men were terrified of being
open and forthright in their private opinions about the gods and their
failure to be diligent in temple worship found many excuses. But I say
the gods have granted us liberty, and because this is so there is no
reason for men to teach what they do not believe to be true.
If, on the one hand, they think that the men whose works they
teach were wise men <and worth reading>, then their role as inter-
preters, indeed as prophets and wise men, would indicate they
should first of all be true imitators of the poets in their piety toward
the gods. But if, on the other hand, they believe our writers to have
been wrong in their belief about the gods, who are most worthy of
our reverence, then let them trundle off to the churches of the
Galileans to expound Matthew and Luke, since the Galileans them-
selves commend a man who refuses to worship at the temple. As for
myself, I can only hope that your ears and tongues might be, to use
your expression, “born again” in accordance with the tastes of those
who think and act in ways pleasing to me.
For all teachers of religion and arts let there be an ordinance to
this effect: Any boy who wishes to attend a school shall not be pre-
vented, for what sense is there in shutting the door of the true path
150 JULIAN’S AGAINST THE GALILEANS

<of knowledge> to boys still too ignorant to know what is good for
them, only to find themselves confused, overpowered <by the rhet-
oric of their masters>, and tricked into following beliefs that were not
those of their ancestors.*”° I admit that it might be more easily done
to cure these teachers as one cures the insane, against their will.
But as we grant pardon for people afflicted with the disease of
insanity, we might agree that the best way to cure the insanity of
Christianity is to teach rather than to punish the afflicted.

2. TO MAXIMUS (361 OR 362)476


This letter, written to the philosopher-theurgist Maximus, is primarily
important for understanding Julian's perception of the role of gods in
ordering fate and the significance of sacrifice and obedience in knowing,
and responding to, the divine will. The life of Maximus is known to us from
the writings of Eunapius, who reports that as a student Julian abandoned
study with Edesius of Pergamon to join Maximus at Ephesus. At least part
of the reason for this transfer of academic loyalty seems to have been Max-
imus’s reputation of receiving his philosophy through oracles. In any event
Julian remained faithful to his teacher, if not to every aspect of his
teaching, until his dying day and the philosopher was at least unofficially
the philosopher royal in Constantinople after 362 and probably sponsored
Julian's initiation into the cult of Mithras. Maximus was at. Julian's
deathbed. Immediately following the death of his patron, Maximus’s influ-
ence began to wane rapidly. Attempts to discredit his teaching as charla-
tanry blended with a vague charge of conspiracy which resulted in his exe-
cution, by order of Valens, in 371. In this letter Julian explains that he is
determined to restore the worship of the gods “in its original purity.”

My mind is a blur of confused thoughts and they well up to stop the


words in my throat, so that nothing can pass from my mouth in
order. Is it a psychic disturbance that causes this to happen, or some-

475. The biographical thrust of Julian’s words in this section deserves to be


underscored; see Athanassiadi, Julian: An Intellectual Biography, pp. 13-51. Cf. Ep.
VII.236A, B, where Julian deplores the teaching of the “utterly ignorant rhetors.”
476. Ep. VIII (Hertlein, 38).
MISCELLANEOUS AND OCCASIONAL WRITINGS 151

thing else?*”” Oh, pray that I can relate what I must tell you in some
sort of sequence—that is, after I have made due offering to the gods
who, in their mercy, have commanded me to write this to you and
may permit us to see each other soon. Now after I was made
emperor against my will, with the gods as my witnesses—that is,
having been proclaimed ruler in the usual way, I led the army
against the barbarians.*’® It was only after three months on this
expedition that I returned to Gaul, and after I arrived I would ask
anyone who would listen to be watchful for a man wearing a
philosopher's cloak or a scholar’s gown <and to> report any such
sighting at once. So I came to Besontio, a village recently improved
a bit, and in ancient times a great city rich in temples and sur-
rounded by strong walls as well as by the River Doub, which nature
adds to its fortification. Besontio rises like a sheer cliff from the
seas—towering in such a way that even the birds find it beyond
reach except in a few places where the river had circled around in
such a way as to create a shoreline.
As I came near to the city a man came along who, judging from
his long cloak and staff, was a cynic, and when I first glimpsed him
I thought, “Ah, it is my old friend Maximus!” When he came nearer,
I thought, “No, not Maximus, but someone who has been sent by
the teacher to speak with me.” But alas! The man was a friend of
mine, not the sort of thing I was hoping for, merely a frivolous
dream. A bit later on I thought that you would be so busy tending
to my affairs that I would scarcely find you beyond the shores of
Greece. Zeus, Helios, and Athene: witness all you mighty gods and
goddesses, how I did tremble for you all the way from Illyricum to
Gaul, how often I supplicated the gods for some sign of your wel-
fare—or rather, how often I asked others to ask the gods your fate
since I myself could bear to hear no bad news about what might
have happened to you. So others enquired for me, and the gods
showed clearly that there would be trouble—though nothing fatal,
477. Julian implies that he is suffering from the “theurgist’s disease,” with the
power of human speech being inadequate to convey the marvels which flow
through his mind.
478. Amm., 20.10: The campaign described took place in 360, when Julian
crossed the Rhine.
B32 JULIAN’S AGAINST THE GALILEANS

nothing to suggest the extent of the wicked plots that ans be exe-
cuted in my absence.*”?
You must forgive my passing over so much that has aeeen to
me—especially I wish you to know that I have received revelation
direct from the gods and was able to escape the death plot against
me that had been perpetrated by my enemies. But escape I did, and
even then I ordered no man’s death, nor did I confiscate anyone's
property, thinking the best punishment was prison for those whose
treachery I discovered first hand. Well, I perhaps should tell you
these things rather than write them down, but I know, dear friend,
you will want the news and be happy to get it.
Know that I worship the gods, I worship the gods publicly, and
all the soldiers returning with me do the same. I sacrifice oxen in
view of all; I offer tons of thanks offerings. And why? Because the
gods have revealed that it is their will that I should restore their wor-
ship in its original purity. | obey them, with absolutely no reserva-
tions. And they promise me return on my labors as long as I am
solicitous of their approval and dedicate myself to them.*®°

3. TO THEODORUS, ON HIS APPOINTMENT AS


HIGH PRIEST AND GUARDIAN
OF THE TEMPLES (363?)**!

The following extract illustrates Julian's concern for the operation of the
temples. Theodorus, known to us only from Julian’s correspondence,
seems to have been a Syrian Neoplatonist and a former schoolmate of
479. The reference is to the parlous state of Julian’s supporters in the East
during this campaign and following his quarrel with Constantius.
480. Julian expresses in this and a number of other letters his conviction that
the gods, especially Helios, had a direct role to play in his confrontation with Con-
stantius, whose sudden death was seen as an act of divine intercession (see Letter X,
‘TvAtav@ Oe@) adumbrated by particular signs; cf. Ep. X, addressed to Eutherius: “I
am alive and have been saved by the gods. Therefore offer sacrifices to them on my
behalf as thanks offerings.” It is especially irritating to Julian that the Christians
have foresworn even thanks-offerings which might be expected in view of his rela-
tively tolerant attitude toward certain churches and teachers, such as Aetius, whom
he addresses (Ep. XV) as an old friend.
481. Ep. XX. 452C-454D (Hertlein, 63).
MISCELLANEOUS AND OCCASIONAL WRITINGS 153

Julian. He studied under Maximus of Ephesus and may have been initi-
ated into the cult of Mithras together with the future emperor under
Maximus's tutelage. The importance of the letter consists in its clear
assertion of Julian's interest in the moral reform of the temple cults and
the priesthood, over which he asserts his rights as pontifex maximus.
While it would be too much to say that Julian is impressed with Chris-
tian concern for the poor, he recognizes charitable work as a means of
encouraging a greater asceticism among the pagan priests. Like Galen,
he regards the ethical practice of the Christians commendable while
holding their philosophical system in contempt.

. lam yielding to your care a trust forwhich I have the highest care
and concern, as it should be to all men everywhere, in the hope that
you will fulfill the task dutifully, and thereby give me comfort and
assurance of the life to come. <For dear Theodorus> I am not a man
who shares the view that the soul perishes before the body dies or
immediately thereafter. I trust no human authority in this matter but
rather believe what the gods have revealed, for is it not probable that
only they have true knowledge of what transpires <after death>, and
is it not certain that this knowledge far outweighs any human spec-
ulation of what is perfectly true. After all, we may be permitted to
speculate about such things, but the gods themselves have no need
for conjecture, for they know all there is to know.
What office do I give to you, you ask? It is the office of governor
of the temples of the East, and with it authority in each city to
appoint priests and determine their benefice. The qualities which
indicate your suitability for this task are primarily justice, kindness,
and benevolence toward all; for we know that a priest who acts in a
hateful way toward his brothers or with irreverance toward the gods
is unacceptable to all men, and that such a man must be publicly
reprimanded in no uncertain terms. I shall lay out in a separate letter
the particular requirements for priests, and this will soon be with
you,*®? while here I mean only to offer a few suggestions.
You will no doubt see the reason in obeying me in this matter,

482. According to Wright (III, p. 54 n. 1), these instructions are contained in


the so-called Letter to a Priest written after the burning of the temple of Apollo at
Daphne in 362.
154 JULIAN’S AGAINST THE GALILEANS

for when it comes to divine things I never act rashly or offhandedly


(God knows!), but always with deliberation and care. Of course I
resist unnecessary change in all of life's affairs, but more especially
am I cautious in matters concerning the gods. The traditions of our
forefathers are always before my eyes, because they are gifts of the
gods themselves; and they would not be of such incomparable value
had they been dreamed up by men.
But my dear Theodorus, in our time these customs and laws have
been sorely tested and violated. Love of pleasure and money has
become an end in itself. Accordingly, we need to look at <these cus-
toms> afresh, with new eyes.
First, then, let us recognize that among us there is a certain indif-
ference about the gods, and that the love of money and pleasure has
suffocated our reverence for the heavenly powers. <How clear this
became to me> when I observed that the adherents of the Jewish
religion would sooner suffer starvation and die than compromise
their faith by eating pork, or animals that have been strangled or had
their life squeezed away. But what of us? We are so indifferent to reli-
gious customs that we disregard the traditions of our forebears,
having forgotten whether any such rule was ever written down. But
the Jews for their part are-God-fearing.**> By this I mean they wor-
ship a god who is the most powerful, most good, and rules the cre-
ated order—a god who, as we recognize, is worshiped by us under
different names.*** And <the Jews> act in a way that accords with
our custom when they do not abuse <our> laws. Indeed, they are in
error about only one thing: they hold their god in such exclusive
regard that they refuse the counsel of other gods, whom they declare
to be the “gods of the gentiles only.”*°° This error they have driven
to such foolish extremes as only barbarians know how. But this
483. The standard designation: 8e00eBeic Svtec.
484. Cf. CG 354b: “And we worship the God of Abraham... . for he is great
and powerful.”
485. The issue of Jewish religious exclusivism on the Christian side is dis-
cussed by Paul, e.g., in Gal. 2-3, where the issue becomes whether Jews hold exclu-
sive title to the God of Israel. Paul's view is that the God of Abraham has been
“extended” to every nation (Gal 3.14) instrumentally, through faith in Jesus Christ.
Presumably what follows in this letter would have been Julian’s discussion of Chris-
tian abuse, as he saw it, of Jewish exclusivism.
MISCELLANEOUS AND OCCASIONAL WRITINGS 155

<error is nothing> as compared to the blasphemous sect of the


Galileans which spreads its disease. . . .48°

4. TO ARSACIUS, A HIGH PRIEST OF GALATIA‘S”


This letter is preserved by the historian Sozomen (5.16), who seems to
think it is particularly important for understanding Julian's attitude
toward Christian benevolence and charitable work. There is no good
reason on the basis of language or content to question its authenticity,
since elsewhere Julian offers grudging praise of Christian social practice
while ridiculing aspects of their belief.

Why is it that the Greek religion does not yet flourish? It is because
those who have the most to gain from its advancement are most
negligent in promoting that splendid worship, that glorious ritual
that puts all other forms of prayer to shame. May Adrasteia*®® excuse
what I am about to say—for it will seem boastful that so much
change could have been expected or even hoped for a short time
ago. These atheists excel in good works to strangers, their scrupulous
attention to the graves of the dead, and the feigned piety they dis-
play in their everyday lives. For my part I wish these habits could be
cultivated by all—not only you but every one of the priests in
Galatia. If they will not conform, then humiliate them publicly or
remove them from their religious office, especially if you witness
them, with their families and slaves, avoiding their duties to the
gods and chasing after the teaching of the atheists.*®’ If this fails,
486. The conclusion is lost, probably expunged by a zealous copyist “because
of some disrespectful reference to Christ” (Wright, III, 61 n. 2) or, more likely,
because it extolled Judaism at the expense of Christian doctrine. The reference to
Christianity as voonua signifies plague, but also passion (thus Aeschylus, Prometheus
225, 265), madness (thus Sophocles, Ajax 338) and disorder in a state (thus Plato,
Laws 906c).
487. Ep. XXII (Hertlein, 49).
488. Nemesis, the “goddess whom none may escape.”
489. Julian’s encouragement of imitating Christian practices is followed
immediately by the warning to avoid Christian teaching; see Gregory Naziazen,
Against Julian, Oration II]; Sozomen, 5.16. The use of aBeotnta. to refer to the Chris-
tians is a judicial aspersion in this instance.
156 JULIAN’S AGAINST THE GALILEANS

then say that a priest may not enter the theater, nor carouse at tav-
erns, nor engage in any trade that is beneath his dignity.
More than this, you should take care to hold those who take
your counsel in high regard—others, throw them out. You should
arrange for hostels to be erected in every city as a sign of our benev-
olence to those in need, and I do not mean Greeks only but those
who most need our help. . . .49° For it is a disgrace to us that no Jew
has to beg, and that every Galilean is ready to provide support for
our poor as well as their own, while men laugh that we cannot
muster aid for our people. Tell those who hold fast the faith of the
Greeks that they must provide for the downcast—offering the first-
fruits to the gods; and tell those same believers that the doing of
good works of this sort has been our practice from time immemo-
rial. Does not Homer have Eumaeus say, “Stranger, I can imagine no
man lower than you, but for my part even if one should come it is
not permitted for me to dishonor a stranger. For all strangers and
beggars come from Zeus. Every gift, however small, is precious.”*7
<As charity belongs by right to us> let us not permit others to
outdo us in such service; for insofar as we are remiss in this way, so
also we show dishonor to the gods. I shall be glad indeed to hear
that you are carrying out these orders.*?”

5. TO PHOTINUS (CA. 362)*

Photinus is mentioned by Julian in Contra Galilaeos (262C) and was


bishop of Sirmium in 351 during Constantius’s time in the city. His reli-
gious views were decidedly controversial and Constantius sought his dep-
osition and banishment (Sozomen, 4.6). Photinus has the distinction of
having alienated both the Arian and the Nicene parties with his theology,

490. In the verses here omitted Julian itemizes the supplies of corn and wine
that are to be allocated throughout Galatia.
491. Odyssey 14.56.
492. The remaining section of the letter contains instructions for dealing with
soldiers and government officials and the reverence due to the Magna Mater (untmp
tov Se@v) in relation to the the temples.
493. Ep. LV (Hertlein, 79), preserved in a Latin version by Facundus Hermia-
nensis (546?); see Neumann, Contra Christanos (1880), p. 5.
MISCELLANEOUS AND OCCASIONAL WRITINGS 157

but his insistence on the humanity of Jesus seems to have won favor with
Julian. In this letter the emperor praises his views in contrast to the ideas
of Diodorus, bishop of Tarsus, who held extreme consubstantiationist
views concerning the divinity of Jesus. Only a portion of this letter, inter-
spersed with its editor's stitches, can be considered authentic (see note).

My dear Photinus, you alone hold what I think is probably true and
come near salvation: that one we believe to be a god cannot be
brought forth from a womb. But Diodorus, the trickster priest of the
Nazarenes, who tries to make a case for this ridiculous theory about
the womb*”* using his flim-flam and sleight of hand <to do it> is as
clever a sophist of the creed of these bumpkins as you are likely to find.
... And if the gods and goddesses and the Muses and Fortuna will only
give me their assistance, I will show that <Diodorus> is a really a
deceiver, an abuser of the laws and customs of the great mysteries and
the mysteries of the gods of the underworld, and that this upstart
Galilean god of his, whom Diodutrus calls “eternal” by preaching some
preposterous fable, has surely been stripped of any divinity ascribed to
him through his humiliating death and burial....<For this same
Diodorus> sailed off to Athens without rhyme or reason, then brashly
seized on the study of philosophy, then literature, and by next
becoming proficient in rhetoric, he trained his vile tongue to utter
blasphemy against the gods of heaven. Ignorant of the mysteries <of
our religion> he drank his fill of the degenerate and stupid nonsense
preached by the ignorant and vile creed-making fishermen. [And for
this he has been punished by the gods for many years, by a gradual
wasting of the lungs that has now spread to his whole body. His face
is now sallow, his body furrowed with scars. But I stress, this <disease>
is not a philosopher's plight, as he wants it to appear when he is
preaching his deceit, but rather a sign of the punishment the gods have
chosen for him in direct proportion to the crime; and now he must live
out his days, painfully and bitterly, a man pale and diseased. |*””
494. Either the view that Mary’s virginity was miraculously preserved or that
she is worthy of the title “god-bearer.”
495. While there is no reason to challenge the central complaint of this letter
as belonging to Julian, the latter portion of the epistle (bracketed) with its obvious
indulgence in the physical description of Diodorus’s disease suggests a sixth-century
Latin provenance, and can probably be attributed to Facundus himself.
158 JULIAN’S AGAINST THE GALILEANS

6. TO HECEBOLIUS*”®

Hecebolius (not to be confused with a sophist of the same name) was a


treasury official in Edessa in northern Mesopotamia. The city had been
a hotbed of heresy, especially gnosticism, since the second century. To
encourage the dominance of the Arian party there, Constantius provided
the sect with the basilica church of St Thomas as a trophy, which had the
effect of further alienating the minority sect of Valentinian Gnostics in
the city. Sozomen (6.1) suggests that Julian was particularly contemp-
tuous of Edessa, whose Christianity seems to have been reinforced by a
persecution of the non-Arian party under Valens (cf. Soc., HE, 4.18). In
this letter Julian is determined to hold the Edessene Christians to their
profession of poverty by rescinding the benefits which his predecessor had
bestowed on the church. The epistle is also significant as being Julian's
clearest expression of the benevolence he feels he has exercised toward the
Galileans. In contrast, Julian asks the recipient to consider the intoler-
ance of the Christians toward the religion of the Valentinians.

I have conducted myself with all the kindness and benevolence the
Galileans deserve, not doing them violence, or dragging them into
the temples, or threatening injury, or coercion of any sort. But the
members of the Arian church, with all the spiteful arrogance that
befits their wealth, have shown only violence to the followers of
Valentinus, and have done things in Edessa of such wanton reckless-
ness that they cannot be imagined in a well-ordered city. So I
decree—because by their most excellent law they are required to sell
all they have and give it to the poor so that they can fly more easily
to their kingdom in the skies—and I order, in order to assist them in
their noble effort, that all the funds belonging to the people of the
church in Edessa be seized and distributed to the soldiers; and I
order that the valuables belonging to the churches be confiscated
and handed over to me. In this way, poverty will teach them good
conduct, but most important, they will not be deprived of that heav-
enly home for which they yearn. I also order the citizens of Edessa
to cease from all quarrels and disputes—for if you do not you will

496. The date of the letter is uncertain, but the “policies” alluded to suggest
the winter of 362. Ep. XL (Hertlein, 43).
MISCELLANEOUS AND OCCASIONAL WRITINGS 159

test my benevolence, and find yourselves, whether by sword or exile,


paying the price for disturbing the peace of the commonwealth.

7. TO THE BOSTRIANS?°”

Bostra is described by Ammianus (14.8.13) as a large fortified city of


Arabia. As in the letter to Hecebolius, Julian here expresses dismay that
the Christians have yet to show gratitude for his lenient policies toward
them. He is especially annoyed at threatening language used by their
bishop, Titus, who has taunted Julian that Christians keep the peace
because the church enjoins them to do so and not because they are
ordered to do so by the emperor.

I would have expected that these leaders of the Galileans would be


more grateful to me than to my predecessor for the way I have gov-
erned the empire, since in his day vast numbers of them were cast
into exile, tried in the courts, thrown into prison—not to mention
the congregations of those called “heretics” who were simply slaugh-
tered—as in Samosata, Cyzicus, Paphlagonia, Bithynia, Galatia, and
the villages where tribes were looted and dispersed. In my reign,
none of this has happened: the ones in exile have been called home.
Those whose property had been confiscated have by my own law had
it restored. But <these Galileans> are so inflamed with madness and
steeped in stupidity that they fault me for refusing to let them act like
tyrants, <for condemning> their conduct toward one another, and
toward those of us who have always revered the gods. They do every-
thing possible to show their arrogance—turning the people toward
rebellion, while they themselves deride the gods with acts of impiety.
As for myself, I do not permit a single <Christian> to be dragged
against his will to our altars. Indeed, I have advised that any man
who wishes willingly to take part in our ceremonies and libations
should first of all perform a ritual of purification and supplicate the
gods who <help us to> avert evil. But far be it from me to want any
of these superstitious folk to partake of sacrifices that we hold sacred,

497. Or, Bosrians; cited by Sozomen 5.15. Ep. XLI (Hertlein, 52).
160 JULIAN’S AGAINST THE GALILEANS

at least until such time as he has purified his soul in the right way,
through prayer to the gods and the purification of his body.
It is obvious that many people have been led astray by the
clergy,4°8 who advocate rebellion against us because we have taken
away certain privileges. It is as if those who behaved like tyrants
before now are not content with the punishments they have
received, but long for the powers they used to enjoy when they sat
as judges, fashioned wills, stole other men’s inheritances and put it
in their pocket, pulled every string to bring on disorder, and to add
fuel to the fire, performed the even greater sin of actually leading the
community into chaos through their wicked designs.
I have, therefore, decided to announce to citizens everywhere
and by this proclamation to make it known that no one may get
involved in the feuds of the clerics, or be persuaded by them to take
stones in their hand and hurl them at those with authority. It is per-
mitted for them to hold meetings whenever they wish and to offer
prayers for their own well-being, as is their custom. But they must —
refuse if the clerics try to persuade them to take up sides on their
behalf in any sort of dispute, for if they do they will no longer be free
from punishment.
I have the city and people of Bostra especially in view because
their bishop Titus and his clergy, according to reports I have seen,
have made certain charges against other Christians. This gives the
impression that when the population began to riot, it was the Chris-
tians who kept the peace and saved the state from collapse. I have
added the very words <of Titus> to my decree, where he dared to
write “The Christians are a match for the Hellenes in numbers, but
they are restrained because of our teaching that no one must disturb
the peace at any time.” This is how your bishop thinks of you. Notice
how he cheapens your behavior by saying it was not your choice but
his words that kept you from violence against your will. So I say, use
this same free will, take your accuser and cast him out of the city.4”
498. The priests and teachers, especially those who are literate or those
engaged in theological dispute.
499. Socrates, HE 3.25 mentions Titus still holding power in Bostra under
Julian’s successor, Jovian. Either Titus was restored in 363 or Julian’s advice was not
followed.
MISCELLANEOUS AND OCCASIONAL WRITINGS 161

But by the same reasoning, live in peace with one another, let no
man be quarrelsome or act unfairly. <To> those who have left the
truth behind, I say do not harass those who give the gods their due
in prayer and sacrifice, as we have been taught to do since earliest
times. And those of you who worship the gods, do not plunder the
houses of those who have erred against truth through ignorance
rather than on purpose. How often have I admonished those who
hold fast to true doctrine not to harm the Galileans or abuse or
insult them but learn to persuade them with reason. For such men
deserve pity more than scorn, since only pity is a sound response to
their sad state. <And you know> that the greatest blessing is this: rev-
erence for the gods, just as the greatest evil is irreverence. And those
who have turned away from the gods in favor of corpses and relics
must pay a price.°°° <By analogy> we suffer in sympathy when there
are people suffering from disease, and we exult with those who are
set free of their sickness by the will of the gods.

8. ON FUNERALS™! (363)
The edict on funerals represents Julian's most explicit, if indirect, attempt
to discourage Christian burial practices. His view that the Christians had
“filled the world with graves and sepulchres,” an allusion to the martyr
cults, the celebration of anniversaries, and the practice of preserving relics
and building shrines at or near the site of the graves of saints, seems to
occasion the ruling. The edict is not directly aimed at Christians but
includes them in its scope. The primary theme of the letter is that death
harmonizes with night and rest, and that the display of the corpse during
daylight is an affront to Helios. Julian ignores, or misses, the Easter sym-
bolism which associated resurrection and sunrise in the gospel tradition.
On the sources of Julian’s teaching, see Eunapius, Life of Iamblichus.
Julian was especially contemptuous of the displays of enthusiasm, often
leading to public demonstrations, at Christian funerals, notably the inter-
ment of the bones of St Babylas in Antioch (Philostorgius 7.8; Soz. HE
5.19; Julian, Misopogon 361B).
500. Julian regards the cult of Christ and the martyrs as a form of perversion;
cf. CG 335B; Misopogon 361B.
501. Ep. LVI (Hertlein, 77).
162 JULIAN’S AGAINST THE GALILEANS

As a duty, I have decided after much reflection to confirm in law the


ancient custom of our ancestors, because when these framers of wise
laws, these ancestors of ours, considered the matter, they held that
the difference between life and death is the greatest of differences.
Accordingly, they taught that each of these states should be governed
by practices and rituals appropriate to it. Now, <they taught> that
death is endless rest—the “bronze sleep” sung by Homer—but
waking life is different—full unequally of pain and pleasure, hard-
ship and prosperity. And so they required that the ceremonies asso-
ciated with the dead should be set apart from the business and
traffic of daily life and conducted privately. They taught that the gods
are the beginning and the end of all things: while we live, we belong
to the gods; when we die, we return to the gods. It is not important
to decide here about such matters, or to know whether both life and
death are in the hands of the same gods or whether certain gods con-
cern themselves with the living and other gods with the dead. Yet
when we consider that the Sun, by being absent or present, is the —
cause of day and night and of winter and summer, we say that the
most venerable of the gods, the one who is before all things and
from whom all things proceed, allotted rulers over the living and
lords over the dead. Therefore, we should give to each of these
domains what is appropriate to them in their kind; and we should
strive to emulate in the working of our daily lives the harmony of
the gods concerning the manner in which things exist.
And so, death is a rest. Night is in harmony with rest. It is there-
fore fitting that the burial of the dead should be conducted at night,
and that we should prohibit the business of burial, or anything asso-
ciated with it, to go on during the day. Consider that <during the
day> people are occupied with all sorts of business, the streets full
of men on their way to the law-courts, or to or from market, or sit-
ting at a stall plying their craft or trade, or visiting the temples to ask
for a sign from the gods. Then some small mob who have just laid
a body on the bier shove into the midst of those who have other
things to do.*” The situation is intolerable in every way! The ones

502. eita ovd« ota cities. ... Julian means the Christians who intrude on
commerce to call attention to their grief.
MISCELLANEOUS AND OCCASIONAL WRITINGS 163

who find themselves shoved aside are filled with disgust—some


because they see it an evil omen, while for others it means they will
not be able to go to the temples until they have washed themselves
of the uncleanness they have incurred. <For we hold> that it is not
permitted to enter the presence of the gods, the source of life and far
removed from decay, after this sort of encounter.
And what can be worse than this? I shall tell you: The doors of
the temples are often open, the very gates of the gods, and it some-
times happens that at the very moment someone inside is pouring
libations, sacrificing or praying, behold! a procession carrying a
corpse wanders by, close to the temple itself, their shrill lamenta-
tions floating even above the altar as an ill omen.
So we must understand that the activities that belong to day and
night have always been separated more than other things. It is for
that reason that burial has been removed from the <activity> of the
day and reserved for night. And it makes little sense to ridicule the
wearing of white for mourning, then to bury the dead in the full
light of day. If the former practice was better because it did not
offend any of the gods, this new practice°®’ can only be regarded as
an insult to all of the gods. For <to bury the body by day> is to
ascribe death to the Olympian gods, and to deprive the gods of the
underworld of their due—if that indeed is what these lords and
guardians of soul are to be called. I do know that the <priests> who
are expert in their knowledge of these practices say that it is right
only to perform the rituals of the infernal gods at night—in no case
before the tenth hour of the day. And if this is the appropriate time
for the worship of these gods, we will not allow another time for cer-
emonies for the dead.
My words should be enough for those who choose to obey. For
those in error, let them apply what they have learned and choose the
better path. But if any man chooses obstinacy, such that he needs
threat or punishment for persuasion, he should know that the
penalty for disobedience will be of the severest kind. He shall not
perform the offices of the dead before the tenth hour of the day, or

503. Of the Christians, but the implication is that pagans have been influ-
enced by Christian practice as well; otherwise the edict would make scant sense.
164 JULIAN’S AGAINST THE GALILEANS

carry the dead person through the city. These things are to be done
only at sunset and before sunrise, reserving the glorious day for pure
deeds and the <worship of> the shining gods of Olympus.
LETTERS CONCERNING
ATHANASIUS AND THE
ALEXANDRIANS

INTRODUCTION

|hree of Julian’s surviving epistles concern Athanasius of


Alexandria, the teacher and bishop whose relentless stand
against the Arian heresy was both a distraction and an annoyance for
the new emperor from the start, as it had been for the Arian sympa-
thizer Constantius before him. During the years of his nominal
tenure of the see of Alexandria (326-373), Athanasius served twenty
years in exile and five times was forced into hiding: once under Con-
stantine, twice under Constantius, once under Julian (362-363),
and lastly under Valens (himself an Arian) in 367. Valens was the
last official patron of the dying Arian cause, whose demise was
ensured by unremitting theological attacks in the east (by the Cap-
padocians) and by western bishops who had encountered its equal
in the form of the modalist and monarchian heresies of the third
and early fourth century. More troubling still was Athanasius’s posi-
tion toward the imperium, which more than any Christian writer of
his day he understood as subsumed under the divine sovereignty
and its hierarchical embodiment, the bishop-theologue.*™ In this
calculus, an emperor like Constantius can be dubbed the anti-Christ
(Historia Arianorum, p. 77) and the Christian bishop has the right of
504. See Letter LVI (Athanasius to Jovian), p. 26.813.

165
166 JULIAN’S AGAINST THE GALILEANS

remonstrance against an emperor. These relatively early assertions of


the episcopal prerogatives will become increasingly decisive in the
contest between church and empire in the fifth century, culminating
in such ruses as the eighth-century forgery known as the Donation
of Constantine, which alleged a perpetual grant of sovereignty to the
bishop of Rome over all the churches of the East as well as those of
the west.
On December 24, 361, the Arian bishop George of Cappa-
docia—a hugely unpopular figure among both the orthodox and the
Arians—was murdered. Initial blame fell on Athanasius who, it was
said, enflamed public opinion against his enemy. The innocence of
Athanasius in the affair is best supported by Julian’s silence in the
matter. While roundly condemning the. attacks as an act of bar-
barism, the emperor—who detested Athanasius and repeatedly
pleaded with the Alexandrians to throw off the superstitions of the
Galileans and return to the faith of their political founder—does not
seem to have blamed Athanasius for the death of George.*® *
Julian early in 362 proclaimed an amnesty for all non-Arian
bishops who had been victims of legislation under Constantius, and
Athanasius, prematurely as it worked out, returned to his see. The
reasons for the amnesty are difficult to discern. It is clear that Julian
had no wish to take theological sides in the debates of the Galileans,
and his contempt for Alexandrian theological orthodoxy as person-
ified in the teaching of Athanasius was consistent with his claim that
he had “given up Christianity” once and for all in 350.°°° On their
return, religious conflict between the orthodox ecclesiastics and
their Arian opponents spread rapidly, making Alexandria the focus
of sectarian violence that would last until some time after Julian’s
death and the solidifying of Nicene orthodoxy.
Christian writers supposed that the reason for the amnesty had
been to create religious war among the Christian sects, but if this is
so there is nothing in the surviving correspondence to suggest cyni-

505. See Socrates, HE 3.31 and Soz., HE 3.7, who also suggest that Athanasius
was innocent and the victim of Arian malice. Philostorgios 7.2, however, implicates
Julian directly in a conspiracy against the Arian bishop.
506. Ep. XLVII.
ATHANASIUS AND THE ALEXANDRIANS Loy

cism on the emperor's part. Rather, Julian, while alert to the possi-
bility of trouble, seems to have regarded this intervention as an exer-
cise of his role as pontifex maximus—if not an attempt to restore the
theological balance of Constantine’s day, then surely as a posthu-
mous rebuke to Constantius and his policies. Julian first scolds the
Alexandrians for their role in the death of bishop George;>” then
writes again demanding that the library of the dead bishop be dis-
patched to him.*° It is thought that the tract Against the Galileans
(composed the following winter at Antioch) was written with
George’s books at hand, though this is not certain, and parts of the
treatise had almost certainly taken shape before Julian received
George’s library.
When word reached Julian of Athanasius’s return to Alexandria,
he immediately ordered the prefect Ecdicius Olympus to expel the
bishop from his see before December 1 (Ep. XXIV). Athanasius com-
plied by October 23, 361 (“It is but a little cloud and will pass,” he
is reported by Sozomen to have said.°°’), but by the autumn of 362
the Alexandrians were calling for the return of the bishop, sending a
legate to Julian with a petition which the emperor ungraciously
refused. Athanasius this time went into hiding near Memphis rather
than into exile and resumed his duties on hearing in 363 of Julian’s
death—which he is reported to have greeted with a smile.

* * *

LETTER 21
JULIAN CAESAR, THE AUGUSTUS, TO THE
PEOPLE OF ALEXANDRIA, <GREETINGS>*”

It may be that you do not honor the memory of the founder of your
city, Alexander, and that you revile the great and holy god Serapis.
But how can it be that you gave no thought to your city, your human
507. Ep. XXI.
508. Ep. XXIII.
509. Soz., HE 5.15.
510. Hertlein, 10.
168 JULIAN’S AGAINST THE GALILEANS

bond, your regard for the decency of your society?" And what of
me, for I must add—you gave no thought to me, even though the
gods, yes your own god Serapis among them—decreed that I should
rule over the whole world. The fate of the offenders should by right
have been left to me: this and only this was the right course. And yet,
anger and passion led you from the right course, as often it does:
“Tossing reason out the door and doing terrible things.”°'*
You managed to restrain your passion at first, but then gave way
to reckless passion. You tossed aside the wise decision you had made,
and yes—you were not even ashamed—the lot of you to do the same
abominable things you are used to seeing done by your enemies.
Tell me in the name of the most holy Serapis, what crime did you
find in George that aroused such hatred in-you*!? The reply, no doubt,
is that he brought Constantius, now of blessed memory, to the boil:
he sent an army into your blessed city, and the captain of this regi-
ment?4 took hold of the sacred shrine of the god, stripped it of orna-
ments and statues, and seized the offerings from within the temples.
You were right to be provoked, and right as well to try to save the treas-
ures of the god from further insult. But Artemius was loathe to accept
this—more afraid of George, I think, than of Constantius—and so
sent soldiers against you illegally, unjustly, and blasphemously.*!> For
511. I.e., in the murder of Bishop George.
512. A fragment of Melanthius quoted by Plutarch, On the Restraint of Anger,
p. 453.
513. Julian’s suggestion that the intolerance of the pagans was as extreme as
that of the Christians is also conveyed by the Church writers, notably Socrates (HE
3.2), who attributes the murder to a mob incensed at George’s attempts to erect a
church on the site of a Mithraic adytum: “The pagans, dragging George outside the
church, fastened him to a camel, and when they had torn him to pieces, they
burned him together with the camel.” The invocation of Serapis seems to support
Socrates’ view—that the populace rather than supporters of Athanasius were to
blame for the murder. In the Arian legend, later circulated throughout the Eastern
church in variants, George is represented as the warrior defending truth against the
dragon-wizard Athanasius.
514. The prefect of Egypt Artemius, executed in 362 by Julian’s command.
515. Sozomen’s balanced view is worth noting: “George compelled both par-
ties to offer worship in the mode he decided and where opposition was made he
enforced obedience by compulsion. He was hated by the rulers because he scorned
them and was giving orders to the officers; and the multitude detested him on
account of his tyranny and because his power was greater than all the rest. The
ATHANASIUS AND THE ALEXANDRIANS 169

it was George who was keeping an eye on Artemius, to keep him, |


think, from acting too moderately, too lawfully—but giving him free
range to be a tyrant toward you.
So, you will say that you had reason to be outraged against
George and for this reason you were entitled to desecrate the holy
city yet again,”'® preferring violence to the verdict of the judges. Had
you chosen the latter course, blood would not be on your hands and
this would not be a case of murder and lawlessness, but a proper
case at law that would leave you clear from guilt in the matter. The
court would have punished an impious man for inexpiable crimes,
and would have taken to task <in the same suit> all those who
ignore the gods. It would have punished those who humiliate
<ancient> cities and communities of men in the belief that cruel
action toward others is the surety of their power.
Please recall that this letter to you is the second, and compare
my words with those I used in the first, written not long ago. With
what words of praise did I féte you then! Even now I want to praise
you, but how can I when you have broken the law? You have torn a
human being into pieces as dogs would tear a wolf, and you raised
to the gods hands still steeped in his blood. And what have you to
say? “George deserved it!” And well might I agree with you—per-
haps he deserved even worse punishment and cruel treatment—and
I might even grant you that, but when you say he deserved it at your
hands, then I will not agree. You have laws, laws which must be hon-
ored by one and all, and preserved. It happens of course that on
occasion an individual will break one or another of these laws, but
on the whole the state is tranquil, well-ordered, because it is recog-
nized that laws were established from the beginning in wisdom and
are not to be treated with contempt.
In any case you are fortunate, men of Alexandria, that this crime
happened in my time, for my reverence for the god and respect for
my uncle and namesake who ruled your city and the whole of
pagans regarded him with greater aversion than the Christians because he prohib-
ited them from offering sacrifice” (Soz., HE 4.30).
516. It is clear that the desecration of the city is Julian’s reason for repri-
manding the Alexandrians as gently as he does, not the punishment of George
whose “inexpiable crimes” he acknowledges.
170 JULIAN’S AGAINST THE GALILEANS

Egypt®!7 causes me to think of you as brothers. If 1 dealt with you in


the way a power that wishes to be respected should—that is, as a
tyrannical and unswerving government would—I could never over-
look the frenzied act of the mob: I would rather root it out and
purge it, like a disease, with the most bitter medicine. But as I have
said in this very letter, I choose instead to give you a milder tonic:
the medicine of caution and argument.
And from these, if you are, as you claim, Greeks at heart, you will
derive greater benefit and be more persuaded. I know and believe
that to this day there remain in your blood the habits and customs,
the noble imprint, of our illustrious heritage. Let this word be pub-
lished throughout Alexandria.

LETTER 23
TO ECDICIUS, PREFECT
(ON THE DISPOSITION OF GEORGE'S BOOKS)*!®

While some men have a passion for horses and others for birds, and
still others for beasts, for my part I have had, from my earliest days,
a passion for possessing books. So it would be foolish of me.to
allow <these books> to be stolen by unscrupulous men who love
money and gold only. Wealth cannot satisfy them, so by evil designs
they plot to steal these as well. Please therefore do me the favor of
locating those books which belonged to George, for so many of
these in his house were on philosophy and many on rhetoric; and
of course many were on the doctrines of the impious Galileans.
These latter <books> I would ask to be destroyed were it not that
along with them some more useful books might be destroyed as
well. So let the search proceed with the greatest care. Give the job to
George's secretary, and tell him that if he searches honestly and well
he will have his freedom; if not, he will be put to the test and tor-
tured. <Let him understand> that I know in many cases what books

517. The elder Julian (cf. Misopogon 365c) held office in Egypt during the reign
of Constantius.
518. Hertlein, 9.
ATHANASIUS AND THE ALEXANDRIANS 171

George possessed—not all, but many, for he lent me some to copy


when I was in Cappadocia (which I returned).

LETTER 24
AN EDICT TO THE ALEXANDRIANS
<REGARDING ATHANASIUS>*"

One who has been banished on so many occasions by so many


emperors issuing so many decrees might have waited at least for a
further edict before being so rash as to return to his own country.
Instead, he has displayed again his tempestuous nature, and treated
our laws as though they had never been written. We have not to this
day extended to the Galileans deported into exile under our blessed
predecessor Constantius the right to return to their churches; rather,
only permission to enter their own countries. But the insolent
Athanasius, as always, with his usual disregard for law, has again
been puffed up, and again grabbed what <they call> the “episcopal
seat.”°?° No wonder the god-fearing people of Alexandria are wor-
ried. We solemnly warn him to leave the city immediately—on the
very day he receives this grant of clemency—or else his punishment
will be all the more severe.*”

LETTER 46 TO ECDICIUS, PREFECT IN EGYPT°2


Even though you have failed me in some matters, you should not
fail me in this request: a report was to have been made on that
enemy of the gods, Athanasius. That was my decree, published some
time ago.°”? My decree was sworn justly before the mighty Serapis,
519. Hertlein, 26.
520. O@pdovoc.
521. In consequence of this edict Athanasius went into hiding outside Alexan-
dria, provoking the language of Letter XLVI.
522. Hertlein, 50.
523. October 24 is the date in which Athanasius went into exile in upper
Egypt; see Socrates, HE 3.14; Soz., HE 5.15. Sozomen also offers this grudging
tribute to Julian’s restraint: “Although Julian was anxious to advance paganism by
LZ JULIAN’S AGAINST THE GALILEANS

and it held that if Athanasius does not leave the city—rather depart
all Egypt before the December kalends, I shall levy on-you and your
cohort a fine of one hundred pounds of gold. You know how slow I
am to condemn; know then how very much slower I am to repent of
what I have condemned. (And in my own hand) It torments me to see
my orders flouted. By the gods, there is nothing that would cheer me
more than to receive word that Athanasius is driven away beyond
the borders of Egypt. The vile old wretch! He has the conceit to bap-
tize even well-born Greek women during my reign—so I| say: Out
with him!°*4

LETTER 47 TO THE PEOPLE OF ALEXANDRIA”?

How is it that you have the nerve to ask for the return of Athanasius?
Is it reasonable that you should ask this of me? Was your founder a
Galilean? Did he violate the law as the Galileans have violated their —
own,°*° for which they now pay the price—the price of defying the
very law they say they live by, the price of innovation and new
teaching? No, it was Alexander who founded your city; it is Serapis
whom you should worship as lord and patron, together with his fair
consort Isis, the thrice blessed lady, the queen of Egypt. < > But you
do not <represent> the city of Alexander, only the diseased portion
that infects the whole and pretends to speak for the whole.>?7
Really, I am overcome with shame, by the gods, you men of
every means, yet he deemed it the height of impudence to employ force or
vengeance against those who refused to sacrifice. ... He did not even forbid them
to assemble together for worship, as he was aware that when freedom of the will is
called into question restraint is utterly useless.”
524. It pained a copyist to read Julian’s reference to Athanasius, so with Chris-
tian fervor the Neapolitanus MS has the following scribble: waxdpiog odtog Kvav
[Lape KOL TploKatdpate napaPata Kai tpiodOAte (“He is a saint, you dog of an apos-
tate—triple cursed and thrice wretched”).
525. Hertlein, 51; lacunae and editorial omissions by a Christian editor (see
note 521 on the Neapolitanus MS) make the translation of the first section problem-
atical. See Hertlein’s notes.
526. That is, the law of the Old Testament; see Against the Galileans 236-37,
306ff.
527. Disease: GAAG 1 vooodv Epos Exrgnierv . . . (437).
ATHANASIUS AND THE ALEXANDRIANS 1/3

Alexandria, when I consider that even a single Alexandrian professes


that he is a Galilean. Is it forgotten that the ancestors of the true
Hebrews were the slaves of the Egyptians in days gone by? But today
the men of Alexandria subject themselves to the race of men who were
conquered by the founder of their city, or rather to their inferiors who
have cast off the traditions of their ancestors and taught you to forsake
your sacred traditions as willing slaves of superstition. And today you
have no memory of the days when all Egypt had communion with the
gods and when all were blessed accordingly. Yet this teaching <of the
Galileans> was introduced among you only yesterday, and what good
has it been to the city? Your founder was a pious man, Alexander of
Macedon, who resembles the Galileans in no way, no, nor any Jews
for that matter—though, to be fair, | must admit that the Jews show
themselves to be far superior to the Galileans.
Was it not Ptolemy the son of Lagos who overcame the Jews°*®
while Alexander himself would have been a match for the Romans
any day. Then there were the Ptolemies who came after Alexander's
time, those who nurtured the city as though she were their very
daughter? what of them? I can tell you, it wasn’t through the
preaching of Jesus that you have come to be famous. Nor was it
through the preaching of the detested Galileans that you learned the
arts of government, in which you excel and which have caused For-
tune herself to grant you favor. Now, finally, when we Romans took
on the responsibility for your welfare, taking you from the Ptolemies
(who had become bad rulers) it was Augustus himself who visited
you and spoke to you in these words: “Men of Alexandria, I absolve
this city of its sins, because I revere and honor the great god Serapis.
And <also I do this> because I honor the people of this great city.
And I forgive you because of the renown of Aureius, my friend.”
Now this Aureius was a citizen of Alexandria like yourselves, a friend
of Caesar no less, a philosopher by profession.*”?
So this is the bounty which, in brief scope, the gods of Olympus
528. According to Josephus, Wars, 1.12.1, Ptolemy I took Jerusalem and then
took numbers of Jews captive into Egypt.
529. Cited by Julian in his letter to Themistius as a stoic philosopher who
turned down a praefecture. He is known to Philostratus and to Seneca as well; in
the latter (Dialogues 6.4) he is said to have been a counselor to the empress Livia.
174 JULIAN’S AGAINST THE GALILEANS

have showered on your city—and I neglect many others I could


name if there were time to recount them all. But more than these are
the many blessings that the known gods bestow on all of us every
day—blessings confined neither to a single race nor to a few chosen
people nor to a particular city,*° but given freely to everyone the
whole world over. How can you deny these? Really, are you blind to
the rays that pour forth from the great god Helios? Are you simply
insensible to the fact that winter and summer are his doing? Or that
plant and animal life take their refreshment from him? What then
do you think of the blessings showered on your city by that great
goddess who is generated by him and comes through him®*'—that
very Selene who is creator of the universe.°?* Oh, but you persist in
refusing to adore any one of our gods, preferring to think that a man
whom you have never seen (nor those before you) ought to be wor-
shiped as God the Word—I mean Jesus. What of the god to whom
the whole of humanity has looked from the very beginning, the god
worshiped and acclaimed by all of us, the god who causes our devo- ~
tion to prosper? I mean Helios, the living and intelligible image of
the invisible god.°? This <unseen principle> is the source of the
good, for he is soul and intelligence itself < >.>4
I beg you, do as I say: return to the straight path, that of the
truth. Accept what I have to say as one who, until his twentieth year,
walked the path <of error> that you now walk. But for twelve years,

530. A reference to the ethnocentrism of the Jews and the Christian exploita-
tion of the theme of having been singled out by God for a special destiny.
531. thy dé €€ avtod Kai nap avdtod Snuovpyov . . . Selene is advertised in lan-
guage deliberately echoing Christian formulations of the son’s generation by the
father.
532. The allusion seems to be to the logos theory of John 1.1-5, which Julian
seems to view as a perversion of the Helios cosmology.
533. Wright's translation diminishes the Neoplatonic thrust of Julian’s allu-
sion: The unseen Helios is voepoc—intellectual (Plato would have said an object of
intellect); whereas the intelligible principle is vontéc, or “known”—a likeness or
image. Julian frequently understands the celestial bodies as living likenesses of the
unseen realities, which can be named differently and indeed venerated differently
from region to region in the Empire (Or. IV.133C).
534. An erasure in the text, probably a Christian copyist objecting to the close
resemblance between Julian's platonizing theory and the doctrine of the Fourth
Gospel.
ATHANASIUS AND THE ALEXANDRIANS 175

thanks to the gods, I have not swerved from the path of truth as I
now know it to be.>*5
May it please you to take me at my word. I should be very happy
if you did. But if you choose to continue on the course of error,
drinking in the superstition and the false teachings of men of no
repute, then at least spare me your craving for Athanasius. Are there
not plenty of his pupils roundabout who can satisfy your itching
ears, bent to hear again his mischievous words? Oh, had I had the
sense to silence not only Athanasius but his wretched school as well!
But as it is there are plenty of his troupe, and you will have no diffi-
culty finding someone to fill his shoes—indeed, I should think that
you could hardly do worse than to choose any man from among the
mobs of the city to take his place, and he will do you at least as
much good as an interpreter of scripture. But perhaps your request
comes to me because you have a particular liking for the old man
Athanasius, or his subtleties (I hear stories, at least, that he a clever
old dog). And if this is the source of the request, then know it was
for being too clever that I banished him in the first place. He is quar-
relsome, and a quarrelsome man is not fit for leadership. Or is he his
own man at all—perhaps rather a puppet taking orders from
someone else—a recipe for disaster, and it looks as though he is
courting disaster openly. It is out of my concern for your welfare that
I sent Athanasius away in the first place and ordered him out of your
city: now I say, Let him be sent clear out of Egypt.
This sentence is to be proclaimed throughout the city of
Alexandria.

535. Julian’s reference to his life as a Christian; see Or. IV.31A, where he
“wishes this time to be forgotten.”
JULIAN AND THE JEWS

ulian’s relations with the Jews have traditionally posed a dilem-


ma for scholars. According to Sozomen (4.7.5) he reversed the
policies of his predecessors, notably Gallus and Constantius, after
being convinced that the Jewish community had been brutalized
and subjected by degrees to a policy of economic harassment.
Gallus, in an action barely mentioned by Socrates and Sozomen,
had responded violently to renegade Diocaesarean Jews for incur-
sions into the territories of Palestine (353AD) by destroying their
towns and slaughtering masses of the insurrectionists. And while
Constantius’s attitude toward Judaism was largely driven by political
necessity, Julian had at least flashes of historical interest in Jerusalem
and the restorationist cause in general. For this reason, as Wright
suggests somewhat anachronistically, “he may almost be called a
Zionist.”°°° He considerably lessened their tax burden, offered (at
first in principle, and then by edict) permission for Jewish sacrifices
to be restored to Jerusalem, and finally promised to rebuild the
Temple and hand the city over to Jewish control. This promise,
which Sozomen understands as part of a larger complex of “artful

536. See his discussion, III, xxi-xxii. Schwarz, Klimek, and Geffcken rejected
the letter as a forgery, and Bidez and Cumont, for completely unpersuasive reasons
enshrined in the European prejudices of their era, omitted it from their 1922 Paris
edition (L'empereur Julien: Ouvres completes, 1.2).

iy
178 JULIAN’S AGAINST THE GALILEANS

designs against the Christians,” was interpreted by pious observers


as an attempt to nullify the prophecy of Jesus (Mark 13.2) that “not
one stone of the Temple shall remain upon the other.” It was not for
love of the Jews, argued the Christians, that Julian’s largesse was
evinced, but rather his hatred for the Christians:
“He hated and oppressed the Christians and showed humanity
and kindness to the Jews. He wrote to the Jewish leaders and patriarchs,
and even the people, asking for their prayers for him and for the
empire. . . . But in so doing he was only trying to strike a blow against
the Christians, who are the Jews’ most determined adversaries. Or per-
haps he thought that by kindness [in rebuilding their temple] he could
persuade the Jews to offer pagan sactifice, for they knew only the letter
of the scripture, and could not (unlike our Christian teachers and the
wisest of the Hebrews) probe the hidden meanings” (5.22.2-4).
Julian’s promise, depending on whose testimony one decides to
credit, fell victim either to natural catastrophe or divine interven- _
tion. Earthquakes rocked the east in the winter of 362-363. The
building contract had been given to a certain Alypius, a Briton,
whose men fled the work following a “shower of brimstone” that
made the project impossible to carry out. Sozomen’s report, the
most sober, does not differ from the legendary accounts given by
(among others) Philostorgius (7.9), Theodoret (3.15), and the
invective of Gregory Naziazen—the last of these being probably the
source of the most fantastic elements of the tradition. Sozomen says
simply that Julian gave the go-ahead for work to begin and that
pagans in the city cooperated with the Jews in the preparation of the
building site owing to a mutual dislike of the Christians. “When
they had removed the ruins of the former building, they dug up the
ground and cleared away its foundation,. . . .<But> on the next day,
when they were about to lay the new foundation, a great earthquake
occurred, and by the turbulence of the earth stones were thrown up
from beneath the ground and many Jews who were at work were
injured” (5.22). Sozomen adds that the Jews were unable to inter-
pret these signs,°*’ and went back to work when the threat seemed

537. Which have been rhetorically adapted to the apocalyptic realization of


Jesus’ prophecy in Mark 13.1-2.
JULIAN AND THE JEWS bes)

to have subsided. It. was then that “fire burst suddenly from the
foundations of the temple and consumed several of the workmen.”
For all its detail, the reference to these “balls of flame” by Ammi-
anus and Gregory seems to be a later Christian gloss of what had
seemed to many Christians at the time the most audacious of
Julian’s attacks on the Christian teleology. That the work was not
started, or postponed, or marvelously interrupted once begun, could
only be seen (together with Julian’s death, which followed only a
few months later) as a miraculous endorsement of Christian
prophecy and fulfillment. In any event, the “worst” did not happen:
the temple was not rebuilt and Jerusalem was not restored to the
Jews. The location of the story in one of the more providential
sequences of Sozomen’s History argues the unreliability of the
detail: it comes as the climax in a series of mirabilia, which include
the story of the destruction of the temple of Apollo at Daphne by fire
from heaven (4.20) and the demolition of a statue of Julian erected
at Paneas (in place of one of Christ) by the stroke of a thunderbolt.
In this succession of miracle stories, the disconfirmation of Jewish
hopes for reclaiming Jerusalem is set alongside Christian confidence
that Jesus’ edict concerning the temple supersedes the Emperor's
promise to the Jews: a panegyric on Christ in his threefold office as
High priest, final prophet, and king of kings.
For all the legendary overlay, however, there is no good reason to
suppose that Julian did not actually intend to fulfill his commitment
to the Jewish community. Much obviously depends on the authen-
ticity of the letter. That such a letter was written by Julian is uni-
formly attested—by Sozomen, Socrates, and later tradition—and
just as vigorously denied by scholars beginning with Bidez and
Cumont in 1898 (1922). Its conciliatory tone has seemed at odds
with Julian’s muscular paganism and frustrations with the Jewish
community expressed elsewhere in his writings. At the same time, it
should be remembered that the use of the Jews as a counterfoil to
Christian propaganda is at least as old as Celsus, and the tone of the
rescript echoes Julian’s perception of Judaism as being more
“authentic” than Christianity—especially in its understanding of
oracles and sacrifice—as recorded in the books against the Galileans.
180 JULIAN’S AGAINST THE GALILEANS

The defense of Judaism extends no deeper than the maxim Veritas


antiqua est (What is old, is true), just as the indictment of Chris-
tianity hinges on the opposite maxim, invoked by Suetonius: it is
new, immoderate, and unapproved superstition.***
The arguments for including the letter in the Julianic corpus are
largely circumstantial, but in the end decisive.°*? The sentiments ex-
pressed in the letter are rightly taken by Sozomen to be typical of
Julian’s desire to belittle Christianity: What he does, he does not out
of love for the Jews but out of hostility for the more aggressive Chris-
tians, whose anti-Semitism seems to Julian a further manifestation of
their impiety, disrespect for tradition, and error-strewn regard for
their origins. The reference in the letter to the God of the Jews as 820¢
bytotog (the most high God, a term sometimes applied to Zeus)
reflects the garden variety hellenistic syncretism that Julian wished to
encourage (or resuscitate, since its equivalent had flourished during
the dying days of Hasmonean rule). And finally, there would be no _
reason for the Jews to invent such a story—or a letter to corroborate
it—since the success of the building project depended on the whimsy
of a man whose tenure in office was uncertain, and whose likeliest
successors would hardly be expected to carry out his programs.
Further, it seems unlikely that historians such as Sozomen
would allude to Julian's project as a matter of record simply in the
interest of documenting the emperor's artful designs against the
Christians if indeed there is not something to what Sozomen has to
offer as an explanation. For this reason, the suggestion that the
emperor wished to demonstrate the weakness of the Christian posi-
tion by “reversing” Jesus’ pronouncement on the temple remains the
strongest reason for seeing the letter as coming from Julian. Some
small proof of this design can be found in Socrates's description of
an earlier report concerning Julian’s treatment of the Christians.
Departing from his original policy, which sought to win friends
among influential Christians in exchange for their continuing vilifi-
cation of the memory of Constantius, Julian ordered that the church
of the Novatians at Cyzicus, which the Nicene zealot Euzoius had

538. Suetonius, Caes., Nero 16.


539. And by Wright, 3:xxii.
JULIAN AND THE JEWS 181

totally demolished, should be rebuilt. The bishop Eleusius was


threatened with the exaction of stiff financial penalties if he failed to
carry out the construction within two months—at his own expense
(3.11). Taxes previously imposed on the Jews on account of their
alleged “impiety” were extracted from the Christians on a scale
determined by their individual possessions in order to finance the
Persian campaigns. In short, there is impressive correspondence
between the benefits accorded the Jews by Julian and the penalties
“transferred” to the Christians in the winter of 362/3 in advance of
the Persian expedition. This will not be the last time that Julian uses
the rebuilding of a preferred enemy’s shrine and remission of taxes
as a way of dealing with a more direct threat to his imperium—in
this case the religious threat to his power and prestige represented by
the growth of the Christian church. As a final point, Julian’s own
commitment is ambiguous in the letter: He does not actually say he
intends to build the temple, but to restore the city of Jerusalem and
to see it repopulated. He says that to do this, he will need to bring
settlers into the city (of what religious predilection he does not say),
so that “all together can once again glorify the most high God,” a
usage sufficiently vague as to invite doubt whether the project
included replacing the Herodian temple. Julian invites the Jews to
perform some duties in exchange for his promise, but all reference
to them has been excised. And as to the time frame for the project,
Julian seems to assume that nothing can be done until his return
from the Persian frontier, which was not to transpire, while
Sozomen’s account assumes that preparation began prior to his
departure in 363.
Taken together, the evidence suggests that Julian was attempting
to curry favor with the Jews in exchange for the gift of Jerusalem.
Courting Jewish opinion had the advantage of punishing the Chris-
tians, or at least doing them philosophical damage, and possibly
encouraging the Jews to come to the aid of the Persian campaigns on
a promissory basis—thus the reference to the folly of a religious tax
imposed on a people who had limited access to their holy places,
and Julian’s willingness to lift it. The memory of this potential out-
rage was vivid in the Christian imagination, but less vibrant to the
182 JULIAN’S AGAINST THE GALILEANS

Jews, who saw the whole affair expended within months after the
letter was written from Antioch.

TO THE COMMUNITY OF THE JEWS>”°

Once upon a time you were slaves, and the most irksome thing
about your slavery was that your lords bent you to their will and
commands, yoked you to their laws, and then made you pay for the
privilege by contributing money to their treasury. I have seen this
with my own eyes, and learned more about it by examining the
records and processes that have been filed against you. But when an
additional burden was about to be placed on your shoulders, I
stopped it, and I ordered those who delight in such obloquy to stop
as well. My desk was full of transcripts of your impiety and accusa-
tions against you—and I threw the lot into the fire so that no one
again might profit from charging you with disloyalty.
My brother Constantius is not entirely to blame for these
reports; rather, the men he had to his table, they are the real cul-
prits—barbarians in mind, ungodly in spirit. I seized them with my
own hands, thrust them into the pit so that no memory of their
wickedness might survive.°*!
My wish for you is that you should prosper. I have instructed
your patriarch, the venerable Iulus, that the taxation imposed upon
you should be prohibited from this time forward: no one should be
able to oppress you through such measures. <I do this> to give you
peace of mind throughout the extent of the Empire, and through
this peace encourage you everywhere to offer heartfelt prayers on my
behalf to the most high God, the creator>4? who has chosen me to

540. Ep. LI (Hertlein, 25).


541. It is unclear what “wickedness” Julian refers to here, unless these are sur-
vivors of the process begun in 354 following the revolt of the Jews in Dio-Caesarea
(Socrates, HE 2.33). If so, there seems to have been some effort to defend the extrav-
agant measures by documenting cases of Jewish disloyalty, a process which Julian
here formally disavows. It may, however, be a reference to a conspiracy against the
Jews fomented by Julian’s brother Gallus, or so the language would suggest.
542. dnuiovpy@ Ged.
JULIAN AND THE JEWS 183

reign by the power of his immaculate right hand. For it is clear to me


that when our minds are troubled, we refrain from raising our hands
in prayer, but when we are free from anxiety, we happily resort to
supplication, praying for my imperial person to God the all-pow-
erful who directs my reign in all ways, according as I wish. < >°43
This you ought to do so that after I have brought the campaign
against Persia to its successful completion I might restore the sacred
city of Jerusalem by my own efforts—the place you have so desired
to see inhabited again—settling men there so that all together, with
you, may glorify the most high God in that place.°*4

543. The syntax is awkward: a list of requirements, possibly stipulating what


is evi-
the Jews must do to earn the restoration of Jerusalem and right of access,
dently omitted. ite
the
544, Kai év avth SdEav 86 pe’ byav 16 Kpeitovi. The project of rebuilding
Temple is not explicit but probably envisaged in the concludin g sentence.
LIVES OF JULIAN

he following early histories (excerpted) of the reign of Julian are


provided here as an epilogue to the effects of his reign. With the
formidable exception of Constantine, no emperor who ruled
between the time of Diocletian and the time of Justinian provoked
such a flurry of biographies. Why this is so is not difficult to imagine:
Julian was seen as the last great defender of the old Roman world, a
world which Christian bishops and intellectuals like Cyril regarded
as having given way, providentially, to the true doctrine of the
Church. Given this perspective it is no wonder that the emperor's lec-
tures on the superstitions of the Galileans, the irregularities of his life,
the ambiguities and strategems he employed to resuscitate the dying
religious world of his ancestors invited serious attention and rebuke.
The first vita comes from the hand of the church historian Socrates
Scholasticus (ca. 379-4392), an interesting personality who had in
common with Julian a love of pagan letters (he quotes, approvingly,
from Sophocles, Euripides, Xenophon, Porphyry, and Libanius,
among others, in his church history). The second and more expan-
sive vita is taken from the church history of (Hermias) Sozomenos,
an author of Palestinian background whose family was converted to
Christianity in the Gaza after beholding a miracle of Hilarion. His
writing covers roughly the period 323 to 439 and thus comprises
some of the same detail as that treated, less expansively, by Socrates.

185
186 JULIAN’S AGAINST THE GALILEANS

SOCRATES SCHOLASTICUS |
THE EDUCATION AND PARENTAGE OF JULIAN> **°

Julian was proclaimed emperor in Constantinople, to which city he


came from the western provinces around the middle of December
361, Constantius having died on the battlefields of Cilicia on the
third of November during the consulate of Taurus and Florentinus.
It is necessary to speak a little about the personality of this man who
was so well known for his learning. In doing this there is no need for
me to affect an aureate style of writing in order to do justice to my
subject’s great rank—just to please his admirers. My purpose in
undertaking this biography is to compile a history of the Christian
religion, and it requires no change in style from subject to subject
for this object to be achieved. So in an orderly fashion I wish to
describe his personality, his birth and education, the way he acceded
to the imperial office, and to do this I must first sketch for you the
essential background of Julian’s career.
The emperor Constantine, who gave Constantinople his name,
had two brothers, Dalmatius and Constantius. These were sons of
the same father by different mothers. ... Now <Dalmatius> had a
son who took his father’s name; Constantius had two sons, Gallus
and Julian. When Constantine the founder of Constantinople died,
the army was quick to put Dalmatius°*° to death: the lives of his
orphaned sons were also in jeopardy. Gallus was saved by serious ill-
ness from being handed over to the same fate as his father. And
Julian was even more fortunate, being only eight years old at the
time and hence considered too young to be a threat. Constantius’s
jealousy seemed to have weakened, and Gallus was permitted to
attend school in Ephesus—where he also had huge landholdings
and estates left to him by his father.
Julian, on the other hand, when he got to be school age, fol-
lowed his studies at Constantinople, indeed even going in ordinary

545. Socrates Scholasticus, HE, 3.1; This translation is based on the edition of
the Greek text by G. C. Hansen, Die Griechischen christlichen Schriftsteller der ersten
Jahrhunderte. Neue Folg, Bd. 1 (Berlin: Akademie Verlag, 1990).
546. That is, the younger brother of Constantius.
LIVES OF JULIAN 187

dress to the palace, where the schools operated in those days under
the supervision of the eunuch Mardonius. He was taught grammar
by Nicocles the Lacedaemonian. He learned rhetoric from Ecebolius
the Sophist, who was known at the time to be a Christian—indeed
his uncle had made strict provision that he should not be taught by
pagan instructors, lest he fall into the superstitions of the pagans, for
at the beginning, Julian was a Christian.54”
His proficiency in letters soon became so famous that it was
rumored he could run the Roman Empire on his own, and this
rumor could be heard everywhere. It so unsettled the emperor that
he decided to send the boy away from Constantinople to Nico-
media, but forbidding him to attend the school of the Syrian
Sophist known as Libanius. This is the same Libanius who had been
driven from Constantinople by a collusion among the educators, to
Nicomedia, where he finally settled and opened a school and wrote
diatribes against the teachers who had caused his exile.
Julian was prohibited, as I said, from mingling with the fellows
of this school because Libanius was a pagan, but being resourceful
he managed to get his hands on copies of Libanius’s lectures, and
these he read with admiration, over and over again. Julian grew ever
more proficient in rhetoric, until one day the Ephesian philosopher
Maximus (not to be confused with the Byzantine Euclid’s father)
arrived to set up shop in Nicomedia—this the same man who was
later executed by Valentinian as a magician and sorcerer.°4* What
brought him to Nicomedia was the fame of Julian. From this man
Julian took not only elementary instruction in philosophy but reli-
gious ideas as well—and something more: the desire to rule the
empire. When it was known that the emperor had got wind of what
was going on, Julian—tossed between ambition and fear—was very
eager to put the rumors to rest. So he began to affect the manner and
style of what he once had been in truth: He had himself shaved, and
pretended to live the life of a monk, publicly reading only the sacred
books of the Christians, but in private following his philosophical

547. Ep. XLVII.434D. It is noteworthy, however, that of the teachers men-


tioned by Socrates, only Ecbolius is averred to be a Christian.
548. Valentinian I (364-375).
188 JULIAN’S AGAINST THE GALILEANS

studies. He even managed to have himself installed as a lector in the


Church of Nicomedia.>4? In this deceitful way Julian managed to
avert the emperor's suspicions. He did this out of fear, but without
abandoning ambition, telling his friends privately that the days were
not far off when he would own the crown.
It was about this time that Julian’s brother Gallus, who had just
been created Caesar, came to visit him on the way to the East, and
when not long after this Gallus was murdered, suspicion fell natu-
rally on Julian. The emperor ordered a guard to be posted to watch
him, but Julian soon found ways to avoid them and managed to slip
from place to place, out of their sight, and to safety. When the
empress Eusebia learned of his situation, she begged the emperor to
leave him in peace and to permit him to’go to Athens in order to
complete his studies in philosophy. So to make short of it, that is
what the emperor did: he called Julian to him and created him
Caesar, also marrying him to his own sister Helen. Then he sent him
off to wage war against the barbarians, for the barbarians whom
Constantius had encountered when they fought as auxiliaries
against the tyrant Magnentius were now on the loose and beginning
to raid the Roman cities: But because of Julian’s age, the emperor
ordered that he should do nothing on his own without consulting
the senior military men.
As it happens, these generals had become lazy with power, care-
less of duty, and as a result the foreign troops grew stronger with
each passing day. Julian, seeing his opportunity, permitted the gen-
erals to continue in debauchery and drunkenness, while insinuating
himself as the true commander of the infantry by preaching courage
and offering a reward to any soldier who killed a barbarian. In this
way he won the army to his side and weakened the enemy. It is said
that once, as he was entering a city, a civic crown which was bal-
anced between two columns fell on his head—which fit it perfectly.
All those present shouted their approval, seeing it as an augury that
one day he would be emperor in his own right. It is also said that

549. avayvootns : a minor office usually assigned to a young man proficient


in public speaking. This would have given Julian adequate subterfuge to disguise his
philosophical activities since he would be publicly in view professing Christianity.
LIVES OF JULIAN 189

Constantius had sent him against the foreigners because he hoped


that Julian would die in the battles. I cannot know for certain
whether this is true, but it seems unlikely to me that <Constantius>
would first have sought an alliance with <Julian> and then, with
harm to himself being the only outcome, seek to have his ally
destroyed.*°° Each must decide the case for himself.
At any rate, Julian complained to the emperor that the generals
had been indolent in the campaign, and this led to Constantius
appointing a lieutenant more sympathetic to Julian’s designs.
Together they waged such an effective campaign against the for-
eigners that they sent him a messenger with credentials to prove that
the foreigners had been ordered by the emperor himself to march
into Roman territories. But [Julian] cast the messenger into prison,
attacked the enemy forces, and quickly overcame them, sending
their king as a prisoner to Constantius.
Upon this success, Julian was proclaimed ruler by the soldiers,
and failing to find a crown at hand, one of the soldiers wound a
chain he had been wearing as a makeshift crown for Julian’s head.
In this way, Julian became emperor. But as to the question whether
he conducted himself after this day in a manner befitting a philoso-
pher, I leave it for you to judge.*™ For Julian did not send an emis-
sary to Constantius, nor find it in his heart to reward past favors with
a recognition of his office. He set about, instead, appointing gover-
nors to the <conquered> provinces, choosing whomever he pleased
for the job. Not only this recklessness, but he began to impugn Con-
stantius everywhere by making it known that Constantius had sent
letters to the foreigners. When this was learned in the marketplace,
Julian had no trouble convincing the inhabitants of the cities to turn
their backs on Constantius and throw their support to him.
At this point Julian had no further reason to wear the mask of a
Christian. He began to style himself pontifex maximus, reopened
the temples wherever he found them, and encouraged everyone to
550. Socrates’ analysis is less fair-minded than simple: Constantius’s motive
for sending Julian against the invaders was that it would have acquitted him of any
direct responsibility for his nephew’s death.
551. Presumably Socrates means in a manner appropriate to someone edu-
cated in diplomacy and the philosophical traditions of statesmanship.
190 JULIAN’S AGAINST THE GALILEANS

celebrate the ancient holidays. And so Julian declared war against


Constantius and had no hesitation in throwing the whole empire
into a state of war, his so-called philosopher's agenda being unat-
tainable except by treason.
By God's grace, however, news of Constantius’s death came just
when Julian had entered Thrace, and the Empire was strangely pre-
served in this way from the calamity of civil war. Julian entered Con-
stantinople in triumph, alert to the need to endear himself to the
population. Accordingly, he did the following: He knew that Con-
stantius had become an enemy of the defenders of the homoousion
faith®>? by deposing their bishops and expelling supporters from the
churches.
By the same token, the pagans were encumbered by laws that
prohibited them from sacrifices to their gods and were eager to have
their temples reopened and the old rites reinstated. Both groups har-
bored angry feelings toward <Constantius>, and the population in
general were passionately opposed to the violence of eunuchs—
especially the rapacious Eusebius, the chief of the imperial bed
chamber. Julian treated each group with subtlety, hypocritical as he
needed to be, patient with others, even aggrandizing some (as he
was always fond of appearing benevolent), but always with an eye to
the interests of the pagans.
First, then, he sought to impugn the memory of Constantius by
publicizing his cruelty toward his subjects. He called the bishops out
of exile and restored them to their seats. Then he commanded that
the pagan temples should be opened without delay. And finally, he
ordered that those who had been made victims of extortion by the
eunuchs should have back their property. As for the eunuch Euse-
bius, he was put to death—not only because of the harm he had
inflicted on others, but because Julian was convinced that <Euse-
bius> had ordered the death of Gallus. And so the body of Constan-
tius he publicly honored with a royal funeral, but the eunuchs, the
barbers, and the cooks he expelled from the palace: the eunuchs
because Julian had resolved not to marry again following the death

552. The supporters of Nicene trinitarianism.


LIVES OF JULIAN 191

of his wife; the cooks because he stuck to a very spare diet,5? and
the barbers because he held that one will do to serve many. And in
addition to these dismissals, he discharged many of the secretaries
who lounged about the palace and raised the salary of those he
retained. He also reformed the system of public transportation by
abolishing the use of mules, oxen, and asses and decreeing that only
horses were to be utilized. Some applauded his austerity, while
others deplored it for bringing the dignity of the imperial office into
question by reducing the amount of pomp and splendor that play
so great a role in enchanting the masses.
It is even said that at night he would sit awake in his chamber
composing orations to be delivered in the senate; in fact, he is the
first and only emperor since Julius Caesar to make speeches in that
body. And to those who were skilled in the literary arts, Julian would
show the most amazing and flattering patronage—especially to pro-
fessional philosophers. One saw countless beneficiaries of this sort
returning from the palace showing off their palliums, many of
whom can only be described as charlatans and sycophants. But
those he honored had one thing in common with the emperor: a
distaste for Christianity and the wellbeing of the Christians. And of
course Julian was no stranger to flattery, having composed a book of
his own, The Caesars, in which he derides the achievements and
characters of all his predecessors, and another act of vanity in which
he attacks the Christians.
The expulsion of cooks and barbers might be seen as an act befit-
ting a philosopher—but not an emperor, for the latter must be
above the influence of jealousy or petty dislikes. Indeed, in respect
of self control and moderation an emperor may be a philosopher.
But when a philosopher tries to imitate an emperor he can only
stumble on his own principles.

553. Socrates will not allow that Julian’s reason for dietary restrictions are
based on Julian’s Neoplatonic view of abstinence, derived largely from Porphyry
(Abst., 3.9.3-3.11.1).
192 JULIAN’S AGAINST THE GALILEANS

SOZOMEN |
A HISTORY OF JULIAN, CALLED APOSTATE*=4
5.1 After the death of Constantius, Julian entered Constantinople.
He had already conquered Thrace when he was proclaimed
emperor. It was said that the demons and fortune tellers had told of
the death of Constantius and his change in fortune even before
<Julian> departed for the east, and while this may have been so, the
life of Julian as emperor was a short one. The power he tasted was
an illusion. I think it is absurd to think that after he heard foretold
the death of Constantius and his own end at the hands of the Per-
sians in combat, he would have chosen death. .. .
After the death of Constantius a great fear of persecution arose in
the church. In fact, the Christians probably suffered more from fear
than from any consequences that would likely have arisen from the
transition; but there had been a long and peaceful interlude. They
had become accustomed to tranquility, and rumors of the emperor's
hatred of Christian teaching spread rapidly, causing them to recall the
misery and tortures their grandfathers had endured at the goading of
other tyrants. It was <also> said that his renunciation of Christianity
was so complete that he underwent the so-called renunciatory rites,
which involve sacrifice and the blood of animals, in order to wash
away his Christian baptism. Both secretly and then publicly he
immersed himself in the observance of pagan rituals and augurs.
One day, it is said, he was examining the entrails of an animal when
he saw the pattern of a cross surrounded by a crown. A number of the
celebrants were terrified at the sign, exclaiming that the symbol pointed
to the victory of Christianity (the circle being the symbol of that which
has no beginning and no end but is of eternal duration); but the chief
celebrant told Julian to interpret the sign as a favorable omen: the
entrails surrounded the symbol of Christianity and impinged on it in
such a way as to suggest that it would not spread or expand, but be lim-
ited by the boundaries of the circle that surrounded it.

554. Sozomen 5.1-6.1 (edited); translation based on the edition of the Greek
text edited by Josef Bidez and G. C. Hansen, Griechischen christlichen Schriftsteller der
ersten Jahrhunderte (Berlin: Akademie Verlag, 1960), Bd. 50.
LIVES OF JULIAN Les

One day, I am told, Julian was plunged into a famous pit in


order to consult an oracle or to participate in an initiatory ritual.
Now these rites are famous for the mechanical devices that are used
to produce effects in the initiates, either to frighten or to overwhelm
them, and on this occasion it was as though ghosts and spirits began
to dance before him. And Julian, overcome with fear, seemed to
forget those who were there with him; and even though he had long
since turned away from Christ to his new faith, nevertheless he
signed himself with the symbol of Christ—just as any Christian
would do when confronted with dangers. As soon as he did so the
apparitions disappeared and the designs were confounded. The
master of the rites was amazed at first, but when he discovered what
had caused the flight of the demons he said the ritual had been pro-
faned. He told the emperor to be brave, to ignore the events, and to
keep his mind away from everything Christian. Then he again con-
ducted the rites. All of this attention to pagan custom worried and
sadden the Christians, as you may imagine—and the more so
because he had himself once been a Christian. He was born of pious
parents, baptized in infancy, and raised according to the tradition of
the church to know the scriptures. He was nurtured by bishops and
Christian teachers. [Sozomen goes on the repeat and expand the
main elements of Socrates’ account of Julian’s education. |
5.3 When Julian became unrivalled in the empire he ordered all
the pagan temples reopened in the east and that those that had
fallen into ruin should be restored with their altars. He set aside
great sums of money for the project of reestablishing the ancient
customs and ceremonies in the cities, as well as the custom of
offering sacrifice.
<Julian> therefore offered libations and offered sacrifice in
public. He bestowed honors of those who performed these duties.
He restored the initiators and the priests, the hierophants and the
keepers of images to their former status. He ratified the laws of his
predecessors regarding various privileges, approving exemption
from military duty and other obligations, such as had been their pre-
rogative under previous emperors. To the temple guardians he
restored their rations, but commanded them to refrain from eating
194 JULIAN’S AGAINST THE GALILEANS

meats, and moreover to lead a life of purity by abstaining from any-


thing which violated custom.
<Julian> also ordered that the nilometer?>? and the sacred letters
and tablets should be housed again in the temple of Serapis, undoing
the edict by which Constantine ordered them to be deposited in a
church. He was in touch with the inhabitants of cities where he knew
paganism flourished, inviting them to state what support they required
<to further their cause>. But toward the Christians Julian showed only
contempt. He refused to visit them or to receive their legates, those
who were responsible for representing their grievances.
When the natives of Nisibis**° supplicated for help against the Per-
sians who were on the very brink of invading Roman territories, Julian
refused all assistance on the grounds that they were wholly christian-
ized. He refused to open their temples and shrines, saying he would
not help them, receive their ambassador, or indeed even enter the city
until he had assurances that they had returned to the old religion.
Similarly he accused the people of Palestinian Constantia of
being attached to Christianity, and made the city subordinate to the
city of Gaza. This same Constantia?’ was once Majuma, a famous
port used by ships coming into Gaza. But when Constantine heard
that its inhabitants had become Christian he created it a city, gave-it
the name of his son and an independent government—thinking
that a city <of this stature> should not be dependent on a city
devoted to pagan rites such as Gaza. Now when Julian became
emperor, the citizens of Gaza took their case against Constantia to
the court, and the emperor himself sat as judge, deciding in Gaza’s
favor that Constantia would henceforth be a dependent town, even
though located some twenty stadia from the its capital city... .
5.4 <So too> Julian erased the name of Caesarea, the famous city
in Cappadocia which is sited near to Mount Argeus, wiping it off the
register of cities, obliterating the name that had been conferred on it
by Claudius Caesar (it was known formerly as Macaza). Julian had
long despised the citizens of this place because of their Christianity
<and because> they had destroyed the ancient temples of Apollo
555. The station used for measuring the waters of the Nile.
556. Nusaybin in Mesopotamia.
557. On the Abora, also called Viransehir.
LIVES OF JULIAN 195

and of their patron deity Jupiter. The temple dedicated to Fortuna


was left standing for a while, but then on Julian's accession it too
was destroyed. When Julian got wind of this, he hated the city with
a passion so intense he could hardly bear <to hear the city men-
tioned>. He hated even the members of the old religion, because he
said—though few in number—they should have defended the
temple of Fortune and suffered what befell them in its defense. So
he ordered the treasury of the churches in the city and region emp-
tied of gold—about three hundred pounds were discovered and
transported to the public treasury along with other property
belonging <to the churches>. Then he ordered all the clergy to reg-
ister with the governor to serve in the military—a service which is
regarded the most arduous and least rewarding in the Empire.
Finally, he ordered the names of all Christians to be recorded,
including women and children, and he imposed taxes on them as
crippling as those already imposed on the <Christian> villages. Then
he ordered temples rebuilt, as quickly as possible, saying he would
not be satisfied and would punish the city further if it was not
done—even if not a single Galilean was left standing. For this name
is the one Julian preferred to the name “Christians,” and there is no
doubt that he would have carried out his pledge had he not died
when he did. <So> it was not out of compassion for the Christians
that Julian at the beginning treated them with greater kindness than
had been the case with previous persecutors. No, rather, he had dis-
covered that their torture brought the Romans no advantage; and he
knew that the Christian religion had increased because of the
courage of those who died in its defense.
So Julian, envious as he was of the strength of the Christians,
tried persuasion and reason to win them over to paganism rather
than the tactics of his predecessors: Fire, sword, drowning, and
burying-alive were eschewed in favor of other means to win them to
his side, because he believed benevolence was a more efficient
weapon than violence.**®

558. Sozomen does not seem to recognize that the measures he is describing
in relation to Palestinian Christians are the opposite of the tactics he otherwise
ascribes to Julian.
196 JULIAN’S AGAINST THE GALILEANS

The story is told that one day Julian was sacrificing in the temple
of Fortuna at Constantinople when Maris, the bishop of Chalcedon,
confronted him and upbraided him as a man of no religion—an
atheist and apostate. In reply, Julian could say nothing by way of
reproach. Maris was old, blind, and led by the hand by a child. But
then Julian with his customary bile turned to the old man and said,
“The Galilean, your God, will not heal you.” And Maris said, “Well,
I thank my God for this blindness; because of it I cannot see the one
who has fallen from our faith.” Julian said nothing but <left the old
man in peace> thinking that by a display of patience and generosity
he would advance the cause of the pagan religion.
5.5 It was for this reason Julian recalled the Christians who had
been exiled under Constantius because of their religious beliefs**?
from exile to return to their cities. He returned the property that had
been taken from them, and called upon the non-Christian population
not to insult or injure them, but to treat them fairly and permit them
to offer sacrifice freely. He said that if they wished—without coer-
cion—to participate in the worship of the temples, they had first to
expiate themselves with purifications, for the pagans believe the
demons must be appeased in this way. But the clergy he stripped of
the honors and immunities and stipends which Constantine had con-
ferred. And he repealed all laws that favored the Christians while rein-
stating their statutory liabilities. This included requiring widows and
virgins (whose poverty caused them to be reckoned among the clergy)
to refund all provision that had come to them from public sources.>
Recall that when Constantine managed the secular concerns of
the church, he made certain that a portion of the taxes raised

559. That is, the anti-Arians.


560. Sozomen’s report should be evaluated in the light of other evidence. In
Ep. XXXVII (to Atarbius, prefect of the district of the Euphrates), Julian writes: “I
swear by the gods that I do not desire the death, beating, or abuse of any of the
Galileans. Nonetheless, I say with conviction that the god-fearing <@e0cePeic> are
to be preferred to them. It is because of these Galileans that everything has been
turned upside down, but through the grace of the gods everything is put right. And
so let us honor the gods, as will all god-fearing people and their cities” (376G,-D).
Julian's standard position is that Christianity is a “disease” and that the persons
afflicted are to be “treated” rather than persecuted for it; see the epistle to the Bos-
trians, above, XLI.438B.
LIVES OF JULIAN 197

throughout the Empire would be reserved for the benefit and sup-
port of Christian priests—protecting this arrangement by a law
which <reenacted after Julian’s death> remains in effect to this very
day. It is said that the exactions of Julian against the clergy were cruel
and meticulous, for the collectors in charge of the process were
required to issue receipts for all the money they gained in the
process, thus proving that Constantine’s generosity was fully repaid.
The emperor's hatred toward the religion was inexhaustible. And
in his spite he did all that is possible to ruin the church. He seized
property, carried away votives, sacred cups, and commanded any
guilty of destroying shrines and altars during the reign of Constan-
tius or Constantine to rebuild them or pay for their rebuilding out
of their own pocket. But because many who were accused under the
inquiry into the disbursement of sacred money were unable to pay,
<Julian> ordered priests and clergy and others as well tortured and
thrown into prison.
So it may be seen that even if Julian shed less blood than other
persecutors of the church—if he was less severe in concocting
painful tortures—he was more cruel in some respects, because he
wanted to do incalculable harm to the church. If he recalled the
priests whom Constantius had banished, he did so because he
hoped they would at once fall into internal strife and feuding,
proving Constantius a fool and vindicating his own program
directed at the failure of the church. He reckoned that his plan
would show his predecessor in his true colors: He would favor the
pagans because he shared their values, but he would also show
mercy to those who had suffered so much in the name of Christ,
because <Constantius> had treated them so unfairly. . . .
5.6 During this time Athanasius, who had been in hiding, heard
of the death of Constantius and returned by night to the church in
Alexandria. His sudden reappearance was nothing short of aston-
ishing: He had managed to escape detection by the governor of
Egypt, who had been ordered by the emperor and the friends of
bishop George to seize him, by concealing himself in the house ofa
virgin in Alexandria. This virgin is said to have been so beautiful that
everyone who saw her exclaimed that she was an apparition never
198 JULIAN’S AGAINST THE GALILEANS

seen in nature. Men pledged to continence would avoid looking at


her, lest the finger-pointers accuse them of having wicked thoughts.
She was at the peak of her beauty—young, unassuming, and tender
—gifts which when added to her natural loveliness enhanced her
charm all the more....
When I look back on the events <of Athanasius’s life> I can only
think that they were directed by God. His relatives might have fallen
under suspicion as to his whereabouts, and might have had trouble
from the authorities if they had been compelled to swear <con-
cerning his hiding place>, but who <including his family> would
have suspected that a priest would have taken refuge in the house of
a beautiful virgin? She bravely took him in, and through her efforts
his life was spared. She became his hostess and servant; she washed
his feet, brought him food, waited on him in every imaginable way.
If he needed books, she arranged for someone to bring them to her.
But all during his long period of hiding, no one in Alexandria had
any idea where he was.
5.7 So Athanasius, who had survived in this manner, seemed to
come out of nowhere, but the people rejoiced to see him and wasted
no time in restoring his churches to him. This meant, however, the
expulsion of Arians from their churches, which in turn meant that
they were required to hold their assemblies in private houses. They
installed Lucius as supervisor of their heresy in place of <their
former bishop> George. George had been murdered in an uprising
of the pagans of Alexandria at the very moment it was known that
Constantius was dead and Julian had been declared sole ruler. They
attacked George first by shouting imprecations, then threatening to
kill him on the spot. George was thrown into prison for a little
while, but the instigators soon returned, rushed at the doors, seized
George and slaughtered him, heaved his corpse onto the back of a
camel after defiling it and subjecting it to unspeakable forms of
abuse before burning it at nightfall. |am aware that the Arians allege
that these things were done to George at the instigation of Athana-
sius. But it seems more likely to me that the perpetrators were
pagans. They had good reason to hate him—more reason indeed
than anyone, for he had ridiculed their temples, insulted the gods,
LIVES OF JULIAN 199

and prevented them from offering sacrifices or honoring the ances-


tors. Moreover, the influence he commanded in the palace caused
the people to despise him, and soon enough they found the flouting
of his authority insufferable.
Meanwhile, a disaster had occurred in Mithrium in an area
where Constantius had erected a church in a desert region and given
it to the see of Alexandria. George set about making way for the
church by clearing the ground in order build an oratory when an
adytum was discovered. <This adytum> was full of idols and instru-
ments used by pagans in their rituals of initiation—strange to
behold, from the standpoint of the workers; worse, the Christians
insisted on making a spectacle of them and publicly parading them
in order to annoy the pagans. At the sight <of the procession> the
pagans formed a mob and attacked the Christians, arming them-
selves with rocks, with swords, with whatever came to hand. They
killed many Christians and in mockery of the faith crucified some
and wounded others.°® This naturally brought work on the church
to a standstill, and the pagans murdered George as soon as they
heard Julian would be emperor. Indeed, the fact is admitted even by
the Emperor, who would have had more to gain if the truth had not
been so compelling, and who would have preferred to blame the
death of George on the Christians rather than the pagans, had not
truth pressed him to disclose the facts. As it is, he wrote a severe
letter>©? to the inhabitants of Alexandria on the subject, in which he
makes it clear where the blame rests. In the letter he condemns but
does not punish; he invites them to attend to Serapis, their tutelary
god, and their founder Alexander, and his uncle Julianus, who was
governor of Egypt as well as Alexandria. It was the <elder> Julian
whose hatred of Christianity was so great and love of the old faith
so strong that he remained a persecutor until his dying day. . . .
5.15 <When Julian heard> that Athanasius was using the church
of Alexandria for divine service and the teaching of catechumens,

561. Sozomen is the sole authority for the narrative of these hostilities. A strik-
ingly similar tale is told by Socrates (HE 5.16) concerning the destruction of a
mithreum at the behest of Theophilus of Alexandria.
562. Ep. XXIX, quoted in its entirety by Socrates, HE 3.3.
200 JULIAN’S AGAINST THE GALILEANS

and that he had converted many from paganism in this way, he com-
manded him to leave Alexandria at once, on pain of death. Julian’s
warrant in this was to say that Athanasius had been banished by
imperial decree under Constantius and had resumed his see without
the approval of the reigning emperor. Julian for his part insisted that
he had never intended to restore bishops exiled by Constantius to
their jurisdictions in the church, but only to their homelands. On
learning that he had been ordered out of Alexandria immediately,
Athanasius addressed the crowds saying, “Take heart: this is a little
cloud that will quickly pass away.” Then he said farewell to the
assembly and entrusted the church to the most courageous and loyal
of his comrades, and he left Alexandria.
At this time some citizens of Cyzicus°® sent a messenger to the
emperor for the purpose of asking favors, and in particular seeking
the restoration of their temples. He praised their efforts and agreed
quickly to their petitions. Soon after, he expelled Eleusius, bishop of
that city, who had destroyed a number of the pagan shrines and des-
ecrated other areas, while building churches and hostels for widows,
as well as houses for virgins, and having urged pagans to abandon
their ancient rites. The emperor then gave word that foreign Chris-
tians were not to enter the city of Cyzicus, fearing that they might
form an alliance with Christians within the city and then mount a
campaign <against the pagans>. There were many sympathizers
among the population who held views similar to the Christians—
men who were wool makers for the state or who traded in money.
They had been permitted by a succession of emperors to remain in
Cyzicus with their families on condition they turn over to the state,
on an annual basis, clothes for the soldiers and a supply of newly
minted coin.
Now Julian was eager to ensure the triumph of paganism what-
ever the effort, but he was not willing to employ harsh measures to
do it; he thought it unwise, for example, to use force or violence
563. As ecclesiastical metropolis of Hellespontus (Propontis) near Erdek, Cyz-
icus had a catalogue of bishops beginning with the first century; Lequien (1, 747)
mentions fifty-nine. A more complete list is found in Nicodemos, in the Greek
“Office of St. Emilian” (Constantinople, 1876), pp. 34-36, which has eighty-five
names. Gelasius, the historian of Arianism, was born in Cyzicus.
LIVES OF JULIAN 201

against those who would not sacrifice>*“*—and besides, there were so


many Christians in so many cities that the emperor would have had
trouble enough just to count them. Indeed, he did not even forbid
them to come together for divine service because he realized that
when freedom of the will is curtailed, coercion is never effective. So
he expelled the clergy and bishops from their churches hoping that
by so doing he would effectively destroy the assembly, as there would
be no one left to convene it. Deprived of clergy to administer the
sacred mysteries, he reasoned, <Christianity itself} would soon
enough disappear. To encourage the decline of the religion, Julian
caused the rumor to be spread that the clergy were fomenting rebel-
lion among the people, and it was this pretense he used to get rid of
Eleusius and his friends—though the city was far from being in rebel-
lion against the emperor. Then Julian called upon the people of
Bostra to expel Titus, their bishop, conveying in advance that he
would hold Titus responsible for any act of sedition—to which Titus
dutifully replied that although Christians were as numerous as
pagans in Bostra, nonetheless they preferred tranquility to conflict
and were inclined to remain quiet and not to rebel. Julian decided
that it would do no good to incite the citizens of Bostra against Titus,
so instead he composed a letter in which he slanderously suggested
that the only thing preventing the Bostrians from insurrection was
the intervention of Titus—not their own peace-loving inclinations.
Then Julian urged them to expel Titus from the city as a traitor.°®
In similar fashion, the Christians were subjected to trials in
many places—sometimes due to commands of <Julian> and at
other times due to the impetuousness of the <pagan> population,
but blame should be attributed in either case to the designs of the
ruler. At no time did he invoke the law against the law-breakers;
instead, he settled for reprimand and scolding and the cheapness of
words, while his actions goaded them on in their desire to hurt the
Church. And so even though we cannot say the emperor persecuted
the Christians, the Christians were nonetheless persecuted, fleeing
from village to village, city to city.
564, Almost no early writer manages to square this information concerning
Julian’s leniency with the desire to style him a persecutor. Cf. Ep. XXXVII.
565. Ep. XLI, preserved by Sozomen.
202 JULIAN’S AGAINST THE GALILEANS

My grandfather was born into a pagan family.°°° His own family


together with that of Alaphion was the first to convert to Christianity
in Bethelia,>°’ a prosperous town near the Gaza, full of temples, and
one which is regarded highly by the people of that country because
of the magnificent design and antiquity of its temples. The most
splendid among these is known as the Pantheon, built on an artifi-
cial mound commanding a view of the city. People say that the place
received its name from the temple and that the original form of the
name was in the Syriac language, later rendered into Greek, but
retaining the meaning that it is a dwelling place for all the gods. Now
it is also said that the families I have mentioned were converted
because of the work of the monk Hilarion. It is said Alaphion was
possessed by a devil, and neither pagan nor Jew could relive him of
his suffering, no matter what prayer or incantation they tried.
Hilarion, however, relieved the man’s suffering simply by calling out
the name of Christ. The demon had been routed, and in thanks- .
giving Alaphion and his whole family were converted to Christianity.
As for my grandfather, he was a man of extraordinary skill, espe-
cially in the exegesis of the holy scripture. He had a wide learning
and was a competent mathematician, and he was loved by the Chris-
tians of the Gaza and Ascalon and the whole region. He was much
in demand by the Christians there because of his superior training
in interpreting difficult passages in scripture. <The family of
Alaphion> does not need the likes of me to praise it, for the first
churches and monasteries built in that country were built because of
this family—and still it produces good men right down to this day.
Even in my youth I knew some of them, but they were then very old.
I will say more about them as my history unfolds.
5.16 The Emperor was grieved to find that his efforts to rekindle
the fire of the old religion were useless, while seeing Christianity
grow more famous day by day. The doors of the temples were wide
open, sacrifice was offered, the old rites were restored in all the cities,
but Julian was not happy. He knew that if he weakened or withdrew

566. It was a matter of pride among Christian intellectuals from the time of
Justin martyr to boast of a pagan background and classical learning.
567. Lit., house of the gods.
LIVES OF JULIAN 203

the pressure <for change> his program would collapse. He was par-
ticularly unhappy that the wives, children, and servants of many
pagans had converted to Christianity, and when on examination he
decided that one of the great virtues of the Christian faith was the
moral life of those who professed it, he decided to introduce the dis-
cipline of Christianity into the temples and to create orders and
grades of ministry: teachers and lectors to give lessons in pagan doc-
trine, and to exhort the people to pray at certain hours. He also
resolved to create monasteries where men and women could lead a
life of philosophical seclusion. And he founded hospitals for the
poor and for the benefit of the resident foreigners, and other chari-
table works. Julian <even> attempted to introduce the Christian prac-
tice of penance among the pagans, related to various kinds of delib-
erate and inadvertent wrongdoing. Another custom he admired in
the Christian discipline was the practice of the bishops giving those
who travelled abroad letters of authorization, commending them to
the care and hospitality of other bishops in all circumstances. These
are the ways in which Julian attempted to graft onto paganism some
of the features of Christianity. But you need not take my word for it,
for I have in my possession the copy of a letter by Julian himself on
the subject where he corroborates these assertions.°® . . .
5.17 Julian’s artful maneuvers were designed to encourage his sub-
jects to move from Christianity to the old faith. He was intent on rid-
ding the state of the Christian religion, but equally intent in refusing
to employ violent means which might prove embarrassing or seem
tyrannical. Nonetheless, he did everything in his power to lead his
subjects back into paganism, and he was especially persistent with the
soldiers, whom he would sometimes address personally but more
often through a commanding officer. Constantine, as I have men-
tioned, had made the insignia for the army the sign of the cross; Julian
commanded that this be discarded and the ancient standard of the
army be reintroduced in order to accustom the troops to the worship
of the gods. He also commissioned a painting to be displayed in
568. Here Sozomen reproduces Ep. XLIX (To Arsacius). While the material for
5.16 is unique to Sozomen, there is no reason to doubt on the basis of style or con-
tent that the letter put forward by Sozomen to document Julian’s concern over the
perceived “moral authority” of Christianity is authentic.
204 JULIAN’S AGAINST THE GALILEANS

public showing Jupiter coming out of heaven and presenting him


with the imperial symbols—a crown or a purple robe; and also of
Mars or Mercury looking intently at <Julian>, rapt in admiring his
rhetoric and his military prowess. He was always certain to depict him-
self alongside the gods, so that in honoring the emperor they would
also be seen to be worshipping the gods. But he abused the older
usages and disguised his purpose from his subjects, thinking that if
they accepted his sovereignty they would also be willing to obey him
in every particular; and if they did not <obey him> he could then say
that they were traitors to the Roman state because they had offended
against custom and the dignity of the emperor. In fact, there were only
a few men who saw through his scheme and refused to venerate the
pictures, and the full weight of the law was brought to bear against
them. The great mass of people in their childish simplicity or perhaps
in ignorance simply accepted the old ways and venerated his image,
and while Julian got no real advantage from this success, he was unde-
terred in his effort to change the religious balance.
Another measure employed by Julian was more direct and less
subtle, and tested the loyalty of many soldiers assigned to the imperial
household. Each yearit is a custom to give a bonus of money to the
troops, and usually the chosen day is an anniversary of some sort—the
birthday of the emperor, the founding of a city. Julian <believed > that
soldiers are naturally slow and simpleminded, and are -attracted to
money; so he concluded that money might be used beneficially to lure
them into the worship of the gods. Fire and incense having been placed
conveniently near the emperor, as each soldier approached for his
bonus according to the old Roman custom, he was enjoined to offer
sacrifice. A few courageously took the gold but refused to comply;
others were so ignorant of the implications of what they were doing
that they blindly followed habit and custom down the path to sin; still
others were so hungry for gold or so afraid of the consequences if they
did not comply with the pagan rites and the test in front of them that
they fell into temptation, when they should have fled.
I am told°™ that a group of soldiers who had succumbed in this

569. This section is an example of Sozomen’s penchant for creating or propa-


gating pious fabrications.
LIVES OF JULIAN 205

way were at table, raising their glasses in celebration, when one among
them happened to mention the name of Christ. “Strange,” another
said, “that you <of all people> should mention Christ when an hour
ago you denied him by throwing incense into the emperor's fire.” At
this they all became agitated and suddenly conscious of the great
wrong they had done; they rushed into the street where they publicly
wailed and confessed their sin, and asked passersby to bear witness
that what they had done, they had done with the hand only and not
the heart—and that from that day they would live as Christians. They
then presented themselves before the emperor and threw down the
gold, asking him to kill them or subject them to tortures on the spot;
for they said that the sin committed by their hand should be inflicted
on their bodies for the sake of Christ. But the emperor, whatever he
may have felt, refused to put them to death, for he wished to deprive
them of the honor of martyrdom. Instead, he rescinded their military
commission and expelled them from the palace.
5.18 Julian felt the same way about all Christians, and he dis-
played these feelings as opportunities arose. Those who refused to
sacrifice to the gods risked having their citizenship revoked, though
blameless in every corner of their lives—Julian would not permit
them to serve as magistrates, judges, officers, or in the assemblies.
He also prohibited Christian children from attending the schools
and from being taught the Greek poets and writers. He detested
Appolinaris the Syrian, a man noted for his skills in the science of
language; as well as the Cappadocians Gregory and Basil, the
greatest orators of the day, and others who were well-versed <in
Greek>, some of whom defended the Nicene cause, and some of
whom were followers of Arius. His real reason for excluding children
of Christian parents from the study of the Greeks was that he knew
such study would lead to greater skills in rhetoric and argumenta-
tion. So Appolinaris used his gifts in order to produce an epic on the
antiquities of the Hebrews from the time of Saul as a substitute for
Homer's poem. <Like Homer>, he divided his poem into twenty-
four parts and assigned a letter of the Greek alphabet to each in turn.
The same writer produced comedies in the fashion of Menander,
tragedies comparable to those of Euripides, and odes like those of
206 JULIAN’S AGAINST THE GALILEANS

Pindar. His themes came from the scripture, based on his extensive
knowledge of the whole, and he <was able to do this> in a remark-
ably short time, producing works that, in terms of <their> style,
poignancy of expression, character, and structure are the equal of the
Greek literature they emulate in diversity and power.°” I would say
that if the writings of the ancients were not held so unfairly in such
high regard, the writings of Appolinaris would be valued as highly.
For this man was a man of copious intelligence, excelling <others>
in many branches of learning, while ancient writers were usually
proficient in only one. <Appolinaris> wrote, for example, a book
entitled “The Truth,”*” against the emperor and pagan philosophy;
in it he shows without appeal to scripture how pagan ideas about
God are falsely deduced, which Julian, upon reading it, exclaimed in
a letter to the bishops, “I read it; I understood it; I condemn it.”?’2
To which the bishops replied, “You read it, but you did not under-
stand, for if you had you would not have condemned it.” .. .
5.19 Julian was now determined to pursue a war against the Per-
sians, and he went to Antioch in Syria. People were saying that while
goods were abundant, they came at a very high price; so out of gen-
erosity the emperor reduced prices for provisions to such a low level
that the vendors fled the city. And so there was scarcity, and for this
the emperor was blamed. People expressed their anger by ridiculing
the length of his beard and the images of bulls which were stamped
on his coins.
Some said he was butchering his people in the same way the
priests were back in business butchering sacrificial animals. When
these slanders were first heard, Julian was annoyed and made noises
about punishing them before he went off to Tarsus. But in time he
grew calm, and answered their ridicule with a finely crafted work
which he called “Aversion to Beards,” which he sent as a present to
570. No judge of literature, Sozomen here parts company with Socrates’
endorsement of Greek literature and philosophy (HE 3.16). Appolinaris was widely
respected by writers of the fourth century for the elegance of his style. Apart from a
few fragments, all of his writings are lost.
571. Sozomen seems to attribute this work mistakenly to this Appolinaris
(Apolonarius); a book by this title was written by Apollonaris of Hierapolis
according to Eusebius, Ecclesiastical History, 1.35.
572. A letter falsely ascribed to Julian.
LIVES OF JULIAN 207

them. In every other respect he treated the Christians of the region


in just the same way he treated them elsewhere, being sure that he
did all possible to promote the spread of the pagan religion.5’3 . . .
5.22 Though the emperor despised and persecuted the Chris-
tians he displayed only concern and showed kindness toward the
Jews. He sent letters to the leaders and elders among the Jewish
people asking them to pray for him and for the good of the whole
Empire. But I am certain he did this not because he had any real
affection for the religion of the Jews, which he knew was the mother,
as it were, of the Christian faith, as both take their authority from
the patriarchs and prophets; rather, in favoring the Jews he intended
to humiliate the Christians, their most stalwart enemies. Perhaps he
thought, as well, to persuade some of the Jews to accept pagan doc-
trines and practices, for most <of the Jews> knew only a little scrip-
ture and unlike the Christians (and the very wisest Hebrews) they
did not discern the hidden meaning <of the sacred books>.
Later events proved this suspicion true, for Julian sent for the
leaders of the Jewish people and adjured them to return to the
observance of the laws of Moses and their ancient customs, but he
was told that as the temple in Jerusalem had been destroyed it was
not lawful for them to <offer sacrifice> in any other place than the
ancestral city—and they had been expelled from <Jerusalem>. On
hearing this Julian donated money from the treasury and com-
manded them to rebuild their temple, and restore the cult of their
forefathers, and offer sacrifices in the old way. The Jews embarked on
the project without giving a thought to the warnings of the prophets,
which said the task could never be fulfilled.”
6.1575... [After Julian arrived at Ctesiphon, which had been

573. The section (5.20-21) ends with a digression on a miracle associated


with the tomb of Bablas the Martyr in Antioch and the temple of Apollo at Daphne
and events surrounding an order to destroy a statue of Christ in Paneas. Following
sections end similarly. Sozomen relates these legends to prove that Julian by this
point in his reign had incurred divine wrath, the climax of which he will see in his
untimely death in the Persian campaign.
574. Sozomen continues with the legend of an earthquake sent by God as a
manifestation of his disapproval of the project. When the workmen regroup, fire
suddenly bursts forth from the foundations and kills the laborers.
575. Sozomen’s account of the death of Julian is pieced together from a
208 JULIAN’S AGAINST THE GALILEANS

made the imperial residence by the Persians, he had his ships


anchored in the Tigris burned and began a land march of several
days, guided by an old man who turned out to be an agent of the
Persian army. With provisions low and the troops demoralized, the
Romans were ripe for attack.]...The Persians chose this as the
moment to wage their assault, and in the confusion an enormous
wind arose occluding the sun with clouds of dust. It was in the midst
of this conflict that a horseman rode swiftly toward the emperor and
drove his lance clear though him. Some say the horseman was a Per-
sian, others say he was a Saracen, and still others maintain he was a
Roman who had become impatient with Julian’s style of leadership
and his foolishness in exposing his army to such danger. The Syrian
orator Libanius, who was Julian’s friend and companion, writes as
follows concerning the incident: “You wish to know who killed the
emperor? I do not know his name. But we do know that the killer
was not a man from among the enemy—for no one came forward to
claim the reward even though the king of Persia proclaimed through
a herald that rewards were to be given to the one who did the deed.
We are thus in the enemy’s debt, for they did not claim the glory of
the action for themselves, but instead have left us to look for the
killer in our own ranks. Those who most wanted to see Julian dead
were those who broke the law and methodically conspired to do
away with him, lying in wait for the first opportunity to do it. What
they wished to achieve was greater liberty than they could find under
his regime, but they were also provoked by the Emperor's religious
devotion to the gods, which they did not share.”
6.2 As Libanius states clearly in the chapter from which I have
quoted, the emperor died at the hands of a Christian.5”° I am sure
this is true. It is also likely that soldiers in the army came up with the
plot against him, since from the time of the Greeks down to today
the murderers of tyrants have won praise for their willingness to risk
death for the sake of liberty and in the defense of country, family,
and friends. So no one performing this bold deed, done for the sake
number of sources; the outline is that of Socrates (HE 3.21), but the account is per-
meated with legend and improbable detail.
576. Libanius says nothing of this; in this independent chapter Sozomen is
working out his own conjectures.
LIVES OF JULIAN 209

of God and religion, is deserving of any blame. I personally have no


other information concerning the men who committed the murder
other than what I have already presented, but the account itself is
accepted by everyone as evidence of the divine wrath which brought
about <Julian’s> death.
If proof is needed, I will relate the story of a vision sent to one of
the emperor's friends who was traveling to Persia with the hope of
joining forces with the emperor. On his way he found himself so far
from any hostel that he had to put up in a church overnight. During
the night, in a dream or vision, he saw all the apostles and prophets
gathered together and lamenting the injuries which Julian had
inflicted on the church, meeting in council to decide what course of
action should be taken. After a long discussion, two men rose from
the assembly, told those remaining to take heart, and departed
quickly—as though they had decided the way to deprive the emperor
of his power. <When he awokes>, the soldier was so troubled by what
he had seen that he did not continue his journey, but waited for news
to find out how the revelation might be fulfilled; so again he fell
asleep, and this time he saw the assembly convened and the two men
who had left the night before sitting again in their places <at the
council table>. And they told the others, “Julian is dead.”°”’

577. Sozomen continues to relate portents, calamities, and miracles associ-


ated with the death of Julian. Socrates ends his more sober report with the informa-
tion that Julian finished his life in the fourth consulate, on the twenty-sixth of June,
the third year of his reign, at the age of thirty-one (HE 3.21). Theodoret provides
the most famous, if unlikely, detail of Julian’s death in his account (Historia, 3.25),
that when Julian saw he was dying he “cupped his hands to fill them with blood
and hurled it at the sun crying, “Galilean, you have conquered.”
an,

/
BIBLIOGRAPHY

Abbreviations of ancient authorities cited frequently in the notes follow the entry.

I. ANCIENT AUTHORITIES

Albinus. Epitoma. Edited and translated by P. Louis. Rennes, 1945.


Ammianus Marcellinus. Res gestae. Edited and translated by J. C. Rolfe. [In
English] London, 1964. (Amm.)
Ammianus Marcellinus. The Later Roman Empire, 354-378. Translated by
William Hamilton. London, 1986.
Aristides. Orationes. Edited by W. Dindorf. Leipzig, 1829.
Celsus. On the True Doctrine. Translated by R. J. Hoffmann. [In English]
Oxford, 1987. (See below, Origen, Contra Celsum [CS].)
Claudius Mamertinus. Gratiarum actio Mamertini de consulate suo Iuliano
Imperatori. Edited by C. E. V. Nixon and Barbara Satlor Rodgers. In In
Praise of Later Roman Emperors: The Panegyrici Latini. Berkeley, 1994.
(Mamertinus)
Codex Theodosiansus. Edited by Theo. Mommsen, P.M. Meyer, and P.
Krueger = Theodosiani libri XVI cum constitutionibus sirmondianis et leges
novellae ad Theodosianum pertinentes. 2 vols. Berlin, 1905. (Cod. Theod.)
Consularia Constantinopolitana. Edited by T. Mommsen. 1892. Berlin, 1961.
Cyril of Alexandria. Contra Julianum (Pro Christiana religione adversus
Julianum Imperatorem). Patrologiae Graecae Cursus Completus, vol. 76.
Edited by J-P Migne. 81 vols. Paris, 1856-61. (PG)
Ephraem Syrus. Contra Julianum. Edited by E. Beck. Louvain, 1957.

eri
oe BIBLIOGRAPHY

Epitome de Caesaribus. Edited by F. R. Pichlmayr and R. Gruendel. Berlin,


1961.
Eunapius Fragmenta Historicorum Graecorum. Edited and translated by R. C.
Blockley. The Fragmentary Classicizing Historians of the Later Roman
Empire. 2 vols. Liverpool, 1983. (Eunapius)
Eusebius. Praeparatio evangelica. Edited and translated by E. Gifford.
Oxford, 1903.
. Vita Constantini. Edited by F. Winkelmann. Berlin, 1975.
Firmicus Maternus. De errore profanorum religionum. Edited and translated
by R. Turcan. [In French] Paris, 1982.
Gregory Naziazen (Nazianzos). Orations IV—V. In Iulianum, I—II. Edited
and translated by J. Bernardi. [In French] Paris, 1983. (Greg., Or.)
. Historica. PG 37.
Eutropius. Breviarum ab urbe condita. Edited and translated by H. W. Bird.
Translated Texts for Historians, vol. 14. Liverpool, 1993. (Eutropius)
Corpus inscriptionum latinorum. Vol. 9. Edited by T. Mommsen. Berlin, 1883.
Gregory of Naziazen. Orationes, IV—V. Edited by J. P. Migne. Patrologiae
Graecae, vol. 35. Paris, 1864. (Gregory, Or.)
Hilary. Liber II ad Constantium. Edited by A. Feder. Corpus scriptorium eccle-
siasticorum latinorum, vol. 60. Vienna, 1916. (ad Const.)
Julian. Contra Galilaeos.Juliani imp. Librorum contra Christianos quae supersunt.
Edited by K. J. Neumann. Leipzig, 1880. (CG) Note: Marginal numbers
in the Greek text refer to the pagination of Spanheim’s 1696 edition of
Cyril of Alexandria's polemic against Julian, Pro Christiana religione, used
by Neumann in his 1880 reconstruction of Julian’s treatise.
. Giuliano imperatore, Contra Galilaeos. Edited by E. Masaracchia.
Rome, 1991.
. The Works of the Emperor Julian, I-III. Edited and translated by W. C.
Wright. [In English] London, 1913-1923.
. Epistulae, leges, poematia, fragmenta varia. Edited by J. Bidez and F.
Cumont. Paris, 1922. (Ep.)
. Lempereur Julien: Ouvres completes, I-II. Edited by J. Bidez, G.
Rochfort, and C. Lacombrade. Paris, 1924-1964.
. luliani imperatoris quae supersunt. Edited by F. C. Hertlein. Leipzig,
1875-1876.
Libanius. Orationes. Edited and translated by A. F. Norman. [In English]
Libanius: Selected Works. 3 vols. London, 1969. (Libanius, Or.)
. Opera, I-XII. Edited by R. Foerster. Lepzig, 1903-1927.
Origen. Contra Celsum. Translated by H. Chadwick. Cambridge, 1953.
BIBLIOGRAPHY 213

. Contra Celsum. Edited by M. Borret. Paris, 1967-76. (C. Cels.)


Orosius. Adversus paganos historiarum libri septem. Edited by Z. Zangemeister.
Corpus Scriptorum Ecclesiasticorum Latinorum (CSEL), edited by the
Academy ofVienna, 1866-1957. CSEL V, Vienna, 1882. (Orosius)
Panegyrici latini. Edited by R A. B. Mynors. Oxford, 1964.
Philostorgius. Historia Ecclesiastica. Edited by J. Bidez. GCS, XXI. Paris,
1913. (Philos., Hist.)
Plato. Opera. Edited by J. Burnet. Oxford, 1900-1907.
Plotinus. The Enneads, I-VII. Edited and translated by A. H. Armstrong.
Cambridge, MA, 1966-1988. (Enn.)
Porphyry. Adversus Christianos. Edited by A. v. Harnack. Berlin, 1916.
(Abhandlung der koen.-Preuss. Akademie der Wiss., Phil.-hist. Kl. I
[fragments].)
Porphyry: Against the Christians: Fragments from the Apocriticus of Macarius
Magnes. Edited and translated by R. J. Hoffmann. [In English] Amherst,
NY, 1994.
Proclus. Theologica Platonica. Edited by H. D. Saffrey and L. G. Westerink.
Paris, 1968.
Sextus Aurelius Victor. Liber de Caesaribus. Edited by F. R. Pichlmayr and R.
Gruendel. Berlin, 1961.
Socrates (Scholasticus). Historia ecclesiastica. Edited by J-P Migne. PG, 67.
Paris, 1864. (Socrates, HE)
. Historia Ecclesiastica. Edited by R. Hussey. Oxford, 1853.
Sozomen. Historia ecclesiastica. Griechischen Christlichen Schriftsteller der
ersten drei Jahrhunderts (GCS) (1891- ). Edited by A. von Harnack and
Theodor Mommsen. GCS 50 Berlin and Leipzig, 1960. (Soz., HE)
Theodoret. Historia ecclesiastica. Edited by G. Parmentier. Berlin, 1911.
(Theod., HE)
Zosimus. Nova Historia. Edited and translated by Francois Paschoud.
Zosime: Histoire nouvelle. 3 vols. Paris, 1971-1989. (Zosimus)

II. MODERN INTERPRETATIONS AND STUDIES

Note: This list represents a selection of works relating principally to Julian


and his religious and social context, and includes works cited frequently in
the notes as well as studies consulted or reviewed but not cited. Standard
journal abbreviations are keyed to the abbreviations list of the University
214 BIBLIOGRAPHY

of Toronto (Journals of Interest to Classicists: TOCS-IN) list: http://www


.chass.utoronto.ca/amphoras/tdata/inform.toc. Otherwise, the full name of
the journal is given.

Alfoeldi, A. “Some Portraits of Julian Apostata.” AJA 66 (1962): 403-405.


Allard, P. Julien l’Apostat. 3 vols. Paris, 1906-1910.
Andreotti, R. I] regno dell'imperatore Giuliano. Bologna, 1936.
Andresen, C. Logos und Nomos: Die Polemik des Kelsos wider das Christentum.
Berlin, 1955.
Arce, J. Estudios sobre el Emperador Flavio Claudio Juliano. Madrid, 1984.
Armstrong, A. H. “The Way and the Ways: Religious Tolerance and Intoler-
ance in the Fourth Century AD.” VChr 38 (1984): 1-17.
Athanassiadi, P. “A Contribution to Mithraic Theology: The Emperor
Julian’s Hymn to King Helios.” JThS 28 (1977): 360-71.
. Julian: An Intellectual Biography. London, 1992; repr. Of Julian and
Hellenism, Oxford, 1981, with bibliographical additions.
Baldwin, B. “The Caesares of Julian.” Klio 60 (1978): 449-66.
Balty, J. “Julien et Apamée: aspects de la restauration de I’hellénisme et de —
politique antichrétienne de l’empereur.” Dialogues d'histoire ancienne 1
(1974): 267-304.
Barnes, T. D. “Poprhyry Against the Christians: Date and Attribution of
Fragments.” JThS 24 (1973): 424-42.
. “Christians and Pagans in the Reign of Constantius.” In L’église et
l'Empire au IVe siécle. Geneva, 1989.
. Athanasius and Constantius. Cambridge, MA, 1993.
Beard, M., and J. North, eds. Pagan Priests. London, 1990.
Benedetti, I. “Giuliano in Antiochia nell’orazione XVIII di Libanio.”
Athanaeum 59 (1981): 166-79.
Bianchi, U., and M.Vermaseren, eds. La soteria dei culti orientali nell’ impero
romano. Leiden, 1992.
Bidez, J. La tradition manuscrite et les éditions des discours de l‘empereur Julien.
Paris, 1929.
. La vie de l’empereur Julien. Paris, 1930.
Blanchetiere, F. “Julien: philhelléne, philosémite, antichrétien.” Journal of
Jewish Studies 33 (1980): 61-68.
Blockley, R. C. “Constantius, Gallus and Julian as Caesars of Constantius
II.” Latomus 31 (1972): 433-68.
Bouffartigue, J. L’empereur Julien et la culture de son temps. Paris, 1992.
Bowersock, G. W. Julian the Apostate. London, 1978.
BIBLIOGRAPHY aha

——.. “Emperor Julian on His Predecessors.” YCIS 27 (1982): 159-72.


. “From Emperor to Bishop: The Self-conscious Transformation of
Political Power in the Fourth Century.” CP 81 (1987): 235-51.
Bradbury, S. G. “Innovation and Reaction in the Age of Constantine and
Julian.” PhD dissertation, Berkeley, 1986.
Braun, R., and J. Richer, eds. L’empereur Julien: de l'histoire a la légende. Paris,
1978.
Brock, S. P. “A Letter Attributed to Cyril of Jerusalem on the Rebuilding of
the Temple.” BSOAS 40 (1977): 267-86.
Browning, R. The Emperor Julian. London, 1976.
Caltabiano, M. “Il comportamento di Giuliano in Gallia verso i suoi fun-
zionari.” Acme 32 (1979): 417-42.
Calza, R. Iconografia romana imperiale da Carausio a Giuliano. Rome, 1972.
Cameron, Av. Christianity and the Rhetoric of Empire. Berkeley, 1991.
Chauvin, P. A Chronicle of the Last Pagans. Cambridge, MA, 1990.
Cinnock, E. J. A Few Notes on Julian and a Translation of His Public Letters.
London, 1901.
Cracco-Ruggini, L. “The Ecclesiastical Histories and Pagan Historiography:
Providence and Miracles.” Ath. 55 (1977): 107-26.
Croke, B. ‘The Era of Porphyry’s Anti-Christian Polemic.” Journal of Reli-
gious History 13 (1984): 1-15.
, and J. Harries. Religious Conflict in Fourth Century Rome. Sydney,
1982.
Cumont, F. Sur l’authenticité de quelques lettres de Julien. Gand, 1889.
Daly, L. “Themistus’ Plea for Religious Tolerance.” GRBS 12 (1971): 65-81.
. “In a Borderland: Themistus’ Ambivalence to Julian.” Byz. Zeitschr.
73 (1980): 1-11.
Di Maio, M., and W. H. Arnold. “Per vim, per caedum, per bellum: A Study
of Murder and Ecclesiastical Politics in the Year 337 AD.” Byzantion 62
(1992): 158-91.
Dillon, J. The Golden Chain: Studies in the Development of Platonism and
Christianity. London, 1900.
Downey, G. “Philanthropia in Religion and Statecraft in the Fourth Cen-
tury.” Historia 4 (1955): 199-208.
_ “Themistius and the Defence of Hellenism.” HThR 50 (1957):
259-74.
Drachman, A. B. Atheism in Pagan Antiquity. London, 1922.
Drinkwater, J. E “The Pagan Underground: Constantius’ II’s Secret Service
and the Survival and Usurpation of Julian the Apostate.” In Studies in
216 BIBLIOGRAPHY

Latin Literature and Roman History, edited by C. Deroux, III:348-87.


Brussels, 1983.
Dvornik, E. “The Emperor Julian’s Reactionary Ideas on Kingship.” In Late
Classical and Medieval Studies in Honor of A. M. Frend, edited by K.
Weitzmann, 71-81. Princeton, NJ, 1955.
Ensslin, W. “Kaiser Julian’s Gesetzgebungswerk und Reichsverwaltung.” Klio
18 (1923): 104-109.
Festugiére, A-J. “Julien 4 Macellum.” JRS 47 (1957): 53-58.
Fornara, C. W. “Julian’s Persian Expedition in Ammianus and Zosimus.”
JHS 111 (1991): 1-15.
Fortin, E. L. Ad Adulescentes in Neoplatonism and Early Christian Thought.
London, 1981.
Geffcken, J. The Last Days of Graeco-Roman Paganism. Translated by S. Mac-
Cormack. Amsterdam and New York, 1978.
Gibbon, E. Decline and Fall of the Roman Empire. Edited by J. B. Bury.
London, 1909.
Gilliard, F “Notes on the Coinage of Julian the Apostate.” JRS 54 (1964):
135-41.
Gleason, M. “Festive Satire: Julian’s Misopogon and the New Year at
Antioch.” JRS 76 (1986): 106-19.
Gregory, T. “Julian and the Last Oracle at Delphi.” GRBS 24 (1983):
355-66.
Guida, A. “Frammenti inediti del Contra i Galilei e della replica di
Theodoro de Mopsuesta.” Prometheus 9 (1983): 139-63.
. Un anonimo panegyrico per l'Imperatore Giuliano: Introduzione, Testo
Critico, Commento. Florence, 1990.
Halsberghe, G. The Cult of Sol Invictus. Leiden, 1972.
Harl, K. “Sacrifice and Pagan Belief in Fifth and Sixth Century Byzantium.”
Past and Present 128 (1990): 7-27.
Hunt, E. D. “Christians and Christianity in Ammianus Marcellinus.” CQ 35
(1985): 186-200.
Kaegi, W. E. “Emperor Julian’s Assessment of the Significance and Function
of History.” Proc. Amer. Phil. Soc. 108 (1964): 29-38.
. “Research on Julian the Apostate: 1945-1964.” CW 58 (1965):
229-38.
King, C. W. Julian the Emperor: Containing Gregory Naziazen's Two Invectives and
Libanius’ Monody with Julian's Extant Theosophical Works. London, 1888.
Labriolle, P. La Réaction paienne. étude sur la polemique antichrtienne du ler au
Vile siécle. Paris, 1934.
BIBLIOGRAPHY Ze

——. “Julian the Apostate.” In Palanque, Bardy, Labriolle, et al. The


Church in the Christian Roman Empire, translated by E. C. Messinger,
1:229-39. London, 1949.
Lacombrade, C. “Julien et la tradition romaine.” Pallas 5 (1960): 155-64.
. “Lempereur Julien, émule de Marc-Auréle.” Pallas 14 (1967): 9-22.
Lane Fox, R. Pagans and Christians. Harmondsworth, 1986.
Liebeschuetz, J. Continuity and Change in Roman Religion. Oxford, 1979.
Lieu, J., J. North, and T. Rajak, eds. Jews among Pagans and Christians.
London, 1992.
Lieu, S., ed. The Emperor Julian: Panegyric and Polemic. Liverpool, 1986.
MacDonald, D. “Another Representation of the Sasanid Triumph Over
Julian.” JNC 28/29 (1978-79): 32-33.
Macmullen, R. Paganism in the Roman Empire. New Haven, CT, 1981.
Malley, W. J. Hellenism and Christianity: The Conflict between Hellenism and
Christian Wisdom in the Contra Galilaeos of Julian the Apostate and the
Contra Julianum of St Cyril of Alexandria. Analecta Gregoriana, vol. 210.
Rome, 1978.
Mau, G. Die Religionsphilosophie Kaiser Julians. Leipzig, 1907.
Meredith, A. “Porphyry and Julian against the Christians.” ANRW II.23.2,
1119-49.
Millar, Ek “Jews of the Graeco-Roman Diaspora between Paganism and
Christianity, AD 312—438.” In J. Lieu et al., Jews among Pagans and
Christians, 97-123. London, 1992.
Momigliano, A., ed. The Conflict between Paganism and Christianity in the
Fourth Century. Oxford, 1963.
. On Pagans, Jews and Christians. Middletown, CT, 1987.
Neander, A. The Emperor Julian and His Generation. Translated by G. V. Cox.
London, 1850.
Negev, A. “The Inscription of the Emperor Julian at Ma’ayan Barukh." Israel
Expl. Journal 19 (1969): 170-73.
Nillson, M. Geschichte der griechischen Religion, II. Munich, 1962.
O'Donnell, J. “The Demise of Paganism.” Traditio 35 (1979): 45-88.
Paschoud, F. “Trois livres recents sur l’empereur Julien.” REL 58 (1980):
107-23.
Pack, E. Staedte und Steuern in der Politik Julians. Brussels, 1986.
Rendall, G. H. The Emperor Julian: Paganism and Christianity. London, 1979.
Ricciotti, G. Julian the Apostate. Rockford, IL [ET]. 1999; Italian original,
L'imperatore Giuliano I’Apostata, Verona, 1956.
Richer, J., ed. L’‘empereur Julien: de la legende au mythe. Paris, 1981.
218 BIBLIOGRAPHY

Ridley, R. “Notes on Julian’s Persian Campaign.” Historia 22 (1973): 317-30.


Rostagni, A. Giuliano l'apostata: Saggio critico con le operette politiche e satiriche
tradotte e commentate. Turin, 1920.
Rothrauff, C. M. “The Philanthropia of the Emperor Julian.” PhD disserta-
tion, University of Cincinnati, 1967.
Sacks, K. S. “Meaning of Eunapius’ History.” History and Theory 25 (1986):
52-67.
Salaman, M. “La conception de l’empereur Julien l’apostat pour la réorga-
nization du monnayage romain.” Wiadomosci Numiizmatyczne 23
(1979): 20-30.
Scheda, G. “Die Todesstunde Kaiser Julians.” Historia 15 (1966): 380-84.
Schwartz, W. De vita et scriptis Iuliani Imperatoris. Bonn, 1988.
Sciolone, S. “Le accezioni dell’appellativo ‘Galilei’ in Giuliano il Apostata.”
Aevum 56 (1982): 71-80.
Sihler, E. G. “The Emperor Julian and His Religion.” In Augustus to Augustine:
Essays and Studies Dealing with the Contact and Conflict of Classic Paganism
and Christianity, edited by E. G. Sihler, 190-217. Cambridge, 1923.
Simpson, W. D. Julian the Apostate. Aberdeen, 1930.
Smith, A. Porphyry’s Place in the Neoplatonic Tradition: A Study in Post-Plo-
tinian Neoplatonism. The Hague, 1974.
Smith, R. Julian's Gods; Religion and Philosophy in the Thought and Action of
Julian the Apostate. London, 1995.
Thompson, E. A. “The Emperor Julian’s Knowledge of Latin.” CR 58
(1944): 49-51.
. The Historical Work of Ammianus Marcellinus. Cambridge, 1947.
Truempelmann, L. “Triumph uber Julian Apostata.” JNG 25 (1975):
107-11.
Vogt, J. Kaiser Julian und das Judentum: Studien zum Weltanschauungskampf
der Spdtanike. Leipzig, 1939.
Wallace-Hadrill, A. “The Emperor and His Virtue.” Historia 30 (1981):
298-323.
Wallis, R. T. Neoplatonism. London, 1972.
Warren-Bonfante, L. “Emperor, God and Man in the Fourth Century: Julian
and Ammianus Marcellinus.” PP 19 (1964): 401-27.
Wilken, R. The Christians as the Romans Saw Them. New Haven, CT, 1984.
Wirth, G. “Julians Persekrieg: Kriterien einer Katastrophe.” In R. Klein,
Julian Apostata, 455-507. Darmstadt, 1978.
j#
iY


he

4g

- >»

a
%
—.

” ie

hy

he’

——>
=
™~


A
4

. Y

,
x

‘ ¥

~ &
7 »

.
og

wi “ad

«a
ae aS
»....
ac

\\
us
J a 7 ~ ;
7 _ 7 . ws
ae 7
aig :
~— ie :
\
=d a
age
ae 7
ro
a ai
_ : ;
7
’ ~~
A
ie ‘
7 7a 4 = ”” ate
,
i

aa
re

:
Psss
J
“a

:
7

ok
.
SveJ

3
.
= —- |

,
"
are

ar
=

rs
=

-:
_

|\
a

4
7

,
We

*
a

;

vi

‘a


op ie keg oe
i _
Pi

ode — _ a a ek ; \ Sy.
oe: Si * ’ . Aaa 7a ao i Peaa rr
BR 160.3 .J85 2004
Julian, 331-363.
Julian’s against the
Galileans

DATE DUE
*

ia

Qr

#47-0108 Peel Off Precsiire Sancitiva


Christian church against the background of
intellectual and religious opposition.:The
translations are supported by a full historical
introduction to the life of Julian and a de-
tailed treatment of his religious philosophy,
including the origins of his understanding of
the Christian faith.
This translation is essential reading for
anyone interested in the religions of late an-
tiquity, the growth of the Christian church,
and the final phase of the conflict between
paganism and Christian teaching.

R. JOSEPH HOFFMANN is the author of


many books on early Christianity including
Porphyry’s Against the Christians. He is
Campbell Professor of Religion at Wells
College, New York, and chair of the
Council for the Scientific Examination of
Religion. He has taught at the University of
Michigan, Oxford University, and the
American University of Beirut.

Photo courtesy of The American Numismatic Society -


Cover design: Jeff Schaller
— v7—* ie Ae ———_—_—

“Julian the Apostate sought to create a kind of ‘Church of Paganism’ as a viab &
native to the Christian religion in which he was reared, but which he repud
take up his ancestral gods and their rites. Thus it is fitting that Julian be trea
Pagan Church Father. That is the treatment given him at last by the erudite R
Hoffmann, whose earlier-reconstruction of Porphyry’s Against the Christi
Celsus’s The True Word allowed these great prophetic anti-Christian voices t
forth again, no longer muffled and gagged by the Christian polemical co 39371
00058
6586
which form alone Mother Church permitted them to survive. Bravo!—both@s
fact of this book and its sterling quality!”
Robert M. Price
Johnnie Colemon Theological Seminary
Author of Deconstructing Jesus and The Incredible Shrinking Son of Man

“Professor Hoffmann in a very solid work has laid the foundation for any future
study of Julian the Apostate. His book is a great achievement, and I will consult
it frequently.”
Dr. Gerd Liidemann
Professor of the History and Literature of Early Christianity
University of Gottingen, Germany
Author of Paul: The Founder of Christianity and Jesus After Two Thousand Years

Julian, nephew of the first Christian emperor, Constantine I, is one of the most fas-
cinating figures in late antiquity. Ruling during the turbulent years of the waning
Roman Empire and the growth of Christianity, his reign inspired both admiration and
contempt. In an attempt to restore the traditional religions of the empire, Julian
argued in Against the Galileans that Christianity is not a universal religion but a local
movement spun out of control.
~.. Noted professor. of religion R. Joseph Hoffmann presents the first modern
English translation of this important work and related writings. Hoffmann’s histori-
cal introduction to Juhan’s life and a detailed treatment of his religious philosophy is
invaluable for anyone interested in the final phase of the conflict between paganism
and Christian teaching.

Sa 2

ISBN L~S510e°L58-7
90000

Prometheus Books
59 John Glenn Drive
iH “=A mherst, New York 14228-2197
9"781591"0 21988"
www.prometheusbooks.com

You might also like