OATHS OF ALLEGIANCE TO CAESAR
The ancients were habitual takers of oaths, both personal and corporate, in
private and in public, and their literature abounds in examples. The social
and religious significance of oaths is of general interest, but for the
historian it is those periods when oaths had a more direct political impor
tance which attract particular attention; and amongst these the oaths of
allegiance sworn to the Roman emperors deserve careful study by students
of the early Empire. We are in a relatively favourable position to do this,
because the considerable literary evidence for the period about vows and
oaths of this type has been paralleled by the discovery of a number of
inscriptions with the actual text of oaths taken in a variety of places in the
provinces of the Empire. A number of questions naturally arise. How much
impact did these oaths of allegiance have on the minds and attitudes of the
civil and military officials who administered them in the emperor’s name,
and on the ordinary citizens who assented to them? Can we form any judg
ment on how many actually participated in such formal ceremonies? What
contribution did the regular oath-taking make to imperial solidarity, when
compared with the rites of the emperor cult, the propaganda of the coinage
and the honorific inscriptions set up in so many public places throughout
the Empire? What signs are there of conscientious objection to the taking of
oaths in general as well as to these particular oaths of allegiance? These are
large questions, and the aim of this article is to review briefly some of the
literary and the documentary evidence for the period and to stimulate
responses to them.
We need not doubt that most Romans took seriously the whole apparatus
of oaths, vows and prayers by which it was hoped that the gods would
reward both individuals and the state with their protection and with pros
perity. For the Republican period Livy provides ample evidence of the dif
ferent types of oaths used amongst the Romans, and the narrative of Book
XXII, at the centre of the Hannibalic campaign in Italy, will serve as an
illustration. At the time prior to Cannae when the Roman forces were sup
plemented by those of the Latin allies, Livy notes that the solemnity of this
juncture in the war was marked by the requirement that the allies should
take a formal oath at the instance of the military tribunes. They pledged
themselves to obey the consul’s command in the field as long as the cam
paign lasted. This action, described as ‘legitima iuris iurandi adactio’,' is
1. X X II.38.
compared with the sacramentum which had sufficed for military discipline
before that time, the voluntary oath of allegiance to one’s commander
which recruits swore when they first joined their unit. Its terms were ‘not to
depart in order to take flight or through fear, nor to retreat from the line ex
cept to recover or obtain a weapon, strike a foe or rescue a friend’. It will be
noticed later that military oaths were an important constituent in the civil
oaths of allegiance in the Empire.
Later in the same Book there is an example of the individual oath. At
Canusium after the Cannae disaster P. Cornelius Scipio burst in upon the
nobiles who were gathered with Metellus to plan their escape from Italy
and, brandishing his sword, swore a solemn oath: ‘I swear from my heart
that I shall not desert the republic of the Roman people, and shall not allow
any other Roman citizen to desert her. If I knowingly swear falsely, may
you, Jupiter Optimus Maximus, afflict me, my house, family and estate
with the worst of deaths.’ Here are characteristic elements—the adjuration
by Jupiter and the curse for perfidy—of the imperial oaths of allegiance
discussed below. But it was the practice of the late Republic which set the
precedents, the increasing political use both of military and civil oaths and a
tendency to combine their functions. The sacramentum remained basic on
the military side; behind the oath of loyalty to one’s commander stood
one’s loyalty to the Roman Senate and people, by whose authority the com
mander held his appointment. A more comprehensive act was the military
coniuratio, whereby a body of men swore obedience to their leader in an
emergency in response to the challenge ‘let him who wants the salvation of
the Republic follow me.’2 In this category may be placed the oath taken to
M. Livius Drusus by his adherents in the Social War in 91 B.C., a command
not sanctioned by the Senate. The terms as given by Diodorus resemble
those of the imperial oaths later: ‘I swear by Capitoline Jupiter, by the
demigods who became founders of Rome and by the heroes who extended
her suzerainty to reckon the same friend and foe as Drusus, and not to spare
my life or the lives of my children or parents if this is not expedient for
Drusus and those who have taken the same oath. If I swear truly, may I
enjoy prosperity; if I forswear myself, the opposite.’3 A further political
extension of the practice was the taking of an oath to abide by a com
mander’s acts by those who wielded less power; thus Sulla as proconsul
obliged Cinna when he took up the consulship to swear loyalty ‘with oaths
and curses’, an action full of portents for the future. Loyalty to an im-
2. Servius o n V erg. Aert. 8.1. F o r the text a n d fu rth e r references see S tefan W einstock, Divus
Julius p .224 η. 1.
3. D iod. 37.11.
OA TH S OF A LLEG IA NCE TO CAESAR
perator was superseding loyalty to the state.4
It was clearly the use by Julius Caesar of the apparatus of oaths, vows
and prayers which led to the imperial practice, Augustus proving himself an
apt pupil in this regard. Although he had earlier used them to advantage, it
was in the last two years of his life that Caesar surrounded himself more
openly with these manifestations of power. Public prayers were offered
annually for his safety, in acknowledgement that Rome’s salus and in
columitas were now centred on that of the dictator.5 When magistrates
entered office they had to swear to support his acta both past and future.
Curiously, we lack direct evidence of a voluntary oath of allegiance sworn
to him at the same time, but because of the regular practice of Augustus and
his successors it is reasonable to believe that this oath was indeed taken, and
that the terms resembled those used by Drusus as mentioned above and, it
seems, repeated in the oaths of the early Empire. Caesar’s name will have
been invoked alongside those of Jupiter and other deities, a most significant
adumbration of the imperial cult. In addition, Caesar fostered the growth
of a more personal cult; oaths were taken by his Fortune (Fortuna or Tyche)
and probably by his Genius. A public sacrifice was instituted in honour of
his birthday. At his funeral, according to Suetonius, Antony got a herald to
read the texts first of the decree which had bestowed on Caesar ‘all honours
human and divine’ and then of the oath taken by all Senators ‘pro salute
Augustus: the Paphlagonian and Samian oaths
As triumvir Octavian along with the lesser magistrates publicly took the
oath to observe Caesar’s acta, but it was at the expiration of the trium
virate, when it was essential to appear as the legitimate Roman leader
against Antony and Cleopatra, that in 32 B.C. he took the boldest step with
the coniuratio Italiae. By this oath he bound to himself (still a faction leader
in reality) the Roman forces of the West, for all these provinces with Africa
were included.7 How voluntary was it, as Augustus later claimed? Syme
comments ‘When an official document records voluntary manifestation of
popular sentiment under a despotic government, a certain suspension of
belief may safely be recommended.” However, the Caesarian veterans
settled in the coloniae formed a solid nucleus, and it may be conjectured
that the municipal senates, perhaps over a period of months, passed resolu
4. W ein sto ck o p .c it., p.222 co m pares S u lla’s a c tio n w ith C a e sa r’s bin d in g o f candidates for
office by a n o a th w hile he w as in G aul. F o r S ulla, P lu t. Sulla 10.6.
C ass. D io. 44.6.
Res Gestae 25.2.
Roman Revolution, p.284.
B.F. HA RRIS
tions incorporating the terms of the oath upon the motion of the resident
clientes of both Caesar and Octavian.
With the coniuratio we are but a short step from the imperial oaths of
allegiance, and it is natural to conclude that its terms, which have not sur
vived, were very similar to those discussed below.9 In all provinces after Ac
tium the legionary recruit took the sacramentum to him personally, whose
imperium surpassed that of all other commanders. Augustus’ name was in
cluded in all the prayers offered by priests on behalf of the Senate and
Roman people, and magistrates together with senators swore to abide by his
acta. 10 On the personal side, libations were to be poured to him at private as
well as public banquets,11 and oaths by his Genius and Fortuna were
extended to the cult of the numen Augusti.
These developments provide the background for the important inscrip
tion from Paphlagonia, discovered at Phazimon-Neapolis and dated 3 B.C.
shortly after the annexation of the area to the province of Galatia. The text
is that of the oath sworn at Gangra, the capital, with the information at the
end of the inscription that along with all the inhabitants of Paphlagonia the
people of Phazimon took this same oath ‘at the altars of Augustus in the
temple of Augustus’.12 The preamble states that the resident Roman
businessmen took it with the provincials.
The oath embodies virtually all the elements found at lesser length in the
other imperial examples—the invocation of Jupiter and all other deities, the
personal pledge of loyalty to Augustus and his house, the total commitment
to protect his interests, reckoning the same friends and foes; the undertak
ing to lay information against potential enemies, and to pursue and punish
Augustus’ foes ‘by land and sea, with weapons and sword’.13 The oath con
cludes with the pronouncement of a solemn curse upon the oath-taker and
his descendants if he fails to fulfil his word. It is noteworthy that ‘Augustus
himself’ is added to the list of deities, and the swearing of the oath at the
Caesareion in each Paphlagonian town shows how the oath of allegiance
and the emperor cult were strongly linked from the beginning. Although
Augustus resisted deification, it is evident that in the East he was counted as
In 1960 the substantial but fragmentary text of a decree at Samos was
9. So P .A . B ru n t a n d J.M . M o o re (eds.), Res Gestae Divi Augusti, p .68.
10. C ass. D io . 53.28.
11. C ass. D io . 51.19.
12. T h e ex tan t texts o f th e o a th s w ere conveniently p rin te d in A pp en d ix 1 o f P e te r H errm a n n ,
Der rOmische Kaisereid (pp.122-6). A tra n sla tio n o f the p resen t one (IL S 8781 = E J 315)
ap p ears in N . Lewis a n d M . R ein h o ld , Roman Civilization II: The Empire, pp.34-5.
13. cf. th e L a tin o f th e A ritiu m a n d S entinum o ath s, ‘arm is belloque internecivo te rra m ariq u e ’ (in fra).
OATHS OF A LLEG IA NCE TO CAESAR
published, which also belongs to the Augustan period.14 Most of what sur
vives is from the honorific decree rather than the oath, but it is clear that
there were processions, sacrifices and a banquet and the oath of allegiance
seems to have been administered by the strategoi at a formally convened
meeting of the citizen assembly. It was resolved also to send an embassy to
congratulate Augustus (on the anniversary of his accession?), its leader to
be ‘the priest of Imperator Caesar Augustus and of his son Gaius Caesar
and of Marcus Agrippa’. Copies of the decree were to be set up on marble
pillars in the shrines of Augustus and of the goddess Hera. Again we
observe the close links between the oath of allegiance and the imperial cult,
which is here extended (as in fact was the oath) to include Augustus’ family.
Tiberius: the Cypriot oath
For the remainder of the Julio-Claudian period we have the evidence of
Tacitus on oaths to add to that of Cassius Dio and Suetonius. The taking of
the oath to Tiberius in A.D. 14 is put by him in the context of ‘servitude’ (at
Romae ruere in servitium consules, patres, eques)15 but he gives the fullest
details we have for Rome—first the consuls swore the oath, then the
prefects of the praetorian guard and of the corn supply, then ‘the Senate,
army and people’. In spite of Tacitus’ deep scepticism about Tiberius’ pro
fessed reluctance to assume power on the death of Augustus, in a later
passage he notes that the emperor declined to accept the oath of loyalty to
his own acta when the Senate tried to enact it,16 on the grounds that the
more powers he was given, the more precarious his real position became.
But he insisted on the recognition of the deified Augustus in this manner
and took that oath himself.17 These two types of oath, in nomen Augusti
and in acta Caesaris are of course closely related, but it is the former, the
pledge of personal loyalty, with which the surviving inscriptions are con
cerned. According to Dio, who gives a much more favourable account of
Tiberius’ motives, the manner of taking the oath of allegiance at Rome was
for a single senator to swear it on behalf of all, with the others assenting by
voice, and the only occasion when they returned (voluntarily) to the old
method of individual pledges was after the discovery of Sejanus’ con
spiracy, as a gesture of reassurance to Tiberius.1* He claims also that
Tiberius would not allow oaths by his Fortuna or the observance of his bir
thday with vows and sacrifices,19 but we may well doubt whether this lasted
P . H e rrm a n n , Annates du Midi, 75, 1960, 70ff. H e p ro p o sed a d a te 6 /5 B.C .
Ann. 1.72. neq u e in a c ta sua iu rari q u a m q u a m censente senatu perm isit.
Ann. 4.42 (rem o v al o f a se n a to r fo r refusing to sw ear).
C ass. D io. 58.17.2.
C ass. D io . 57.7-8.
more than the first short period of his principate.
This is the literary context for an important inscription discovered near
Palaipaphos in Cyprus in 1959 and published by T.B. Mitford the next
year.20 It contains most of the text of the Cypriot oath of allegiance to
Tiberius in A.D. 14. Amongst the deities invoked are Aphrodite ‘our own
Maiden’, Apollo and the Dioscuri, with Augustus’ name alongside them (as
the descendant of Aphrodite) and also Rome. The Cypriots swear to obey
and to worship Tiberius and all his house, to reckon the same friends and
foes and intend (it appears from the fragmentary ending) to vote divine
honours to Rome, Tiberius and ‘the sons of his blood’. One interesting
detail is the space left after the second occurrence of Tiberius’ titles,
presumably for the later insertion of ‘Autokrator’ (Imperator), from which
Mitford concludes that the oath was to be regularly administered. Here is
more evidence of the deification of the living emperor which was natural to
the Hellenistic cities of the East.
Gaius: the Assos and Aritium oaths
Both Dio and Suetonius testify to the efforts made by Gaius upon his
accession to emphasise his autocratic power, in spite of His scrupulous
adherence to Augustan precedents in some areas. He annulled Tiberius’ will
and refused to have oaths sworn to his acta, but included his own with those
of Augustus.21 More than this, he inserted the names of his sisters not only
in the public prayers for his welfare and that of Rome, but in the oath of
allegiance itself. Senators once more swore the oath individually.22 Accord
ing to Suetonius, Gaius made it compulsory to swear by his Genius, on pain
of execution, and he encouraged the worship of his numen and even himself
swore by the numen of his dead sister Drusilla.23
Whatever the accuracy of this in detail, there is no mistaking the general
picture, and it is appropriate that the documentary evidence of oaths
accords with it. The decree and oath of A.D. 37 found at Assos in the Troad
show the same features as those already cited, with some interesting varia
tions in the text and setting.24 The decree is on the motion of the demos as a
whole, but stood in the names of the Council and the resident Roman
businessmen as well as the assembly. Like the rest of the Roman world,
Assos is eager to set eyes on the god, ‘now that mankind’s most joyful age is
established’, and has appointed a loyal embassy (consisting of four Greeks
20. JR S 50, 1960, 75ff. See also his c o n trib u tio n ‘R o m an C y p ru s’ to H . T em p o rin i (ed.)
Aufstieg und Niedergang der rdmischen Welt II. 7.2 esp. p .l3 4 7 ff.
C ass. D io. 59.3 .4 , 9.1-3.
C ass. D io. 59.13.1.
Gaius 27.3, 24.
Sy//.797 = S m allw ood, Documents . . . 33; Lewis an d R einhold, p .87.
OATHS OF ALLEG IA NCE TO CAESAR
and a Roman) to cement further the links created in person by the visit of
the emperor’s father Germanicus with his young son twenty years previous
The oath itself is short and to the point. It invokes Zeus, the deified
Augustus and Athena, with the pledge of loyalty to Gaius and his house,
keeping the same friends and foes, with the usual curse at the end upon per
jury. Included in the embassy’s duties were the offering of public prayers
and a sacrifice while in Rome to the Capitoline Jupiter.
Our only complete Western example of the oath of allegiance comes from
Aritium in Lusitania and is also dated to the beginning of Gaius’s principate
(in fact, two months after his accession).25 Instead of the consular names
for the year as at Assos, it begins with the name of the Roman governor
before whom the oath was sworn—C. Ummidius Durmius Quadratus; a
precise date and the consular names are at the end, together with those of
the local duoviri. What the Western oath may lack in fulsome Greek flattery
and in explicit worship of Gaius, it makes up in its thorough-going terms.
Gaius’ enemies will be pursued until they have paid the due penalty, the
salus Caesaris is to surpass that of their own families, and if a false oath is
sworn Jupiter and the deified Augustus with all the gods are to take
revenge, the language repeating the fortuna and incolumitas familiar from
the imperial prayers.26
We may here mention the largely parallel Latin text found at Sestinum in
Umbria.27 Its dating is uncertain, and must rely on the comparison of the
language with other oaths, particularly of course the Aritian.28 Some
scholars assign it to Gaius, others later because of the terminology which is
broader, or the variations may be of local origin and not representing a
development. The following expressions are included—‘I shall not cease to
hate or bestir (his foe) until he shall have paid the penalty for his crime of
parricide . . . his enemies I shall reckon as implacable foes . . . their im
pious and wicked (counsel) . . .’ The Sestinum text, fragmentary as it is, at
least reminds us that the oaths of allegiance underwent changes of form due
to time or locality or both.
Claudius to Trajan: the later literary evidence
In order to assess the importance of oaths of allegiance over a longer
period in the early Empire, it is useful to sketch the evidence after Gaius.
Claudius showed a restraint similar to that of Tiberius in the acceptance of
ILS 190 = S m allw o o d , Documents . . . 32; Lewis a n d R ein h o ld , p .86.
‘(m e) ex p ertu m p a tria in co lu m itate fo rtu n isq u e o m nibus fa x in t’.
C IL xi.5998a.
See H e rrm a n n o p .c it., p p .52-4, A . von P rem erstein , Vom Werden und Wesen des Prinzipats, p .4 6 ff.
honours and in the development of the emperor cult. As incoming consul
for AD 42 he took the oath in acta Augusti along with the senators, but did
not permit its extension to his own; three years later he is recorded as ‘hav
ing sworn all the usual oaths’, but reverted to earlier practice in having one
representative of each magistracy swear on behalf of his colleagues.29
Claudius’ attempt to forbid his deification in the East is illustrated by the
Letter to the Alexandrians, but the habit of swearing by his Genius or For
tuna continued alongside the observance of the cult, and in the eyes of many
provincials any fine distinctions between the emperor’s near and actual
divinity must have been meaningless. As for Nero, despite his proclaimed
observance of Augustan models in AD 54 he soon exploited all the means,
oaths of allegiance included, of strengthening his despotic power.
More interesting for our purpose are some of the events of the Civil War
following his death. In the struggles for power amongst the provincial com
manders the taking of the oath assumed a new importance. In the case of
Galba, the oath-taking by the legions of Upper and Lower Germany did not
take place until early January of 69, by which time Vitellius who held the
latter command was himself plotting for the supreme power. The taking of
the oath was protracted and difficult, with only ‘a few voices from the first
ranks’ responding and the rest holding back in the hope that some would
boldly reject it. The legions in Upper Germany did go further; refusing to
swear the oath to Galba but not wanting to be seen throwing aside their
respect for the Empire, they took it in the ‘forgotten names’ of the Senate
and people of Rome. This provided Vitellius’ supporters with their oppor
tunity and he was saluted as imperator.30 There was a special irony in the
events in Egypt three months later, when Vespasian, who was now weighing
his own chances, was so strongly supported by his troops that as he read out
the sacramentum and offered all the prayers for Vitellius’ success they
responded with complete silence. Vitellius will have been informed that the
oath had been duly administered, but on the first of July Vespasian was
saluted in Alexandria and the troops in Egypt took the oath to him.31
The sacramentum became a political weapon in such a turbulent period,
but it was only one part of the apparatus of oaths, vows and prayers by
which acceptance of imperial power needed to be displayed. With Pliny,
governor of Bithynia-Pontus under Trajan c. AD 110-2, we have the fullest
literary account for the period, at a time of relatively stable conditions in
the provinces. Pliny’s careful performance of his ceremonial as well as his
administrative and judicial duties is recorded in the tenth book of the Let
ters, those ad Traianum. Ep. 35 reports the sollemnia vota on behalf of the
29. C ass. D io. 60.1 0 .1 , 25.1.
30. T ac. Hist. 1.55-7; S uet. Vit. 8.
31. Hist. 2.74, 79; S uet. Vit. 15, Vesp. 7.
OATHS OF A LLEG IA NCE TO CAESAR
emperor and his house at the beginning of Pliny’s second year of office;
these were the prayers offered on 3 January in Rome and the provinces, and
Pliny seems to have been in Nicomedia, the capital of Bithynia at the time.
Sherwin-White plausibly suggests that a meeting of the koinon (provincial
council) may have coincided with this.32 Trajan acknowledges, and the same
exchange is made in Epp. 100-1 a year later from Pontus, probably at
Amastris. This time soldiers as well as provincials participated. Pliny also
reports his celebration of the emperor’s birthday, no doubt with the
customary festivals (Epp. 17a, 88-9).
A more important date in the province was Trajan’s dies imperii, 28
January, and Pliny reports for two successive years his faithful execution of
duties (Epp. 52-3, 102-3). These were firstly the celebrations which included
the prayers pro incolumitate (also offered at the beginning of the month)
and more significantly the administration of the oath of allegiance, the text
of which will have been similar to those which survive for the JulioClaudian period. Both soldiers and civilians took the oath, with enthusiasm
Pliny claims.33 How many provincials actually took the oath? The Assos
inscription of A.D. 37 assumes that the local magistrates administered it to
the citizens en masse, and we may conjecture that in Pontus an assembly
was also convened for this purpose.34 In other centres, where Pliny was not
present in person, the oath may well have been taken at meetings of the
Council, on behalf of the inhabitants.
Objections to oath-taking: the non-conformists
Within this elaborate structure of imperial allegiance, indications of any
protests either in principle or in practice against oath-taking are naturally of
interest. In classical literature there are occasional passages which show that
oaths in general were thought to be unnecessary between men of honour;
for example, Sophocles makes Oedipus respond to Theseus’ assurance of
his pledge by dismissing the need for such an oath—his word alone will suf
fice.35 It is not surprising that the Pythagoreans, with their stress on the
brotherhood and on moral strictness at every level, should have forbidden
the practice. In the imperial period Plutarch raised the issue of why the
priest of Jupiter was not permitted to take an oath. At its lowest, the reason
may have been to avert the danger of a perjured priest officiating at impor
tant public rites, but more likely it was that the word of such a man, if of
anyone, was worthy of trust and there was something ill-omened about the
32. The Letters o f Pliny, p .613.
33. Ep. 52. P raeiv im u s et com ilitensibus ius iu ra n d u m m o re sollem ni, eadem provincialibus
certatim p ietate iu ran tib u s.
34. P lin y ’s d escrip tio n does n o t fit a council m eeting only: cf. S herw in-W hite, o p .c it., p .634.
35. Oedipus Coloneus 644-51.
curse with which oaths usually ended.36
In Roman Stoic thought we find some expressions of dislike for oaths.
Epictetus advised against them, and Marcus Aurelius commended sober
language, saying that the man ruled by the divine element within him
needed neither oaths nor witnesses to his word.37 But this is far removed
from the realities of imperial rule, and we may be sure that Marcus knew
that explicit oaths of allegiance were necessary signs of the fidelity of his
legions. There is one important example, however, of a Stoic senator refus
ing to participate in the annual oath-taking, Thrasea Paetus. In the famous
passage beginning, ‘Nero, after his murder of so many eminent men, finally
yearned to extirpate Virtue itself by putting to death Thrasea Paetus and
Barea Soranus’, Tacitus summarises the case brought against him by the
informer Capito Cossutianus: at the beginning of the year Thrasea avoided
the customary oath, he did not attend the public prayers and had never, in
spite of belonging to a priesthood, offered a sacrifice for Nero’s welfare. In
sum, he was guilty of deliberate withdrawal and schism.38 It was inevitable
that Nero allowed a trial for maiestas to follow and that, at this point in his
despotic rule, the death sentence was passed.
But this case was somewhat isolated on the Roman side, at least in the
non-conformity with oaths and prayers. We may judge that Thrasea’s
objection was not to oaths of allegiance in principle, but rather to the
unworthiness of Nero and the hypocrisy of the senatorial class in this as in
so many respects. His protest was based on political and moral grounds.
Oaths of allegiance brought more fundamental problems for those
religious minorities which not only refused divine honours to the emperor
but rejected the Graeco-Roman deities in toto, namely the Jews and Chris
tians. Although at this period the Jews practised a religio licita in the eyes of
Rome, which was not true of the Christians, it is obvious that the ter
minology of Gentile oaths as well as their content presented problems of
conscience. In regard to the general issue of recognition of the emperor and
the official cult, the Jews for their part were willing to go to considerable
lengths, and it is to the credit of the Romans that they generally respected
the limits of Jewish tolerance. In the Temple at Jerusalem a sacrifice (of an
ox and two lambs) was offered twice each day ‘for Caesar and the Roman
nation’, and special offerings were made at the accession of Roman
emperors.39 Besides this, Jewish synagogues frequently displayed in their
forecourts tributes and dedications to the emperors.··0 But emperor-worship
Quaest. Rom. 44.
E p ict. Ench. 33.5, M .A u r. Med. 3.5.
Jo sep h u s, Bell.Jud. 2 .10.4; P h ilo , Legatio, 23, 45.
40. Leg. 2, In Flacc. 7, 49.
OA TH S OF ALLEG IA NCE TO CAESAR
was not required of them, and on those occasions when the oath of alle
giance was administered we may presume that it was in terms not offensive
to the Jewish conscience. Josephus records that when Vitellius the governor
of Syria was staying in Jerusalem with Herod the tetrarch, upon receiving
the news of Tiberius’ death he administered the oath to Gaius (an ominous
vow) to the populace.41 Orthodox Jews, of course, were bound by the
Mosaic teaching on oaths and their rightful use, and by the Roman period a
large number of refinements taught by the rabbis were observed by the
It remains to outline the problem of oaths and prayers to the emperors as
they affected Christians, towards whom the Roman authorities showed no
such deference. Here we find a basic confrontation which is historically very
instructive and leads into the larger questions of church and state in the later
Empire. The Christian attitudes to oaths were rooted in the teaching and
practice of Jesus, and in the Hebraic tradition which also manifests itself at
various points in the New Testament.42 Jesus placed himself in strong
opposition to contemporary Jewish practices in oath-taking, which allowed
many distinctions between what was binding and what was not; ‘if anyone
swears by the temple, it is nothing; but if anyone swears by the gold of the
temple, he is bound by his oath.’ Oaths by the temple or by the altar or by
heaven in effect, he taught, called God himself to witness.43 The core of his
teaching was to forbid his disciples to use oaths of any kind, because of the
seriousness of using the divine name and attributes and the danger of false
oaths: ‘let what you say be simply Yes or No; anything more than this
comes from the evil one.’44 But it is noteworthy that at his trial, when the
high priest ‘adjured him by the living God’ to say whether he was the Christ,
he did not reject the demand because of the oath, but accepted the solem
nity of the charge and affirmed his messiahship.45
In the letters of Paul there are direct echoes of the most solemn Jewish
oaths, but more commonly what may be called asseverations rather than
oaths, and the early Christian communities saw no conflict between this and
the teaching of Jesus.46
Our first external evidence of one kind of Christian oath comes from
Pliny’s correspondence ad Traianum, in connexion with the trials of groups
of Bithynians accused of professing this faith. Some who denied adherence
to Christianity described to Pliny the weekly services, at the first of which
F o r exam ple, in th e Ep. to the Hebrews 6.13-18, 7.20.
e.g. 1 Thess. 5.27, Phil. 1.8, Gal. 1.20, 1 Cor. 1.23.
B.F. HA RRIS
they met at dawn, sang a hymn to Christ as God, and bound themselves by
an oath (sacramentum) ‘not to commit some crime but to abstain from
theft, robbery and adultery, from breach of trust and from denying a
deposit when required to restore it.’47 It is better to take the ‘oath’ here in its
plain sense of a pledge than to link it with baptismal renunciations or with
the eucharist,48 and to believe that its terms were quite simple after the man
ner of a Pauline asseveration.
As a test of adherence, Pliny had obliged the defendants to repeat after
him an invocation of the Roman gods, to offer wine and incense before Tra
jan’s image, and to curse Christ’s name. This amounted to much the same
as the customary oath of allegiance to the emperor by the names of the
gods, and, it was recognised, could not be assented to by a genuine believer.
Christians were regarded as ‘atheists’, non-believers in the Graeco-Roman
pantheon, and it was this charge together with the refusal to swear by the
Genius of the emperor which was central to the recorded trials of the second
century. Thus when Polycarp, bishop of Smyrna, appeared before the pro
consul of Asia the heart of the Roman’s urging was ‘Swear by the genius of
Caesar; change your mind; say “ Away with the atheists” ! . . . curse
Christ.’49 Similarly in the case of the Scillitan martyrs, Saturninus told them
‘We swear by the genius of our Lord the emperor, and pray for his safety,
as you also ought to do.’50 It was the refusal to swear that brought execu
It may finally be mentioned that at the end of the century Tertullian dealt
at length in his works with these questions of conscience for Christians
under the Roman rule. There was no reticence about ‘offering prayers for
the safety of our emperors to the eternal, true and living God’, since it was
through his choice the Caesars came to power and to him that they owed a
duty to rule justly and effectively.51 But Tertullian was perfectly clear about
the limitations—Christians could not worship the emperor nor participate
in the offering of sacrifices to him;52 to swear by Caesar’s Genius was to
treat him as divine, and it was not possible to set aside lightly the invoca
tions in oaths of the gods of Greece and Rome in whom they disbelieved.
This would be ‘a collusion of faith with idolatry’, and Tertullian extended
the principle to situations where tacit consent, either in verbal adjurations
or in the written word, would amount to a denial of Christianity.53 He did
F o r d iscussion, see S h erw in-W hite, o p .c it., pp.702-7.
Mart. Poly carpi 9.2-3.
Passio Sanct. Scillit. in Act. Mart.Sel., p p .22-7.
Apol. 30.1, 33.1; cf. Rom. 13.1-4.
Apol. 10.1 cf. O rigen, Contr. Cels. 8.55-67.
De Idol. 20, 23.
OATHS OF A LLEG IA NCE TO CAESAR
not shrink also from applying the same considerations to service in the
Roman army; he commended the Christian brought to trial for refusing to
wear the chaplet and receive the imperial largess, because of the associations
with pagan rites which included vows and sacrifices to Jupiter.54 In the work
De Idololatria, although frequently indicating the presence of Christians in
the legions, Tertullian wrote most plainly about the conflicts in loyalty;
‘there is no agreement between the divine and the human sacramentum.’55
It is clear that while there is some evidence elsewhere of objections to
oath-taking on moral and political grounds, it was within the Christian
movement during this period that the question of swearing allegiance to
Caesar became most acute, on primarily religious grounds.
The texts of the oaths themselves are of course somewhat fortuitous sur
vivals, and the other evidence traversed above has been illustrative only, but
some initial responses can be considered to the broad questions raised. The
swearing of oaths of allegiance became firmly established under Augustus,
after Caesar’s use of them as one of the main symbols of his power; the oath
was a vital indicator of the perpetuation of the principate. On the military
side, it was a natural extension of the individual sacramentum, and the
legions were well aware, especially in times of upheaval, that no princeps
could consider himself safely in power until tidings had come from the pro
vinces that the oath had been taken. Its annual renewal was part of the con
tinuing pattern of one man’s imperium. Every commander who dictated the
oath was reminded whence he drew his own authority.
As for the civilian oath-taking, we are presented with a variety of settings.
At the start of each year in Rome the ceremony was an important one, in
which the magistrates most conspicuously, but also the senators as a body,
confirmed their loyalty, not daring except in isolated cases to fail to con
form. From Tacitus’ description we might conclude, unless there was con
trary evidence, that in Rome the citizen assembly took the oath, at least in
the early Empire. In the provinces there was no doubt some variation in
practice. The Assos inscription assumes a meeting of the ecclesia and in
Paphlagonia formal sessions will have been convened at the shrines of
Augustus; we note that resident Romans are mentioned in both cases. At
Samos the chief magistrates appear to have presided, and the same public
assembly may perhaps be assumed for Palaipaphos in Cyprus. Similarly at
Aritium in Spain, where either the propraetor himself, or the magistrates in
his presence, is likely to have presided. At Assos and Samos an embassy of
54. De Cor.
55. De Idol. 19.
congratulation is voted at the same time. We can of course only guess how
many attended such formally convened meetings, but it is certain that the
members of the local council will have been prominent, ready to display
their loyalty to Rome, and it is perhaps likely that as meetings of the whole
citizen body gradually fell into disuse, the councils assumed more and more
a representative role in such matters. The magistrates in particular would
know that the Roman governor required punctilious performance of these
Compared with the ‘passive’ evidence of the Roman power, the
numerous inscriptions recording civic decrees involving Rome and the
emperor and the use of the coinage for imperial propaganda, oaths of
allegiance must have left a more active and vivid impression on the mind.
Many of course were themselves inscribed on stone or bronze, but they
reflected something that the council or assembly had physically done on a
public occasion, and the actual terms of the oaths were uncompromising
and memorable. Likewise with the imperial cult; the many inscriptions
reflected things the citizens regularly engaged in and witnessed—the vows
and prayers and sacrifices at the altar of the Roman ruler.
It cannot have been easy therefore to disengage oneself openly from all
recognition of the temporal power and the divinity of the emperor. When,
in the case of the Christians, such non-conformity was based on fundamen
tal principles and could readily be linked with suspicions of shameful prac
tices in their own cult, we can see how exposed they were to general
unpopularity, and to prosecution when public emotions ran high against
them. Thus the oath of allegiance, like the rites of the imperial cult, remain
ed a yard-stick of conformity with the regime. The Roman rulers knew their
value and ensured their observance.