Roman Involvement in Anatolia, 167-88 B.C.

Author(s): A. N. Sherwin-White
Source: The Journal of Roman Studies, Vol. 67 (1977), pp. 62-75
Published by: Society for the Promotion of Roman Studies
Stable URL: http://www.jstor.org/stable/299919
Accessed: 03/10/2010 01:21

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=sprs.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Society for the Promotion of Roman Studies is collaborating with JSTOR to digitize, preserve and extend
access to The Journal of Roman Studies.

http://www.jstor.org
 ROMAN INVOLVEMENT IN ANATOLIA, I67-88 B.C.*

By A. N. SHERWIN-WHITE

I. THE PROTECTORATE, I67-133 B.C.

The defeat of Perseus at Pydna, the destruction of the Macedonian kingdom, and the
contemporary humbling of the Seleucid monarch Antiochus Epiphanes at Alexandria by
an insolent Roman parvenu, are reasonably taken to demonstrate the absolute supremacy
of Rome over the Hellenistic kingdoms of the Orient. At the same time the state of Rhodes
suffered a drastic reduction of power through the removal of its mainland territories, and the
king of Pergamum was severely snubbed, because both were believed to have favoured a
negotiated settlement of the Macedonian war. Henceforth, in the consensus of modern
opinion, the kings of Anatolia were puppets on a Roman string. This follows the theme
song of Polybius-that Rome had acquired the mastery over all the parts of the civilized
world, and in common opinion men had no choice but to listen to the Romans and obey
their instructions.1
Yet, when examined on this assumption, the actions of the Anatolian kings, from Pydna
to the death of Attalus III, are most surprising. The kings of Bithynia and Asia, Pontus and
Cappadocia, do not exactly behave as though they were under the active hegemony of a
foreign power.2 The rivalry of Prusias of Bithynia and the Attalids over the control of
Galatia provides a touchstone. Shortly after Pydna, when Eumenes was in deep disfavour,
a Roman mission was sent to investigate the Galatian raids of which he was complaining. At
a conference, from which he was excluded, the Romans failed to bring the Galatians to heel,
or else, as Polybius suspects, made no attempt to do so.3 Yet afterwards Eumenes did not
hesitate to restore order in Galatia by his own military efforts. The worst that the Senate
then did was to request him to withdraw his troops after the pacification, and to order that
the Galatians, who had been left in independence twenty years before, after their conquest
by Manlius Vulso, should remain autonomous, provided that they kept within their borders.4
This was the strongest Roman intervention in Asian affairs for many years. Prusias of
Bithynia tried to exploit the situation by a series of missions to Rome that alleged aggressive
intentions of Eumenes in Galatia. Hostile Roman missions to Pergamum failed to prove
anything against Eumenes, though they tried their hardest, and Prusias got nothing for his
pains.5
The succession of Attalus II in 159, an old favourite in certain quarters at Rome,

* This paper, which I delivered as my Presidential which is comprised in the tail-piece of the whole
Address to the Society in June I976, contains a work (e.g. 38. I. 5; 9. 6-8; IO. II-I3 II. I; I2. 3;
summary of opinions formed during extensive work I6. 9; I8. 7-8). See further the comments of F. W.
on the oriental policy of Rome from I67 to 50 B.C. Walbank, Historical Commentary on Polybius I (1957);
The notes contain the source references and a basic 40-2, 129-30, 30I.
bibliography. The whole field of studies down to 2 Modern accounts of the foreign policy of the
1950 is covered by the Notes in the remarkable second Anatolian kingdoms after Pydna are somewhat
volume of D. Magie, Roman Rule in Asia Minor selective and discontinuous. Magie, op. cit. I, covers
(I950), which provides a compendium of knowledge most of the field briefly in his narrative text: 26-33,
and critical discussion to that date, since when not so Pergamum; 193-4, Pontus; 201-3, Cappadocia;
much attention has been paid to this subject, 3I5-I8, Bithynia. P. V. M. Benecke, CAH viii,
especially by Anglo-American scholars, except when 279 f. is remarkably brief. M. Rostovtzeff, Social and
stirred up by epigraphical discoveries, or in connec- Economic History of the Hellenistic World (I94I), does
tion with politics at Rome. not study political relationships in detail, but has
1 It is the theme of the introductions to Books I strong and influential views about them. E. Will,
and 3, put specifically in i. I. 5; 2. 7; 3. I. 4, and Histoire politique du monde hellenistique ii (I967),
reappears in i. 63. 9-64. i, in 6. 2. 3, in the introduc- 319-24 is penetrating and independent, but perforce
tion to the discussion of the Roman polity, and at omits much. E. V. Hansen, The Attalids of Per-
the end of the epilogue, 39. 8. 7. The qualification in gamum 2 (1971), gives a full account of the activities
3. 4. 3, 'it seemed to be generally agreed ', is not to of Pergamum, but the interpretation is mostly
the theme of Roman invincibility, which he asserted derivative. E. Badian, Foreign Clientelae (1958), 99 f.,
unequivocally in I. 2. 7, but refers to his opinion that is brief.
to evaluate the position of Rome one must not stop at 3 Pol. 30. 3. 6-8; Livy 45. 34. 10-14.
Pydna but consider how she behaved to her subjects 4 Livy 45. 44. 21; Diod. 31. 14; Pol. 30. 2;
in the following generation, since ' great successes 30. 28; For contemporary records of his victory,
can bring great disasters if states do not use their including OGIS 299, 763, and its importance, see
powers wisely'. Polybius' point of view appears very Magie, op. cit. II, 766, n. 63.
clearly in his interpretation of the Achaean revolt, 5 Pol. 30. 30. 7; 31. I. 6; I5. IO; 32. 1-2.
 ROMAN INVOLVEMENT IN ANATOLIA 63
brought a change of front. Prusias miscalculated seriously by making open war on
Pergamum in 156, for reasons that are obscure. His invasion was successful, penetrating
deep into Attalid territory. First reports of this at Rome were dismissed as Attalid fabrica-
tions. But when it became clear that Prusias was indeed the aggressor, the Senate sent
repeated missions to stop the war.6 Prusias proved remarkably contumacious for a man
who had once performed a proskunesis to the whole Senate in formal session. He refused to
accept the Roman conditions, and threatened their envoys with violence. Attalus meanwhile
assembled strong forces, but held them back on Roman advice. Finally the Roman envoys
cancelled the formal treaty between Rome and the Bithynian kingdom. Though this was
not a declaration of war, Prusias lost his nerve. Peace was made in 154. Prusias had to pay
reparations to Attalus, but both sides surrendered occupied territory, and Prusias retained
his kingdom.7 Though the ultimate authority of Rome is evident, it is astonishing that these
events took place, if the Anatolian kings after Pydna considered themselves to be merely
the puppets of Rome. The brevity of the evidence often makes it difficult to judge situations.
But the discredited Eumenes had not hesitated to attack the Galatians without Roman
approval. Prusias, though he had failed to make headway while Eumenes was alive,
expected to be allowed a free hand against his favoured successor, and was not easily
deterred.
A few years later, Attalus acted with similar independence against Prusias. A con-
spiracy was set afoot in 149 by an agent of Attalus at Rome with the crown prince of Bithynia,
young Nicomedes. Nicomedes sailed to Pergamum. Attalus invaded Bithynia on his
behalf, and Nicomedia opened its gates. No prior Roman approval was sought, and a
belated Roman mission failed to prevent the replacement of old Prusias by young Nicomedes.
Afterwards Attalus coolly claimed that Prusias had violated the previous settlement made
by the Romans.8 But he had done nothing except to delay the repayment of reparations,
and Rome did not withdraw its support from him.
Then there is the remarkable affair of the intervention of Attalus in the Cappadocian
succession. Cappadocia had come into the Roman orbit after Magnesia. Ariarathes IV, the
ally of Antiochus Megas, made his submission to Manlius. Vulso and became a formal ally
of the Roman people. He had his son and heir educated at Rome, who on his succession as
Ariarathes V was at pains to secure his formal recognition from the Roman Senate.9 When
his kingdom was harried by Galatian raiders, he gained the support of a Roman mission
that was travelling through Cappadocia to Syria, and offered it every assistance in its Syrian
assignment.'0 Yet when Ariarathes was ousted from his throne in 18 by his brother
Orophernes, in conjunction with the Seleucid king Demetrius, who was persona non grata
at Rome, Ariarathes sought in vain for help from the Senate in his restoration. The Senate
merely recommended the division of the kingdom between the brothers. Ariarathes,
returning disconsolate from Italy, was restored to his throne by the armed intervention of
Attalus, and he showed his gratitude by helping Attalus in the war with Prusias.11 There
was a curious incident when Attalus and Ariarathes conjointly ravaged the territory of the
free state of Priene, to compel the restoration of certain monies to Ariarathes. Priene
appealed to Rome as the guarantor of her freedom, but the Senate took no firm action.12
The Senate seems remarkably unconcerned in this decade about events in Anatolia,
though there were no great wars in Europe or Africa to distract its attention between i67 and
the troubled years from c. I50 onwards. Its occasional interventions were frequently
ineffective. The initiative that prompted such intervention came from the parties themselves
seeking Roman diplomatic support. Galatia only became an issue at Rome when Prusias or

6 Pol. 32. IS-$6, cf. App., Mithr. 3; Diod. 31. 35; II

All this is relatively well-documented: Pol.
OGIS 323, 15-22, confirming Polybius. Cf. also 32. IO-I2; Diod. 31. 32-32b; App., Syr. 47;
Pol. 33. I. I-2; 13. 4-5. Justin 35. I. 2. Cf. Magie, op. cit. II, 1097, n. 9, on
7Pol. 33. 12-13; App., Mithr. 3. the discrepant version of Livy, Ep. 47: 'a senatu
8 Pol. 36. I4- App., Mithr. 4-7; Strabo 13. 4. 2 restitutus est '.
(624). Cf. OGIS 327, celebrating ' Attalus and those 12 For the war against Prusias, Pol. 33. 12-13.
who marched with him against Prusias and besieged For Priene, 33. 6. 6-8, and the fragmentary OGIS
Nicomedia'. 351, in which the Senate merely instructs a Roman
9 Pol. 31. 3; Diod. 31. I9. 7; Livy 42. 19. 3-6; magistrate to write to the kings. Magie, op. cit. 1,
29. 4. I17, ignores Polybius' statement that the Senate
10 POl. 31. 8, 32. 3. ' paid no attention '.
64 A. N. SHERWIN-WHITE

Eumenes acted as prompters. The Roman role was limited to maintaining the status quo of
each party at the end of an affair, even when Prusias or Orophernes was manifestly in the
wrong. Status quo rather than balance of power is the key-note. The kings themselves
behaved as though there was still room for acts of aggrandisement which Rome could be
induced to countenance. Remarkable light is cast on the situation by a letter of Attalus II
to the High Priest of the temple state of Pessinus, about a warlike enterprise.13 Attalus
explains how he consulted his advisers when planning an attack on an unnamed enemy-
probably the Galatians. All agreed with the king except one, who insisted that nothing must
be done without consulting Rome. This had not occurred to Attalus or the others. After
many days of discussion they finally agreed to do nothing without the Romans, because 'if
I succeed it will renew the envy and suspicion which they felt for my brother Eumenes, and
if I fail it will be utterly disastrous. So we decided to seek Roman approval . .. then if we
fail they will help and protect us '.
This letter demonstrates the very opposite of what it is often taken to show. What is
remarkable is not the dependence on Roman approval, but that this factor had been entirely
forgotten by the king and all but one of his advisers, who were not easily persuaded to change
their minds. Magie speaks of Attalus' fixed policy of always seeking Roman approval.14
But this was far from the only time that Attalus took his own line. The king acted as if they
believed that Rome, after instilling a proper respect into them after their various mis-
demeanours, had no positive interest in the direct management of Anatolia ' this side of
Taurus
Still less was Rome concerned in this generation with the lands beyond the Halys. The
kingdom of Pontus does not impinge upon the affairs of the other kings nor they on it.
No missions arrive from Pontus in Rome or sail from Rome to Sinope in the historical
record from I81 to I24.15 At an uncertain date a formal connection was made between
Rome and Pontus, and Appian notes that Mithridates Euergetes was the first king of
Pontus to assist in a Roman war, when he sent ships to the siege of Carthage.16 There was
as yet no Pontic problem to disturb the calm of Anatolia.
What then was the purpose of the erratic Roman supervision of the Asian kingdoms?
Magie, who reflects modern opinions very fairly, takes it for granted that the kings were
subservient to Rome.'7 They lacked any real independence, and managed Anatolia in the
interests of Rome. Pergamum in particular formed a strong buffer between Rome and the
Seleucid empire. Magie finds no variation in the situation down to the annexation of the
Pergamene kingdom. The weakness of this view is that it greatly underestimates the extent
and the persistence of the freedom of action that the kings enjoyed, or believed they could
enjoy. It assumes that Rome had an abiding interest in Anatolian affairs, and that the
eventual establishment of direct control of all Anatolia was inevitable, because that is what
eventually happened. But the second century saw a growing military commitment of

13 OGIS 3I5, VI, with Dittenberger's notes his alliance with Rome. Two coins (Recueil2 i. nn.
(= Welles, Royal Correspondence,no. 6i). 2-7) and an inscription (Inscr. Delos I555) combine to
14 For Hansen, op. cit. (n. 2), 132, the letter shows identify him as the brief successor of Pharnaces
that Attalus was determined to keep the friendship of C. I59; cf. Magie, op. cit. II, I090 nn. 46, 48, 49, and
Rome, and nil ultra. Magie, op. cit. I, 280: ' Attalus, J. A. 0. Larsen, Cl. Phil. 5I (I956), I57 f., against the
whose policy it was to be guided in such matters by older identification with a son of Mithridates
Roman wishes ', with reference to the beginning of Eupator c. 8o (for which see Dittenberger ad loc.).
his reign; ibid. 27, the decision of the letter ' was a 17 cf. Magie, op. cit. I, 20, on the Attalids (and
distinct step towards closer relations with the n. I4 above); 202, the Ariarathids; 3I5-I7 on
Senate'. He adds that Attalus engaged in military Prusias II and Nicomedes II. Rostovtzeff, op. cit.
activity in regions where Roman interest was not (n. 2), 80I-3, 827, regards Pergamum and Bithynia
concerned, but does not probe beyond actions at as reduced to vassalage. This all goes back to
Selge and Attaleia in Pamphylia. Mommsen, History of Rome III (894), 234 f. The
15 In i81, the intransigent Pharnaces, during his subtle Will considers this a confused period in which
war with the Pergamene coalition, which eventually Rome failed to exert her power consistently, though
defeated him, sent an emissary to Rome, but paid no the kingdoms were dependent on her grace and
heed to a Roman commission: Livy 40. 20. I; Pol. favour, op. cit. (n. 2) II, 302, 3I2 f., 320. For Hansen,
23. 9. I; 24. I. 1-3, 5. i. The first recorded mission op. cit. I41, both Eumenes II and Attalus II were
after Pydna is that implied c. 124 by Gellius, NA i i, vassals. For Th. Liebmann-Frankfort, see n. i8
io, which is followed belatedly by that of I03-2 below. Only R. B. McShane, The Foreign Policy of
(see n. 6i below). the Attalids of Pergamum (I964), I90, briefly denied
16 App., Mithr. io. In OGIS 375 (ILS 30) that the Attalids were ' subservient' to Rome in
'Mithridates M.f. Philopator Philadelphus ' records Asian affairs.
 ROMAN INVOLVEMENT IN ANATOLIA 65
Roman manpower in southern Europe: the conquest of Cisalpine Gaul, the defence of the
Alpine frontier, the prolonged Spanish wars, the Balkan entanglements that followed the
annexation of Macedonia, and the increasing involvement in Transalpine Gaul, to which must
be added the African problems, the elimination of Carthage and later entanglement in
Numidia. These problems increasingly absorbed the military manpower of Rome in the
second half of the century, so that there was nothing left to spare for building an oriental
empire, and every inducement to leave well alone east of the Aegean.l?a There was a period
of relative quiescence in the fifteen years following the defeat of Perseus, but at that time the
Senate was still unwilling to annex provinces even in European Greece. Beyond the
Aegean, so far from seeing an increase in Roman intervention, this is the time when, as we
have seen, the Anatolian kings were particularly free to practise their mutual antipathies.
A more sophisticated version of Roman policy in Anatolia as a planned protectorate is
given in the many writings of Therese Liebmann-Frankfort, who has systematically surveyed
Roman oriental policy from Apamea to the conquests of Pompeius.18 For her, Rome
carefully constructs a rampart of states between the lands under Roman control and the
Seleucid power, so that Italy should not be exposed to a direct military confrontation with
the Seleucid empire. The 1'reaty of Apamea effectively fixed the demarcation line at the
Taurus watershed and the Halys river. So the rampart originally consisted of the kingdom
of Pergamum and the state of Rhodes, both greatly enlarged by the donations of I89,
together with the Galatian confederation subdued by Manlius Vulso. To these Roman
diplomacy rapidly added the kingdoms of Bithynia, Pontus and Cappadocia, thus vastly
enlarging the rampart.
This is a buffer state theory. But a buffer state operates, as in the classical examples of
Armenia in the Roman empire or Afghanistan in British India, by interposing an obstacle
between two hostile powers. Either through its own strength or through the difficulty of its
terrain, or both, it impedes effective military action between the two principals. That was
not the situation between the Roman state and the Seleucid kingdom in the time of its
strength. Rome did not directly control any territory adjacent to Anatolia down to the
annexation of Macedonia, and Rome did not need buffers for her own protection after the
great victories of Magnesia and Pydna. The only Roman concern was that the Seleucids
should not be able to seize the rampart itself, as Antiochus Megas had done with con-
siderable ease before his war with Rome. But if the Senate thought that the Seleucid power
was a dangerous rival of Rome, they cannot have also believed that the small states of
western Anatolia could oppose it by themselves.
The ancient sources that discuss the reorganization of the Asian kingdoms after
Magnesia do not talk about defence but about management. There was a great deal of
vacated territory and something had to be done about it. In Polybius' account the problem
is discussed by Eumenes and the Rhodians, in their role as advisers of Rome, as one of
control.19 It is delicately assumed that Rome did not wish to take over the direct government
of any Anatolian territory, and it is a question of dividing the spoils between the great allies
of Rome: ' as at a rich banquet there is enough for all, and more than enough'; ' Rome has
become the mistress of the world and needs no extension of material resources .20 The
Rhodians concede that the kingdom of Eumenes should be fattened 'to ten times its
previous extent '.21 This view is repeated in the Roman tradition by Sallust's report of
Mithridates' gibe that the Romans turned Eumenes into the watchman of their conquests:
'post habitum custodiae agri captivi '22 The function of the kings was primarily not to
defend Anatolia but to manage it.
This helps to explain the relative indifference of the Senate to the machinations of the
kings. The Senate supervized and controlled its interests beyond the Adriatic by the
despatch of missions-legationes-composed of experienced senators. These would address
]7 See n. 75a below. So, e.g., when Attalus II restores Ariarathes, it must
18 Th. Liebmann-Frankfort, La frontie're orientale be on the advice of Rome (114-5), despite PoI. 32.
dans la politique exte'rieure de la R4publique romainie 10-12.
(I969), puts her views together. The theme is worked I1 Pol. zi. I8-23, cf. Livy 37. 52 -4.
out in ch. i-2. The kings are turned into ' satellites ', 20 Pol. zI. 7-9; 22. 13; 23. 2-5.
' absorbed ', and even ' integrated ', as an alternative 21 Pol. 2I. 22. I5.
22 Sall., Hist. iv, fr. 69. 8.
to annexation or permissive expansion. Cf. IOI f,
Bithynia; 103 f., Pergamum; Io8 f., Cappadocia.
66 A. N. SHERWIN-WHITE

the kings as occasion demanded. One cannot help noticing in Polybius and Livy that where
the great military powers are concerned, when there is trouble with Macedonia or the
Seleucid king, the Senate shows great energy and takes the initiative, despatching zealous
commissionersto 'observe affairsin Macedoniaor Syria', and briefed with ultimatums to
suit the situation.23When the young prince Demetrius escaped dramaticallyfrom Rome to
Antioch, and dethronedthe Romannominee, all the alarumbells rang. The experiencedand
authoritativeTiberius Gracchus was sent (Polybius says) ' to look at things in Greece, to
keep an eye on the other kings, and to watch developmentsin Syria I.24
That was a special crisis. Mostly, in the generation after Pydna, the managementof
affairsinside Anatoliaproceededratherdifferently. The senate came to hear of incidents in
the kingdoms when contesting parties sought Roman intervention. The rivalries of the
kings were strong, and usually sufficedto keep Rome informed. But the Senate, as we have
seen, was frequently caught out by events, and able to intervene only after a fait accompli to
restore the status quo. The Roman envoys did not always seem to inspire great awe, and in
Anatolian contexts the Roman tail could be twisted, without dire consequences, even by
such a one as Prusias.25
The scope of Roman supervision was limited by the weakness of the legatio as an
institutionalform. It was neither regularnor permanentnor ubiquitous, and it lacked any
territorialbase and executive substructure. In great crises a strong envoy could secure the
immediate execution of senatorial policy without the support of armed force, though the
more strikinginstances are in the context of Syria. That the Senate and its envoys so often
left events in Anatoliato take their course, or intervened belatedly or ineffectively,suggests
that the Senate did not take a very serious view of Roman policy in Anatolia down to the
death of Attalus III.

2. THE INIhERITANCE OF ATTALUS

The annexation of Pergamene Asia is one of the more surprising turns in Roman
history, though historiansgenerallytake it for granted.26So far the annexationof permanent
provinces outside Italy had been the by-product of great wars with majorpowers, either to
prevent the recoveryof a beaten enemy, as by the annexationof Sicily and southern Spain,
or to secure the final elimination of the recalcitrant, as in Africa and Macedonia. No
necessity, real or imagined, required the annexationof Asia in I33. Attalus II had shown
his customary vigour down to his last years when he effectively defended his Thracian
possessions single-handedagainstthe assaultsof the ferocious Diegulis.27 Within five years
his successor, Attalus III, died young and childless, aged about thirty-six. By a remarkable
will he left his kingdom to Rome. No source reveals his motives. Magie, expanding a
notion of Mommsen, speculatedthat Attalus, recognizingthe pervasivedominationof Rome
in the East, reckonedthat only direct Romancontrol could maintainpeace in Anatoliaagainst
the ambitions of the local kings or against the sort of disorders that followed his death.28
Such a view ignores the effectivenessof the previous reign, and would be more convincing
if Attalus III had lived twenty years longer. Attalus, son and nephew of a sexagenarianand
octogenarian,had no reasonto expect an early death in the fifth year of his reign, or ultimate
childlessness. His will should be interpretedby other instruments of the same kind.
The first of these was the will of king Ptolemy Euergetes of Cyrene, of which a copy

23
cf. e.g. the frequent missions to Macedonia, Even H. M. Last, CAH Ix, I03, thinks that there was
Achaea and the Anatolian kingdoms leading up to the no hesitation, though there should have been.
war with Perseus: Livy 42. 17. I; I9. 7-8, 26. 7-8, Hansen, op. cit. 148, 'the only logical course he could
37;24 Pol.
45. '-5. follow'. Will, op. cit. (n. 2) II, 350, as usual is more
31. 7-II.
15. puzzled.
25
cf. nn. 6-8 above, and the notorious criticism by 27 Diod. 33. I4,15; Strabo 13. 4.2 (624). OGIS 330
M. Cato of the mission that failed to save Prusias in may date this to 145, cf. Dittenberger ad loc.,
149, Pol. 36. I4. 4-5. If factionalor family interests Hansen, op. cit. (n. 2), 139.
underly some of the ambiguities of Roman behaviour, 28 Th. Mommsen, op. cit. (n. 17), III, 278. His
this only emphasizes the absence of an overriding view prevails; cf. especially Magie, op. cit. I, 32.
public interest. McShane's notion, op. cit. (n. i6) 194, that Attalus
26 cf. V. Chapot, La province romaine proconsulaire wanted to end the tension between Roman and
d'Asie (1904), 5, 9, 10-I ; Magie, op. cit. I, 31-2, Pergamene power is hardly intelligible.
147; II, 780 n. 9I; Rostovtzeff,op. cit. (n. 2) iI, 807.
 ROMAN INVOLVEMENT IN ANATOLIA 67
survives.29 By this, Ptolemy in 155 left his kingdomto the RomanPeople in the event of his
death without an heir. The provision stemmed from the feud between Ptolemy and his
brother Philometor, king of Egypt. The basic intention was to make it profitless for
Philometor to murder his brother. The last such will was that by which Rome acquired
Bithynia on the death of the much harriedNicomedes IV in 74. There is a strong hint that
it contained a similar clause: Sallust recordsthe allegationthat the Roman inheritancewas
invalidbecause a son of Nicomedes survived.30Nicomedes, twice expelled from his kingdom
by Mithridates,had every need of such a device. Attalus could have felt the same necessity
for protection against the ambitions of his rival kings or the malice of his subjects, who are
represented as cordially detesting him.3' It need not have been the primary intention of
Attalus that his kingdom should pass into the hands of Rome. Nothing prevented him in
the course of time-if childless-from adopting an heir among his kinsmen.32The Romans
were far from objecting when a king of Numidia adopted a bastardas son and joint heir in
these very years.33 And Attalus himself had been a late and possibly illegitimate child.34
Little is known about the contents of the will, except that it granted civic freedom and
territoryto the city of Pergamum,as a well-known inscription has revealed.35It may have
provided similarlyfor all the Greek cities, because the tribune Ti. Gracchuswas proposing
to deal in some way with all the cities of Asia, and the Epitomeof Livy recordsthat Asia was
due to become free, when the revolt of Aristonicus supervened.30Attalus might have had
in mind the condition of the cities of southern Greece after the recent establishmentof the
province of Macedonia. A largenumberwere free states, exempt from proconsularjurisdic-
tion and interference,and in some cases from Roman taxation,while even the subject cities
were left very much to their own devices in internal affairs.37But whatever the intentions
of Attalus, they were set aside by the consequences of the revolt of Aristonicus.
The Roman decision about Asia should have been related to the strategic situation
beyond the Taurus. Though this was a time of peace inside Anatolia, new dangers had
arisen in the Seleucid zone during the past decade. The Parthian power had driven the
Seleucids out of their satrapiesbeyondthe Euphratesby I4I, and the effort of Demetrius II
to recover them had ended in a disastrous defeat in Media in the next two years. When
Antiochus Sidetes renewed the attempt in his Median campaign of 130, it too ended in
disaster,and the Seleucid empire was reduced to a local kingdom in Syria.38 This situation
had thus arisenbeforethe death of Attalus, and had been proved irretrievablebeforethe final
organizationof the Roman province. But there is no sign that the Roman governmenthad
appreciatedthe implicationsof these events. Roman informationabout the orient had last
been refreshed by the mission of Scipio Aemilianus to Asia, Syria and Egypt in 139.
Scipio distinguished himself by a total failure to estimate the new forces at work beyond
the Euphratesat that very time. Instead he was unduly impressed by the wealth of Egypt,
and warned the Senate against dangersfrom the effete Ptolemaic monarchy. Strabo, in an
account of Scipio's mission, probablyderived from Posidonius, stresses the Roman neglect
of the Parthianfactor at this time.39

29
SEG Ix 7. Cf. Will, op. cit. (n. 2), II, 305 f., for a OGIS 338.
survey and bibliography. 36 Plut., Ti. Gracchus I4. 2; Livy, Ep. 59: 'cum
30App., Mithr. 7I; B.C. I. III; Livy, Ep. 93; testamento Attali regis legata populo Romano libera
Eutropius 6.6. Cic., de leg. agr. 2. 40, 50, confirms. esse deberet.' For discussion and bibliography cf.
Sallust, Hist. iv. fr. 69. 9: ' Bithyniam Nicomede Will, op. cit. (n. 2), 351 fL; Magie, op. cit. I, 32-3,
mortuo diripuere (sc. Romani), cum filius Nysa quam II, 78o-I nn. 92-4. It is possible that the bequest was
reginam appellaverat genitus haud dubie esset '. He limited to the ager regius and the townships other
does not allege that this was a recognized and than Greek cities.
legitimate son; cf. ibid. ii. fr. 71, for the rejection of "7 RE i, i9o f. Rostovtzeff, op. cit. (n. 2) II, 749 f.
his claim. and III, 1506, n. I5, for bibliography. Cf. the notable
31 Diod. 34. 3; Justin 36. 4. I-3. Cf. Strabo letter of the proconsul after a local revolution at
14. I. 39 (647), with Hansen, op. cit. (n. a), 144, Dyme, SIG3 685 (= Sherk, Roman Documents, no.
n. 55. 43), with Paus. 7. I6. io.
32 Magie, op. cit. II, 778 n. 87 minimizes the 38 For a recent survey see Will, op. cit. (n. 2) IT,
numbers of relatives. That Attalus remained 342 f.
unmarried for some years after the death of Berenice 39 Diod. 33. 28 a 2-3; Strabo I4. 5. 2 (669). For
does not mean that he had no intention of remarriage. the mission cf. A. E. Astin, Scipio Aemilianus (I967),
33 Sallust, BJ 9. 3. I27, 177. Liebmann-Frankfort, op. cit. (n. I8), I33,
34 For the birth of Attalus cf. Magie, op. cit. II, 772 oddly thinks the disappearance of menaces to north-
n. 76. Hansen, op. cit. (n. 2), 47I f., on the possible west Anatolia now made it possible to create a
implications of Pol. 30. 2. 5, 33. i8. 2. province.
68 A. N. SHERWIN-WHITE

The Roman Senate was in no hurry to take up its inheritance in Asia. Attalus had died
(it seems) in September I34, and the tribune Ti. Gracchus had proposed the allocation of
the treasures of Attalus before his own death in July or August I33.40 But the Senate did
no more in I33 and I32 than to despatch a commission of five senators to prepare for the
organization of Asia. No magistrate with inmperium was sent to take the province under
direct government in these two years, despite the fact that a rebellion spread through Asia
with increasing intensity after the publication of the will. Instead, the Senate applied the
method by which order had been kept in Anatolia since the settlement of Apamea. The local
kings of Bithynia, Pontus and Cappadocia were invited to crush the rebellion, and the free
states also lent a hand, with some success, since the flotilla of Ephesus was able to destroy
the ships of Aristonicus.41 But by the end of I32 the despatch of a consul and a Roman
army was seen to be necessary. The timetable is instructive. Time was further wasted by a
quarrel between the consuls of I31 over the assignment of Asia, and it was not till some
three years after the death of Attalus that a Roman consul arrived in Asia.42 It is clear that
the Senate was quite unprepared in I33 for the establishment of direct rule across the
Aegean, although the ending of the Spanish war had freed their hands for intervention
elsewhere.
It is true that an inscription which records the instructions given by a senatorial decree,
in the last quarter of an uncertain year, to the ' magistrates being despatched to Asia ' has
been attributed to the year I33, because it ratifies arrangements made by Attalus at Pergamum
up to the eve of his death.43 But the concluding lines of the text show that this was a later
confirmation of an earlier decree to this effect: present and future governors are instructed
to alter nothing in respect of those arrangements; hence the otherwise mystifying use of the
plural ' magistrates '. So this text belongs to the period after the campaigns of Perperna and
Aquilius, when permanent arrangements are being made for the province, and ambiguities
are being cleared up.
The territorial arrangements made after the revolt of Aristonicus reveal the senatorial
attitude towards Anatolia. The area of the kingdom was drastically reduced to form the new
province, by the grant of the regions of Great Phrygia and Lycaonia to the kings of Pontus
and Cappadocia respectively.44 These grants underline the absence of any strategic interest
in the annexation. Through Lycaonia, an immense region of infertile steppes and salt
desert, there passed the highway that led from the Aegean coast of Asia through the Cilician
Gates to Syria and the Euphrates.45 Its cession to Cappadocia indicates that Rome had no
Syrian preoccupations at this time.
The fertile uplands of Phrygia form the hinterland of maritime and Lydian Asia. They

40 The date of the death of Attalus III is commonly the reallocation of the consular province by a lex
attributed to spring or summer I33 with unjustified that instituted a direct and open election (sur-
confidence (e.g. Magie, op. cit. II, 78I, n. 94). The prisingly), at which P. Crassus was preferred to
back-dating of the Roman era on republican cistophori Scipio Aemilianus, although Crassus, famous for civil
to September I34 seems to suggest an earlier date virtues, lacked any military ability and as pontifex
(CIL I2. 2, p. 76I-2: Kubitschek, RE I 637). The maximus should have not have left Italy (Gellius, NA
sole other criterion is the length of reign given by I. I 3. I o; Livy, Ep. 5 9).
Strabo I3. 4. 2 (624) for Attalus II, twenty-one years, 43 OGIS 435. Though it would suit my view to
and Attalus III, five years, to be dated from c. March take the magistrates for commissioners, the term
I59, when inscriptions indicate that Eumenes II was a-rparlyol can only mean praetores in a public
still alive (Hansen, op. cit. I27). It is not clear that document of this date, pace J. Vogt, Atti del terzo
Strabo's years are completed regnal years. If there congresso int. epigr. gr. Lat. (I959), 45 f., whose
is an overlap, the fifth year of Attalus III, counted argument that the phrase eiS 'Aoa1av,TopEvopEvot
from I39-8, must end in 134. J. Carcopino was must refer to legati is contradicted by the Cnidian
perhaps right about the sunstroke and the summer and Delphian texts of the Piracy Law, JRS 64
but wrong about the year, Autour des Gracques2 (I974), 204, col. iv 9-IO; FIRA2 I, 9. B. 28-9. The
(I967), 34 f. date may be any year after I33 when both consuls
41 Strabo 14. I. 38 (646) distinguishes the stages of were out of Rome in September and October, since a
Roman intervention carefully, pace Magie, op. cit. praetor, otherwise unknown, presides. Cf. Magie,
II, I037 n. io. The mission of five arrives after the Op. cit. II, I033, n. i; Broughton MRR I, 496;
kings and cities have taken action against Aristonicus T. Drew-Bear, Historia 2 (I 972), 75.
and before the despatch of the consul of 131. So too 44 Justin 37. x.2; 38.5.3; App., Mithr., 57. The
Liebmann-Frankfort (op. cit., T40) observes the text of Justin 37. I. 2, which has Syria for Phrygia,
tactical delay. Eutropius 4. 20, Orosius 5. I0. I-2 also adds Cilicia to Lycaonia, which makes no sense,
are less exact. unless one boldly amends it to Pisidia. Cf. Magie, op.
42 Strabo loc. cit., Livy, Ep. 59. Cic., Phil. ii. i8, cit. II, I044 n. 28; A. H. M. Jones, CERP2, I3I.
for the quarrel. This led to a iudicium populi and to 45 Magie, op. cit. I, I25 f., 276-7.
 ROMAN INVOLVEMENT IN ANATOLIA 69
were of strategic interest in the second century, before the rise of an aggressive power in
Pontus, only for the control of the brigand chiefs of Galatia, whose power was broken long
before I33. By making Phrygia over to Mithridates Euergetes, the Senate showed itself
indifferent to the growth of Pontus, which must have already secured some form of control
over Galatia, since this lies astride the routes from Pontus to Phrygia. The Senate did,
however, retain within the province the southernmost sector of Phrygia containing the road
centre of Apamea, from which the central highway takes its departure towards Lycaonia,
since one of the military roads that Aquilius built in Asia in I29 reached Tacina, between
Laodicea and Apamea.46 But the roads of Aquilius, which included the remarkable luxury
of a coastal road linking the Aegean ports, seem to have been designed, so far as they are
known, not for external operations, but to secure control of the regions in which the revolt
of Aristonicus had centred.47
The original province was limited in its extent northward by the survival of the free
states of Cyzicus, Lampsacus and Ilium, which technically separated the province from the
Hellespont and the Propontis.48 In the south it is doubtful whether at this date the province
extended south of the Maeander. The cities of Caria and Lycia, when delivered from
Rhodian control in I67, were not added to the kingdom of Pergamum, and appear to have
retained their free status down to the first Mithridatic war. The evidence is somewhat
ambiguous, but the Senate seems not to have exploited the excuse provided by the revolt of
Aristonicus to extend the province in the south beyond the territory of the former kingdom.49
So Rome was not greedy of territory at this time. Certainly the financial classes were
not involved in the original decision to accept the inheritance of Attalus, because it was not
until I 23 that, as is well known, the lucrative farming of the tax-collection of Asia was
transferred from a local function to the control of the Roman publicans. Strategically Asia
was treated as a dead end for the next thirty years. The governors of Asia down to I02 had
no known military functions, and it is uncertain whether they had any legionary forces at all
under their command. While the European provinces of Rome were the scene of perpetual
frontier warfare and aggrandisement, the praetors of Asia lived in unbroken peace. It is
significant that no consular army was sent to Asia from I 29 to 87.50 When the praetors of
Asia were first involved with Mithridates, they are found operating with armies formed from
Asiatic levies, and according to the laconic sources have 'few Roman soldiers '.51 The
analogy is with the praetors of Africa, who likewise from I46 to the Jugurthan war fought
no campaigns, and whose territory likewise was similarly safeguarded by the loyalty of
adjacent kings.
The sole break in this peaceful regime down to the nineties neatly tests the rule. This
is the campaign of the praetor Antonius in ioz against the Cilician pirates. We know very
little of what Antonius did, but a great deal about why he did it.52 Strabo explains how the
pirates of the eastern Mediterranean ranged unchecked after the breaking of the Seleucid
power by the Parthians, and the text known as the Piracy Law of ioi-ioo, now extended by

46
cf. Magic, op. cit. II, I042 n. 26, 1048 n. 39; 49 See A. N. Sherwin-White, ' Rome, Pamphylia
CIL I2. 2, 646. There is no other direct evidence at and Cilicia I33-70 B.C.,' JRS 66 3 n. 6.
(I976),
this date. By II3 Pisidian Prostanna was within the 50 If Q. Mucius went to Asia after his consulship,
province, cf. Inscr. Delos I603; Magie, op. cit. II, as E. Badian argues, Athenaeum N.S. 34 (1956), 104,
ii6I, n. I2. it was not a regular assignment lege Sempronia,
47 Magie, op. cit. I, I57-8; II, I048, nn. 39-40. because he was at Rome after the campaigning season
For the milestones, CIL I2. 2, 646-5I. A new stone of 95,when he vetoed his colleagues' triumphus (Asc.
from the Burdur region adds nothing substantial; cf. I4C; Cic., de Invent.2. I I I). See now B. A. Marshall,
Annual Report of the British Institute of Ankara Athenaeum N.S. 54 (I976), I I7, against Badian.
1975, 10. 51App., Mithr. II, I7, 19: Cassiusand Aquilius in
48 Cyzicus, a city state with extensive territory, was go-89 raise a great army of ' Phrygians and Galatians '
free c. 133 (IGRR iv, 134,11. i8 f.), and reappears as from 'Bithynia, Cappadocia, Paphlagonia and
free after the first Mithridatic war (Plut., Luc. 9. I; Galatia '. Cassius also had a small army, but no
App., Mithr. 73; Diod. 38/9. 8. 3; Strabo I2. 8. II legions are mentioned. Cf. Memnon, FGrH 434,
(575-6); Magie, op. cit. II, I I I I n. 4). Lampsacus, (22), 7, 'with few Romans '; Justin 38. 3. 8, ' Asiano
free earlier and never part of the Pergamene kingdom, exercitu instructos '. So too L. Sulla assisted the
became provincial after that war: Livy 43. 6. 8-io; restoration of Ariobarzanes ' with few troops of his
SIG3 591; Cic., Verr. 2. I. 8I; Magie, op. cit. II, 947 own but eager allies ', Plut., Sulla 5. 7. For the
n. 5I. For Ilium cf. Magie, ibid. 950, n. 6o; whether possibility that he was the regular propraetor of Asia
its freedom is earlier than Sulla's settlement is not see my discussion, op. cit. (n. 49), 8-9.
52 ibid., 4-5, with bibliography.
certain. See Strabo I3. I. 27 (594-5); IGR iv, 194,
cf. Jones CERP2, 60-3, 86-7.
70 A. N. SHERWIN-WHITE

the new fragmentsfrom Cnidos, publishedin JRS I974, revealsthe concernof the governing
classes at Rome over this new problem. The law treats it as a matter for joint action. The
kings and rulers of the maritime states-Rhodes, Cyprus, Cyrene, Syria, Egypt-are all
urged to co-operatein the suppressionof piracy. In the new text from Cnidos the proconsul
of Asia is instructed to explain that this was the reason why the People have now made
Cilicia a provincialzone.53 The author of the bill goes out of his way to assure the oriental
powers that there is no aggressive intention behind the suppression of a menace that
threatened all alike. The Romans did not at this time-in my opinion-establish a new
province inappropriatelycalled Cilicia in the regions of Pamphyliaand Pisidia, as is some-
times suggested.54The wording of the law should mean exactly what it says. The praetors
of Asia are instructed to operate against Cilicia, the mountainous coastal region where the
pirates had their strongholds. There was no need for a second province in Roman Asia.
The existing praetorshipof Asia was a virtual sinecure at this time, with no military and
few civil duties. The praetor could very well take on the task of suppressing pirates, for
which the province of Asia could supply the means-a local fleet, materialresourcesand a
naval base in Pamphylia, which was an outlying sector of the inheritance of Attalus.55
Hence the operationsof Antonius in no way changed or were meant to change the balance
of power in Anatolia. So too with the revision of the territorial donations of I29. The
Senate removed Phrygia and Lycaonia from the successors of Mithridates Euergetes and
AriarathesV.56 No reasonsare revealed. Some snub to the independenceof the kings may
have been intended, as in the removalof Cariaand Lycia from Rhodian control in I67. Or,
more probablyat this time, it resulted from the notorious intrigues of the Romanpublicani
to extend the limits of their operations.57This was not an act of imperialaggression. Rome
reassumedwhat had alreadyfallen to her by the inheritanceof Attalus.

3. THE AMBITIONS OF MITHRIDATES

If Romanpolicy had not altered,other events were radicallytransformingthe situation.

From about I I5 onwardsthe young king MithridatesEupatorset about expandingthe small
kingdom of Pontus into an empire of the lands beyond the Halys.58 His first operationswere
in areas beyond the Roman horizon. He secured control of the fertile lands around the
eastern Euxine, in the Crimeaand the coastlandsof the southern Ukraine, where he broke
the power of the Scythian paramountchief called Scilurus, and gained the hegemony of the
old Greek cities. East of the Straits of Kertch he gained the fertile valleys of Phasis and
Colchis, and consolidatedhis gains by annexing the intermediatemountain land of Little
Armenia, which separated Colchis from eastern Pontus. These events are approximately
dated to Mithridates' early years in the brief passages of Strabo and Justin that record
them.59 Only once is it indicatedthat Rome paid any attention,when Memnon recordsthat
the sons of Scilurus managed to send a mission to the Senate, which duly requested
Mithridatesto return their principalitiesto them-without any effect at all.60
Mithridates finally turned his attention to Roman Anatolia, west of the Halys, in an

aa Strabo 14. 5. 2 (668-9). For the new text of the 57 cf the dispute, in 129 or ioi, in the s.c. de agro
Piracy Law, cited as ' Lex Cnidia ', see M. Hassall, Pergameno, Greenidge and Clay, op. cit. 278; R. K.
M. Crawford, J. Reynolds, JRS 64 (1974), I 95 f., and Sherk, Roman Documentsfrom the Greek East, no. 12;
for the text from Delphi, ibid. and FIRA2 I, I2I, no. 9. cf. H. B. Mattingly, AJ3P 93 (1972), 412 f.; at Priene
Whether this is one law or two laws of the same date is c. 98-i, Inschr. Priene, no. III, 1.II2 f.; at Oropus in
here immaterial. For Cilicia, Lex Cnidia III. 30-40, 73, Syll.8 747, 24-30, and Ilium, ILS 8870. Cf. the
completing Lex Delphica B. 7-8. new evidence for publicani and annexation in Thrace,
" See my detailed discussion op. cit. (n. 49), 6-8, Lex Cnidia Iv, 5-I 8; also, later, Memnon, FGrH
and 4, n. i.. Add that the Livian tradition of 434, (27), 5-6.
Eutropius 6. I. I and 3. I succeeds in distinguishing 58 Th. Reinach, Mithridate Eupator (I890), 49-I o6,
between Cilicia and Pamphylia in the context of is still the basic reconstruction, enlarged only for the
pirate wars. Crimea by M. Rostovtzeff, CAH IX 225 f., and
55 cf. Sherwin-White, op. cit. (n. 49), 2-3. summarized by Magie, op. cit. I, i95 f.
'6 For Phrygia, App., Mithr. 57; Justin 38. 5. 3, 59 Strabo 7. 3. I8 (307), 4. 7 (312); 12. 3. (54
with OGIS 436, which may date its removal to iI9 28 (555); cf. also 2. I. I6 (73); Justin 37. 3. 2 and
or I I6 according to the restoration of the name or Prol. 37. For the record of Diophantus, Syll.3 709.
names of the presiding magistrate(s). Cf. T. Drew- '0 Memnon, FGrH 434, (22), 3-4, cf. Reinach,
Bear, op. cit. (n. 43), 79 f. For Lycaonia see Lex op. cit. (n. 58), 95-6.
Cnidia III, 22 f., where its resumption precedes the
law of ioi-ioo.
 ROMAN INVOLVEMENT IN ANATOLIA 7I
operationcharacteristicof the diplomaticmethods of the past two generations. Some time
before ioI, in concert with Nicomedes of Bithynia, he partitioned the principality of
Paphlagonia, the fertile zone of valleys and uplands formed by the coastal mountains
between the easternboundaryof Bithyniaand the Halys. The Senate, on receipt of protests
from the Paphlagonians,sent a mission which the two kings succeeded in bamboozlingby
technicalities and bribes.6' They remainedin effective possession for the next six or seven
years.62 Meanwhile both kings coveted the territory of Cappadociain the south-eastern
quadrantof Anatolia. Since the death of the great AriarathesV about I30, a succession of
minoritiesand regencieshad weakenedthe ruling dynasty. Mithridateshad secured indirect
control of Cappadociabefore the Paphlagonianaffair,throughthe marriageto AriarathesVI
of his sister, who was now regent for her young son. But Nicomedes invaded the kingdom
and marriedthe lady. Mithridates retaliatedin the interest of the boy prince, Ariarathes
VII, who eventuallyproved too independentfor his liking. Hence anotherinvasion, and the
notorious assassinationof Ariarathesby Mithridatesin person.63This was followed by the
installationof a pretender, a young son of Mithridates,whom we call AriarathesIX.
These remarkableevents passed without any attention from Rome until, severalyears
later (as will emerge), Nicomedes, outwitted and outgunned, appealedto the Roman Senate
in the name of yet another fictitious claimant. He was soon followed by a countermission
from Mithridates,claimingthat his man, AriarathesIX, was a true Ariarathid.64The Senate
at last took cognisance. But when? The dates are vital for appreciationof the Roman
attitude. But they are difficultto fix. Justin's OrientalHistory is most imprecise, and the
known regnal years of the Cappadociancoinage, though definite, do not have a certain
starting point.65 Two data seem to be agreed amongst scholars-that AriarathesVII was
still alive, unliquidated, in the year 1Ioi-1o,66 and that his successor, the false Ariarathes,
reigned for five years before his reign was interrupted.67Hence the appealof Nicomedes to
Rome was not before 97-6. So we cannot explain the Senate's neglect of the upset of the
balance of power in Anatolia by attributing it to the great crisis of the Cimbric Wars of
103-I. That may explain the Roman remissness over Paphlagonia. But Rome was seldom
less occupied with warfarethan in the early nineties.
The Senate finally acted with some decision, yet once more it sought a compromise
solution. Both claimantsto the Cappadociandiademwere disallowed,and the Senate made
the remarkableproposal that Cappadocia should be given freedom: that is, that the
aristocracyor feudal lords should rule the country without a king. At the same time,
Nicomedes was required to vacate his part of Paphlagonia, from which he had never
withdrawn, and Paphlagoniaalso was declared free. Justin states that this was done to
placate Mithridateson the largerissue of Cappadocia.68The declarationof freedom was an
inappropriateapplicationof the device that Rome had used effectively in the past century
when anxious to secure the independence of the Hellenistic city states of Achaea and Asia.
But Paphlagoniadid not consist of efficient leagues of self-governing cities. The Senate
could not have done more to demonstratethat it preferredweak diplomacyto armed inter-
vention in Anatolianaffairs,despite the possession of a broad territorialbase in the Asian
province. But this was the last evasive decision in the old style.

61 Justin 37 4. 4-9. The traditiolnal date c. 104 "4 Justin 38. 2. 3-7. I omit the vain attempt of the
depends on connecting Paphlagonia with the murdered king's brother (Ariarathes VIII) to expel
embassy of Mithridates to Rome known from Diod. the false Ariarathes, ibid. I-2.
36. 35, one or two years before the second tribunate 65 The Cappadocian coinage has been resurveyed
of Satuminus. Cf. Reinach, op. cit. 95 f.; G. Daux, and recatalogued by B. Simonetta, Num. Chron. I96I,
BCH 57 (1933), 8I. 9 f., with some modification of the data on which
62 Nicomedes still held Paphlagonia at the time of Reinach established his chronology, cf. n. 67 below.
the replacement of Ariarathes IX by Ariobarzanes in 6" The date results from OGIS 353 and Inscr.
c. 96 (n. 67 below), Justin 38. 4. 6-7. I omit con- Delos 1I576, 1902, cf. G. Daux, op. cit. 8i f.
sideration of Mithridates' occupation of Galatia at 67 Coins record his regnal years 2-5 (but not 6), I2,
this time, according to Justin loc. cit., for lack of I3 and 15, Simonetta, op. cit. I8. All scholars seem
supporting evidence. Possibly this refers to the to take the break after ' five ' to mean that Ariarathes
territory of the Trocmi beyond Halys, in which IX was expelled in or after his fifth year and restored
Mithridates eventually built Mithridation (Strabo not later than his twelfth year (c. 90-89). This is not
I2. 5. 2 (567)). affected by the radical arguments of 0. Morkholm
Justin 38. I, with Mennon, FGrH 434, (22), 1,
63 about other aspects of the coinage, Num. Chron. I962,
elucidated by Reinach, op. cit. 97 f. Cf. Magie, op. 407 f.; I964, 21 f.; I969, 26 f.
cit. I, 203, and his notes. 68 Justin 38. 2. 6-7, with Strabo 12. 2. II (540).
72 A. N. SHERWIN-WHITE

The Cappadocianbarons did not approveof the Senate's solution, and with its consent
installed one of themselves, a certain Ariobarzanes,as king. We now touch on a minor
controversythat does not greatlyaffectthe interpretationof events. Ariobarzaneswas either
reinstated, after a brief reign and a sudden expulsion, or originally installed, as Professor
Badian would have us believe, by the armed assistance of the Roman propraetorLucius
Sulla.69 Sulla, in 96 or 92, invaded Cappadocia,which was held at the time by the baron
Gordius, who had long acted as the agent of Mithridates in Cappadocia. Gordius was
supportedby Armeniansoldiery, not by Pontic troops, and Sulla's army consisted mostly of
Asiatic levies.70 This was not an act of war between Rome and Mithridates, who had
prudently withdrawn his puppet from Cappadociabefore the arrival of Sulla,71 but the
suppression of a rebel Cappadocianin the name of the legitimate king, Ariobarzanes. But
it was the first time since i88 that a Roman army of any sort had intervenedin the dynastic
quarrelsof the Anatoliankings. Plutarchcomments that the underlyingpurpose of Sulla's
mission was to check the expansion of Mithridates,who was seeking to double the power
that he alreadypossessed. Such languagehad not been used in an Asiatic context since the
humbling of Antiochus Epiphanes at Alexandriaseventy years earlier.
Mithridateswas foiled for the time but not deterredfrom his western ambitions. On
the death of Nicomedes of Bithynia about 93, he set on foot an intrigue to replace the new
king, Nicomedes IV, by his bastard brother, a man known surprisingly by the name of
Socrates the Good. There was yet another arbitrationby the Senate, which rejected the
claims of Socrates.72 Still undeterred,Mithridatesremoved Nicomedes by the more direct
method of supplying Socrates with a private army for the task, while his own generals
expelled Ariobarzanesfrom Cappadocia.73These events of 9I or 90 bring us to the great
enigma of the First Mithridatic war. Action follows action. A Roman emissary, Manius
Aquilius, arrivesto support the praetor of Asia in the restorationof the two kings. They
raise a large army of Asiatic levies but have no substantialRoman forces. Mithridatesdoes
not resist the restoration,but when Nicomedes, driven on by Aquilius, raids his territory,
Mithridates lodges strong protests and again expels Ariobarzanes. The Roman leaders
now organizea general invasion of Pontus from three directions. This act of open war led
to the rapid destructionof the three armies and the expulsion of the Romans from Asia.74
We are faced by a double puzzle. On the Roman side, how did their leaders in Asia allow
themselves to be drawninto a war with Mithridatesfor which their preparationswere quite
inadequate,at a time when Rome was in the throes of the Social War in Italy? And, what
emboldened Mithridatesto believe that he could secure decisive militaryvictory where the
Seleucids and the Macedonianshad failed?
For Mithridates there is both a political and a military explanation. We take it too
readily for grantedthat the Romans were invincible. Their superioritywas not so great in
the nineties as it appearswith afterknowledgeof the campaignsof Sulla and Lucullus. The
reputationof the Roman legions had been tarnishedsince the great victories over Hannibal
and the Hellenistic kings. Protracted wars against tribal barbariansin Spain, Gaul and
Macedonia had been markedby spectaculardefeats and the annihilationof several Roman
armies. Only the elderly Gaius Mariushad shown notable military talent in the last forty
years. In Asia itself men rememberedhow it had taken three successive consuls to suppress
the peasant revolt of Aristonicus; quite recently the insignificantJugurthahad baffledthe

69
The arrangement of E. Badian (Athenaeum N.S. 72
Granius Licinianus 35. 30 (F), badly summarized
37 (Ig59), 279 f., reprinted in his Studies in Greek by App., Mithr. io; Memnon, FGrH 434, (22), 3.
and Roman History (I964), 56 f.) implies that the Cf. Magie, op. cit. I, 207, Ii, IO99 n. I9. Reinach,
Senate took a strong line with Mithridates in op. cit. (n. 58), II4, did not know the Flemisch text
Cappadocia from the start, using force instead of of Licinianus, which still leaves much obscure.
diplomacy. I have criticized it in a forthcoming article 7 App., Mithr. Io; Justin 38. 3. 4.
to appear in CQ I977. For the older view, cf. Magie, 74 This summarizes the story of App., Mithr. I I-I 9.
op. cit. I, 2o6, following Reinach, op. cit. (n. 58), The Livian epitomators (Ep. 76; Florus I. 40. 3-6;
105. Eutropius 5. 5; Orosius 6. 2. 1-2) are very thin, with
70 Plut., Sulla 5. 6-7 is the principal source, with a different emphasis, omitting the role of Aquilius
Livy, Ep. 70; App., Mithr. 57. almost entirely, which reappears briefly in Justin
71 Justin 38. 5. 6, confirmed by App., Mithr. 38. 3. 4 and 8, Memnon, FGrH 434, (22), 7. Cf.
57,
not noticed by Badian, op. cit. (n. 69). n. 86 below.
 ROMAN INVOLVEMENT IN ANATOLIA 73
Romans for seven years in Numidia.75 By the nineties Roman manpower was gravely
stretched. On four European frontiers, in Spain, Transalpine Gaul, north Italy and
Macedonia, tribal pressures, both external and internal, required the constant presence of
legionary armies, and were apt to explode into major wars at any moment. The logistical
situation had greatly changed since the time of the Hellenistic wars, when northern Italy
was the only serious preoccupation elsewhere.75a Mithridates could reckon that Rome had
little to spare for a war in Asia.76 Rightly: in the twenty-five years of warfare with
Mithridates down to 66, Rome was unable to spare more than five legions at any one time
for the oriental war.77 This was about the standard consular army of the second century,
when two Roman legions were regularly matched by a similar force of Italian cohorts, for a
consular command.78 Mithridates could calculate that, even without the diversion of the
Social War, the military problem was manageable.
Then there is the naval factor. Mithridates had an immense naval superiority which
gave him control of the seas. The great Roman fleets that had opposed the Carthaginians
and the Greek kings had been allowed to disintegrate. The Romans were reduced to
dependency on the local flotillas of the Greek cities, which, except for the Rhodian con-
tingent, were a scratch collection.79 But Mithridates set about building a great fleet, which
in 89 contained some 300 decked vessels, and was manned by trained men from Egypt and
Phoenicia.80 With this fleet he was able to sweep the seas clear of Roman men-of-war, so
that Sulla's army was unable to cross the Aegean or even to attack the main base of the
Pontic army in Euboea, and Roman reinforcements were harried in the passage of the
Adriatic.81 It was only the tactical and technical superiority of the veteran legions of Sulla,
hardened in the bitter fighting of the Social War, that defeated the armies of Mithridates,
which fought in Greece with remarkable courage and persistence. The fair-minded source

75 cf. the speech attributed to Mithridates by V. Ilari, Gli Italici nelle strutture militari romane
Pompeius Trogus in Justin 38. 4-7, on the theme (I974), I67 (with Strabo 4. i. I. added); P. A.
'Romanos posse vinci', citing Aristonicus (6. 4), Brunt, Italian Manpower (I971), 427 f. For the
Jugurtha (6. 6), Cimbri (4. 15). consular commands from 150 on, see T. R. S.
75a From I90 to i68, according to Afzelius' Broughton, Magistrates of the Roman Republic2 I,
evaluation of the detailed evidence of Livy, from under each year. For consular armies see n. 78 below.
eiglht to ten legions, with their allied complement of For praetorian commands, Livy 39. 30. 12; 40. 36. 8;
five to eight thousand men apiece, were regularly 4I. 5. 6-7, 2I. 2, suggests a norm of one legion with
deployed in the two Spains, north Italy, and in some socii, which frequently escalated by the retention of
years Sardinia, in consular and praetorian commands. legions from year to year.
The figure rises to twelve legions during the oriental 76 cf. Justin 38. 4. I6: ' etiamsi singula bella
wars, which required armies of four legions in some sustinere Romani possint, universis tamen obruantur
years, found in part by cutting down the garrison of ut ne vacaturos quidem bello suo putet ', a view that
north Italy, while two legions remained around Rome Trogus or his source evidently found tenable.
as a short-term strategic reserve. This figure, with 77 Sulla has six legions in Campania in 88 (App.,
the Italian complement, gives the maximum potential B.C. I. 57), and takes five legions and some extra
of Roman manpower under the traditional system. units to Greece and back (Mithr. 30; B.C. I. 79). In
After the termination of Livy, statistical information Greece he acquires an extra legion mysteriously in 86,
disappears. A few isolated figures suggest that the probably withdrawing it from Macedonia, Plut.,
standard consular and praetorian armies remained Sulla I5. 4-5; Memnon, FGrH 434, (22), I2.
much the same in size down to 9I. Pressure did not Valerius Flaccus took out only two legions, which
abate after I50, when the African war required five Sulla left behind for Murena (App., Mithr. 5i, 64).
consuls out of six from I49 to 147, and renewed L. Lucullus took out only one legion to add to four
troubles in Spain took two consular armies each year, legions then stationed in Asia and Cilicia, Plut., Luc.
under consuls and proconsuls, from 143 to 134. 7. I, 8. 4; App., Mithr. 72. Aurelius Cotta seems
Meanwhile Macedonia became a praetorian commit- not to have a consular army in addition to his fleet.
ment from 146. Between 125 and I20, the conquest After infantry losses of three to five thousand men at
of Transalpine Gaul occupied four consuls, con- Chalcedon, little was left to join Lucullus: App.,
jointly in some years. These overlapped with two Mithr. 7I; Plut., Luc. 8. 2; Memnon FGrH 434,
consuls operating in Sardinia (126-2) and Nearer (27), 7-8.
Spain (I23-c. I2I). Macedonia required consular 78 Pol. 6. 19-20, 22. 4, 6, 26. 3, 7 f. Cf. A. Afzelius,
attention from I14 to I07, overlapping with the op. cit. (n. 75a), 34 f., 62 f.; V. Ilari, op. cit. (n. 75a),
Numidian war from I I I onwards, and with consular ch. vi; P. A. Brunt, op. cit. (n. 75a), 68i f.
commands in north Italy in at least I I 3 and I 09. The 79 cf. my discussion op. cit. (n. 49), 4-5, nn. 9-14.
Numidian command in turn overlapped with consular Sulla, arriving in Greece in 87 without a fleet after
activity in Gallia Transalpina from 107 to 105, the surrender of the Asiatic flotilla in 89, was confined
when two armies were on foot in Gaul and a third to land operations in Achaea until his quaestor
in reserve in north Italy. Thus the sole inter- Lucullus retumed, in the winter of 86-5, with a naval
vention in Asia ( 31-29) fell in a rare quiescent force collected with difficulty from Syria, Rhodes and
period: the annual Roman requirement from 125 Pamphylia: App., Mithr. I7, 29, 33, 5I; Plut., Luc.
onwards could hardly be less than nine legions. See, 2. 2-3, 3. I-3, 4. 2.
for the period before I67, A. Afzelius, Die rdmische 80 App., Mithr. I3.
Kriegsmacht (I944), 47 f., 62 f., 78-9. For 146-IO1, 81 APP., Mithr.
5I-
74 A. N. SHERWIN-WHITE

of Appian went out of his way to commend them for their first successes: ' though few in
number they mastereda much more numerous enemy, not through accidents of terrain or
the mistakes of the other side, but by the merit of their generals and the courage of the
soldiery'.82 They were unlucky to come up against the legions of Sulla, trained by the
professionalismof the Social War to a level that had seldom been reachedbefore. Against
lesser men, notablythe legions of Murenain 83, and those of Triariusat Zela in 67, they won
impressive victories.83
Militarily the prospects were good. But there was more to it than that. Mithridates
might well take the view that the Romans drew a clear distinction between their interest in
European Greece and their commitment in Anatolia. In Macedonia, since the occupation
of I48, they had maintaineda majormilitarypresencethat escalatedto meet foreign dangers,
and the proconsulshad steadily extended their control over the neighbouringtribes.84 But
in Asia the presence and the policy of Rome had been static. There was still no serious
military establishment. The resumption of the districts of Lycaonia and Phrygia-taken
back impartiallyfrom the kings of Pontus and Cappadocia-was no more than a return to
the original terms of the inheritance of Attalus. The recent interventions in Cilicia and
Cappadociawere concerned with the maintenanceof the status quo.85 Mithridates might
well conclude from the management of Roman policy in Anatolia during the last half
century, and from his own experience of it during the last fifteen years, that the Romans
were by no means committed to the expansion of their empire east of the Aegean or even
to the maintenanceof their position in Asia at all costs. He was, as we have seen, a great
intriguerand a great negotiator,sharpto press his advantage,but quick to withdrawfrom a
dangerousposition before reaching the point of no return, always the man to prefer half a
loaf to no bread. That he seized the golden opportunity of 89 to destroy the provincial
levies of Cassiusand Oppius, and that he invaded Macedoniaand Greece in the summer of
88 when the Romansfailed to despatcha consulararmy againsthim, does not mean that he
expected to retain all the territoriesthat he had so swiftly occupied.86 But the greater his
success in Europe, the stro-ngerwould be his bargainingposition for a final settlement.
Unfortunatelyhe misjudgednot only the effectivenessof the Romanwar machineat the
moment but the spirit of contemporaryRoman imperialism. The springs of foreign policy
were altering in these years. It is not just that the Senate had belatedly recognized the
dangerousgrowth of the empire of Mithridates, and was now determined to check it, or
that a Roman consular,accordingto a somewhatapocryphalanecdote,had privatelywarned
Mithridateseither to make himself the equal of Rome or silently obey her commands.87All
that was standardRomanpolicy and propagandain Asiatic affairs. Eumenes of Pergamum,

82App., Mithr. I9. At Chaeronea the sources his rejection of one of its statements. He ignored the
criticize the tactics of Archelaus, but not the valour order of events in the detailed narrative of Appian
of his men, who crack only in the final rout: App., (Mithr. I7-2I), which places the warfare in Anatolia
Mithr. 42-4; Plut., Sulla I7. 9-I9. 8. Their skill in and the occupation of Asia before the consular elec-
siege warfare was outstanding, App., Mithr. 34-7, 40. tions of 89 and the assignment of Asia as a consular
83App., Mithr. 65, 89; Plut., Luc. 35. I-2. The province; while the siege of Rhodes, mopping-up in
loss of twenty four tribunes and one hundred and fifty Lycia and Paphlagonia, the despatch of Archelaus to
centurions at Zela indicates a major disaster, even if Achaea and his clash with the proconsul of Macedonia,
exaggerated by the friends of Pompeius. are set in the year of Sulla's consulship (88), when
84After increasing trouble with the Scordisci from political events at Rome prevented the normal
I I8 onwards (SIG8 700), Macedonia became a departure of the consul with his army for the cam-
consular province from 114 to c. 107: Livy, Ep. 63, paigning season. Appian's order of events makes
65; Florus r. 39. 4-5; ILLRP I 337; Fasti Triumph. much better sense of this protracted series of
for Io6. After the campaign of the praetorian campaigns, but the matter needs discussion else-
T. Didius c. 102-I, the territory of the Caeni was where. Livy's Epitomes and the other subsidiary
annexed, cf. Lex Cnidia IV, 5-30. sources are susceptible of various interpretations;
85 Above, p. 69 f. so too the amended Olympic date in Mithr. 17.
86 The alert will notice a revision of the chronology Orosius (5. 19. 2.), using Livian compendia, was
of Reinach for the beginning of the war, hitherto justly puzzled about the year 88: 'utrum abhinc
unchallenged despite the difficulties that it creates primum coeperit an tunc praecipue exarserit
(cf. Will, op. cit. II, 398-9). Reinach, op. cit. 112 f. (bellum).'
attributed the campaigns of Mithridates in Anatolia, 87 Plut., Marius 3I. Marius in 99-8, during his
the siege of Rhodes and the invasion of Achaea to 88, unofficial visit to Cappadocia and Galatia religionis
when the crisis of the Social War was passed, so that causa, addresses Mithridates thus. Too much has
in go-89 Mithridates, taken by surprise, 'missed the been made of this ' secret history' by R. J. Luce,
bus.' Reinach based his dates on the serial order of Historia 19 (1970), I62 f., following E. Badian, op.
events in Livy, Ep. 76-9, despite its ambiguities and cit. (n. 69), 279 f.
 ROMAN INVOLVEMENT IN ANATOLIA 75
Antiochus Epiphanes, the Rhodians, Demetrius, and Prusias of Bithynia, had all received
a dose of that medicine in their time.
What is new is the attitude of the Roman proconsuls in Asia in 89 to the notion of
aggrandisement. Appian gives a startling account of the final breach between Mithridates
and the Roman leaders. First, they force Nicomedes, after his restoration, to provoke
Mithridatesby plunderinghis territory,and then, after rejectinghis protests, they organize
the three-fold invasion of Pontus, in an act of open war for which Appianthree times firmly
assertsthat they had no authorityat all from the Senate or People of Rome.88 It is difficult
to discount this tradition. The first part of it, the enforcementof Nicomedes, can be traced
back from Appian, and other late sources, to the Roman annalists of the late Republic,
through a brief allusion in the Historiesof Sallust (which did not cover this period), when
Sallust attributedto Mithridatesthe charge: 'me. . per Nicomedem bello lacessiverunt'.9
It was not the policy of the Roman Senate, amid the turmoil of the Social War in Italy,
wantonly to stir up a majorwar in Anatolia. They expected, rightly, that a minor show of
force would secure their purpose. Hence the despatchof two praetorsto Asia, with modest
forces and the advice of a consularcommission, to support the restorationof the two kings
in the style of the previous operation of Sulla.90 But what emerges in Appian is the first
manifestationof the aggressiveimperialismof the individual army commandersof the late
Republic, notably documentedby Lucullus' invasion of Armenia,the Caucasiancampaigns
of Pompeius, the campaigns of Caesar in Transalpine Gaul and Crassus' invasion of
Parthia. This was not the officialpolicy of the Roman Senate, whose objective, fulfilled by
Sulla in the Peace of Dardanus,was merely to confine Mithridatesto his native kingdom.91
The emergence of this new attitude in the context of Anatolia was fatal to the ultimate
expectations of Mithridates. Once the ambitions of military men had turned towards the
East, the possibility of negotiatedsettlements must disappear,and the old Roman implaca-
bility, shown of old to the Samnites and to Carthage,would assert itself in a new environ-
ment, as in the end it did, through Lucullus and Pompeius. But I would insist that the final
developmentshould not be assumed as the operativefactor in the mind of Mithridatesor of
the Roman Senate before its first manifestationin 89.

St. John's College,Oxford

88 App., Mithr. I5, I7, I9. authorized against Socrates (Justin 38. 5. 8): ' regem
89 Sallust, Hist. iv fr. 69. io. Cf. Florus I. 40. 3; Bithyniae Chreston in quem senatus arma decre-
Dio fr. 97; Justin 38. 5. 10. verat '. There is not one word about war with
90 Oppius, whose provincia covered southern Mithridates. When Pelopidas, in the prolonged
Phrygia and Lycaonia, was probably intended, like negotiations with Aquilius and Cassius, eventually
Sulla, to restore Ariobarzanes to Cappadocia (though proposes that the Senate should be consulted, the
he is not named in App., Mithr. i I), while Cassius in Romans promptly dismiss him and organize their
Lydia and northern Phrygia was well placed to assist offensive without referring the request to Rome,
Nicomedes in Bithynia. Cf. App., Mithr. I7, 20. App., Mithr. i6-I7. Earlier they pressed Nicomedes
This may be the first occasion of the division of the and Ariobarzanes to attack Mithridates' territory
provinces, cf. my discussion op. cit. (n. 49), 9. precisely because they lacked direct authority for this
91 cf. Justin's technical language (38. 3. 4): themselves, App., Mithr. iI: 'and so to provoke
'decernitur in senatu ut uterque in regnum restitu- Mithridates to war, because the Romans would
antur, in quam rem missi M'. Aquilius etc.', con- support the kings as allies if they were at war.'
firming App., Mithr. i I. The use of force was