future subjects, nor did it govern its empire as a militarized ruling class
controlling ethnically and culturally distinct populations.
Modern studies of any aspect of how the Roman Empire worked
go badly astray when they underestimate the role of personal power
as compared to the power of the state. One scholar notes that in his
exhaustively detailed history of Judaea under Roman rule, Josephus
hardly discusses the Roman state and does not seem to understand
the concept of state power.37 The army was by far the largest institu-
tion of the Roman state; but it was social relationships, and not mainly
the army, that knit the empire together. Much of “Roman” rule was
done by local aristocracies, petty kings, chiefs, “big men,” and large
landowners acting out their own agendas and bringing in the Roman
army or the Roman government when it suited them. Taxes were col-
lected by local agents, and many local governing institutions continued
to operate. Parties to local feuds and rivalries turned to their Roman
governors to settle disputes—this was the essence of Roman law and
of Roman provincial government, not edicts and occupation. Even in
places where indigenous institutions were transformed by their contact
174 Mattern
with the Roman Empire, many people never saw an official representa-
tive of Rome apart from the occasional soldier, and in some provinces
even these were rare. The traffic in favors and injuries that governed
Roman social relations governed the empire as well.
One could go even further and say that to describe Rome’s prov-
inces as distinct territorial regions is to oversimplify. Although Roman
law and policy recognized the province as an administrative unit, each
under a governor of senior senatorial rank, areas like Sicily, Gaul, or
Judaea did not have a single uniform relationship with Rome. Rather,
each was a network of communities and individuals with a unique set
of relationships to the Roman state, represented by the Senate and the
emperor (or, in the republic, the “Senate and the People of Rome”),
and to individual Roman aristocrats. That is true even where Rome
ruled provinces directly; but at all times we hear of a bewildering array
of petty kings and local chiefs allied to Rome and considered part of
its empire. These are complicated points, which I illustrate with two
examples.
It would raise few eyebrows among scholars of Roman history to
say that in Judaea, the Romans supported a friendly king, Herod the
Great, for several decades until his death in 4 BCE. However, it would
be more accurate to say that first Julius Caesar, then the tyrannicide
Cassius, then Mark Antony, and eventually the emperor Octavian sup-
ported Herod and that Herod supported each of these men in turn.38
In the turmoil that followed the assassination of Julius Caesar in 44
BCE, even though Herod’s father owed his Roman citizenship and
status as king to Caesar, Herod accepted troops from Cassius in re-
turn for a presumed alliance. When Mark Antony defeated Cassius,
Herod switched allegiances and offered his support to Antony. When
the Parthians invaded Syria and Judaea and deposed its high priest and
set up their own nominee, Antony supported Herod in his efforts to
get them out and had the Senate formally declare him king of Judaea.
Herod and his army also took part in Antony’s unsuccessful campaigns
against Parthia that occurred at the same time. Finally, when Octavian
defeated Antony at Actium in 31 BCE, Herod made a famous pilgrimage
to Octavian to switch sides once again. After he convinced the emperor
that he would be as loyal to him as he had been to Antony, Octavian
Counterinsurgency 175
acknowledged his friendship. He also gave Herod possession of certain
cities in Palestine that the Romans at the time considered subject to
them or to the deceased queen of Egypt.
Herod supported his Roman friends with troops when they asked
him to, and he was a promoter of their interests in other ways. But in
return, Herod received the means with which to defeat his dynastic
rivals and enlarge his kingdom. His story inextricably entwines Jewish
politics and Roman civil war with regional and international politics. It
is largely through these complex relationships that the Roman Empire
managed foreign and internal threats, which in this system are not al-
ways easily separable; and the system worked better to the extent that
all parties benefited.
After Herod’s death, his kingdom was divided among his four sur-
viving sons until 6 CE.39 At that time, amid civil unrest, the Romans
deposed Archelaus and established as the prefecture of Judaea a part of
Herod’s kingdom that included Jerusalem. But Herod’s sons continued
to rule the rest. In 41 the emperor Claudius gave Herod’s grandson,
Agrippa I, sovereignty over the whole kingdom once ruled by Herod.
After Agrippa I died in 44 CE, Claudius made most of that territory sub-
ject to a Roman procurator of equestrian rank, except that Agrippa I’s
son, Agrippa II, continued to rule part of Galilee. This is the situation
in the province we know most about; whether its history was more
complicated than that of other provinces is not known. But clearly a
definition of empire that mainly relies on direct military or bureau-
cratic control fails to capture the essence of the situation. Only one
that takes account of dynamic relationships among Rome the state,
individual Romans, and local elites can capture it.
The most accessible window onto a province long under direct Ro-
man rule—that is, subject to a Roman governor—are the speeches that
record Cicero’s prosecution of its corrupt governor Gaius Verres in 70
BCE. At that time Sicily had been part of the Roman Empire for nearly
two centuries. Cicero’s orations Against Verres reveal the complexity of
Rome’s relationship with Sicily and of the social ties that connected
the Roman ruling class to its indigenous, hellenized urban ruling class.
Most of the province paid one-tenth of the grain crop as tax. The
contract to collect the tax was sold in Sicily, to local corporations; this
176 Mattern
was, or was at least perceived to be, the same system that prevailed
under the last king of Syracuse, Hiero II, who ruled during the first
two Punic Wars. Throughout his long speeches Cicero refers to the
“law of Hiero” with reverence, as an institution for which disrespect
amounted to gross misrule. Besides this, five cities were exempt from
taxation; two paid on their own without contracts; and the contracts
to collect taxes from some cities were sold at Rome, an arrangement
less advantageous to them.40 The terms of each community’s rela-
tionship to Rome reflected the circumstances of its participation in
the First Punic War or in subsequent conflicts, or its relationship to
individual Roman patrons. The Claudii Marcelli and Cicero himself
considered themselves patrons of the province as a whole; Cicero calls
the Sicilians “allies and f
riends of the Roman people and close con-
nections of myself.” Some individual cities, such as Segesta, Syracuse,
and Messana, had relationships with specific aristocratic families.41
And individuals from among Sicily’s hellenized elite also enjoyed spe-
cial relationships with Roman senators that continually surface in the
course of the trial; the one most commonly mentioned is hospitium.
It is an especially damning sign that evidence against Verres can be
extracted, on cross-examination, from his own hospes, Heius of Mes-
sana, or that he presided over the unjust conviction of another of his
hospites, Sthenius of Thermae.42
There was no Roman army in Sicily. When Verres needed shock
troops to carry out his extortionist schemes, he called on the slave
guards of the local Temple of Aphrodite at Eryx, whose normal job
was to protect the temple treasury.43 He took kickbacks from local tax
corporations and from feuding aristocrats prosecuting their enemies
in his court.44 Seeking redress, Sicilian individuals went to Rome and
cities sent delegations pleading their cases to their patrons and connec-
tions in the Senate, before whom he would be tried.45 Verres’s rule was
corrupt and rapacious, but it was only his social connections inside and
outside Sicily that allowed him to get away with it. Cicero makes no
references to insurgency in this period and characterizes the Sicilians
in paternalistic terms as a docile, childlike people.46 A more plausible
explanation for the low level of insurgency in Sicily in Cicero’s time—
if this characterization is accurate—is the density of the connections
Counterinsurgency 177
between the local ruling class and the Roman aristocracy, which Ci-
cero’s speeches illustrate very well.
What, then, is insurgency, and what is counterinsurgency? One way
to view insurgency, resistance, and banditry is as attenuated areas or
holes in the network of social relationships that linked the empire to-
gether and bound it to the senatorial aristocracy and to the emperor.
In other cases one of the nodes in the network—a petty king, a Roman
aristocrat, an auxiliary commander—might yank its strings in a new
direction, activating a new set of connections at cross-purposes with
the dominant ones. Foreign relations, local politics, and rivalries inter-
nal to the Roman ruling class worked inextricably together or against
one another.
The Romans negotiated diplomatical y with petty kings, tribal chiefs,
bandits, and nomads.47 They paid subsidies and made treaties. They
granted citizenship, titles, or military support. They formed an infinite
variety of personal connections that linked the Roman ruling class to
local aristocrats and strongmen. When this social network failed, as it
often did, they ruled by force. They occupied territory, in a few cases
very densely. They waged major wars against rebels and took pride
in defeating them with their superior discipline, tenacity, and military
engineering. They fought bandits with patrols, posses, and occasional
military campaigns. They terrorized rebel ious subjects with harsh re-
prisals. Whether these latter measures worked is difficult to say. The
Roman Empire endured a long time, but no era was free of insurgency
and banditry. My argument, however, is not about the efficacy or inef-
ficacy of the Roman army; it is that the military aspect of insurgency
and counter insurgency, and of empire itself, is only the tip of the ice-
berg. The Romans ruled because their social relationships reached
every where—or at least, they reached far. Those relationships could be
manipulated by anyone. Postulating a tense and dynamic network of
relationships in which al actors vigorously pursue what they believe to
be their own interests—a network that may have gaps and holes, and
in which alternative networks only loosely connected to the dominant
one also operate—might be the most effective way to envision empire.
In its foreign policy, the United States faces problems similar to
those of al ancient and modern empires. In particular, the occupation
178 Mattern
of overseas territory is always expensive and difficult, and the “ruling”
power always governs as a tiny minority. It has become fashionable in
the last decade to look to the Roman Empire for lessons applicable to
modern times. Some of this has seeped outside the universities and the
Beltway and into popular culture.48 Fundamental economic, technologi-
cal (imagine the Romans with nuclear weapons), demographic, and so-
cial differences between the modern and premodern worlds make this
lesson-seeking a very chal enging activity, and not everyone agrees that
the analogy is appropriate or the scholarly endeavor justified. I myself
have expressed skepticism on this question.49 But when I am asked to
comment on the practical lessons of Roman history, my response, with
these caveats, focuses on the critical role of social institutions in holding
the Roman Empire together. The Romans ruled because, as a col ective
“state” and as individuals, the ruling class’s network of dependencies,
favors owed, and negotiated relationships extended everywhere. Where
the Roman social network did not extend, or where part of the ruling
class chose to deploy its own network against the interests of another
part, there trouble arose. Rome succeeded because it drew on, or built,
a common social and cultural language with the elites of the territories
subject to it, and because many powerful elements of its subjects’ popu-
lations found it in their best interest to recognize Roman authority. The
nearest modern paral el may be the “global vil age” created by telecom-
munications technology, financial institutions, free trade, and the con-
sumer tastes and interests that link international communities today. A
focus on shared economic and cultural interests rather than on ideology
is a promising direction for foreign policy in the future.
Further Reading
The subject of revolt and insurgency has received inadequate attention from scholars.
Two articles by Stephen Dyson, “Native Revolts in the Roman Empire” ( Historia 20
[1971]: 239–74) and “Native Revolt Patterns in the Roman Empire” ( Aufstieg und Nieder-
gang der römischen Welt [Berlin: Walter de Gruyter, 1975], 2, 3:138–75), are stil important.
On banditry, the work of Brent D. Shaw, especial y his classic “Bandits in the Roman
Empire” ( Past and Present 105 [1984]: 3–52), has been most influential. Benjamin Isaac has
done much to cal attention to the function of Rome’s army as an occupying force, con-
trol ing the population and policing for banditry and other smal -scale threats: The Limits
of Empire: The Roman Army in the East, 2nd ed. (Oxford: Clarendon, 1992). On the Jewish
Counterinsurgency 179
revolt, the most influential study is by Martin Goodman, The Ruling Class of Judaea: The
Origins of the Jewish Revolt against Rome, a.d. 66–70 (New York: Cambridge University
Press, 1987); and now see his more general work, Rome and Jerusalem: The Clash of Ancient
Civilizations (New York: Knopf, 2007). On personal power in the Roman Empire, impor-
tant works are Richard P. Sal er, Personal Patronage under the Early Empire (Cambridge:
Cambridge University Press, 1982); Brent D. Shaw, “Tyrants, Bandits and Kings: Personal
Power in Josephus” ( Journal of Jewish Studies 49 [1993]: 176–204); the essays col ected in
Patronage in Ancient Society, ed. Andrew Wal ace-Hadril (London: Routledge, 1989); and
Fergus Mil ar’s The Emperor in the Roman World, 2nd ed. (Ithaca, NY: Cornel University
Press, 1992). J. E. Lendon, Empire of Honour: The Art of Government in the Roman World
(Oxford: Clarendon, 1997), is critical to understanding how personal power operated.
Erich S. Gruen’s The Hellenistic World and the Coming of Rome, 2 vols. (Berkeley and
Los Angeles: University of California Press, 1984) revolutionized our understanding of
Roman imperialism by refocusing attention on the institutions and political struggles
of its future subjects, showing how only a thorough understanding of these can explain
how Rome became involved in a region and the shape that its domination took. Other
important works on nonmilitary aspects of Roman imperial control are Greg Woolf,
Becoming Roman: The Origins of Provincial Civilization in Gaul (Cambridge: Cambridge
University Press, 1998), and Clifford Ando, Imperial Ideology and Provincial Loyalty in the
Roman Empire (Berkeley and Los Angeles: University of California Press, 2000).
notes
1 On population, see most recently Walter Scheidel, “Demography,” in The Cam-
bridge Economic History of the Greco-Roman World, ed. Walter Scheidel, Ian Morris, and
Richard Saller (Cambridge: Cambridge University Press, 2007), 45–49. On the size of
the army, for a summary of arguments, see Susan P. Mattern, Rome and the Enemy:
Imperial Strategy in the Principate (Berkeley and Los Angeles: University of California
Press, 1999), 82–83. On taxes as a fraction of GDP, for a summary of arguments see
Elio Lo Cascio, “The Early Roman Empire: The State and the Economy,” in Scheidel
et al., The Cambridge Economic History, 622–25. On subsistence levels and per capita in-
come, Jongman argues that overall per capita income was relatively high for antiquity,
though very low by modern standards (Willem M. Jongman, “The Early Roman Em-
pire: Consumption,” in Scheidel et al., The Cambridge Economic History, 592–619), but
Makers of Ancient Strategy: From the Persian Wars to the Fall of Rome Page 28