Version classiqueVersion mobile

Reconsidering Roman power

Katell Berthelot

The dynamics of power

The rulers ruled

Greg Woolf


This paper considers a series of problems that emerge from conventional uses of the term ‘power’ in discussing the Roman world. Drawing on critiques and reformulations of the concept by social theorists it explores the operation of power in two spheres, first at the centre of the empire where emperors, aristocrats and other courtiers competed for influence, and second in the provinces where identity politics and economic interests intersected in different ways. It argues for conceptualizing power in terms of webs of influence, networks and complexes of relationships and suggests some advantages to such an approach.

Texte intégral

Power at the Centre

  • 1 De Ste Croix 1981, p. 380-381.

1About half way through his masterpiece, The Class Struggle in the Ancient Greek World, published in 1981, G. E. M. de Ste Croix launched an attack on "the seriously false notion that there was a necessary and deep-rooted conflict between the emperor and ‘the Senate’ or ‘the aristocracy’".1 He cites as an instance some remarks made by Keith Hopkins in a paper on elite mobility in the Roman empire published in Past and Present 1965, "which speaks again and again of ‘tension’, ‘conflict’ or ‘hostility’ between the emperor and the senatorial aristocracy collectively […] even of the emperor’s ‘battle against aristocrats’, and of all the emperors as ‘necessarily engaged with the aristocracy in a struggle for power’".

  • 2 De Ste Croix 1981, p. 503, giving the title to Brent Shaw’s brilliant review: Shaw 1984.

2De Ste Croix concedes that serious revolts were almost always led by members of the aristocracy and that the emperors were occasionally obliged to curb the corrupt behaviour of senatorial governors if only to safeguard their own interests. But he objects that however badly relations deteriorated between any individual emperor and the aristocracy, the response was never to abolish the monarchy but simply to replace the tyrant with a new emperor. From early in the principate the Senate did not even try and choose the next emperor, and tolerated any ruler who treated them with tact. Emperors, according to de Ste Croix "could rely on the support of the senators as a class". Indeed, as Class Struggle progresses it becomes clear that it is the aristocracy (or perhaps the propertied classes), not the emperors, who are the real villains of the piece. The final pages criticise Peter Brown for writing that in late antiquity "the prosperity of the Mediterranean world seems to have drained to the top". Rather than the metaphor of ‘drainage’, de Ste Croix prefers "to think in terms of something much more purposive and deliberate – perhaps the vampire bat".2

  • 3 This chapter was written before the appearance of Harris 2016 which pursues an analysis of the rel (...)

3There is more at stake here than the conventional Marxist tendency to play down conflicts of interests among the propertied classes and to stress instead the shared interests of the wealthy and the parlous condition of the classes whose labour supported them. Nor is this just a matter of perspective – that from close up the power struggles of court and senate are more visible, but that from a distance they seem trivial compared to the centuries-long war of attrition between rich and poor. The disagreement illustrates the difficulties that arise from the use of an under theorized notion of ‘power’. Any discussion of the real locus of power struggles presupposes a consensus about what counted as power in antiquity, and that consensus clearly does not exist. This essay is intended as a contribution to rethinking Roman Power.3

  • 4 Excellent studies in this vein include Syme 1958, Millar 1965, Wallace-Hadrill 1983, Roller 2001.

4The disagreement also raises the question of perceptions of power. Hopkins, like others before and since, based his case for conflict, tension and hostility between emperors and aristocrats mainly on written testimony produced by members of the senatorial and equestrian orders.4 Were Tacitus, Suetonius and Dio wrong to claim that real power was at stake in the high politics of the empire? If so, how can we use their testimony to examine Roman Power? But if they were correct – and they were after all very close to what they described – then how should we relate short-term conflicts at the centre to the long-term processes through which wealth drained (or was sucked) to the top? Perhaps there was almost no connection and what we have is a detailed and often thrilling account of essentially trivial events that had little or no consequences in the long term. The anxiety is Braudelian. Did the history of events make any contribution to structural history, or did the latter simply provide a context for the former. Is Roman Power a single field of enquiry?

5The contrast of timing is significant. De Ste Croix’s approach invites us to think about structural violence, and about the essential stability of Roman Power. If Roman Power was not about sustaining social hierarchy through law, ceremony, fiscality and violence then what was it about? Yet the second approach takes us closer to the tortured heart of imperial government, and the experienced paradox of the Roman élite that despite their spectacular eminence they might be subjected to degrading subordination and terrifying risk. As a class their position may have been almost unassailable. Yet as individuals the most prominent figures were often also in the most precarious of positions. This included the emperors themselves.

Power at the edge

  • 5 Steel 2001, Richardson 2008.

6Similar issues arise in respect of the provinces, where Roman power seems at first sight easy to describe – as the power exercised by Romans over non-Romans – but on reflection is more difficult to pin down. The idea that power was something exercised by Romans over others seems to cohere with the language that Cicero and others employed to describe the obligations laid on provincial governors.5 Terms like imperium evoke transitive verbs in which Romans are actors and provincials objects of domination.

  • 6 Braund 1982.

7Yet when considered more carefully relations of power turn out to be more complicated. The extension of Roman influence in the east during the last two centuries BCE often involved not the removal of local power structures, but their co-optation to serve Roman influences. Most areas continued to be ruled by kings, but the territories allocated to each king might be altered, divided or reallocated on the order of commissions of legati or later the command of individual generals with plenipotentiary authority such as Lucullus and Sulla, Pompey and Antony. Some kings were replaced, and others sustained only by Roman support.6 We might understand all this as the expansion of Roman power – as Polybius certainly did using the term arche to describe it – yet few Roman officials were regularly to be found east of the Adriatic before the reign of Augustus, and few actual orders were issued to the friends of Rome. Power was the power to command, whether or not it was exercised. For most of the time it was visible only in the recognition that other peoples made of the hegemony exercised by the senate and people of Rome, by their attempts to ascertain or anticipate the Romans’ desires and to comply with them, by the referral of difficult issues to the Senate or its representatives. Roman power, in other words, was reflected in a new context for the exercise of local power, in a set of additional constraints on and redirections of authority.

  • 7 Drinkwater 1978, Burnand 2005-2010.

8The expansion of Roman power in the west sometimes took similar forms, proceeding through the recruitment of local rulers (for instance in the Alps, in Britain, in north Africa and beyond the northern limits of Rome’s provinces). Kings persisted as Roman agents in some of these areas well into the first century CE. There were savage wars of conquest, yet in these cases too Roman power was often extended by others. Caesar’s campaigns in Gaul were carried out by legions largely raised in Italy, assisted by Gauls from both Cisalpina and from the Provincia Transalpina. During the Gallic War more and more allied tribes were drawn into the conflicts. When hostilities ceased, they left behind systems of indirect rule, in which former allies exercised control over the rest. This was a less eirenic redistribution of authority than occurred in the east. The death count in the eight years of the Gallic War probably came close to a million, entire populations were dispossessed and many were enslaved. So much bullion was removed that gold and silver coinages collapsed overnight. Many of Caesar’s allies immediately found themselves redeployed to triumviral battlefields. Those who returned from the wars had a lot of rebuilding to do, and the first generation were necessarily fierce Caesarians.7

  • 8 Brunt 1976, Braund 1982.
  • 9 Haynes 2013.
  • 10 Woolf 1997, Mattingly 2006.
  • 11 Mattingly 2004.

9The new order was not however the rule of the Roman people over barbarians. Most of those who exercised power in the new order, were at first non-Romans and later ‘new Romans’ recruited from the conquered. As they and their successors were enfranchised and incorporated into the military and political élites of the empire some would exercise power over ‘old Romans’. The key role of local élites in the conquest and government of the west has long been recognised,8 as has the contribution of non-Roman soldiers to both Republican and early imperial armies.9 Studies of cultural change now try to move beyond the opposition of Romans and Natives, and seek to describe more tangled webs of agency through which the new order, political and cultural, was created and sustained.10 It follows that there was no single Roman or provincial experience of Roman power.11 So much is agreed, yet the implications for Roman Power have not been spelled out. We can no longer describe Roman Power as the power that Roman citizens or the Roman people collectively exercised over non-Roman Others. We need to understand as pure ideology phraseology such as the opening of the Res Gestae:

  • 12 Rerum gestarum divi Augusti, quibus orbem terrarum imperio populi Romani subiecit, et impensarum q (...)

Below is a copy of the achievement of the deified Augustus, by which he made the world subject to the power of the Roman people and of the expenses he incurred for the state and people of Rome.12 (trans. Cooley 2009)

10What language like this mystifies is the rule of a militarized and politicised set of propertied classes over everyone else, and there were non-Romans and Romans among both the rulers and the ruled.

Power as capacity

11Historians often write about power as if it is a finite resource, a possession that may be gained, lost or shared in what is a zero-sum game. Both de Ste Croix and Hopkins wrote as if this were the case, Hopkins contending that emperors and their peers struggled for control of power, and de Ste Croix arguing that these conflicts were insignificant because they simply sought to redistribute power within the same broad class, a class which was in the meantime accumulating more and more of it at the expense of others.

12This shared assumption is different from the position that regards power as essentially the capacity to bring about a desired state of affairs, to impose one’s will over that of others. The centurion who in Matthew’s Gospel meets Christ at Capernaum and asks for his slave to be healed expresses this clearly when he declines Jesus’ offer to visit and says he simply seeks a word of command:

  • 13 Matthew 8.8-9: καὶ ἀποκριθεὶς ὁ ἑκατόνταρχος ἔφη· Κύριε, οὐκ εἰμὶ ἱκανὸς ἵνα μου ὑπὸ τὴν στέγην εἰ (...)

Lord, I am not worthy to have you come under my roof. Instead, just say the word and my servant will be healed. For I too am a man under authority, with soldiers under me. I say to this one, ‘Go’ and he goes, and to another ‘Come’ and he comes, and to my slave ‘Do this’ and he does it.13

13We sometimes distinguish this sort of power to control or command from a power to forbid or prevent, like the power of Roman tribunes to obstruct the action of magistrates by declaring intercedo and to stop the legislative action of the people by saying veto. Yet arguably these are two sides of the same coin, the power to command and the power to forbid both being modes by which one person imposes his or her will on another.

  • 14 Lukes 1974.

14This sort of view can be usefully elaborated. Steven Lukes distinguished a one-dimensional view of power which focused on the behaviour of individuals as they compete in decision-making to impose their will on the future; from a two-dimensional view which takes into account the way interests shape the way decisions are openly discussed; from a three-dimensional view (which he advocates) which takes into consideration the interests of the different groups involved, their latent conflicts and the ideological means through which these are obscured and mystified.14

15Lukes’ emphasis on contexts, relationships and complexes of interests and interested knowledge resembles in some respects Michel Foucault’s view of power as ubiquitous, and of systems of power as inseparable from regimes of knowledge. He applied this approach not only to disciplinary institutions such as the asylum and the school, but also to the power relations implicit in discourse and science and also bio power, the means by which individual bodies are subjugated and controlled. Roman Power reimagined in these terms would describe a set of relationships that encompass all the texts we read and all the images we look at, as well as all the social relations we study. In terms of the question of perspective what this means is that ancient writings cannot comment on power relations from outside them but are themselves produced by and producers of power. That includes Polybius’ narrative and analysis of Republican imperialism and Tacitus’ and Dio’s account of the relations of emperors, court and senate. This is not the place to explore the power relations sustained by these particular strands of discourse – all penned by powerful individuals highly conscious that they were ruled even as they themselves exercised power. But it is worth remembering there are no dispassionate or innocent voices.

  • 15 MacMullen 1966.
  • 16 Pippidi 1976, Giovannini 1987, Woolf 1993, Shaw 2000.
  • 17 Saller 1982.
  • 18 Price 1984, McCormick 1986, Wörrle 1988, Benoist 2005.

16This is not the place either to explore claimed incoherences within Foucault’s writings on the subject, nor the differences between his view and that of Lukes and other social theorists. Roman Antiquity is probably not the best place from which such critical projects might be launched. Yet collectively these debates help us move ours along. Social theory discourages us away from treating power as a scarce resource possessed and used by individuals. Power is a not a thing that may be shared or monopolised. Social theory directs our attention instead to the relations and structures and systems through which societies, cultures and polities are constituted. This does not mean all previous work needs to be discarded for the idea is anticipated in several studies. The idea that power was ubiquitous in imperial society fits quite well with some strands of work on resistance to Rome. A good example is Ramsey MacMullen’s Enemies of the Roman Order which set out – from very different theoretical foundations – to map not the exercise of Roman power but the points at which it encountered opposition or resistance.15 The examples he brought together stretched across the social spectrum from Caesar’s murderers and the senatorial opponents of first century tyrants to astrologers, bandits, prophets and rebels. Other surveys of resistance and opposition have produced equally heterogeneous catalogues.16 Equally Richard Saller’s argument that patron-client relations did not subvert the Roman order of the Principate but constituted it coheres well with a network approach to Roman power.17 The same applies to a range of studies exploring ways in which ceremonial reproduced and naturalized relations of power.18 This list could easily be extended.

Webs of power

17Members of the senatorial and equestrian orders living under the emperors exercised power and were themselves subject to it. Many occupied key positions around what are conventionally seen as centres of power viz. the city of Rome, the imperial court (wherever it happened to be at the time), and occasionally the camps. Perhaps we should say they occupied places near the centres of a number of webs of power, and exercised much of their influence through webs or networks in which they were themselves enmeshed or entangled.

  • 19 Mann 1986.
  • 20 Hall 1985, Gellner 1988.

18The advantage of thinking in terms of multiple webs of power is that it captures the variety of kinds of power that were exercised. Michael Mann suggested that all societies are structured by overlapping webs of four kinds of power, viz. political, ideological, economic and military. For Mann the different historical trajectories of otherwise similar societies are largely shaped by differences in the way these various kinds of social power are articulated with each other.19 From one society to another one form might be dominant, and one kind of power might reinforce another or oppose it. Historians of élites are well aware that in some societies the military leadership is drawn from socially and economically prominent groups (as was the case in the early Roman empires) but that in other times and places there was more of a separation. As late antiquity advanced divisions opened up between the political élite based around the court and the economic élite in the western senate, while Christianization brought other religious élites into focus. This sort of analysis has been a powerful tool for some comparative history.20

19Even in the Principate there are examples of tension between different kinds of power, despite the empire’s evident centralization and hierarchization. Imperial freedmen and aristocrats, for instance, depended on different sources of power, had different aims and succeeded through different strategies. Those tensions led to some incoherence and weakness in the system as a whole, but they also empowered emperors adept at playing off different groups against each other. If we apply Mann’s categories to Rome we might differentiate between the political power exercised by magistrates, legati, praefecti and procurators and by the senate as a whole; the economic power all members of the ordines necessarily exercised as a precondition of membership; the military power enjoyed by a small subset during the short periods of their career in which they commanded units of legionaries or auxiliaries; and the ideological power that sustained their eminence and naturalized the order in which they felt they had a central place.

20Power in one sphere was sometimes convertible into power in another. Notoriously there was no powerful class of financiers whose interests might be opposed to military or intellectual elites presumably because the political system made it easy to transfer economic into political capital, and because the wealthy had no recourse apart from politics to protect their interests in the absence of strong institutions like banks and independent courts. Even when there were gradations of dignity, as between the two aristocratic orders, there seems to have been no sharp rift between elites in outlook or interests. Some individuals or families had a better chance of winning high office or priesthoods for themselves, but some families included individuals pursuing different routes to power and alliances between members of different elites - aristocrats and imperial freedmen for instance, or governors and provincial notables - were made easy by patronage, local patriotism and common military service.

  • 21 Hopkins 1983 on different kinds of aristocrat. On status dissonance, Hopkins 1961, 1978.

21Yet there were areas of incoherence and tension. The influence exercised by those equestrians who held the great prefectures, or simply were amici Caesaris, shows that different forms of power were not completely reducible or convertible to each other. Some positions – in the inner circle of amici or at the head of an army – offered unusual risks and possibilities. Other senators were clearly content to remain wealthy, eminent, well born and perhaps more cultured than their peers without exercising much political influence. Hopkins referred to these as ‘the grand set’ as opposed to ‘the power set’ and elsewhere exploited the notion of status dissonance in his discussions of social mobility.21 These incoherences between different networks of power were manifested in social mobility, in occasional political crises like those of the reign of Claudius, and in a complex discourse in which ethics of withdrawal, engagement, resistance and idealism were rehearsed.

22Let us return to Mann’s four categories of social power. Military power followed from political office and was in practice limited. Only in rare instances would armies of the first two centuries CE follow their commanders into civil war. There are no instances when a Roman army marched on an established emperor before the third century CE. Political power involved exercising influence over several constituencies, commanding respect from fellow senators, influence over emperors and making alliances with other power brokers inside and outside the senatorial aristocracy. Patronage captures some of this activity but not all of it. A capacity to build fama and amicitiae covers more. Practical rhetorical skills clearly remained important whether directed at emperors, fellow aristocrats, or military and civic audiences. Economic power was necessary for all, but far more important for some than others.

  • 22 Benoist 2005, Price 1987.
  • 23 Várhelyi 2010, Martin 1985.
  • 24 Garnsey 1970.

23Ideological power is more complex. The essential conservatism shared by all members of the ruling class and many citizens beyond them sustained the eminence of the ordines themselves. This conservatism protected their position but perhaps it also inhibited their capacity to find new ways of acting. As far as we can tell no aristocrat ever had an ambition greater than to succeed in pursuing traditional goals (or to replace the current emperor). Fundamental to this was an even wider belief in a hierarchical order, one in which individuals and families were ranked by dignitas. This order was represented in and reproduced through many media, from the arrangement of seating in the theatre to the costumes and order of precedence at imperial consecrationes.22 The collective dignity of the senate, and the individual worth of its most distinguished members, was particularly visible in the many public rituals at which senatorial priests presided throughout the year.23 Their presence on the Ara Pacis shows how early this became a means of presenting the distinctive sacral authority of the ordo. When meeting as the senate and probably at other times, senators were intensely aware of the sequence in which they might speak. Further means of ranking: patricians stood above plebeians, consulars before those who had held the praetorship, older before younger members and so on. The visibility of the emperors and senators on public occasions reinforced a public knowledge of these distinctions and stabilized them. The efficacy of this ideological structure of power – Lukes’ third dimension of power – is illustrated by the fact it seemed natural to jurists and others to treat witnesses and criminals differently in Roman courts depending on their legal status.24

  • 25 For different versions of this claim: Crone 1989, Yoffee 2005, Ando–Richardson 2017 and perhaps al (...)
  • 26 Talbert 1984.
  • 27 Matthews 1975.

24Naturally there were shifts over time. It is widely believed that over the first three or four centuries CE the relations of power shifted in ways that benefited the emperors and the propertied classes. Emperors and other rulers faced severe limits on their capacity to impose their will on others in all periods. Preindustrial monarchies have in general often been characterised as weak, successful only in achieving very limited goals, sustained largely by their interdependence with other institutions.25 In the case of Rome these other institutions included the city, the family, and the propertied classes themselves. Yet in late antiquity the Roman state became a little stronger and more authoritarian. The richest became richer, ceremonial became more and more elaborate. Conversely the political and military influence of senators became weaker from the late second century on. The senate itself de facto lost influence in many areas to the imperial courts – often far from Rome for years on end.26 Edicts replaced senatusconsulta as sources of law, embassies attended the senate less frequently, praetorian prefects and their staffs increased their influence (but how would we measure it for certain?). Relations between senate and court became periodically more fraught.27 Yet throughout this period of political weakening, the ideological power of the senatorial elite remained untouched, so that courtiers continued to seek senatorial titles throughout late antiquity. Mann’s analysis offers us a way to theorize these changes in terms of changes in the relative strength of different kinds of power, and shifts in location of the centres of the different webs of power.

Where the webs grow thinner

25Thinking about Roman power in terms of relationships and networks, as strands in a cluster of not quite coincident webs, is a powerful tool for reimagining the imperial centre or centres. It is much superior to accounts based on the idea of power as a finite resource and evades disputes about whose power struggles mattered and whose were trivial. Does it also offer advantages for thinking about areas more marginal to the competing imperial projects of the political, economic, military and social élites? To the provinces in other words.

26Some linkages are obvious and have already been mentioned. Relations of patronage are a case in point both long lasting civic patronage blurring into euergetism and the short term contingent alliances between governors and local notables. Military power extended all the way down from senatorial generals mostly biding their time but occasionally thrust into the limelight of history to humbler figures like the centurion at Capernaum.

27There are some additional advantages in thinking of power in the provinces in terms of webs of influence. One concerns the prose literature we still consult for Roman views of provincials and vice versa. Because our historical sources are themselves products of struggles for legitimacy and influence, we should read texts as much for their ideological slant than as source of information about the realities of provincial government. This can be a rather revealing process. For example in the body of texts that describe episodes in provincial rule from the perspective of the governors (texts such as Tacitus’ Agricola, Pliny’s tenth book of letters, Dio on the Boudicaan revolt and so on) there is very little sign that members of the political-cum-military élite ever produced sophisticated analyses of the imperial system that might be compared in subtlety to the reflections on the city-state produced by Greek writers centuries before. Perhaps the same ideological forces that ensured the continuing social eminence of the ancient ordines despite their political decline, had the effect of shoring up traditional understandings of the social order.

  • 28 Swain 1996, Whitmarsh 2001, Jones 1971.
  • 29 Lavan 2013.
  • 30 Steel 2001, Ferrary 1988, Richardson 2008. Some attempts at a synthesis in Woolf 2012.

28Something similar applies to provincial views of the centre of power. Neither Dio Chrysostom’s speeches peri basileias nor Aristides’ panegyrics compare with Machiavelli’s Il Principe as an analysis of imperial autocracy. Greek literature offers many views, some explicit some implicit, of the Roman Empire but they are mainly cast in terms of ethical rather than political theory.28 One reason for this, I suggest, may be that the elite continued to think and write about power mostly in terms of webs of personal relations. Myles Lavan has shown how a language of empire was developed in part by extending the metaphorical range of terms relating to domestic servitude.29 If so then their account of Roman power in action actually coheres quite well with the ways social theory encourages us to describe it. Ethics provide rules of play for these structured interactions, a guide to how to patronize and how to seek support. This sort of perspective offers some advantages over the sometimes anachronistic use of territorial notions of province and empire criticised in recent studies of the language of empire. 30 It also closes the gap between the behaviour of members of the empire’s élites when in the provinces, and their behaviour in Italy and Rome. There is not much sign that in the imperial period that gap was felt, in the way it was in nineteenth century empires that often designated the colonies and dominions as spaces in which quite different moral rules applied.

  • 31 Edwards 1993, 1997.
  • 32 Wallace-Hadrill 1983, Roller 2001.

29Roman historians and satirists turned their moralizing gaze on emperors and senators in much the same way irrespective of where they were exercising their power.31 The conduct of individual emperors was a source of endless fascination in all genres whatever stage they performed on: Nero in Rome and Nero in Naples were the same as Nero in Greece.32 No doubt to ancients an approach to empire focused on conduct and relationships seemed pragmatic. Arguably more ethical and governmental problems have been created in more recent times by the notion that India, the Congo or Tasmania were lands in which different moralities applied.

  • 33 For some recent work on the identity politics of the early imperial period see Swain 1996, Woolf 1 (...)

30If we reject territorial descriptions of Roman Power as largely anachronistic, and if we recognise the language of the Res Gestae as largely ideological how should we approach the exercise of power where the imperial webs grew thin? One starting point would be to see imperial systems as means of controlling people (more than land or territories) by binding them into networks of obligation and compliance. We need to avoid the ideological trap set by an imperial rhetoric that divides the world into Romans and Others. What the imperial state attempted to do was to mobilize human labour – as soldiers farmers, traders, slaves and so on – to achieve its objectives. And the state was an instrument employed by the propertied classes. During the first century CE, many wealthy foreigners were converted into Romans of course, and in some varieties of Latin community it probably was the case for a few generations that (new) Romans ruled over (old) Aliens. Enfranchisement rewarded those who had thrown their lot in with the imperial project. It also converted citizenship from a mark of descent into a sign of privilege, shaping ideas of identity and obligation in the process. The attention that many subject peoples – or at least their elites – gave to fashioning and celebrating local identities becomes easier to understand as a result.33 New understandings of the empire, of its place in the world and in history, and of the place within the empire of particular cities, families and individuals all formed part of a new imperial order of knowledge. That new knowledge was the Foucauldian accompaniment of Roman Power, created and sustained by relations of power, made plausible by their operation, and in turn encouraging imperial-minded conduct and a respect for the universal legitimacy of Roman Power.


Ando – Richardson 2017 = C. Ando, S. Richardson (ed.), Ancient States and Infrastructural Power: Europe, Asia, and America, Empire and After, Philadelphia, 2017.

Benoist 2005 = S. Benoist, Rome, le prince et la cité : pouvoir impérial et cérémonies publiques (Ier siècle av.-début du IVe siècle apr. J.-C.), Paris, 2005.

Braund 1982 = D. Braund, Rome and the friendly king: The character of the client kingship, London, 1982.

Brunt 1976 = P.A. Brunt, The Romanization of the local ruling classes in the Roman empire, in D.M. Pippidi (ed.), Assimilation et résistance à la culture gréco-romaine dans le monde ancien. Travaux du VIe Congrès international d'études classiques (Madrid, septembre 1974), Bucharest-Paris, 1976, p. 161-173.

Burnand 2005-2010 = Y. Burnand, Primores Galliarum. Sénateurs et chevaliers romains originaires de Gaule de la fin de la République au IIIe siècle, 4 vols., Brussels, 2005-2010.

Cooley 2009 = A.E. Cooley, Res Gestae Divi Augusti. Text, Translation, and Commentary, Cambridge, 2009.

Crone 1989 = P. Crone, Pre-industrial Societies. New Perspectives on the Past, Oxford, 1989.

de Ste. Croix 1981 = G.E.M. de Ste. Croix, The Class Struggle in the Ancient Greek World. From the Archaic Age to the Arab Conquests, London, 1981.

Derks – Roymans 2009 = T. Derks, N. Roymans (ed.), Ethnic Constructs in Antiquity: The Role of Power and Tradition, Amsterdam, 2009 (Amsterdam Archaeological Studies).

Drinkwater 1978 = J.F. Drinkwater, The Rise and Fall of the Gallic Julii, in Latomus, 37, 1978, p. 817-850.

Edwards 1993 = C. Edwards, The Politics of Immorality in Ancient Rome, Cambridge, 1993.

Edwards 1997 = C. Edwards, Self-scrutiny and self-transformation in Seneca's Letters, in Greece & Rome, 44-1, 1997, p. 23-38.

Ferrary 1988 = J.-L. Ferrary, Philhellénisme et impérialisme : aspects idéologiques de la conquête romaine du monde hellénistique, de la Seconde Guerre de Macédoine à la Guerre contre Mithridate, Rome, 1988 (BEFAR, 271) (réimpr. Rome, 2014).

Garnsey 1970 = P. Garnsey, Social Status and Legal Privilege in the Roman Empire, Oxford, 1970.

Gellner 1988 = E. Gellner, Plough, sword and book: The structure of human history, London, 1988.

Giovannini 1987 = A. Giovannini (ed.), Oppositions et résistances à l'empire d'Auguste à Trajan, Vandœuvres-Geneva, 1987 (Entretiens sur l'Antiquité classique, 33).

Goldhill 2001 = S. Goldhill (ed.), Being Greek under Rome: Cultural Identity, the Second Sophistic and the Development of Empire, Cambridge, 2001.

Gruen 2011 = E. Gruen (ed.), Cultural Identity in the Ancient Mediterranean: Issues and Debates, Los Angeles, 2011.

Habinek 1998 = T. Habinek (ed.), The Politics of Latin Literature: Writing, Identity and Empire in Ancient Rome, Cambridge, 1998.

Hall 1985 = J. A. Hall, Powers and Liberties: The Causes and Consequences of the Rise of the West, Oxford, 1985.

Harris 2016 = W. V. Harris, Roman Power. A Thousand Years of Empire, Cambridge, 2016.

Haynes 2013 = I. Haynes, Blood of the Provinces: The Roman Auxilia and the Making of Provincial Society from Augustus to Diocletian, Oxford, 2013.

Hopkins 1961 = K. Hopkins, Social Mobility in the Later Roman Empire: The Evidence of Ausonius, in Classical Quarterly, 11, 1961, p. 239-249.

Hopkins 1978 = K. Hopkins, Conquerors and Slaves: Sociological Studies in Roman History I, Cambridge, 1978.

Hopkins 1983 = K. Hopkins, Death and Renewal: Sociological Studies in Roman History II, Cambridge, 1983.

Janniard – Triana 2006 = S. Janniard, G. Traina, Sous le concept de « romanisation ». Paradigmes historiographiques et perspectives de recherche, in MEFRA, 118-1, 2006, p. 71-166.

Jiménez Díez 2008 = A. Jiménez Díez, Imagines hibridae. Una aproximación postcolonialista al estudio de las necrópolis de la Bética, Madrid, 2008 (Anejos de Archivo Español de Archeologia).

Jones 1971 = C.P. Jones, Plutarch and Rome, Oxford, 1971.

Jones 1999 = C.P. Jones, Kinship Diplomacy in the Ancient World, London, 1999.

Lavan 2013 = M. Lavan, Slaves to Rome: Paradigms of Empire in Roman Culture, Cambridge, 2013.

Lukes 1974 = S. Lukes, Power: A Radical View, London, 1974.

MacMullen 1966 = R. MacMullen, Enemies of the Roman Order, Cambridge, Mass., 1966.

Mann 1986 = M. Mann, The Sources of Social Power. I. A History of Power from the Beginning to A.D. 1760, Cambridge, 1986.

Martin 1985 = S. Martin, The Imperial Senate: Images of Power, in Journal of Roman Studies, 75, 1985, p. 222-228.

Matthews 1975 = J.F. Matthews, Western Aristocracies and Imperial Court, A.D. 364-425, Oxford, 1975.

Mattingly 2004 = D. Mattingly, Being Roman: Expressing Identity in a Provincial Setting, in Journal of Roman Archaeology, 17, 2004, p. 5-25.

Mattingly 2006 = D. Mattingly, An Imperial Possession: Britain in the Roman Empire 54 BC-AD 409, London, 2006.

Mattingly 2011 = D. Mattingly, Imperialism, Power and Identity: Experiencing the Roman Empire, Princeton, N.J.-Oxford, 2011 (Miriam S. Balmuth Lectures in Ancient History and Archaeology).

McCormick 1986 = M. McCormick, Eternal victory: Triumphal rulership in late antiquity, Byzantium and the early medieval west, Cambridge, 1986 (Past and Present Publications).

Millar 1965 = F. Millar, Epictetus and the Imperial Court, in Journal of Roman Studies, 55, 1965, p. 141-148.

Millar 1977 = F. Millar, The Emperor in the Roman World, London, 1977.

Pippidi 1976 = D.M. Pippidi (ed.), Assimilation et résistance à la culture gréco-romaine dans le monde ancien. Travaux du VIe Congrès international d'études classiques (Madrid, septembre 1974), Bucharest-Paris, 1976.

Price 1984 = S. Price, Rituals and Power: The Roman Imperial Cult in Roman Asia Minor, Cambridge, 1984.

Price 1987 = S. Price, From Noble Funerals to Divine Cult: The Consecration of Roman Emperors, in D. Cannadine, S. Price (ed.), Rituals of Royalty: Power and Ceremonial in Traditional Societies, Cambridge, 1987, p. 56-105.

Richardson 2008 = J.S. Richardson, The Language of Empire: Rome and the Idea of Empire from the Third Century BC to the Second Century AD, Cambridge, 2008.

Roller 2001 = M.B. Roller, Constructing Autocracy: Aristocrats and Emperors in Julio-Claudian Rome, Princeton, N.J., 2001.

Saller 1982 = R.P. Saller, Personal Patronage under the Early Empire, Cambridge, 1982.

Shaw 1984 = B.D. Shaw, The Anatomy of the Vampire Bat, in Economy and Society, 13-2, 1984, p. 208-249.

Shaw 2000 = B.D. Shaw, Rebels and Outsiders, in A. Bowman, P.  Garnsey, D. Rathbone (ed.), Cambridge Ancient History, IX, The High Empire, A.D. 70-192, Cambridge, 2000, p. 361-403.

Steel 2001 = C.E.W. Steel, Cicero, Rhetoric and Empire, Oxford, 2001.

Swain 1996 = S. Swain, Hellenism and Empire: Language, Classicism and Power in the Greek World, AD 50-250, Oxford-New York, 1996.

Syme 1958 = R. Syme, Tacitus, 2 vols., Oxford, 1958.

Talbert 1984 = R.J.A. Talbert, The Senate of Imperial Rome, Princeton, N.J., 1984.

Várhelyi 2010 = Z. Várhelyi, The Religion of Senators in the Roman Empire: Power and the Beyond, Cambridge, 2010.

Wallace-Hadrill 1983 = A. Wallace-Hadrill, Suetonius: The Scholar and his Caesars, London, 1983 (Classical Life and Letters).

Whitmarsh 2001 = T. Whitmarsh, Greek Literature and the Roman Empire: The Politics of Imitation, Oxford-New York, 2001.

Whitmarsh 2010 = T. Whitmarsh (ed.), Local Knowledge and Microidentities in the Imperial Greek World, Cambridge, 2010 (Greek Culture in the Roman World).

Woolf 1993 = G. Woolf, Roman Peace, in J. Rich, G. Shipley (ed.), War and Society in the Roman World, London, 1993, p. 171-194.

Woolf 1994 = G. Woolf, Becoming Roman, Staying Greek. Culture, Identity and the Civilizing Process in the Roman East, in Proceedings of the Cambridge Philological Society, 40, 1994, p. 116-143.

Woolf 1997 = G. Woolf, Beyond Romans and Natives, in World Archaeology, 28-3, 1997, p. 339-350.

Woolf 1998 = G. Woolf, Becoming Roman: The Origins of Provincial Civilization in Gaul, Cambridge, 1998.

Woolf 2012 = G. Woolf, Rome: An Empire's Story, New York, 2012.

Wörrle 1988 = M. Wörrle, Stadt und Fest im kaiserzeitlichen Kleinasien. Studien zu einer agonistischen Stiftung aus Oinoanda, Munich, 1988 (Vestigia, Beiträge zur alten Geschichte, 39).

Yoffee 2005 = N. Yoffee, Myths of the Archaic State: Evolution of the Earliest Cities, States, and Civilizations, Cambridge-New York, 2005.


1 De Ste Croix 1981, p. 380-381.

2 De Ste Croix 1981, p. 503, giving the title to Brent Shaw’s brilliant review: Shaw 1984.

3 This chapter was written before the appearance of Harris 2016 which pursues an analysis of the relation between internal security and external violence in the longue durée.

4 Excellent studies in this vein include Syme 1958, Millar 1965, Wallace-Hadrill 1983, Roller 2001.

5 Steel 2001, Richardson 2008.

6 Braund 1982.

7 Drinkwater 1978, Burnand 2005-2010.

8 Brunt 1976, Braund 1982.

9 Haynes 2013.

10 Woolf 1997, Mattingly 2006.

11 Mattingly 2004.

12 Rerum gestarum divi Augusti, quibus orbem terrarum imperio populi Romani subiecit, et impensarum quas in rem publicam populumque Romanum fecit, incisarum in duabus aheneis pilis, quae sunt Romae positae, exemplar subiectum. Cooley 2009, p. 58.

13 Matthew 8.8-9: καὶ ἀποκριθεὶς ὁ ἑκατόνταρχος ἔφη· Κύριε, οὐκ εἰμὶ ἱκανὸς ἵνα μου ὑπὸ τὴν στέγην εἰσέλθῃς· ἀλλὰ μόνον εἰπὲ λόγῳ, καὶ ἰαθήσεται ὁ παῖς μου·9καὶ γὰρ ἐγὼ ἄνθρωπός εἰμι ὑπὸἐξουσίαν, ἔχων ὑπ’ ἐμαυτὸν στρατιώτας, καὶ λέγω τούτῳ· Πορεύθητι, καὶ πορεύεται, καὶ ἄλλῳ· Ἔρχου, καὶ ἔρχεται, καὶ τῷ δούλῳ μου· Ποίησον τοῦτο, καὶ ποιεῖ.

14 Lukes 1974.

15 MacMullen 1966.

16 Pippidi 1976, Giovannini 1987, Woolf 1993, Shaw 2000.

17 Saller 1982.

18 Price 1984, McCormick 1986, Wörrle 1988, Benoist 2005.

19 Mann 1986.

20 Hall 1985, Gellner 1988.

21 Hopkins 1983 on different kinds of aristocrat. On status dissonance, Hopkins 1961, 1978.

22 Benoist 2005, Price 1987.

23 Várhelyi 2010, Martin 1985.

24 Garnsey 1970.

25 For different versions of this claim: Crone 1989, Yoffee 2005, Ando–Richardson 2017 and perhaps also Millar 1977.

26 Talbert 1984.

27 Matthews 1975.

28 Swain 1996, Whitmarsh 2001, Jones 1971.

29 Lavan 2013.

30 Steel 2001, Ferrary 1988, Richardson 2008. Some attempts at a synthesis in Woolf 2012.

31 Edwards 1993, 1997.

32 Wallace-Hadrill 1983, Roller 2001.

33 For some recent work on the identity politics of the early imperial period see Swain 1996, Woolf 1994, 1998, Whitmarsh 2001, 2010, Mattingly 2011, Jones 1999, Jiménez Díez 2008, Janniard–Traina 2006, Habinek 1998, Gruen 2011, Goldhill 2001, Derks–Roymans 2009.


Institute of Classical Studies, London,

© Publications de l’École française de Rome, 2020

Licence OpenEdition Books

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search