How was power exercised in the Later Roman Empire (c. ad 300–600)? Answers to this question usually highlight what is, besides Christianization, the most obvious change of the period: the centralization of government, visible in the expansion of the imperial bureaucracy and, so it is presumed, the increasingly distant and hieratic nature of imperial authority. Bureaucrats and emperors are familiar figures in accounts of the period, and related themes such as legislation, ceremony, and social class are high on the scholarly agenda. In this book, I approach the question from a different angle and start out from the smallest unit of social life, namely social interactions (i.e. exchanges between two or more individuals). By closely studying four such interactions (petitions, parrhesia, intercession, and riots), I seek to render explicit the norms that govern them. In turn, these norms reveal three central features of Late Antique society: personal interaction as the privileged locus of power; the hierarchical nature of society; and justice as its ultimate aim and foundation. Thus, I argue, the petitioner accosting a governor in public should be as central to our understanding of what power was in the Later Roman Empire as is currently the handling of petitions by the bureaus of the imperial administration.

In this opening chapter, I offer an outline of my methodology and how it helps to identify personal interaction as central to Late Antique society, and I locate the notions of hierarchy and justice within Late Antique thought. I start with a brief sketch of the scholarly background against which this book positions itself.

Beyond Bureaucracy

The question of how power was exercised in Late Antiquity usually is answered with reference to the emperor and to bureaucracy and legislation as his manifestations. The two main traditions in recent scholarship on Late Antiquity, the institutionalist approach of Arnold H. M. Jones and the cultural history of Peter Brown, tend to concur in viewing both emperor and bureaucracy as central to any understanding of the Later Roman Empire. “The Later Roman Empire was before all things a bureaucratic state,” Jones wrote in his masterly study of the period, whilst “both in theory and in the practice of the constitution, the emperor’s powers were absolute.”Footnote ¹ Whilst the inspiration that Brown drew from social anthropology sometimes has clashed with the constitutionalist history of Jones, the opening lines of Power and Persuasion in Late Antiquity share the focus on the emperor and his bureaucracy: “[This book] attempts to describe the expectations with which upper-class subjects approached the emperor and his representatives, to ward off the cruelty and fiscal rigor that characterized the government of the Roman Empire in the fourth and fifth centuries a.d.”Footnote ² The bureaucratic beast and the capricious emperor, both capable of bringing large-scale violence, seem to have kept the elite awake at night. As put by Brown in a memorable overstatement, “a tide of horror lapped close to the feet of all educated persons.”Footnote ³ There are good reasons for this focus, as the transition from “soft” to “hard” government, with greater central control through the bureaucracy, is indeed one of the characteristic transformations of Late Antiquity.Footnote ⁴ The change seems uniquely visible through the two major collections of legal texts by Theodosius II (408–50) and Justinian (527–65), the Codex Theodosianus and the Codex Justinianus.

As the Roman Empire was a monarchy, the emperor unavoidably will loom large in any account of this period, but his depiction as all-powerful and potentially dangerous owes much to the influence of older scholarship, including Gibbon and Mommsen, and it already was substantially nuanced by Brown himself in Power and Persuasion. Absolute power did not need to contradict the rule of law materialized in legislation and courts, whilst the expanse of the Empire and slowness of communication made it hard to fully steer what imperial representatives did in the provinces. Further, as Brown argues, the elite deployed a discourse of paideia that subjected the emperor to its expectations of civility. More recently, Christopher Kelly has shown brilliantly that bureaucracy and emperor did not always sing from the same hymn sheet, whilst work on the legal evidence, inspired by legal anthropology, has warned against overestimating the impact of imperial law to the detriment of local rules and customs.Footnote ⁵ In German-speaking scholarship, the notion of acceptance (Akzeptanz) has acquired great currency. The concept emphasizes that the emperor did not acquire legitimacy by being appointed by a particular procedure, but that he had to be accepted as such by various status groups (army, people, elite, clergy).Footnote ⁶ As acceptance could be withdrawn, the emperor constantly had to keep these groups happy to avoid being challenged by a usurper. More recently, scholarship has underscored the socioeconomic transformations of the Later Roman Empire, and the entanglement of emperor, imperial administration, and aristocracy caused by the introduction of a new gold coin, the solidus, which was uniquely accessible to the new aristocracy of service created by Constantine the Great. Preserving the tax basis was a crucial concern for any emperor of the Later Roman Empire, not least in view of military expenses, and its disruption a crucial factor in the loss of control over the West.Footnote ⁷ Taking all of these nuances into account, scholars would agree that the institutional and associated economic transformations of the half-century under Diocletian (284–305) and Constantine the Great (306–37) fundamentally reshaped the way Late Antique society functioned.Footnote ⁸

Whilst the focus on emperor, bureaucracy, taxation, and legislation is justified given the nature of the Later Roman Empire and our evidence for it, it is not the only way to approach a society. Ultimately, a society is made up of numerous social interactions between individuals, and it is through such interactions that political decisions are enacted or resisted. Such a change of perspective is, interestingly, in tune with how Late Antique observers represent their own society. As noted by Peter Brown, “to the ancients, how individuals acted mattered far more than did the structures within they acted.”Footnote ⁹ Institutions, which feature prominently in modern analyses, rarely take center stage in our sources. Rather, Late Antique writers assess more often how individuals acted and interacted. Even at the heart of a key source for the administrative workings of the Empire, namely the Theodosian Code, we find personal interaction: 95 percent of all constitutions in the Codex Theodosianus are letters, often to praetorian prefects. Edicts – that is, general pronouncements directed to everyone – were a rare form of legislation.Footnote ¹⁰ It seems, then, worthwhile to see what can be gained by starting our analysis of the Later Roman Empire from its most basic unit: personal interaction.

From this shift of focus follows an emphasis on individual agency. The approaches just outlined are largely indebted to what in sociological terms can be called objectivism, that is, a focus on social structure to the detriment of subjective agency. Although there are scholars who successfully negotiate this gap, not least Peter Brown, interpretations of the Later Roman Empire still often focus on structures (e.g., bureaucracy) and social groups (e.g., the senate) and hence tend to privilege certain categories of evidence, in particular legislation.Footnote ¹¹ Functionalist explanations, in turn, describe social phenomena with reference to the role they play in ensuring the long-term stability of society as a whole and hence pay little attention to individual actors and their subjective discourse.Footnote ¹² Discourses on power then appear as directed from the center and instrumentalized for ensuring legitimacy for that center.Footnote ¹³ Further, acceptance theory has focused on how stereotypical actions established channels of communication between social groups and is in this respect explicitly indebted to the French sociologist Pierre Bourdieu. Yet his notion of habitus refers to how actors have internalized social structure, and it is understood to determine what individuals do on particular occasions.Footnote ¹⁴ Such objectivist approaches have yielded important insights into the functioning of the Later Roman Empire. Nevertheless, when explaining action, it leaves little room for the intentions of the actors, intersubjective meanings, and the space individuals carve out for themselves in society. I shall, therefore, zoom in on the room for individual agency and, importantly, argue that the exploitation of that room shaped power relations between the participants.

The proposed shift toward social interaction and individual agency invites us to read the sources differently. Some scholarly explanations of ancient social phenomena, based on modern theories, are separated from ancient discourses, which are classed as ideology and have at best a justificatory function for “real” power relations. In Richard Finn’s seminal overview of Late Antique almsgiving, for example, thirty pages separate his discussion of the social function of almsgiving, interpreted in the light of modern sociological assumptions, from his account of the Late Antique ecclesiastical discourse on, and “ideology” of, almsgiving.Footnote ¹⁵ But if the “ideology” of almsgiving was the dominant discourse in Late Antiquity on the topic, shouldn’t we expect it to have at least some relation to the intentions of the agents and their practices? This points to a larger methodological issue. Following objectivist assumptions, historians sometimes presume they can separate descriptions of social interactions in the sources from the normative expectations attached to these interactions. The former then provide the material with which the historian builds his picture of Late Antique society, and the latter is analyzed as discourse or ideology.Footnote ¹⁶ Yet can we really separate both? For, without reference to intentions and normative expectations, a description risks being misunderstood: When my son hits his brother, he is doing something, but in order to know what this something is (playing pirates, having a row, saving him from a wasp), we have to refer to rules (of playing pirates, of settling a disagreement, of caring for one’s sibling) and intentions. A similar, Late Antique example is given by the fourth-century African bishop Evodius of Uzalis, who, in a discussion of the role of the will in sinning, points out that there is a difference between someone who goes to prison as a punishment for some crime and someone who goes freely into prison to visit someone else. It is the intention that creates the difference between two otherwise-identical actions.Footnote ¹⁷ Few observers would confuse the two actions, yet the example shows well that one cannot simply separate intentions, which evoke certain rules, from actions and treat the former as subjective and the latter as objective. Without a complete understanding of the intentions and moral expectations that underpin descriptions of actions in the sources, we cannot fully understand these actions.Footnote ¹⁸ Norms and intentions are integral to understanding social phenomena, and we should read our sources as reflecting that intertwining of fact and value.

Methodology

To operate the shift toward social interaction, individual agency, and the integration of fact and value, I base myself on three interlocking concepts: social role, practice, and virtue. The first two are modern in nature, whilst the ancient notion of virtue helps to show how they can be made to apply to Late Antiquity. All three foreground normative expectations regarding social interaction and thus delineate room for individual agency within the context of social structure.

Social roles have been defined as “a set of normative expectations attached to a social position.”Footnote ¹⁹ Indeed, actions (reading a book with my son) refer to roles (being a father), which are shaped by a set of shared rules (reading is good for children). The notion of rule does not mean that there is an explicit set of prescriptions for each social role, which would allow the role to be performed by simply following them. Rather, it means that actors in a particular society can identify what right and wrong ways are of performing this role, without necessarily being able to explicate in advance what a role entails in every given circumstance.Footnote ²⁰ One can, hence, be a good or a bad father, or a good or a bad professor, which are not fixed identities but have to be actualized in each instance: One can generally be a good son and behave badly on occasion. Moreover, circumstances change, meaning that no instance of the performance of one’s role is identical. Socialization, enhanced by suitability for performing a role, helps us to acquire a sense of what a role entails at a particular moment. Practical reasoning and judgment are needed to perform it well in every instance, including in exceptional situations. These qualities are also required because of a further complicating factor. Individuals always assume more than one role: I am a son, a father, a professor, and so on. These roles can conflict or can be seen to do so, for example, if my son were to enroll as a student of mine.

This brief account allows us to draw out two important features of social life that social role theory draws attention to. Social roles need to be rendered concrete in specific social interactions. Most interactions are codified to a greater or lesser degree, with certain ones, such as public ceremonies, being explicitly scripted, whilst others – think of family dinners – are only implicitly so. Many interactions hence obey a broad regularity, but they cannot be described in a mechanistic way that excludes the individual and the context from the analysis. Indeed, as the maxim “crisis reveals character” implies, exceptional circumstances and hence irregularities in how a role is to be performed may reveal whether one is suitable for that role. In other words, we can abstract a social role from its many situational instantiations, but we cannot reduce the latter to the former.

A similar observation applies to the relationship between, on the one hand, descriptions of social interactions in which a role is performed and, on the other, the normative expectations attached to these interactions. Rules are intersubjective, but we encounter them in accounts that are the subjective interpretations of observers of the social interaction. Such accounts are often shaped by personal interests, intentions, and emotions. Rules can, thus, be formulated, but they remain abstractions from concrete social practice. If, as Anthony Giddens formulated it, “a discursive formulation of a rule is already an interpretation of it,”Footnote ²¹ historians can only seek to abstract normative expectations from a wide array of accounts of social interactions. In fact, every statement about what a good emperor is, to take an example from Antiquity, implies an interpretation of, and engagement with, the normative expectations of that particular social role. The precise expectations may be dependent on the position of the observer: What an emperor is expected to be for a soldier may well be different from what a farmer expects. Yet, however partisan one’s judgment is of the emperor, it will still be shaped by intersubjective expectations of what an emperor should be.

If social role theory focuses on the agent and their social role, practice theory zooms in on what is actually done. Basically, practices are the ways in which we relate to the world around us, and in and through which the social system is constituted. The term practice, rather than action, shifts attention away from the conscious agency of the individual and draws attention to the fact that my interaction with the world, as in the example of being a father, is shaped by my body, rules, language, and understanding. Whilst I preserve my agency, it is not just my agency that counts, as the focus on practices invites us to think about the impact of, for example, space on what I do. As argued by Theodore Schatzki, practices are organized by understandings, rules, and teleoaffective structures. Understandings can include both practical reasoning (knowing how to do something) and general understanding (the shared meanings attributed to a practice); rules are the normative expectations about what a good way of acting is, whilst teleoaffective structures refer to intentions and emotions.Footnote ²² By understanding practice in this way, Schatzki likewise reacts against the tendency visible in objectivist strands of sociology to separate the analysis of actions from the meanings attributed to them by the actors. The latter then tend to be understood as opinions and beliefs, studied as psychological states or ideology, but not given any real significance in explaining actions. Given the fact that they are coconstituted by social interaction and by intersubjective meanings, practices reflect the wider social system. As argued by Anthony Giddens, whose sociology has affinities with practice theory, interactions reproduce and shape the system, and hence the social system can be grasped in these interactions.Footnote ²³ Close attention to episodes thus is a method to grasp the features of the social system, in particular a focus on episodes of similar practices that reveal wider features

The emphasis on practices as a basic unit of social life helps to bring into focus something that remains implicit in the concept of social roles as I have set out because of this concept’s focus on one individual agent who is acting: A particular practice, initiated by someone else, may draw a person into a type of social interaction and activate a social role. For example, in the next chapter we will see how an individual could, by doing and saying certain things to a superior, put himself in the position of a petitioner, even on occasions that were not allocated to receiving petitions on the part of the superior. Petitioning can, then, be understood as a practice that awakens normative expectations for both sides in the exchange and thus has both inferior and superior step into particular roles. The four case studies at the heart of this book relate to such practices, in which we can see social roles being activated.

A central notion in ancient ethics, virtue is absent in modern accounts of social role and practice. Nevertheless, its ancient understanding ties in well with these modern theories. However diverse, ancient ethics generally can be labeled eudaimonistic, identifying happiness (eudaimonia) as the highest good in a person’s life.Footnote ²⁴ Happiness is not to be understood in a modern utilitarian way as the maximization of gratifying experiences, nor deontologically as the result of the fulfillment of moral precepts; nor is it a momentary good feeling. It is the choice of a way of life that is most appropriate to oneself as a human being, as part of a society and as an individual. Although ancient authors commonly do not use the notion of a social role, the idea of a way of life comes close. A way of life obviously means to act, which, according to ancient thought actions, is guided by virtues. Conversely, virtues are expressed in actions, and one can hence judge someone’s character by looking at their conduct. Virtues are, then, dispositions to act correctly in a given case, whatever its particularities.

When applying social role theory and practice theory to Late Antique sources, it is thus good to know that these will employ a language of virtue in discussing social interactions. Yet ancient ethics also offers us three insights that are an important supplement to my modern concepts, namely the expectation that one matches one’s natural capacities with one’s social role; the supposition that the performance of one’s role offers a window on that person’s moral character; and a strong awareness of the situational nature of actions and judgments on actions. To grasp this better, I briefly turn to two Roman treatises entitled De officiis, one composed by Cicero in 44 bc and the other by Ambrose of Milan in the 380s ad. Cicero’s work is based on the work of the Stoic Panaetius (c. 185–110/9 bc), whereas Ambrose heavily relies on Cicero. The term officium (“duty”) translates what the Stoics called kathekon, the “middle” or ordinary virtuous action which contrasts with supreme virtue called kathorthoma. As such, both treatises are concerned with how one can be virtuous in ordinary social life.

In De officiis, Cicero distinguishes two pairs of personae that shape one’s life.Footnote ²⁵ The first two are shared human nature and individual endowments. To these rather general categories is added another pair that comes closer to the modern notion of social role: what befalls us by chance and what we choose. Examples of the former include social positions, such as political office and wealth. The fact that these are ranked under “chance” shows that Cicero presupposes that one enters into a preexisting society. One may chance upon a particular role, such as being a politician, because one’s father was a leading statesman. What is one to do if one has to step into such a social role? Either one assumes the role and tries to live up to what it demands, or one chooses a different social role that is better suited to one’s talents. As noticed by Christopher Gill, within such a framework the crucial problem of the modern self, authenticity, does not present itself. Modernity posits an authentic self behind the public persona and understands social roles at best as in tension and at worst as in conflict with the authentic, inner self.Footnote ²⁶ By contrast, in Antiquity, the social role and the self are to be aligned. Obviously, one was aware that a disjunction between natural disposition and social role could occur. One example is the theme of the “true bishop” versus the “official bishop” in Late Antiquity: There may be bishops who are not ordained, whilst some who are ordained as bishops are not truly bishops.Footnote ²⁷ In other words, some individuals have all the virtues of a bishop but not the role, whilst others have the position but lack the suitability for the role. Equally, ancient thinkers like Cicero and Ambrose are aware of the possibility of dissemblance, that is, of the pretense of playing a certain role without really having the concomitant virtuous disposition, and that of ineptitude, that is, taking up duties that go beyond one’s nature.Footnote ²⁸ These are, however, identified as moral shortcomings: The lack of alignment is not the human condition, as modernity has it.

It is difficult for ancient thinkers such as Cicero to imagine the problem of authenticity because one’s character, and hence one’s virtues, are supposed to become visible in the actions performed.Footnote ²⁹ Therefore, Cicero and Ambrose pay much attention to the public nature of actions. One is being engaged with, looked at, and judged. Witnessing an action always also means judging it: Is it a proper action, in this context and for this person? As a consequence, the gaze of others is an important element in enforcing the proper performance of the social role. As Ambrose put it:

The fact of being looked at and judged curbs both dissemblance and conceit: In order to avoid losing face, one will not assume a role one cannot properly perform. In Ambrose, the gaze of God, mediated through one’s consciousness, also provides a check on immoral behavior.Footnote ³¹ Praise, in turn, is ideally supposed to be a consequence of virtue and may spur one on to continue on the good path – but one should not desire praise for its own sake.Footnote ³² Learning to make moral progress, in turn, can be taught by seeing good examples and learning how to act in similar (but never identical) circumstances.Footnote ³³

Doing well and judging well is not easy, however, for social relations are never entirely scripted, and, just as in social role theory, action is situational and dependent on the context (the agent, their social position, and the other person(s) involved, etc.) in which it takes place. Cicero takes responses to unanticipated things to be the best test for the truthfulness of one’s moral disposition. As we shall see in Chapter 6, the responses of an emperor to unscripted events during a ceremony are described in Late Antique narrative sources as test cases for his aptitude to reign. The situational nature of action is most evident in Cicero’s discussion of liberality and Ambrose’s Christian counterpart, almsgiving: How much one gives depends on the moral character of the receiver, their (former) social position, and their social proximity to the giver.Footnote ³⁴ Almsgiving is thus not simply a matter of giving the poor something to eat. The proper response may be to give more to someone who has lost their fortune in a fire than to someone who squandered it on parties. As put by Ambrose, “we also need to consider, in every action, what is appropriate for particular personalities, circumstances, and stages of life, and what is best suited to the abilities of each of us as an individual.”Footnote ³⁵ Bringing the right judgment to bear on a particular action can, therefore, be difficult. In narrating an episode from his own life, when he controversially sold church vessels, Ambrose shows that in circumstances when the audience seems startled and disgruntled one needs to be able to justify oneself and to explain why this particular deed was the proper thing to do.Footnote ³⁶ In cases where the interpretation is not clear-cut, argument and dispute will thus surround the action.

Social role theory and practice theory invite us to see social interactions not as singular instances of individuals interacting but as instantiations of social rules that point the way to social structures and to general features of a particular society. This implies also that we read accounts on particular social interactions not just as ideological reports by particular authors but as context-specific representations that tap into common understandings and rules. The ancient ethics of virtue shares many of the emphases of these modern theories on how general rules and principles are instantiated in particular actions and deploys the language of virtue to describe these. Its understanding of public actions as reflections of the individual’s qualities and demonstrations of their suitability for the social role generates a highly judgmental public gaze, which we encounter in the Late Antique sources.

Hierarchy

In this book, I argue that two concepts are central to understanding Late Antique society: hierarchy and justice. The second one is unsurprising, as justice was in ancient thought the virtue governing social relations. Yet its Late Antique inflection has specific features, some of which are due to the other notion: hierarchy. In Late Antiquity, most social interactions were between unequal individuals. Social hierarchies are mostly understood as the unequal distribution of wealth and power, but, as this section argues, this needs to be complemented with attention to the moral expectations inherent in a hierarchy: In Late Antiquity, the superior was supposed to be better. If we explore this expectation that moral and social hierarchies coincide, we shall see that it is the very condition on which social interactions rested in Late Antiquity, thus shaping the way power was exercised in this period.

Late Antiquity was a hierarchical society, as all ancient societies were. More generally, social hierarchies are a feature of any society, including today, however, much Western societies proclaim equality. Nevertheless, Roman society was explicitly hierarchical, and self-consciously so. At the bottom of society, the free and unfree were strictly set apart, with the rest of society made up of a pyramid of senators, knights (equites), civic elite (decuriones), and lower groups, such as freedmen. It was possible to change social rank, even if the concentration of wealth, culture, and political power in the hands of the elite ensured a high degree of stability.Footnote ³⁷ Some of these groups disappeared in Late Antiquity, such as the equites, whilst Later Roman emperors rewarded the highest posts in the imperial administration, which had grown significantly under Diocletian and Constantine, with senatorial status. As the stipends for officials were paid in the new gold coin, this created an aristocracy of service that through its access to gold was able to transform itself into a landed aristocracy.Footnote ³⁸ Within this enlarged elite group, distinction was sought after, and a complex system of rank titles developed, built around the basic ranks (from low to high) of vir clarissimus, vir spectabilis, and vir illustris. The aristocratic competition for status that ensued could be steered by the emperor. Numerous laws fix, for example, the order in the imperial adoratio, that is, the paying of homage to the emperor, and rearrange the legal privileges attached to each grade.Footnote ³⁹

Besides this system of ranks in the upper echelons of society, a second feature of Late Antique society is the distinction between honestiores and humiliores, the honorable and humble members of society. Whereas previously all free citizens had in principle enjoyed the same legal rights, from the second century ad onward two groups, commonly labeled the honestiores and the humiliores, were treated differently in legal terms, with the most striking difference being that the former were protected from corporal punishment and torture. In Late Antiquity, the notion of infamia, which earlier had indicated a limitation of one’s capacity to act legally as a Roman citizen due to ill repute, transformed into a punishment for the elite. Infamia was a nightmare, for it meant the lowering of one’s existimatio (reputation) from honestior to vilis (humilior) and hence the loss of legal privilege in addition to the loss of reputation. As noted by Peter Riedlberger, it could be the equivalent of the death penalty for lower-class individuals, implying social death. By contrast, infamia could not be a punishment for someone lower on the social ladder who did not belong to the honestiores – they were tainted by infamia anyway.Footnote ⁴⁰

A third feature of Late Antiquity is the development of an ecclesiastical hierarchy. By the end of the third century, the model of a single bishop directing a local clergy, with priests, deacons, and a host of lower clergy, had spread to most places. From the fourth century onward, a hierarchy of bishops developed, broadly mapped onto the provincial structure, with the first in rank being a metropolitan who resided in the provincial capital. Conflicts about rank abound, from the jockeying for position between bishoprics (most prominently Rome and Constantinople) to disputes about precedence within local churches between priests and deacons.Footnote ⁴¹

In the light of these developments, scholars agree on the increasingly hierarchical nature of Late Antiquity.Footnote ⁴² Emphasis is laid on the socioeconomic and political dimensions of hierarchy, understanding it as social inequality and unequal access to power. What receives less attention is the fact that normative meanings were associated with social distinction. If treated at all, they tend to be rejected as ideologies serving to make inequality palatable to the subordinated and to secure the standing of the elite. Scholarly discussions of hierarchy therefore tend to be negatively colored and implicitly or explicitly contrasted with the positively connoted concept of equality.Footnote ⁴³ Let us therefore take a closer look at how social hierarchies were understood in Late Antiquity and what this meant for social interactions.

Affirmations of social equality are not uncommon in Roman and Late Antique thought, but they tend to go hand in hand with the affirmation of hierarchy as the normal condition in society. Broadly, two tendencies can be observed. First, it could be underscored that individuals were equal qua human beings – that is, as far as regards their human nature. While Stoicism is known for affirming the equality of all human beings, all taken to be citizens of the cosmic city, this is in the first place a statement about our shared human nature. Yet the Stoic view of individuals as social beings, especially as found amongst the Roman Stoics, emphasized that we find ourselves in a hierarchical society, as we saw in our discussion of Cicero’s theory of social roles. Even in Epictetus, who was formerly enslaved, we encounter the thought that it is normal to expect a slave to have more vices than someone who is not a slave.Footnote ⁴⁴ Similarly, Christian scriptures emphasize the equal participation of every believer – men, women, slave or free – in Christ.Footnote ⁴⁵ Augustine saw social hierarchy, or at least the basic distinction between free and unfree, not as an original part of God’s creation but as His response to humanity’s sins.Footnote ⁴⁶ Thus, from this perspective, equality is located outside of human society, which is hierarchical in nature.

According to another view point (which need not exclude the first), social equality could be achieved within society in particular circumstances. One particular social bond, namely friendship, was traditionally understood to entail equality in social position and virtue.Footnote ⁴⁷ This does not mean that friendship is only possible between social equals: Friendship means the temporal suspension of hierarchical difference, allowing friends to speak out against each other freely and share their thoughts. Ideally, this perfects the virtue of both friends, whose flaws are complemented and corrected by the other’s qualities and who can become one.Footnote ⁴⁸ This particular bond explains the attractiveness of friendship as a social category. For Plato and Aristotle, generalized friendship was central to their understanding of an ideal state.Footnote ⁴⁹ Seth Schwartz has argued that Jewish writings depict relations amongst Jews as those amongst friends, just as Christians did, thus putting them equal footing. This served to contrast their own communities with the dominant social interaction of hierarchical reciprocity.Footnote ⁵⁰ In Late Antiquity, a rhetoric of friendship could serve to clothe acts of parrhesia, whereby criticism of the superior was justified by the inferior posing as the ruler’s friend.Footnote ⁵¹ But however important friendship was in ancient discourse, we must realize that it owed this status to the fact that it marked an exception to the general understanding of society as hierarchical.

Thus, leaving aside idealized presocietal conditions and unique social bonds, society was thought of as hierarchical in Late Antiquity. Indeed, the notion of a society itself was taken to entail the concept of hierarchy. A comedy possibly dating from the fifth century, Querolus, describes how after a social revolt and the disappearance of Roman power, the people north of the Loire are living according to the “laws of the nations,” with no signs of rank remaining: “peasants even plead in court and private men are judges. All is allowed there.”Footnote ⁵² Seeking to rise above one’s station could undo society, lamented the fourth-century Antiochene orator Libanius in response to those who thought it was good to seek social advancement by acquiring imperial office and thus abandoning their duties toward their home town: “Goodness and happiness consist in a man’s maintenance of his station, whatever it may be.”Footnote ⁵³ Of course, there is much hyperbole in his claim that society will fall apart if one does not stick to one’s position. What matters for us here is that Libanius can make such an argument at all, that it thus seemed plausible to imagine society as made up of a set of hierarchical roles which one had to respect. A more positive expression of the same idea is the notion that every individual, however mean their role, is necessary for the overall order. With his sense for rhetorical surprise, Augustine argued that even pimps and prostitutes were a necessary building block of societal order.Footnote ⁵⁴

Hierarchical difference was, however, not merely thought of as sociopolitical distinction. In fact, Late Antique sources align with the contention by the indologist Louis Dumont that hierarchical differences in the Indian caste system were morally charged and, more specifically, that the superior is supposed to be better than the inferior. Dumont emphasized, in a somewhat Hegelian fashion, how a hierarchy structures society through a series of oppositions that are synthesized by the higher in the opposition. For example, in the story of Adam and Eve, the male is superior to his opposite, the female. By being the origin of the female, the male in turn encompasses both male and female and unifies them. Being superior means also better incarnating a value – in the case of India, purity – with the Brahmins representing the highest purity. One may wish to question Dumont’s understanding of the caste system and his use of structuralist oppositions, but this should not deflect us from the essential insights he revealed. A hierarchy is morally valuated and places higher demands on the superior than on the inferior in terms of the actual realization of the value that they incarnate.

A number of Late Antique sources confirm the ideal coincidence of social and moral hierarchy. Pseudo-Dionysius the Areopagite underscores in the sixth century that the superior possesses the qualities of the inferior – a formulation reminiscent of Louis Dumont’s interpretation of hierarchy.Footnote ⁵⁵ In his treatise on duties, Ambrose of Milan states:

Ambrose is considering the conditions on which a person can give sound advice. This is, by its very nature, an asymmetrical situation: The person who seeks advice puts himself into a relation of dependency and inferiority toward the person giving advice. Advice can only come from someone who is “higher than oneself,” the literal translation of supra me debet esse, rendered in the preceding translation as “he has to be a better man than I am.” The translation cited highlights the moral meaning, but in supra me the social meaning also rings through.

The same assumption undergirds the concerns we encounter about misbehaving Church and state officials. We have seen that they had an exemplary function, providing models of conduct for others. If an official had a morally flawed character, this risked vitiating the entire society, as was depicted in stark colors for the Christian clergy by a fourth-century Roman cleric going by the modern name of Ambrosiaster:

As this passage shows, believing that a social hierarchy should coincide with a moral one does not rule out the possibility that superiors misbehaved or that they could be criticized for doing so – even if we shall see that this was not easily done. Indeed, within a monastic context, failure by a superior to live up to his role could be more severely sanctioned than the same deed by someone lesser in rank.Footnote ⁵⁸

One may be tempted to understand all of this as a hegemonic ideology, strengthened by being embedded in a system of morality, and seeking to protect the exalted position of some to the detriment of the others. Nevertheless, precisely the moral charge of social relations provides opportunities for social advancement and for the challenge of superiors. We have seen that in Late Antiquity people were aware of the fact that virtue and social role did not always coincide, although they ideally should. The display of virtue thus could lead to social advancement, for example, by the election of a virtuous ascetic to the position of bishop. One could therefore pretend to be virtuous or display real virtue in an improper, ostentatious way, in order to advance one’s career. The widespread chastisement of ambition and praise of humility in Late Antique sources are rooted in this expectation of sincerity as well as in the recognition that virtue carried one higher. Yet humility, as a recognized sign of true virtue, therefore also risked being instrumentalized.Footnote ⁵⁹ Unsurprisingly, then, the praise of humility as a virtue, spreading even to the emperor, goes hand in hand with a profound suspicion of particular acts of humility, as is most visible in prescriptive treatises on monasticism.Footnote ⁶⁰ Our sources are constantly on the lookout for sincerity.

Although the assumption that a superior was of a higher moral standing than their inferior could insulate the superior from criticism, it also generated higher expectations for them, which could be exploited by the inferior. This is the dynamic that the chapters on petitions and parrhesia will chronicle. Once one petitioned someone in authority, one appealed to their virtue of justice, limiting the range of possible responses to the petition. Similarly, as tolerating parrhesia was an imperial virtue, the espousal of the stance of a free-speaking individual was a way of forcing the emperor into a particular social role which carried specific normative expectations. Precisely the meanings and expectations inherent in the social roles allow individuals to mediate the hierarchical difference and, in some instances and usually momentarily, to bridge it or to reverse the asymmetry between superior and inferior in their favor – and thus briefly exercise power over the superior.

Moral and Legal Justice

The previous section has underscored that we should think of hierarchical relations also as moral ones and hence as being governed by virtues. Within social interactions, all virtues play a role, but ancient authors agreed that justice was the prime virtue governing society, as justice took into account not only one’s own eudaimonia but also that of others. I shall first illustrate how closely the virtue of justice was tied into Late Antique thinking about society and then argue that the virtue of justice was considered to be only imperfectly realized in positive law. In other words, moral justice was often opposed to legal justice. In fact, all the social interactions that I study in this book sought a form of justice that could not be offered by institutions and positive law.

In an often-cited passage from the City of God, Augustine states that kingdoms without justice are nothing more than bands of robbers:

When such a robber band expands over a territory, it will end by calling itself a kingdom. The passage occurs in the context of Augustine’s criticism of Roman imperialism and the idea that Roman military conquest had brought happiness to its subjects. The crucial words in Augustine’s quote are the first two: Without justice, there is no society. That idea was built on very common notions widespread in ancient literature. In his Republic, Cicero defines the people (populus), that is, the constitutive element of the Roman state, as “an assemblage of a multitude associated in an agreement with respect of justice and a partnership for the common good (iuris consensu et utilitatis communione).” He also draws the comparison of a state lacking just laws to a robber band. Justice, then, was included in the definition of the people and hence of the state.Footnote ⁶² Drawing on historical arguments, Cicero states that kings were chosen from the virtuous so that the people could enjoy justice. Kingship was thus understood as standing in a relationship of justice to the people, a notion that we encounter in Late Antiquity too, as I will detail in Chapter 5.Footnote ⁶³

Besides Cicero, another source of inspiration for Augustine was the early fourth-century Christian teacher of rhetoric Lactantius. He composed a lengthy defense of Christian doctrine, entitled Divine Institutes. Later, he produced an epitome of this vast, seven-book work. In both works, he presents his views on justice, which develop Greco-Roman ideas within a Christian context. In the Epitome, he starts his discussion by noting that Scripture and philosophers agree that “man is born for justice,” citing Cicero in support of the idea.Footnote ⁶⁴ Then he continues:

Given our created nature, we owe religio to God and caritas toward man, as our fellow rational being. Religion falls under wisdom, here understood as the highest possible knowledge, whilst love (caritas) is a virtue. Justice (iustitia) therefore covers both piety toward God and care for one’s fellow humans. Later in the Epitome, Lactantius ties this view into his understanding of salvation history. Ignorance of God logically entails that the tie of justice toward one’s fellow humans is dissolved. Violence is the consequence, which in turn demands positive law. But then people start doing evil secretly. This negative cycle is interrupted by God, who sends His Son to remind us of the true order of the world.Footnote ⁶⁶ The way Lactantius roots good interpersonal behavior in a proper understanding of God is distinctly Christian in outlook. His use of justice to cover one’s attitude toward both God and men allows him to use a stronger, Christian term for the virtue guiding social interactions, namely caritas, meaning love or affection. Yet its fundamental presuppositions are not specifically Christian. The fifth-century Neoplatonist Hierocles, to give but one example, also sees justice as the virtue that defines interpersonal relations, to the exclusion of human–animal relations, and also considers it ultimately to be rooted in humanity’s orientation toward the divine.Footnote ⁶⁷

In sum, then, the notion of society was a normative one in Late Antiquity, in the sense that justice was seen as a condition for its existence.Footnote ⁶⁸ The justice that lies at the roots of society is, however, a virtue that cannot be reduced to justice understood as the application of positive law, decreed by state institutions. Indeed, in his On Duties, Ambrose underscores that formal obligations are subordinate to moral duties: If the social relationship changes, it may be proper, for example, not to live up to a legal obligation or not to respect one’s oath. Similarly, the Neoplatonist philosopher Sopater (fourth century) stresses that a judge should take the character of the offender into account when determining a punishment rather than strictly following the stipulations of the law.Footnote ⁶⁹ The practices that I study in this book often aim explicitly at correcting the wrongs that risk being done by applying positive law.

A good illustration of this opposition between justice as a virtue and justice as positive law can be found in Late Antique descriptions of the emperor as “living law.” In a speech addressed to the emperor Constantius II (337–61), Themistius describes the emperor as the apex of the social hierarchy and thus the fount of justice for society. This means, of course, the creation of positive law, for which the Late Antique emperor was the only source. Yet, the justice given by positive law is depicted as seriously limited, not to say flawed:

Themistius here relies on the assumption that it is the emperor’s role to imitate the divine, which was, whatever one’s religious convictions in this period, identified with the Good. The emperor, at the top of the social hierarchy, had to imitate God – for example, as put by Themistius elsewhere, through justice and philanthropy.Footnote ⁷¹ The notion of the image was not restricted to the emperor: Every individual in the chain of command was an image of one’s superior. Themistius emphasized that every official is a small image of the emperor and that people conclude from their contact with officials what the original is.Footnote ⁷² A consequence of this idea was that positive law, enacted by the emperor, was ideally a translation of divine law, that is, the divine precepts for the world and society. But positive law is a blunt instrument. Even when leaving aside the possibility that emperors lack in virtue and thus could enact unjust laws, it was ill suited to do justice to the situational nature of social interactions. Further, positive law has to be applied, which can be done badly. Thus, the emperor has also the role of ensuring that someone who commits a crime by error or misfortune is not punished in the same way as someone who consciously committed wrongdoing. By righting the wrongs inherent in positive law, he provides an example of true justice to all his subordinates. This conception allows Themistius to play on a double meaning of the law: The emperor is the law, that is, he embodies divine law, which is due to his position at the apex of the social hierarchy, whence he has to relay justice on earth. At the same time, he is above positive law (“the laws”), in that he, as the ultimate source of positive law in the Empire, has the possibility to adapt positive law to the divine ideal. Only if one disregards this conceptual background to Themistius’ argument can one claim that he attributes total sovereignty to the emperor, as if he did not have to respect existing law.Footnote ⁷³ Rather, for Themistius, it is an essential part of the social role of the emperor to ensure that true justice is enacted in each individual case, whereby true justice is a virtue (and hence situational) and not just the application of positive law.Footnote ⁷⁴

Statements about the emperor being law embodied may seem justifications of absolute rule. Some of these statements may have served such a purpose, but it is more appropriate to see them as rooted in a hierarchical conception of society, whereby the virtue of justice, understood to be modeled on divine justice, was expected to regulate social interactions. In the remainder of this book, I shall be using the term “moral justice” to indicate this form of justice, whereas I use “legal justice” to indicate justice as enacted by positive law.Footnote ⁷⁵ The discourse about the emperor as “law embodied” or “not bound by positive laws” brings out the tension that existed between both forms of justice. Being the source of both legal and moral justice, the social role of the emperor was inhabited by the tension between both. In Chapter 4 we shall encounter emperors who, in a single gesture, condemn to death and exercise mercy.Footnote ⁷⁶ This is a dynamic in the exercise of power that we fail to grasp if we focus predominantly on institutions and positive law.

Conclusions

The remainder of this book shall render more concrete how social interactions in Late Antiquity are performances of justice. Four interlocking aspects will be explored. First, the importance of personal interaction as a locus for social power. The expectation of virtue rendered social interactions occasions to demonstrate one’s suitability for one’s social role. That suitability, in turn, gave one authority to perform one’s role and hence to exercise power. This does not preclude the exercise of power through other means such as institutions, but I do argue that we do not attribute enough importance to personal interaction in our understanding of the exercise of power in Late Antiquity.

Second, the situational nature of social interactions. What it means to perform such a practice well is conditioned by these expectations, within a particular context (circumstances, other participants, space, time, etc.). The judgment on a practice and on the character of the person performing it is further shaped by the intentions, convictions, and (in ancient language) virtues of the reporter. Thus, whilst tapping into the same common understandings of a practice shared across society, different reports can vary widely in their judgment on particular instances. As we shall see, Late Antique actors were often aware of the expectations and dynamic of a particular social interaction and sought to exploit these to their own benefit, for example, by choosing the right occasion and thus creating the context that would allow a positive outcome for oneself. In other instances, individuals tried to avoid being drawn into a particular practice, for example, by not appearing in public to avoid being petitioned.

Third, social interactions reveal the character and virtues of the actors involved. The methodological individualism of social roles, which focuses on how a particular person performs a socially defined role, aligns with the perspective of our sources, which focus on the character of the person acting. The focus on character is, however, not merely a culturally determined blindness for structure. The concern that our sources have with the possible disjunction between social role and virtue and their suspicion of legal justice when compared to moral justice indicate something stronger: the belief that institutions are secondary to the performance of virtue and that institutions can and need to be overruled to ensure justice in society.

Fourth, justice as the central virtue. Unsurprisingly, justice will be the central concern across the four case studies in the following chapters. As are all powerful ideals, it was as attractive as it was ill defined, and much depended on the concrete instance to see what justice meant. Crucially, from my perspective, the normative expectations that are an integral part of social interactions were available to all, that is, to inferior and superior alike, thus locating power on both sides of the social interaction. In some scholarship, there is a tendency to assume that the values shaping ancient and medieval society were defined by the elite and then projected onto society at large. This usually goes hand in hand with the idea that power is exercised top-down.Footnote ⁷⁷ Yet, as we shall see, the expectations implicit in particular social interactions, such as petitions, could consciously be activated by the inferior, who was, however, himself also subjected to these expectations. My emphasis on the normative expectations inherent in particular social interactions should not be taken to mean that Late Antique society shared a single set of moral convictions. On the contrary, the importance of the situational nature of action and interpretation underscores difference and conflict. In Chapter 4, we will see that the practice of intercession was especially performed by clergy and monks and that it was the scene for a conflict between an ecclesiastical and a secular moral code. Yet, on a deeper level, both agreed on the fundamental value of justice, whilst the two codes conflicted as to how this was to be exercised. Disagreement and conflict on what the fundamental values of a society entail are not just a feature of modern, fragmented societies but are inherent in any society.Footnote ⁷⁸

These four chapters present only a small slice, albeit a fairly well-documented one, of social interactions. The aim of this book is to get into focus general features of the late Empire. For that reason, my selection of practices is skewed toward those that tie in with larger debates about the Later Roman Empire. Petitions, for example, have been the key type of evidence in discussions about the responsive nature of imperial action. The supposed demise of parrhesia in Late Antiquity has been seen as a sign of the decline of free speech in an authoritarian society. Violent practices of intercession by monks have been adduced as evidence for problematic Church–state relations. Popular riots, finally, have been interpreted as caused by the decline of institutions of representation and as signals of social instability. I hope to show that, approached from a different angle, these practices invite us to reconsider these debates. As a consequence, I do not explore the potential of my approach to other, less public, sectors of society, such as households and monastic communities. I also do not address questions of gender, age, or ethnicity. These would add nuance and complexity, but the questions of political power that shape my interest in this book imply a preference for sources on the male elite. I have also refrained from offering a developmental or regionally diversified account. The four selected interactions had an earlier life in the High Empire and an afterlife in the kingdoms and empires that succeeded the Late Roman Empire, but this would have meant incorporating much additional material into what is already an extensive source basis, and my focus is also not so much on tracing the history of these practices as on uncovering what they reveal about the Later Roman Empire.