In the towns and cities of the Roman world, over the first centuries AD, the law was alive. It was something that people thought with, something that they wrote about, copied, excerpted, edited, fought over, joked about, imitated, and parodied; something, in other words, that they invested with meaning and animated with narratives. By the end of the fourth century AD, it was no longer so: it was a specialized technology, held in books, and propounded by experts. A great separation had taken place: law became divorced from society; it became something to be applied to subjects, recalcitrant or not, and measured on the basis of its efficacy. The Judgment of the Provinces is about this transformation: when, why, and how it happened, and with what effects on provincial and metropolitan culture more generally, and political and rhetorical culture more specifically.

***

The main thread of this story runs through the eastern half of the Roman Empire: a vast and heterogeneous swathe of land running from Greece and Macedonia through Asia Minor, into Syria and the Levant, and down to Egypt and Cyrenaica (see Figure 1.1). Within this densely populated world, a frenzy of hermeneutic activity took place. Christians, Jews, and polytheists all told stories of the righteous individuals confronting imperial governors in court. They laughed at the lawyerly logic-chopping of the crafty slave Aesop. They recopied administrative documents into petitions and used them to make claims on representatives of the imperial state. Among these populations, the language of law makes its appearance in texts both humble and learned, urban and rural. And yet the texts themselves are only the ripples on a surface, attempts to fix in writing a move or gesture in a larger conversation about the nature, content, and legitimacy conditions of the world their writers inhabited – the world of empire. As such, law was the starting point for a distinctly provincial political theory.

Figure 1.1

Map of the Eastern Roman Empire.

Barbara Levick, High Tide of Empire (Cambridge: Cambridge University Press, 2023), p. 9.

These provincial interpreters were interested in law not as an abstract system, but as a practical manifestation of authority. They engaged the courtroom as fact and symbol, interpreting legal interactions through their local understandings of power, of legitimacy, and of the ideal relationship between legal rules, administrative practices, ethical demands, and rhetorical performance. In other words, within this imperial context, law was vastly more than mere nomos (a mode of declaring a statute or a binding rule). It was also logos (a mode of speaking and reasoning), aisthēsis (a mode of sensing and feeling), and erōs (a mode of longing and desiring).Footnote ¹ It was a register of speaking about problems that was deeply normative, generative of thinking and feeling, and constitutive of communities.

How are we to explain this historical horizon? Why did law come to be among the most important discourses through which subordinated people sought to explain the nature of empire, of administration, and of community?Footnote ² Who conceived it, in which ways, and to what effect? Why, more generally, are empires so productive of legal interpretation?Footnote ³ What effects did these discourses have on the imperial center? And why, eventually, did the center try mightily to make this culture go away? These are the questions that structure what follows.

***

Before attempting an answer – even a brief answer – to these questions, it might be useful to gesture at some ways that contemporary scholarship has framed the problem of law in the Roman world more generally, if only to give a sense of what this book is not. The goal of writing the legal history of the Roman Empire, and even of the Roman provinces, is hardly new. Some reason for this is simply the profusion of legal evidence, both from the metropole and from the provinces. Rome was, in an important sense, the great regulator: Roman elites settled disputes between communities, including communities over which they did not formally rule; they wrote constitutions for subject communities like municipalities; passed laws and decrees regulating the legal status of communities that were subject to them; exercised jurisdiction over their own citizens and others; and granted legal privileges to some and imposed disabilities upon others. We know of this legal activity not least because, in a great number of cases, communities that were the objects of such regulation took it upon themselves to carve the words of such declarations of law into stone, thus memorializing such legal discourse in permanent media.Footnote ⁴ In other cases, the records of such attempts at regulation come through the sheer luck of their having been written on papyri that have been preserved in dry climates. Periodically, too, the traces of such regulation were written down in texts aspiring to the status of “literature” and passed through the medieval manuscript tradition. Given the vagaries of this process, in which the great majority of evidence was surely lost, the preservation of substantial amounts of legal regulation surely speaks to its more general importance among the communities of the Roman world.Footnote ⁵

Law was implicated in Rome’s attempts to regulate, structure, and describe its empire from the earliest period of Roman expansion. Roman magistrates announced, in solemn idioms, what “the rules” were to be, to whom they were going to apply, and by whom they were going to be enforced. At times the Romans announced rules to other peoples because those peoples had requested that Rome settle a dispute that occurred among them or between themselves and their neighbors or requested that a privilege be granted or confirmed. But often this process of announcing rules to a population was done at the barrel of a gun (or at swordpoint, if one likes), in the wake of conquest, when Roman magistrates simply dictated to other communities how they were to be structured. It ought to be added that a key part of these Roman regulatory schemes involved Roman magistrates declaring that communities ought to be independent and autonomous – that they use their own laws or some version thereof, rather than those of the Romans.Footnote ⁶

Three impulses motivated these decisions. First, Rome often lacked the desire, and at times lacked the capacity, to involve Roman magistrates in the day-to-day affairs of subject cities. Second, as a matter of policy, Rome rewarded the most loyal of allied cities with the right to self-government. Third, Rome was reluctant to impose Roman legal frameworks on peoples and territories that were not juridically “Roman” but rather, in the terminology of Roman jurists, peregrini or “foreigners.” According to the Roman jurists, the Roman ius civile applied to Roman citizens alone. This claim was truer in theory than in practice, but it performed the important ideological work of marking the uniqueness of Rome vis-à-vis its subjects. As a result, only a small proportion of the inhabitants of the eastern Mediterranean would have held Roman citizenship before AD 212, when the emperor Caracalla granted citizenship to all free inhabitants of the empire.Footnote ⁷ Nonetheless, even in these acts of refusing to provide substantive rules to many communities, Roman magistrates were essentially legislating for them constitutionally, and thus, in a sense, governing their empire through law. What is more, Rome attempted, through the language of law, to regulate its own magistrates when they acted abroad. Such men, it was well understood, might prove politically dangerous.Footnote ⁸

This profusion of legal acts has not failed to excite scholars, not least because it would appear that, from such acts, a relatively straightforward story of Roman legal practice can be told across the first four centuries AD. Earlier, relatively doctrinal scholarship sought to think of this flurry of legislative action as providing an aperçu into Roman imperial structures along the lines of a more contemporary, rationalized constitution with various actors exercising well-defined powers and offices. The more recent tendency has been to see these various pronouncements within an arc of historical development.Footnote ⁹ This is surely correct and salutary, at least in part.

Within this arc of development there were tendencies towards both fragmentation and homogeneity, both acting in productive tension. The pragmatic demands of governance tended to produce fragmentation. Though Roman magistrates produced texts written in authoritative language that purported to regulate behavior from the Republic onwards, the texts themselves were often ad hoc responses to particular problems and challenges. The edicts of governors tend to address highly localized problems: a famine here, the abuses of detachments of soldiers there, and so forth. This is hardly surprising: the Mediterranean itself, and the eastern Mediterranean in particular, was geographically, linguistically, and politically fragmented. Egypt might require a particular rule, Bithynia another. The patchwork of political units across this landscape further complicated matters: some areas were filled by cities that enjoyed varying degrees of independence, whereas others were crisscrossed by imperial properties that presented different regulatory challenges. In some areas, the imperial government had placed colonies of Roman citizens who had particular status and privileges; in others, the territory was imagined to be completely unurbanized, and thereby subject to different types of rules. While at the ideological level Roman administrators and provincial subjects might have been committed to the proposition that imperial law may have suffused these areas, in practice it did so unevenly, and in some places not at all. Law thus tended to mark regional and political difference, rather than bringing together diverse territories under a single constitutional regime.Footnote ¹⁰

By the same token, however, there were patterns to such responses, and they become more pronounced during the first three centuries AD as the advent of emperors and a nascent imperial bureaucracy began to homogenize some aspects of the Roman imperial system. Some of this stability emerges from ruling ideology: Roman administrators often sought to project a sense of conservatism in their legislation, sometimes to the extent of simply copying whole paragraphs of earlier texts. Such was the case, for instance, in crafting constitutions for municipalities, the majority of which existed only in the Roman west. In other cases, stability and consistency emerged from the brute necessities of governing a vast territory: Roman administration depended largely on a governor, who divided the territory into a series of conventus centers – locations to which he would ride circuit, to hear cases within his jurisdiction and to check on matters that required his oversight. These visits were necessary as a matter of management, not least because they operated to oversee processes of taxation and material extraction. They also had important ideological and symbolic value, for jurisdiction was expressive of Roman sovereignty in a time of peace. It should be added that the conventus districts did not follow the contours of local cultural, geographic, or linguistic divisions. By dividing the province, the conventus districts tended to group dissimilar people together and in so doing attempted to collapse local modes of distinction. Finally, we might add that in all territories Roman rule was imagined to operate hierarchically: the governor was the chief authority over his province, and was in charge of subordinate officials; but he was also subject to the emperor, who could be appealed to under particular circumstances.Footnote ¹¹

Overall, this tension between homogeneity and particularity seems mostly to have been productive, rather than problematic. Though hard to quantify, it appears that large numbers of subjects settled their disputes in provincial courts. From Egypt, the province that affords us the most granular documentation, we see not only a flurry of petitions to both local and imperial officials, but also a flurry of correspondence between those who travelled great distances to have their cases heard, and those who stayed back, awaiting the results. Moreover, this movement across the landscape continued despite the personal status of the individual complainants, most of whom were not Roman citizens. Cities, in turn, competed to host the governor’s assizes, a badge of their importance vis-à-vis their neighbors.

If in fact the penetration of legal institutions was spotty, the ideological importance of such institutions nonetheless coaxed them into the intimate disputes even of country towns and villages. It bears adding, of course, that we get only rare glimpses of those disputes that people chose not to settle in formal venues – because they thought them unacceptably alien, because the cost of litigation was too high, or simply because they did not speak or did not care to speak the official language of such a venue. Still, at the ideological level, this flurry of litigation was guaranteed by the emperor himself, who was imagined to be available to his subjects to rule even on matters approaching the terribly mundane, both in person and via written correspondence. But one sees the deep penetration of legal institutions in ways independent of litigation: one can see it in the migration of legal forms for structuring official documents, in the development of archival practices, or in the development, against the background of Roman law, of forms of contractual and property rights that resulted in burgeoning trade. Legal forms also operated, in some communities, as indices of identity, which might urge us to consider this shared legal culture as an index of “Romanization.” Even if such a term makes us anxious, the picture that is revealed is nonetheless that of an empire suffused with law and legal practice.Footnote ¹²

***

The picture I have been painting thus far is of legal history in the institutional register. It operates from a central assumption, which might be termed legal positivism: that law is, primarily, a practice of the state – and in particular, a practice of the state in which binding commands are issued to guide the practice of either a subordinate official or a non-state end-user. Accordingly, in institutional legal history one begins by drawing a map of the universe of state pronouncements (be they pragmatic and local, or ambitious and universalizing), or even a map of state capacities to make such pronouncements, and then measuring the extent to which that map adequately describes practice “on the ground.” How, that is, law enforcing institutions operate, or how institutions affect the behavior of non-state actors.

A few things might be said in favor of such an approach. To begin with, this method at times produces empirical findings of great significance: it matters, for instance, to know the extent to which local institutions continued, over the first three centuries AD, to produce legislation and to what extent, if any, the shape of those institutions changed.Footnote ¹³ Second, insofar as institutional history creates maps of complex processes, maps of this sort are often useful. They simplify messy realities by abstracting salient details from a universe of empirical noisiness. It does indeed matter to have a sense of which organs were – or thought themselves – capable of constraining the behavior of others, even if they were, in actuality, limited in their ability to actualize such constraints.

Third, and on related grounds, it might make sense to begin with the authoritative pronouncements of the imperial Roman state (and those authorized to speak in its name). It possessed the power to enforce its commands with violence. It often did so in a way that was regarded, by other Romans in positions of power at least, as final. The importance of violence ought not to be underrated, and more will be said about it in the course of what follows. For now it will suffice to note that violence is often at the foundation of systems of incentives (and on some accounts law simply is a system of incentives).Footnote ¹⁴ More to the point, Roman administrators were perfectly content to deploy violence to enforce their dictates when they deemed it necessary.Footnote ¹⁵ The possibility that someone would suffer material loss (of money, of status, or of flesh and skin) does with regularity influence behavior. It is a good reason to give these sorts of state commands an important role in crafting our maps of provincial legal practice.

Its advantages to one side, it might also be said that the institutional account of Roman legal history rests on two assumptions, each of which will be interrogated in what follows. First, the institutional account assumes that law consists primarily in commands, and second, that the goal of such commands is ultimately enforcement. A third assumption, as well, has sneaked in, one characteristic of the “law and society” literature that began to gain currency in the second half of the twentieth century: that the goal of writing legal history is essentially that of measuring the gap between edict and enforcement, between “law on the books” and “law in action.” The purpose, ultimately, is to assess the health and strength of the state.Footnote ¹⁶ It is here that what follows parts ways with the institutional account of Roman legal history.

***

We might raise several critiques of this model of writing legal history. The first is methodological. Attempts to measure state efficacy in incentivizing the application of its own rules only have salience as part of a comparative endeavor: that is, to understand the effectiveness of Roman state pronouncements only makes sense in comparison to the effectiveness of other states in enforcing their pronouncements. Yet in many institutional histories of the Roman world, this explicit comparison is simply omitted. The result of such inquiries is thus often unsatisfying.Footnote ¹⁷ No more satisfying are institutional histories that seek to sidestep this problem by claiming a more restricted purpose, namely, of simply showing how things “actually” worked (normally in comparison to an ideal, articulated or not).

But there is also a second, perhaps more consequential objection: on what grounds do we privilege the state as the source of law in the first place? In other words, to what degree ought our histories of law in the Roman world be founded on an account of legal positivism – that is, the claim that law is, and just is, what the state declares it to be? Or even upon a modified version of legal positivism, one adapted to the study of empires, that law ought to be understood as the command both of the imperial metropole and of subordinate units?Footnote ¹⁸ I would suggest that such an account is, at best, only partial.Footnote ¹⁹

Consider the following example. At the start of his second-century textbook, the Institutes, the Roman jurist Gaius attempted to delimit the field of study by anticipating a student’s question, namely, where was one to go to learn the Roman civil law (ius civile)? To begin with, Gaius explains, there was not one single source of law (ius); Roman civil law consisted, rather, in multiple sources of authoritative discourse (iura):Footnote ²⁰

Gaius continues from here, defining the nature of each of these, and giving the background that a student would require: thus, a statute is said to be “what the populus decrees or orders,” a plebiscite is what the plebeians decree or order, and so forth.

We need not detain ourselves over the well-known problems that this text poses, not least that there are rival attempts at defining the content of the ius civile by other authors, namely Cicero and Pomponius, crafted under different circumstances and pressed into doing different sorts of work. Nor do we need to worry about what seems to be a sleight of hand: that Gaius’ account of the ius civile is substantially positivist and written, relying heavily on official orders coming from the state, while in point of fact much of ius was oral – so much so that what was written down by jurists hewed closely to this tradition, copying much of its oral style. Jurisprudence is another puzzle in Gaius’ account. He gestures at it in his claim that the ius civile also includes the responsa of jurists – the replies of learned men – but for this to be accurate we would have to take “responsa” in a much broader sense than Gaius seems to intend.Footnote ²¹

We should instead focus on the “frame” that structures this sort of definitional activity. Gaius is not stating a fact: that is, that Roman law just is these six things – for in fact it was more. He is rather engaging in an act of deixis: he is pointing a student, his imagined reader, to a body of materials, materials that have been elevated to the status of texts that deserve commentary and study.Footnote ²² And as with all such acts of deixis or pointing, defining one body of materials as deserving of study does two further pieces of work. First, it operates to create community: those who comment on these texts are discussing the ius civile. Second, and relatedly, by creating community in one place it operates to exclude other, rival understandings of the relationship between texts and communities: those who comment on these texts (and not those texts) are discussing the ius civile. He is, in effect, issuing an invitation: join me in discussing these texts, follow my lead in learning how to discuss them, and you will join me in a learned community – with all of its attendant privileges. This act of pointing cannot be separated from values, or from status. It is also political: these texts, the invitation suggests, constitute us as a political community. To join in their interpretation is to partake of our history and our values, and to share in, or at least become habituated to, our deepest assumptions about optimal social and political ordering.Footnote ²³

Gaius’ account of the ius civile can be contrasted to another example of deixis. This one comes from a text known as the Mekhilta de-Arayot, a Rabbinic commentary on selections from the book of Leviticus that was later preserved in some editions of Sifra, another commentary on Leviticus which was edited or compiled from yet earlier materials in the Roman provinces sometime probably in the third or fourth century.Footnote ²⁴ Like Gaius, the authors of this passage worried about the content of the law, and worried more particularly about the relationship of the law as written in the Torah to the authoritative pronouncements of the Roman state.Footnote ²⁵ It is a text that worries, in other words, about the nature of community, and of order, in relation to law. It took up these questions in its discussion of Leviticus 18:3, “and in their statutes you shall not walk”:Footnote ²⁶

To make sense of the biblical text, the authors of the Sifra seek to understand the nature of “statutes”: what does it mean that one must not “walk” in them? A separate biblical prohibition had already covered religious acts: child sacrifice, augury, and the casting of spells. So as not to replicate this prohibition, the authors perform an act of redefinition: statutes are not just written prohibitions, but rather customs (nimosot, from the Greek nomos, both “custom” and “law”) – things that suffuse daily life and serve to define communities. To avoid “walking” in “their statutes,” one must avoid theatres, circuses, and arenas; one must similarly avoid certain types of personal grooming, such as avoiding wearing one’s hair qomi (presumably from the Greek for long hair: komē). For Rabbi Meir, these things, taken together, can be captured through an act of deixis: these, he asserts, are the ways of the Amorites and are thus prohibited.

“Statute,” on this reading, is more than the command of the state. It is what archaeologists might refer to as a cultural assemblage: it encompasses architecture and urban space, hairstyle and dress. The things that make them them are the things that we must avoid; since the text posits a we that is constituted by statutes (by Torah), so by deduction there must also be a they, similarly constituted.Footnote ²⁷ It ought immediately to be added that the we posited by this text is a rhetorical construct with only a tenuous relationship to the realities of Roman Palestine.Footnote ²⁸ But the author goes farther, and in so doing manages to capture something essential about the experience of pluralism in an empire: a sense of longing for the culture of the dominant, either by noting an absence in one’s own culture (“They have statutes and we have no statutes”), or by judging alien practices to be aesthetically superior (“But theirs are more beautiful than ours!”). Law, on such a reading, is far from just positive law declarations: it is also the object of aesthetic contemplation, indeed even of erotic desire.Footnote ²⁹ It is in this context that the reference to the “evil instinct,” the yetzer ha-raʿ, makes sense: the desire for the laws of others here is imagined as a sort of base if not sexualized yearning, to be countered with authoritative, Biblical injunctions, to which the reader is directed through another act of deixis.

Both Gaius and the Sifra can be read as attempting, through their own idioms, to define the category of Roman law. For Gaius it is restricted and positive, whereas for the Sifra it is extensive and manifested in things as diverse as hairstyle and entertainment. It will not do, at this juncture, to suggest that the Sifra was simply incorrect or tendentious in claiming such an extensive territory for “their” laws. To begin with, Roman jurists and administrators themselves at times imagined that legal practices, and legal statuses in particular, were just parts of one’s habitus, one’s way of being in the world.Footnote ³⁰ The institutions of the city – theaters, circuses, and arenas – were, among the Greek speaking populations of the empire at least, the signs of civilization, and they might easily be imagined as the physical instantiation of law, a reflection of privileged status, and a source of collective pride in one’s political community. The Sifra, then, gives a fairly workable definition of how a substantial part of the population of the empire imagined the law of the empire: as a habitus, a package of traits that determined how one ought to live. What is more, such a definition would have made sense to a Roman, even to a Roman jurist, who would also have recognized that, even among communities of juridically “Roman” status, there would necessarily be variation in local customs, which would have to be accounted for in the absence of positive law.Footnote ³¹

The point, then, is not that one definition is tendentious and the other is not. They both are. We might say that both Gaius and the Sifra, and many other texts besides, engaged in struggles to define the content and boundaries of law. To privilege one account a priori – normally, in modern literature, the account of Gaius – comes at the cost of missing this broader hermeneutic struggle. It also, if perhaps quietly, tends to give priority to the Roman imperial state, simply, so far as I can tell, because it claimed to be the final authority on what sorts of violence against subjects might be taken as authoritative. As such, it leads scholars to take institutions – courts, law-creating organs, personnel, even rules and statutes themselves – as presumptively authoritative and binding because they emanate from the power of states to actualize their claims, either through persuasion or violence. To be sure, the best institutional histories can show, often in great detail, that a given institution might work better or worse, persist or transform, and so on. But the ways that institutions operate as claims upon spaces is often elided.

But institutions never just are; they are the product of particular human desires to organize a space or a practice in a particular way. They are therefore also the product of normative claims – that justice is done here, at this time, and in this way not because that is simply how it is done, but because that is how it ought to be done. Institutions, one might say, complete the act of deixis in a particular way. They attempt to instantiate or reify a particular act that defines law as one thing but not the other. To be sure, this often works: people can operate through institutions because they find their claims persuasive or profitable, or because they have been induced – or terrified – into compliance. They can naturalize the existence of some institutions, to the point where competing ones might be perceived as unnatural or abnormal. They can even be convinced that an is and an ought are identical: that the way that “we” solve problem X is just the way problem X is solved.

But when we see such claims we ought to imagine that behind them sit – or at some point sat – processes of struggle, competing acts of deixis, and alternative modes of imagining that have been effaced or made illegitimate. In other words, for every Gaius, there was also a Sifra. If Roman historians have tended to privilege Gaius and not the Sifra, we ought to ask why. And we cannot assume that the answer is just, “state violence,” or even the forms of authority that follow on the back of such violence. We need instead to interrogate the category, ask what it can and cannot do, to whom and under what circumstances, and with what sort of effects for governance more generally.Footnote ³²

***

Legal history in the institutional register has a further shortcoming. In its acts of mapping, it does a very good job of reconstructing certain phenomena that admit of ready reconstruction: the existence of courts, the careers of officials, the edicts of emperors, and so forth. These things tend to be especially amenable to mapping precisely because they claimed official status and were thus amenable to memorialization. But it does a rather poorer job of explaining how law came to be a register for creating meaning in the first place. Yet throughout the Roman world, discussions of law – along with discussions and stories of trials, of judges, and of other “legal” phenomena – came to serve as loci of meaning, as generative of theories of governance and of empire more generally. That law came to serve as a mode of meaning-making and political theorizing, largely independent of legal institutions themselves, calls for explanation.

We might consider two examples. Both are what we could call legal storytelling or even mythmaking: foundation stories that use legal action to point towards other phenomena of government and governance, and in so doing serve to make sense of, and to make claims upon, the world of empire. Mythmaking of this sort was a common practice throughout the empire, and the stories range from brief vignettes to elaborate, multi-stage narratives, encompassing a wide range of genres.Footnote ³³ I shall have more to say about such mythmaking in what follows, but for present purposes I want to single out two texts to give some sense of the ways law is generative of meaning in other spheres, and to probe the broader consequences of this form of meaning-making. The first text comes from the learned second-century florilegium of Aulus Gellius, who recounts a story about the emperor Hadrian, confronted in Rome with a difficult case:

One can read Gellius’ story in two ways. One can read it through lawyerly, institutional eyes, as a “case”: in this sense, we might say that it is a story in which a legal outlier (an unquestionably virtuous woman with an exceptionally long pregnancy) comes before an emperor, similarly renowned for his legal innovations and good character. Faced with an exception to the rule, the emperor chooses not to apply the law as written, originally, by the decemviri (the authors of the Twelve Tables, the original source of Roman statutory law). Rather than subordinate the demands of justice to a body of ancient writing, the emperor can show his innate sense of justice by extending the law.

Such a story is meant to be exemplary, to serve a pedagogical purpose. It insists that a good emperor will behave in a similar way; it also reminds readers that it is only emperors – rather than others – who have the temperament and moral capacity to exercise such a prerogative. Read along these lines, as an example of legal mythmaking, the story fits comfortably within a larger framework in which good emperors, through their ability to recognize virtue, contribute to the body of law itself.Footnote ³⁵ On such a reading, this vignette works to explain the internal development of the legal system. This system develops through the intervention of virtuous men like Hadrian who seek to achieve substantive justice by transcending the confines of written tradition – but only when necessary. On such a reading, Gellius’ story is the kind of legal myth that one would tell a law student: it seeks to domesticate the actions of an emperor by providing an ethical framework that explains them. The emperor changed the law because he was a moral actor faced with empirical realities that demanded remedy. More importantly, it explains these interventions as a necessary and natural part of jurisdiction more generally.

But there is another way to approach this story. We might begin by noting that stories of miraculous fecundity seemed to attach themselves to the emperors, Hadrian in particular. Jurists repeatedly recounted these stories – an odd fact, since jurists were in general loathe to comment on such sorts of paradoxa. Multiple or unusual births, we might say (with apologies for the pun), served as a fertile locus for thinking through some particular aspect of imperial power. So far as the jurists’ stories go, they all turn on the relationship between the natural and the unnatural and the extent to which law might be made to account for the latter.Footnote ³⁶ This is, to a certain extent, what is going on in this case: the question is whether the emperor ought to change the law itself in order to account for the outermost possibilities of the human capacity for parturition.

But Gellius adds an important twist. The reason for changing the law was not because the emperor had an innate sense of justice or equity, but rather because he had done his reading. Because he went to consult the “ancient philosophers and physicians,” he could effect a change in the fabric of law: accordingly, he ruled (statuere) that birth was indeed possible in the eleventh month. Note that, on Gellius’ telling, Hadrian is no longer ruling on the particular case of the woman – that is, he is not announcing that, for legal purposes, a pregnancy in the eleventh month after the death of a husband will simply be counted as, or deemed to be, the product of a marital union. Rather, he rules “that it was possible for children to born in the eleventh month as well.” Law, on this reading, does not expand or change simply from the personal good (or ill) will of the sovereign; it operates, ideally, in the light of “scientific” learning external to it, and defers to learned consensus.

Such a story is a myth of legal legitimacy. It uses a moment of judgment to comment on the conditions of law’s legitimacy vis-à-vis other, competing modes of learning. Gellius told it in the context of a larger discursive system in which similar stories were told and retold, to competing purposes. Within such a system, stories of law are never purely descriptive. They are always interested, ideological, and eventually political. Like the acts of deixis to which I referred earlier, they isolate particular moments that they deem to be exemplary acts of lawmaking, and then narrate these moments to teach a broader lesson about the foundations of legal and political authority more generally.

For much of the population of the Roman world, I shall argue, these sorts of myths or stories were the most important ways to frame these larger issues. Lacking a contemporary social scientific vocabulary to talk in the abstract about institutions, institutional performance, and legitimacy, the inhabitants of the Roman Empire deployed the narrative, the exemplum, and the myth as tools to make sense of their political, legal, and institutional worlds. It should quickly be added that this is no shortcoming on their part: among the most powerful ideas of history have been parables, fables, or aphorisms. Narratives, as every historian knows, focus the attention on salient matters and make them vivid; storytelling (and the additional interpretive endeavors – the commentaries – that stories invite) is a powerful tool for explaining the world in which one lives and for imagining the world in which one might.Footnote ³⁷

I should like to take a very broad view of such mythmaking in what follows. These myths could be communicated in texts, and indeed this is the way that contemporary scholars encounter them. But they did not have to be. Often such stories were performed in public spaces, displayed as tableaus for inquisitive audiences, for the purpose of teaching lessons. Sometimes, as scholars like Maud Gleason and Kathleen Coleman have made clear in the case of capital punishment, these tableaus were meant to inspire a sense of horror and dread – a fear of the consequences of not respecting legal authority – as well as the satisfaction of knowing that it was ultimately someone deserving who was being so violently disposed of.Footnote ³⁸

But myths can be performed in many ways besides, and throughout the world of the Roman provinces, they often were. My second example of the ways law comes to serve as an important symbolic resource comes from a text, Philo’s Against Flaccus, purporting to describe a public ritual in the city of Alexandria in the early first century AD. The background to the story is the worsening relations between the population of Jews resident in the city of Alexandria, and the Alexandrian citizens themselves. A catalyst for the riots that soon followed was the visit to Alexandria of Herod Agrippa (Julius Agrippa), the tetrarch in charge of northeast Galilee. Though the deep causes of the tension between the Jews and the Alexandrians were complicated, in this case, the arrival of a Jewish monarch in Alexandria was particularly problematic because it raised the specter of a realignment of the relations between the two communities. The Alexandrians had been treated badly in the course of the Roman conquest of Egypt, some sixty years before, while the Jews had been given privileges. In the intervening years, a modus vivendi was worked out, where the two had agreed to tolerate each other, but probably only at arm’s length. Agrippa’s arrival threatened to upset the balance.Footnote ³⁹

According to Philo, Alexandrian motives were far simpler: the Alexandrians were incensed “that a Jew had become a king,” and began to mock Agrippa publicly. Part of this came in the form of mimes and jests but eventually it took the form of a public performance, an inversion of Jewish sovereignty:Footnote ⁴⁰

Like Gellius’ account of Hadrian, this too should be read as an attempt to make sense of a moment of judgment. But here the context is different: rather than express their understandings through written texts, the Alexandrians recreate, as a public spectacle, the external trappings of royal judgment to highlight their defective content. Their “king” is not a king but a madman; his clothes are not royal but ragged; his scepter is not granted but found. Nonetheless, they make a point of greeting him, submitting cases to him, and petitioning him about the affairs of their community. On Philo’s telling, the stagecraft required the efforts of a significant number of members of the community, all in on the act: the bodyguards carrying sticks, the petitioners, and the multitude who shouted to him in mock Syriac. All colluded in crafting a scene of the inversion of royal jurisdiction. No doubt such a performance was also designed to engage an audience, and to send a clear message to anyone who might have been either indifferent to, or perhaps even pleased by, the imminent arrival of Agrippa.

Whereas Gellius focuses on the substance of the decision reached (and the process for reaching it), the Alexandrians focus instead on the “frame” which makes jurisdiction possible. Such a frame, of course, would have been immediately understandable to anyone living in the Roman provinces: the advent of the emperor or the governor, accompanied by formalized greetings, was an affair similarly symbolized by the hearing of cases. To submit to the jurisdiction of another meant to acknowledge the other’s superiority and sovereignty. For the Alexandrians, with their wry spectacle and linguistic play, the point was to use this scene to mock Agrippa’s ambitions to sovereignty more generally, and to highlight that any Jews who welcomed him as a monarch needed to be put in their place. On their logic, handing a petition to a Jewish monarch was essentially the same as handing one to the local derelict. Moreover, the very presence of such a person in their city – even in spite of his close connection to the emperor – was an affront to their dignity, and to the dignity of their governor (Aulus Avilius Flaccus).

Jurisdiction thus came to serve as a potent locus for thinking about the nature of community, and the nature of the hierarchies within a particular community or between particular communities (in this case, an Alexandrian community on top, Jews and others on the bottom). To rephrase my claim just slightly: enacting a story – a myth – about law through the degraded body of Karabas was, for the Alexandrians, a way of enacting their understandings of the ideal relationship between imperial sovereignty and local communities more generally. By a performance of jurisdiction, a political problem was laid bare and made tangible. In this case it was done through play, a ludic act binding together the Alexandrians as a community, producing the very thing it depicts and absorbing the participants and audience within it. As they play, the participants are absorbed in the moment and engaged pre-rationally; the experience is cathartic, even transcendent. It is also pedagogical, teaching a moral lesson – and offering a threat – to those not in on the act. If nothing else, law here comes to be thick with meaning.Footnote ⁴¹

It might be objected that in pairing Gellius’ Hadrian – thoughtful, learned, and fair – with the riotous Alexandrians, I compare apples to oranges: that there just is a difference between the substantive ruling of an emperor and the frenzied mockery of sovereignty, and that the one really is law (a formal ruling on a particular case) and the other really is not. To be sure, one (the ruling of Hadrian) was written down and placed in some sort of an archive. The other was, by its nature, ephemeral. But while this may be, the broader objection – that combining the two texts makes an error of categories – privileges, a priori, a definition of law that is coterminous with what the state says it is, and in so doing begs the question. As I shall argue in what follows, the locus of legal creativity – of what Robert Cover once called jurisgenesis – was nowhere near as secure, in the first three centuries AD, as jurists, ancient or modern, might wish it to be. Like all pre-modern states, the Roman Mediterranean was prolific of sovereigns.Footnote ⁴²

Finally, what is “real” is not the same as what is “official.” The Alexandrians, at least, were quite aware of this. Others were as well. In a world suffused with official documentation, and where official documentation often made claims about reality, there were nonetheless voices of doubt about the potential for mismatch. Thus Lactantius, in describing the last days of the emperor Valerian, could make a sophisticated critique of the mismatch between reality and official documentation:Footnote ⁴³

But one could go further and say that, insofar as it resides in and is expressed through language, law has to be understood as sitting in relation to a broader world of rhetoric and narrative. To recognize something as law relies upon stories, often unarticulated or enthymematic, about the history of government, the nature of power, and the conditions of legitimacy that make a ruling recognizable.Footnote ⁴⁴ Hadrian may be framed as making a statement of law, a simple claim about what will be accepted as true, ceteris paribus. But to know that it is law requires understanding him within a historical genealogy of imperial power in Rome and being able to hear his statement about the possibilities of parturition as being spoken within a register that renders his reading binding and decisive. To hear him making law, in brief, requires that we see him emplotted within a broader, historical, political and ethical narrative that enables us to make sense of his behavior as introducing a sensible reform rather than malignly barking imperatives.Footnote ⁴⁵ The Alexandrians, of course, had their own narrative – a similarly political and ethical story about the nature of community and sovereignty and justice, and about the relation between monarchy and local communities. Such a story had its own cast of heroes and villains, and contained its own understanding of the relationship between sovereign power and community consent. For the Alexandrians (and many others besides, as I will argue), law did not consist in sovereign commands or rulings. Law rather emerged through dialogue between subjects and rulers. And to engage in dialogue with a man like Agrippa was to engage in dialogue with a lunatic off the street; like wrestling with a pig, nothing redeeming could come from such an interaction.

Accordingly, if we choose to elevate to the status of law one (Hadrian) and not another (the Alexandrians), then we have acquired the burden of explaining this choice. To explain it on the grounds that one narrative is “official” and the other not is, to my mind, an exercise in circularity. It also does violence to the sources (for Gellius himself makes no claims to report “official” acts). Perhaps more to the point: both of these texts deserve to be paired, since they are both acts of legal exegesis that seek, through their narratives, to comment on what they took to be the decisive texts of the law: the decision of Hadrian, in one case, and the person of the (purported) sovereign, in the other.

What counted as “legal” was a moving target in this age. There was no single text, only commentary. Or rather, there was a proliferation of acts of deixis that pointed to some texts and tried to persuade others to raise them to the status of the texts that deserved commentary. I mean for this to be taken in the strongest, most anti-foundational sense possible. When we seek to write the legal history of the Roman provinces in the first few centuries AD, we are faced with a world that is radically decentralized in comparison to modern systems of law (which are anyhow nowhere near as centralized as people might hope), a world in which there was no universal consensus on what ought to count as law, and no Ur-texts. It was commentary all the way down, as competing understandings of law pressed against one another, animated stories and myths, and produced a struggle over institutions and their meaning more generally.

***

With that said, people nonetheless were keen to make claims, and to try to find foundations. This leads to a further point: any account of how law operates has a theory of political order at its base. Arguing about the nature of law, at least at this level of generality, raises foundational questions of political order: who are we, who are you, and what rules guide the interactions between us? These questions of politics emerge from questions about law and will be a fundamental part of what follows.

The inhabitants of the Roman world were divided as to what answers they should give to these basic questions, and through their storytelling they engaged in what deserves to be called political debate. Over the course of the first four centuries AD, they imagined law according to four major groups or patterns, four ways of conceiving “law as.”Footnote ⁴⁶ To conceive of “law as” involves a way of imagining the foundations or anchors of the legal action more generally; it involves a commitment to a particular vision of politics. But it also involved the sort of deixis that I flagged earlier: a set of assertions that there were, in fact, foundational texts that ought to be raised to canonical status. These stories often echoed one another, even when they did not overtly acknowledge competing viewpoints. But the echoes are sufficiently closely patterned that, when we can identify them, we ought to take them as salvoes in a broader political conversation.

In what follows, I identify four such patterns. In the Roman world, law could be imagined as documents, dialogue, ecstasy, and books. The first two groups (documents and dialogue) for imagining law sought to embrace its potential: for those who imagined law as anchored in documents and dialogues, legality offered a rhetoric – a repertoire of ways of speaking – that enabled speakers to imagine themselves in some sort of better future. (Rhetoric, as Aristotle reminds us, is always about what we should do, and how we should understand; even when explaining the past, it always smuggles in normative claims about the future.)

Part I (Text) begins with the world of archives and documentation and their relation to civil litigation. What I will call, faute de mieux, a “subaltern” class of provincial subjects, believed that law was founded upon the text of official documents that could be recovered from local archives. For these litigants, law was not so much a rationalized structure as a running list of privileges and disabilities, each of which could be expanded or contracted to fit the needs of a given occasion. The consequence of such a belief, however, was that it produced a politics of archival knowledge, one which the Roman government had difficulty controlling. Archival knowledge was highly localized and heavily dependent upon modes of reasoning advanced by subjects themselves. Those who saw law as anchored in documents conceived law as a running list of privileges and disabilities; activating those privileges depended on provincial subjects offering to imperial officials an account of how best to use and read archival sources. Through these readings, they argued that the state could be bound by the quotation of language. They negotiated rulership in the same way that they sought to bind gods and neighbors through incantations and curse tablets, except that in the legal arena they had an advantage: the sources from which they quoted were local in nature.Footnote ⁴⁷

While the litigants of Part I were skeptical of the idea that law was a rationalizable structure, others disagreed, especially when confronted with the high-stakes world of criminal judgment. Thus, in Part II (Dialogue) I turn to the criminal courts, the most highly visible acts of Roman justice in the provinces, and institutions designed to impress upon provincial consciousness the power of the Roman state. However, such acts by the Roman state were not simply deemed virtuous or not; rather, by writing about criminal trials (and re-writing the stories of such trials), provincial populations reimagined these rituals of sovereign justice as moments of dialogic encounter between subjects and governors. By framing law as a dialogue (rather than simply an act of sovereign will), subjects demanded that, as a condition of its legitimacy, law’s violence be founded upon logos – upon reasoning, expressible through language, rather than upon acts of spectacular violence.

The face-to-face encounter might provide a testing ground for provincial ideas about legitimacy, but it might also be adapted to other purposes as well. Thus, stories began to circulate about other characters who confronted the powerful through the same rhetorical moves they had developed in the courtroom. Thus, in Chapter 5 I turn away from the courtroom to show how dialogue – the testing of the judgment of the powerful in face-to-face confrontations – became a way of thinking through claims about domination more broadly. Chapter 5 pays the most attention to trying to pick apart this very subaltern political theory.

In response to these claims, counter-ideologies emerged, which I discuss in Part III. Some members of both Jewish and Christian communities began to experiment with escapist ideologies of legality, imagining a world in which genuine legality – and a properly ordered polity – might be constructed independently from the administrative machinery of the empire (Chapter 6). The courtroom similarly terrified the elites of the empire (Chapter 7). It could easily reveal that they were, in fact, no better than those whom they dominated: both could bleed, and governors often took pleasure in showing that certain elites were, in fact, capable of suffering punishment and degradation. Accordingly, these urban elites sought to avoid law’s judgments by crafting an ideology of elite male sociability that encouraged the avoidance of such spaces. They argued instead that conflicts ought to be resolved through extra-legal practices, and in particular through the use not of the legal register of speaking, but of a different register altogether. The elegant speech that they generated – in reaction to legal speaking – was, in turn, connected to a package of techniques of the body, which marked men as sharing in elite culture, or paideia.

Paideia was a category that enabled such elites to judge others simply by observing their bodies, and to determine whether they were noble or base, worthy of punishment or deserving of sociability. In short, paideia denoted not just “education,” but a system of physical and mental training by which such men could settle disputes, negotiate with imperial powers, and discern the character of those they dominated. Hence, among a number of urban elites – and among some sub-groups of Jews and early Christians – we find emphasis on law as a practice that demanded ekstasis, in the sense of the Greek word: as requiring transcendence, an ability to stand apart from the quotidian.Footnote ⁴⁸ For such men, this transcendence would constitute “real” law, something that necessarily differed from law as practiced in the quotidian administrative paper-pushing that otherwise characterized the Roman world. These ideas of law as ecstasy, however, were not sui generis. They emerged as a reaction to a world where the courts hummed with business.

These elites additionally began to outline a concept that would compete with the broader provincial understandings of the nature of legal rationality: thauma. We might provisionally define this as “wonder” or “amazement.” As is fitting, such a category sits uncomfortably with legal traditions, which almost universally prize transparent, easily reconstructed chains of reasoning. Thauma, by contrast, is an affective category: someone experiencing thauma is placed outside of the space of everyday reasoning, and brought, as it were, to a higher plane. The wondrous, like its cousin, the sublime, transcends the normal experience of reasoning and marks itself as operating in a space apart, governed by unique rules. According to the ideology of these elites, achieving this affective state would suspend the normal administrative functions of a bureaucratic state and bring men, otherwise in conflict, into a state of deep, affective interconnection. As such, “amazement” came to operate as an extra-legal, if not anti-legal category – one that was, by virtue of its capacity for emotional supercharging, superior to law itself. It came to be a model for how an empire ought to work more generally.

One important takeaway from the distribution of these legal ideas across the landscape is that their use does not map easily onto many of the normal axes of analysis through which we divide the population of the Roman Empire as a whole. That is to say, it is not the case, for example, that Christians only thought of law as documents, whereas Jews only thought of law as dialogues; or that urban people thought of law as ecstasy, while rural people thought of law as documents, and so forth. The way one imagines law, I shall suggest, is best taken not as an index of some sense of class, religious, or ethnic belonging, but as a mode through which one imagines one’s relation to empire, government, and community more broadly. Thus, I shall show that we can find Christians and pagans, wealthy and weak, all imagining law, for instance, as dialogue; we can likewise see a convergence between Greek hyper-elites and rhetoricians, thinking about law as ecstasy, with some trends among early Christians, who made similar imaginary moves (albeit to different ends). Thus, my terms “elite” and “subaltern” must be taken with a sharp qualification: I use them not to refer to actual social status (many of my “subalterns” were, in fact, quite wealthy, and of high status), but to the ways in which individuals and communities imagined themselves in relation to imperial power – whether they felt themselves fundamentally insiders or outsiders.

Parts I–III of this work are devoted to explaining the ways in which claims about law served a political function. The diachronic arc of my story picks up primarily in Part IV. Roman jurists and later on Roman emperors, starting in the late second and early third centuries AD, came to be increasingly interested in thinking of law primarily as books. This was not an effect of the quality of jurisprudence, or even of the growth of bureaucracy or the spread of citizenship. It was, rather, a reaction to the challenge of governing a diverse and fractious populace with multiple and diverging commitments to, and understandings of, law and its foundations – the protagonists, that is, of Parts I–III. Jurists and emperors reacted to the claims of such people by trying to monopolize hermeneutic authority over law. In collaboration with jurists, emperors reimagined law not as a set of practical solutions to problems that emerged from concrete interactions with subjects, but instead as something ideally abstract, codified, and certain, a finely textured, self-replicating web capable of an increasing scope of constraint. They also imagined it as something fundamentally distinct from, or even superior to, politics.

It was through this process of pushing back against local understandings that the Roman state transformed itself from a diverse and heterarchic polity into something more rationalized and hierarchical – into something that we would recognize today as a bureaucracy, accompanied by a rule of law. This process, while hardly linear, was substantially completed by the end of the fourth century. In other words, to the extent that the Roman state moved in the direction of making law a codified and rationalized “system,” it did so not because it had any intrinsic preference for such a thing, but because it sought to place limits on the number of voices that could contribute to the body of law as a whole.

As a result, beginning with the legal revolutions of the Severan period (AD 193–235), and extending until the late fourth century, we see not the great opening up of the Roman system (by granting citizenship to all of its inhabitants, or by rationalizing provincial administration), but rather its great closing. It was a moment – or rather a succession of moments – in which the state tried to monopolize authority to create and enforce law, to achieve hegemony over legal interpretation, and to put an end to political argument.Footnote ⁴⁹

The effect of this government reaction, I shall argue, is an important historical horizon not only for the Roman state, but for the way that we conceive law more generally. In this act of seeking to close off alternative modes of imagining law, the Roman state developed the kernel of a distinction that haunts us today, namely, the distinction between law “in the books” and law “in action” – the latter always a derivative and defective enactment of something ideally possessed by experts.Footnote ⁵⁰ Two outcomes resulted. First, this process ensured that the voices that had previously constituted law could only now be seen as reflecting it. The second result was a fetishization of legal knowledge and with it, a perverse form of lawyer-worship that casts a long shadow over legal history even today. Namely, the belief that jurists, in the words of Ulpian of Tyre, are uniquely placed to “attend to justice, profess knowledge of the good and the fair, divide the fair from unfair, and separate the permitted from the illicit, inclining to make men good not only by making them fear punishments, but also by urging them towards rewards – practicing, I should say, true philosophy, rather than the imitation of philosophy.” For Ulpian, this special talent made the jurists not just learned men, but rather something like priests. It also earned them a place close to imperial power.Footnote ⁵¹ The historical roots of this vast and arrogant claim shall be taken up in what follows, with attention to the historical processes – and their attendant acts of violence – that made it not just sayable but also convincing.