Version classiqueVersion mobile
OpenEdition Books

Religion, Institutions and Society in Ancient Rome

John Scheid

Religion, Institutions and Society in Ancient Rome

Inaugural lecture delivered on Thursday 7 February 2002

John Scheid
Traduction de Liz Libbrecht

Texte intégral

1Mr Administrator,
My dear colleagues,
Ladies and Gentlemen,

2The Romans never departed from their custom of paying homage to the gods every time they started a new activity. Our predecessors in these premises, like the first royal lecturer of Latin Eloquence, Barthélémy Latomus born in the Duchy of Luxembourg, were accustomed to celebrating the founding king of the Collège de France. How could we not subscribe to traditions such as these when, like me, we are concerned in two respects? I will therefore start by respecting the rite, in paying homage to the Republic, and I have two reasons to do so. Firstly, I would like to express my gratitude for the liberality with which it has welcomed me for nearly twenty-five years, and for the generosity with which it hosts many foreign researchers and lecturers at French universities. As early as 1789, in Éclaircissements sur le Collège royal de France, Abbot Jean-Jacques Garnier stressed that the Collège’s agenda should be international in scope, and that it should recruit its professors without any distinction of citizenship or origins. When I look at the rows of my colleagues, I see that this aim has been more than fulfilled, both here and in French academia in general. In the age of European construction, this type of openness warrants attention and praise. We have come a long way since the first decades of the Collège and the animosity aroused in 1534 by the nomination of Barthélémy Steinmetz or Lemasson, known as Barthélémy Latomus.

3Secondly, I would like to thank the Republic and the Assembly of Professors of the Collège for having created a Chair devoted to Ancient Rome and to the religion and institutions of a dead civilization. Nowadays, such generosity cannot be taken for granted. Many people in France and Europe consider that the study of this type of civilization is not a priority. It is therefore not rare to see countries and universities closing rather than creating research and teaching positions in this field as budgets are cut and trends change. Yet the importance of the role of Chairs and teaching devoted to Antiquity cannot fail to appear obvious to anyone who considers the recent history of Europe and the world.

4This is not only a matter of defending what is known as the study of the Classics in secondary education and of insisting on the quality of this training. It is true that Latin and Greek make it easier to learn French, to understand the grammar and to reason. But let’s be fair, if French is taught properly it can render the same service. The importance of training in the Classics lies elsewhere. Because it forces one to look beyond national borders, it constitutes an ideal access to the common cultural heritage of Europe and far beyond. The Roman Empire, let’s not forget, stretched from Gibraltar to the Black Sea, from Great Britain to the Sahara.

5As it has been the case in Europe, the Middle East and elsewhere, for the preceding two centuries until today, historical heritage is a major issue in identity-related conflicts – with dramatic and generally tragic consequences, as we know. Since the 19th century at least, historical and religious heritage has been closely linked to the identity of states and nations, which, in their origins and founding myths, project that which they define as their essence. In this quest for a “pure” identity, sites and documents from the past could be assigned a highly dangerous function. To avert this ever-present risk of a pernicious exploitation of history, states cannot afford to do without a scientific institution capable of providing relevant and, as far as possible, accurate answers to these issues. More precisely, this heritage requires specialists who can manage and pass on the sites and texts composing it, and who have the means to control the use to which they are put. But it also requires the training of graduates who, in their profession, are able to understand the discourse of these experts.

6The study of dead civilizations is therefore, in my opinion, an essential tool for maintaining control over assertions concerning the identity of both one’s own community, country or continent, and those of others. You will not be surprised if I tell you that the history of religions from the past – or religions in the past – has the mission of playing a key part in this control. By opposing sectarian discourses with the universal weapons of history, philology and anthropology, in short, the entire arsenal of science and reason, the history of religions of the past enables us to deflate modern myths, and not only those of others but also our own. It allows us to identify the projection, in the imaginary past, of the “origins” of nationalist, religious or racist fantasies, and to disarm exaggerated interpretations of the sacred texts. Within nations inherited from the 19th century, ancient history can help to deconstruct the representation that nation states sometimes create of their past, by showing that despite their apparent proximity, their “ancestors”, often simply assumed to be so, were as distant from the current society as the inhabitants of the antipodes, and hardly resembled the image assigned to them. It enables us to challenge the “Greek miracle”, the “Roman genius”, the “Germanic superiority”, or the Hegelian dialectic professing that religions and history tend towards Christian monotheism. Closer to us, it puts into its rightful place the irreducible “Gallic character”, the underpinnings of which have been brilliantly dismantled by Christian Goudineau – who has paid me the immense honour of supporting my candidature in this institution.


7It is probably about all this that you were thinking, dear colleagues, when you created a Chair devoted to the Religion, Institutions and Society of Ancient Rome. How, starting from there, can I fulfil the mission that you have imparted to me?

8Roman history has existed for a long time in this institution, since Léon Renier, René Cagnat, Stéphane Gsell, André Piganiol and Paul Veyne, to mention but a few prestigious names that, along with Barthélémy Latomus, can put to shame anyone who claims to teach publicly in this place. Is the idea then simply to carry on their teachings? Of course not. Each generation writes its own history, and you, dear colleagues, are expecting today’s Roman history to be different from yesterday’s.

9“History exists only in relation to the questions that we ask of it”, said Paul Veyne in this very place twenty-seven years ago. And in his inimitable style, he wondered what questions we should therefore ask. As he saw it, the historian’s profession had two main aspects: erudition and conceptualization. “Any historiography” he said, “depends on the questions that it sets out to answer, and the documents available to it”. Of these two pillars, Paul Veyne clearly chose to rely on conceptualization. Roman history seemed, to him, to lend itself to that particularly well, in so far as it “makes us step out of ourselves and compels us to explicate the differences that separate us from it”. Moreover, this history “is so poorly documented” that it “gives rise to ingenuity, which in turn spawns a new richness”.

10The researchers of my generation followed this proposition with enthusiasm, for steeped as they still were in classical culture, they saw new avenues opening before them in a field that sometimes seemed in danger of ossification. The inventory of differences was an incentive for a comparative approach, an important part of the teachings of the Collège de France and of the spirit of the 5th Section of the École Pratique des Hautes Études, where I took my first steps in academia and whose teachers, staff and students I see here with emotion this evening. Georges Dumézil had already retired but his work, as well as the conversations taking place in the chaos of his office, made me attentive to “similarities against a background of differences”, as he put it. He also taught me to be wary of the aberrations of certain comparative practices, which he firmly denounced. The intimate collegiality of the École Pratique, where the spirit that Marcel Mauss had breathed into it still lingered, was an incentive to pursue questions in the comparative laboratory. These comparative studies taught us to ask the right questions and to work with operative concepts. When, in his inaugural lecture, Georges Dumézil rose up against the superficial facility with which certain historians of religions were content, he was paving the way for what was not yet called religious anthropology but that he was already practising without naming it. At the Collège and at the Hautes Études, among others, my generation learned to manipulate the concepts and techniques of religious anthropology, under the supervision of Jean-Pierre Vernant, Pierre Vidal-Naquet, Charles Malamoud and Marcel Detienne, and last but not least Paul Veyne in relation to Roman studies. It was clearly in these seminars that I acquired the taste for rites and learned to question myself on myths. Antiquity gradually took on a new relief. It appealed to us all the more in so far as it moved away from us and became less familiar. We started questioning ourselves and bringing to light our prejudices concerning the universality of ideas and behaviours. Evolution itself seemed to us less straightforward and mechanical, for beneath superficial changes emerged structures – those structures so dear to Claude Nicolet – which change more slowly and in more complex ways than aging as such.

11Over the past thirty years, the new conceptualization, based on social anthropology and sociology, has profoundly transformed history and become renowned outside of France. It is no exaggeration to say that this is now how history is made. For a historian of Rome the situation nevertheless calls for some reserve. The generation that lived through the transformation of history as a discipline, under the impetus of new forms of enquiry, was perfectly comfortable in the second province of history, that of erudition. Paul Veyne himself was and still is a master in this respect. Like him, the historians of Rome were mostly students of Hans Georg Pflaum. In his seminars at the École Pratique, and under the roofs of the Île Saint-Louis, he introduced us with paternal severity to the methods of Roman epigraphy and the mysteries of prosopography. With awe we suddenly transcended the bounds of literary sources, to contemplate the structures of the Roman Empire by examining political and administrative careers sequentially. Historians of Republican Rome continued these studies under Claude Nicolet who, in the room next to the one where Hans Georg Pflaum officiated, reconstructed the structures of the pre-Augustean res publica and explored the multiple facets of the citizen’s role. This was an altogether more complex domain in that it required that all the resources of erudition be implemented before a new series of questions, adapted to the public law and institutions of the Roman state, could be formulated. Finally, on the other side of the landing, at the 5th Section of the École Pratique, Robert Schilling taught the vocabulary and spirit of Roman religion, based on detailed analysis of the texts. The generation of historians who matured in the seventies are therefore Janus-faced: one traditional, if you will, and rooted in the sources; the other drawn to the new methods of studying those same documents.

12What is the current situation? I fear that its evolution is cause for concern. Twenty-seven years ago Paul Veyne could say that in his day not much Latin was learned. What can we say today? Who could claim that classical languages are alive and well in France and in Europe? In spite of the support that it receives, the teaching of this discipline has been going through difficult times for over two decades. We are compelled to question the demagogic rage consisting in abolishing the study of languages even though they remain indispensable for anyone who has some interest in our historical heritage. And there is no point in maintaining this teaching if it becomes purely superficial or symbolic. Just as is the case for living languages, teaching worthy of the name exists only if the student effectively learns the languages.

13Historians are necessarily worried about the situation that they have inherited. How can one study Greek and Roman history without knowing ancient languages? How can one ask the right questions if one is incapable of verifying the facts, and especially if one is not able to derive concepts from sources that have not always been translated? Thirty years ago, historians of Rome could draw on their reading and classical culture to reconstruct the appropriate context for asking the right questions. Once they have lost the ability to use ancient languages, the history of Rome is very likely to resemble medicine in Molière’s day. Without the documents to support them, the best concepts remain inoperative.

14But let us not carry on lamenting in the festive context that brings us here together this evening. I would simply like to draw attention to a serious problem that is threatening the discipline itself and not only ancient history. Insistence on the primacy of documents brings us back to the very origins of the Collège de France. The mission of the first Chairs of classical languages was, as you know, to make the Writings intelligible in the agitated context of the 16th century. In his inaugural lecture on the humanities, Barthélémy Latomus pointed out that many passages, “from the divine Writings […] remain unintelligible and […] we cannot comment adequately on any of their parts without knowledge of the humanities”. I am afraid that this fact and this mission remain unchanged, even if nowadays it is not only the biblical text that has to be made intelligible but, more generally, all classical sources.


15Nothing is more deceiving than the study of past religious practices, and nothing is more instructive. Whether they practise a religion or are atheists, historians today have their minds formed, or deformed, by fifteen or sixteen centuries of Christianity. And our inclinations continue to tend instinctively towards whatever conforms to this culture. One of the most compelling examples is the way in which the votive rite practised by the Romans has been interpreted.

16As we know, the Christian vow consists of a promise which may or may not be linked to a request, but which is in any case immediately binding, whether God fulfils the wish expressed by the believer or not. The same did not apply to Roman religion. A vow was a contract with a divinity, or at least a conditional promise, stated before witnesses, generally in writing, then sealed and put in a place of worship or in archives. This contract stipulated precise conditions and set a term. If the historian is content to note merely the occurrences in texts where vows are formulated or fulfilled, he or she runs the risk of a misinterpretation. The great German jurist Alfred Pernice thus sought to show, a century ago, that the Roman vow was immediately binding. Yet one need only research and read in detail the votive stipulations known to us, to see that that was not the case. The vow became binding only once the condition had been met. Several testimonies even indicate that if the divinity failed to satisfy the stipulated condition, the vow was cancelled and a new vow substituted for it. The principle of the Roman vow does therefore not correspond to the do ut des “I give so that you give”, but to the da ut dem, “give so that I give”. An anecdote will illustrate the specific nature of this practice. Under the Empire, it was customary on 3 January of every year for the different authorities of the Roman state to celebrate public vows for the emperor’s safety. They were supposed to start by fulfilling the vows made the previous year, before formulating new vows for the year to come. In January 101 AD, then again in January 105 AD, when the Emperor Trajan had to go off to war against the Daces, in current Romania, public vows regularly formulated for the prince’s safety did not include fulfilment; they were limited to the expression of a new votive promise. By deciding not to make sacrifices previously promised, the celebrants wanted to make the threat weighing on the prince and the Empire manifest. Faced with the ineffectiveness of their request, they broke the votive contract that they had concluded on 3 January of the previous year with Jupiter, Juno and Minerva, and formulated a new contract: da ut dem, “give so that I give”. Previously, during the Second Punic War, the Romans had already on several occasions postponed the fulfilment of an important vow that had been made for five years, in 217 bce, at the time of the confusion following the military disaster at Lake Trasimene. They decided to defer the fulfilment of their promise for as long as the gods had not satisfied the clauses of the vow demanding Roman victory over the Carthaginians and the Gauls. This happened in 195 only, after the celebration of Scipio’s triumph over the Carthaginians in 201, and then those of Gnæus Cornelius Cethegus, Quintus Minucius Rufus and Marcus Claudius Marcellus over the Insubri, Cenomani and Ligurian Celts in 197, 196 and 195, when the pontiffs finally recommended that the Senate fulfil the vow made in 217. The gods had given; the gods received.

17Historians cannot discover all this with concepts alone. They will find it only by tirelessly examining all the available documentation. Let us continue with the example of the vow. By highlighting the specific characteristics of the Roman vow we end up formulating new questions. Take for example the issue of the cultural practices of a Roman people from Palmyra, who in the 2nd and 3rd centuries frequented a place of worship in Rome, in Caesar’s Gardens in Trastevere. We see the way in which these people juggled with Roman votive rites and the Palmyrian religious principles, well known to my colleagues Javier Teixidor and Michel Tardieu. When they expressed themselves in Latin, these Palmyrenians addressed Roman-type wishes to their national gods because they were Roman citizens and Roman ritual practices were theirs too. Yet when they transposed the inscription of the Palmyrenian rite next to the Latin text on the altar, they limited it to the dedication. In other words, they banned the vow and its contractual aspect, because a formulation of this nature was unthinkable in the framework of the relations that these men maintained, as Palmyrenians, with their national gods. This example enables us to pinpoint the dual religious culture of Roman citizens from the provinces of the Empire, as well as the double life of their national gods, who were also Roman on the one hand and Palmyrenians, Greeks, Africans or Gauls on the other.

18In ancient history, drawing on sources has the function that experimentation has in what are known as the exact sciences. In fact, researching ancient history consists in constantly re-running experiments which necessarily involve a limited number of elements, and which rely on highly variable parameters. Even if the researcher has had many, and sometimes renowned, predecessors, the results will inevitably fail to appear, so complex and insufficiently studied are the sources.

19We could assume that the Romans’ funeral rites are well-known. The sources are numerous: literary, epigraphic and archaeological, and over the past twenty years all the aspects have been discussed and interpreted at conferences and seminars. I thought that I had solved the problem of the sacrificial rites that take place around the tomb during a funeral and then during the annual celebration of the dead. Sacrifices celebrated the day of a funeral seemed to be similar to those that the family celebrated every year at the tomb, during the celebration of the dead in February, or on other commemorative occasions. By re-examining this subject, I realized however that the texts raised other questions.

20As is generally the case in literary or epigraphic descriptions evoking the most common rites of the period, testimonies are as numerous as they are allusive. The historian is therefore compelled to exercise extreme precision in exploiting rare and relatively explicit documents. Imprecise work, and especially work that is second hand and based on the conclusions of other historians, can only lead to a dead-end. Even with excellent concepts, no one could hope to obtain results by reinterpreting a colleague’s interpretation.

21Only a return to the sources can lead to new and better questions. When I was analyzing the subject of funeral rites, I focused more on the general framework of the ceremony than on specific rites that took place near the tomb. The questions were not poor but the approach was still too superficial and the view too general. I had to go into detail. It was my research on the sacrifice that made me sensitive to a number of data concerning funerals rites, to which I had not done justice until then.

22I was thus able to propose a more precise reconstruction of traditional Roman funeral rites, as they were practised at the beginning of the Christian era. When the deceased was incinerated, a sow was sacrificed to the goddess Ceres. This victim was shared between the goddess, the deceased and the family in mourning. Such sharing was possible only because, during this funeral phase, the status of the deceased and that of the family were not yet permanent. In other words, the dead person was not yet part of the Dii Manes, that is, the divine entity encompassing all the deceased that the family considered as its own. Thus, the sharing of food could bring together the deceased, his or her parents, and a divinity such as Ceres who, incidentally, was not an infernal divinity. This sharing nevertheless took place according to very specific modalities. The goddess received the part of the sow reserved for her, on an altar; the deceased’s share was placed on the pyre and burnt with the body; and the parents consumed their own share at a table laid next to the tomb. This was the beginning of an eight-day period about which we know relatively little. We understand that during this waiting period the deceased and his or her parents symbolically moved towards their respective statuses. The outcome of this transition was announced on the ninth day by two other sacrifices. The first was intended for the Manes of the deceased who had definitively joined this collective deity and could therefore receive a sacrifice. In line with the infernal character of these divinities the sacrifice consisted of a holocaust. The entire victim was burned on a pyre on the ground – a type of offering characteristic of the worship of the Dii Manes. On certain occasions, during anniversary commemorations for the deceased, a holocaust was the only rite that was celebrated. The wonderfully detailed inscriptions prescribing the funeral honours that the young princes of the Julio-Claudian dynasty ought to receive are fine evidence of this. Like the ninth-day rites, and along with the holocaust offered to the deceased’s Manes, the sacrifices celebrated every year on the occasion of the commemoration of the dead, in February, also included a sacrifice for the Dii Penates, that is, the divinities who protected the family. The meat was then eaten on site, by the living.

23The problems have by no means all been solved. We can presume that this reconstruction will itself be reviewed when a researcher approaches the same sources from a different angle one day and notices elements that currently still elude us. But we have to acknowledge that he or she would not be able to go much further in the explanation of the essentially literary and epigraphic sources that I have just mentioned. After twenty years of successive revisions, corrections could hardly improve on the current conclusions, especially as regards Rome at the beginning of our era, whence the vast majority of the available testimonies come. The situation is even more critical for later periods, and for the rest of the Empire. Sources are rare and allusive, and it would be hazardous to directly transpose data on the city of Rome to all the cities of the Roman world. To get beyond this reconstruction, we would need to turn to other types of data, especially those likely to be yielded by archaeology, a discipline that I had the good fortune – another favour of the Republic – of being able to learn and practice at the French School at Rome.

24The archaeological fieldwork that Christian Goudineau called for one evening in 1984 has, indeed, been entirely renewed over the past thirty years and fruitful dialogue has been established with history, notably religious history. Among others, funerary archaeology has made spectacular progress. Excavations of necropolises now use the most modern techniques of archaeological fieldwork, and collaborate with specialists in organic and vegetal vestiges. A host of new questions is thus arising. What was done to the corpse before incineration? How was it placed on the pyre? Was there a difference between the funeral of a man and that of a woman? What offerings accompanied the body, and how were they positioned on the pyre? What did these offerings mean about the sacrificial funeral rites? Is there a relationship between the written sources and the practices to which archaeology bears witness? How was the corpse positioned in the funeral urn and in the tomb? Are there traces of a regular funeral cult celebrated around the tombs? How did this ritual evolve in space and time? Were the rites the same for ordinary funerals? And so on. So many relevant questions have been asked, or rather have progressively been brought up by this new archaeology of rites. Only a very precise analysis of the material vestiges will effectively enable historians to go beyond the purely literary approach to funeral rites – provided, of course, that they know what they are looking for, and that the burials are well preserved.

25Seen from the outside, ancient history can thus give the impression of stretching unduly the scope of the few available ancient sources. That is not inexact. But it is precisely this ceaseless return to the same sources that constitutes the interest of ancient history. For this rumination makes one think. Whether they have a classical education or not, all Westerners have the impression that they know the Greek and Roman world. In fact, that is not so at all. The volume of documents is infinitely smaller than that of other periods of history, and often even the evidence on a particular issue is limited to a few pages of exploitable sources. But how do our colleagues who have an endless mass of sources work? Are they not also forced to make choices? Does the only difference not lie in the fact that scholars of the Classics cannot make choices? Their corpus of sources is largely pre-established, but fortunately not entirely. Despite the limits that restrict their freedom of movement, every time researchers take a closer look they discover something new that no one before them had seen in its entirety. All in all, their situation is fairly similar to that in which ethnologists find themselves when they return to visit a tribe that has already been studied by a predecessor, and find it very different to their colleague’s descriptions of it. This is how ancient history constitutes a science in the making and not a museum of received ideas. By its way of proceeding, it puts out a warning that can be beneficial to all. It highlights the dangers stemming from the impression of familiarity that a culture close to us gives, and denounces the facileness of superficial syntheses. In their daily lives, teachers, researchers and citizens alike operate with general ideas which are often, let’s admit, exaggerated or at least approximate, because these are inspired by emotion, ideological choices or even intellectual laziness. From this point of view, ancient history has the virtue of encouraging mistrust, and erudition is assigned a mission that is not limited to filling in footnotes.

26Yet the interest in sources, as fundamental as it may be, is not enough. This is not simply a matter of properly understanding a detail in a text or institution. As I noted when I started, without concepts there is no history, even with sources. But where are these concepts to be found?

27A recent trend, probably due to the weakening of education in the Classics, has turned researchers towards concepts elaborated by disciplines that study more recent periods. My aim is in no way to deny the advantages of these concepts and disciplines. The analysis of the dynamics of acculturation consequent to the colonial conquests can, for example, afford fruitful points of comparison. Research on the construction of the large nation states also makes very useful concepts available to the ancient historian, for assessing the evolution of the Roman state. Things get out of hand however when concepts, often forged by researchers studying civilizations and societies very different from those of the Ancients, are transposed without any precaution onto Roman data. This brings us back to the dangers of ill-mastered comparative studies, formerly denounced by Georges Dumézil. Research then risks creating pseudo-tools of investigation and explication that will never be fit to produce an acceptable historical conclusion.

28It is true that every generation shapes its own explanatory models. To stick to Roman history, the prosopography fascinated us in the seventies since it seemed to afford immediate access to the very structures of the Roman state and Roman society. This is partially true. With hindsight, I find however that it is true primarily for the reconstruction in previous decades of the Roman social and administrative framework. Reference books were published and lists of dignitaries of all kinds and all ranks were drawn up. These admirable works have constituted the tools that have been indispensable to historians since then. Yet, as is often the case in scientific adventures, it is less the answers to the questions initially asked that allow for the advancement of science, than the unintentional offshoot of the research itself. Although his main aim was to reconstruct the “meta-history” of societies that spoke Indo-European languages, in the process Georges Dumézil compiled the first religious anthropology of the Romans.

29The results produced by prosopographic research concerning the Roman elite provide us with another example. I have personally encountered the problems raised by the Roman religion by seeking to reconstruct the membership of sacerdotal colleges, starting with the Arval Brethren and then, with one thing leading to the next, the other colleges, with the intention to draw a number of comparisons. From the start it seemed essential to define a sacerdotal function that was very distant from our conception of it, because closer to the jurist’s and the magistrate’s function than to that of the priest, as we see it. But over and above the rational approach that my research was taking, it was the discovery of extraordinary and as yet unexploited sources – in short, a chance encounter – that prompted me to choose Roman religion as my field of research.

30Like any historian, I was nevertheless constantly confronted with my own prejudices and concepts. When it comes to religion, our minds are formatted in such a way that we talk of Religion, in the singular, as if it existed. To paraphrase Paul Veyne, religion exists no more than does history. It is all the questions that the historian addresses to the sources that construct religion as a historical fact. And the religion of the Greeks, Romans and Gauls will be what historians define with their questions. Yet, while religious practices are universal, their variety is infinite. And history lies in this infinite variety, not in the vague universality of behaviours. It is an illusion to hope to obtain a tangible scientific result by working with our own religious concepts. That is primarily why, for a long time, historians of religions were hardly convincing, compelled as they were, to a greater or lesser extent, to distort the sources in order to impose anachronistic points of view on them. It was far better to make close readings of testimonies and to study religious institutions and their own logic. That was the approach of the German Georg Wissowa, whose manual has lost none of its worth in a century. Even though certain facts are now outdated, the whole remains valid and requires only a few revisions in light of new documents. His synthesis continues to be helpful because Wissowa sought the spirit of the Romans’ religion in their religious institutions rather than in his own conception of the religious. It was in these terms, in any case, that in 1912, in the preface to the second edition of his textbook, he defended his reconstruction of Roman religion against those who argued that this set of practices, this Roman juridicism and ritualism, did not constitute a religion. In the eyes of a generation fed on religious anthropology, Wissowa’s choice seemed, on the contrary, excellent.

31Only the attentive exploration of religious practices in written texts and archaeological sources enables us to get closer to the heart of Roman religion and to identify the conceptual framework needed to study it. Rather than considering it from his or her own conception of the religious as a decadent, delinquent and manipulated religion, the researcher is gradually able to distinguish its true nature. He or she discovers that it is a ritualism based on an oral tradition, itself regularly updated by sacerdotal jurisprudence, in other words, by the adaptation of oral custom to precise cases. To understand this religion of rites, it is not necessary to speculate on the soul or metaphysics. When they practised their religion, the Romans were concerned with neither the survival nor the safety of their soul. They were celebrating rites intended to guarantee the well-being on this earth of the community to which they belonged. Not that metaphysical questions were of scant interest to the Romans, but they never raised them in the practice of their religion. Instead, they did so in an intellectual context outside of religious practice, like the Moderns who read the philosophers or the physicists of matter to reflect on the mysteries of the universe and creation, while adhering to a particular religion.

32We need to understand that rites per se have no meaning other than that attributed to them by their primary signification. The Roman sacrifice was effectively, and remained, a food offering. Rites nevertheless took on meaning with the comments that the Ancients constantly applied both to complex sets of rituals and to isolated segments of rituals. The lack of any orthodoxy and the absence of sacred books and of a single religious system meant that questions raised by rites, and consequently their answers, were countless, multifarious and sometimes even contradictory. Ultimately, this was of little importance, for in most polytheistic ritualism the rite constantly needs to be called into question in order to be kept fully alive.

33Once the obstacle created by our own conceptions of the religious has been removed, once we have rendered plausible the hypothesis that, until a relatively late Antiquity, the majority of Romans had little concern for the salvation of their soul when they practised their religion, a way opens up to a more reliable approach to Roman religion. This consists in imitating the Ancients and in giving rites the most probable meaning, the meaning which seems to justify the majority of practices and thus to reflect the feelings of most Romans. The logic underpinning behaviours and priorities is then revealed, religious roles take on meaning, debates and affairs become intelligible. And from then on it becomes possible to reconstruct a history of what we call Roman religion.

34It is in this sense that the study of this religion can be of interest to the historian and the educated man. It confronts him or her with an aspect of religious behaviours – ritualism – that has largely been repressed in Europe since the 16th century. In so far as the Romans are at the origin of many of our republican ideals, their wish to rationalize both the human and the divine, that is, the real and that which transcends it, necessarily concerns us. In this way a path opens up for reflection on the subsequent evolution of these ideas, after Christianization.

35It is indeed undeniable that this ritualism constituting the foundations of their religion was closely linked to the republican idea that the Romans had of their world. Religious rules and customs, born perhaps at the same time as the institutions which had evolved over time into the Roman Republic, developed in parallel, although not always chronologically in phase. We note, thus, that two main principles governed public religious life as well as institutional life: the distribution of religious power, and the assertion of the citizen’s freedom. From the beginning of the Republic, towards the 5th century bce, religious functions were shared between the magistrates, the Senate and the priests. In this way, religious power could never become the exclusive property of some of the people. The desire to avoid any uncontrollable concentration of religious power continued to manifest itself throughout the following centuries. The most important priests lost some of their power throughout the republican period. The sacerdotal colleges opened first to plebeians, before their recruitment was finally carried out via a public election. The reaction of Sylla, who in 82-81 bce abolished the election of priests and restored the traditional system in which they were co-opted without any external control, clearly shows how unbearable the people’s influence on the choice of priests, however superficial it may actually have been, could seem to a dictator. Since the high priests, especially the pontiffs and oracles, had the means to influence political decisions directly by their ability to challenge the legality of laws and elections, the dictator sought to preclude any opposition by closing sacerdotal recruitment again. However, on the surface, this conservative blamed the malfunctions of the republic partially on recruitment by popular approval. Either way, the best evidence that the evolution of these sacerdotal institutions corresponded to the republican spirit lies in the fact that after the Syllanian experience, the former recruitment rules, which had been the outcome of a secular evolution, were restored. At the beginning of the Empire the new master of Rome was to find other ways of making the priests submit to his power.

36The second grand republican principle governing Roman religion concerned the control of religious initiatives and decisions. In fact it was a corollary of the first one. From the 3rd century bce. at least, all decisions pertaining to religion had been taken by the magistrates and the Senate, sometimes after consultation with the priests concerned. And the stringent rigour of public rituals related without any doubt to the fundamentally republican nature of this practice. The legitimate partner of Roman gods was not the individual, the Roman citizen, but the Roman people as a whole, in other words the Roman state. Worshipping was done by those who represented it, on its behalf and for its benefit. The authorities thus wielded strict control over cultural acts. What better way to practice surveillance than by codifying worship, like other public procedures? From an institutional point of view, ritualism protected the public act and interest against all subjective influence.

37Through the rite, the act and the intention became objective and common to all.

38The fact remains that this codification was possible only owing to the appeal that ritualism was known to have had for the Romans. Otherwise, one could indeed talk of corruption, assuming that this term has any meaning at all in history. The very principle of Roman ritualism, as evidenced in their ritual behaviours and the myths or debates concerning them, was embedded in the idea that the citizen’s freedom was preserved by the primacy of ritual obligation. While implicitly and explicitly recognizing the omnipotence of the Immortals, the Romans always tended to reduce their relations with them to an exchange of homage and services compatible with the human status. The divinities could not demand anything from their human partners other than gestures, as humans were incapable of directly contemplating them or of moving beyond relations that had been codified for a long time and were governed and controlled by the magistrates. To be sure, the Romans explored other types of relations with the gods, relations that highlighted their own relationship of inferiority and their ineluctable submission to the divine power; or relations that were supposed to enable their soul to escape from its terrestrial body and to rise up to the gods. Significantly, however, these experiences never took place in the context of collective religion. Finally, for the elite, ritualism conceived of in this way was a guarantee against failure, a precautionary principle if you will, as the magistrates always acted with the gods’ endorsement and in close collaboration with them. A failure had a religious cause above all; in other words, it was based on a ritual mistake that had aroused divine anger and thus prevented the gods from lending their support. The political responsibility of the author of a decision was therefore moved to a different level, which allowed for and justified any kind of reversal. In fact, apart from immediate justifications, Roman ritualism elaborated during the Republic represented an endeavour to rationalize collective life and the world with a view to controlling them both. And contrary to what we might be tempted to assume, this effort did not disappear with the establishment of the imperial regime at the beginning of the Christian era.


39The history of imperial institutions changed the day when, between 1866 and 1874, the German Wilhelm Henzen discovered and published the epigraphic protocols of the Arval Brotherhood. On these marble tablets, among other liturgical accounts, the historian found the description of religious services commemorating the various acts of investiture of the Julio-Claudian and Flavian emperors. Despite the shortcomings, they contained not only dates but also, to everyone’s surprise, the fact that, contrary to Roman historians’ affirmations, the new emperor neither took nor received all of his powers by virtue of a single decision by the Senate. The procedure was more complex. Not only were the executive powers conferred separately on the prince, but they were also conferred, at least until Domitian, by the comitia or popular assemblies. No one had any illusions as to the true nature of the popular sanction of powers thus conferred on the emperor, if only when considering the solemnizing and consensual nature that the Roman comitia always had, as Egon Flaig’s analyses have shown. But that is not the important point. It lies in the fact that a century after the imperial regime was founded by Augustus, in 27 bce, the Roman authorities still deemed it necessary to mobilize what remained of the popular assembly to maintain the fiction that it approved the choice of an emperor, and conferred on him his main powers, one after the other. Here the specialist of divine functions and powers cannot help but wonder whether there was not a structural impossibility for the Romans to conceive of a single, all-encompassing and complete power, as the Christians later granted to their God. Just as the power of Jupiter that stemmed from a juxtaposition of complementary powers and effects necessitated the intervention of divine partners such as Victoria, Iuventas, Terminus, Fides, Venus, Ops and many others, so too was the omnipotence of the emperor designed and conferred in the form of a set of mutually complementary powers. The exploration of the Romans’ implicit theology that enables us to understand their practice of polytheism finds an unexpected ally here in the institutional study of the beginnings of the Roman Empire.

40The same religious sources also shed new light on the powers of the first Roman emperor. Careful study of all the decisions that he had the opportunity to take in religious matters shows that, despite his crushing victory and the undeniable popular support that he enjoyed, the Emperor Augustus had his hands partially tied by the pontifex maximus Lepidus. Captive and exiled since 36 bce, the Lepidus lived until 13 bce, and, in keeping with custom, continued to hold his function. Augustus, in his eagerness to follow tradition, tied his own hands by refusing to accept the Great Pontificate that the people had offered him after his victory. In so doing, he postponed the implementation of a series of measures that would have served his cause and his image. However, independently of the fact that he found therein a means for displaying his own piety, Augustus furthermore refrained from intervening because there were a number of domains in which he could clearly not act with full legitimacy without the approval or formal intervention of the pontifex maximus. Moreover, prosopographic analysis clearly shows that the affair was premeditated. From the thirties, the future Augustus had neutralized the pontifical college by ensuring that only those men loyal to him were elected, at the expense of those who were less so and whom he directed towards other priesthoods. Rather than radically reforming the recruitment of public priests, Augustus preferred to secure privileges enabling him to control the colleges from the inside. After the death of his old foe Lepidus, Augustus ensured his own triumphal election as pontifex maximus on 6 March in the year 12 bce. He immediately thereafter launched the religious initiatives that he had been postponing for eighteen years. They were thus all the more spectacular and enabled him to revive the image of the restorer of traditions that he had given himself at the end of the civil wars, in 27 bce. Apart from the immediate political facts, Augustus’ paradoxical behaviour also gives some indication of the powers of the pontifex maximus. We have little knowledge about these powers but do know that the position was eminent. Augustus’ calculated tactics highlighted the impossibility in which the uncontested master of Rome found himself to accomplish a number of acts without this pontifex maximus’ approval and participation. While the image of the emperor’s omnipotence was somewhat reduced, the weight of the pontifical function was increased.

41A last example, which takes us beyond the borders of the city of Rome, illustrates how fruitful the exploitation of religious sources can be. As the ties between religion and politics, between holy and public law were indissoluble in Rome, the study of any religious monument is an excellent means to approach the institutional and social changes induced by the intervention of Roman imperialism. The founding of colonies and municipia or, better still, the transformation of alien cities into municipia or colonies under Roman law, had inevitable repercussions on the institutions and the religious scene of the cities concerned. The few municipal institutions that survived enable us to outline these transformations. In Italy they generally consisted of the integration of former cities into a Roman institutional framework that was more or less restrictive. The inventory of places of worship in ancient Italy that I am doing with a group of colleagues – and that the means provided by this institution will enable us, I hope, to make available to everyone on the Web –, is gradually revealing the fact that, in the early days of the Empire, these changes were the outcome of an evolution that was initiated much earlier. More direct interventions, however, signal the persistence of suspicions, or at least, the Roman authorities’ desire to appropriate Italic places of memory. Thus, from the time of Sylla, and especially under Augustus, major supra-regional Italic places of worship were transformed into Roman colonies or allocated to colonies. In former “Barbarian” lands, changes were even more obvious. By grouping the Gallic peoples into Mediterranean-type cities with a chief town, a territory and a single government, the Romans gave them unity and permanence that they had probably never known. And in the colonies or municipia, these Gallic peoples acquired a set of entirely new religious, legal and political rules and practices. In this respect too, the imprint of Rome – that is being studied in another collective project in which I am participating – is clearly evident in the religious vestiges. Most of the public inscriptions of these regions concern religion, not to mention the fact that the most well preserved public buildings are generally places of worship.

42Archaeology is once again set to play a major role in this inquiry. A place of worship like Jbel Oust, 31 km south of Tunis, that I had the honour and pleasure of exploring with colleagues for the École française de Rome and the Institut national du patrimoine tunisien, makes palpable the problem of the border-sanctuaries and the role that they played in Roman municipal order in Africa. Hence, exploring these documents and identifying the changes that took place in the religious landscapes of the cities can enable researchers here or elsewhere to understand the modalities of the integration of these peoples into the Roman Empire. Such integration was possible mainly in the colonies and in the so-called Latin municipia, for which we know the questions that need to be put to the sources. Once the movement has been launched, the experience will have to spread to the comparison with those ciuitates that remained foreign, peregrinae.


43Once again, we see that the inquiries that I wish to undertake and to pursue in greater depth over the coming years, both alone and with others, are based on the detailed study of available sources. In ancient history it is only in the details and the concepts that they force it to devise that the research can hope to get close to past reality. To be sure, we are talking about micro-analyses and not a full picture of the Roman world. Ancient history is compelled to work with such microcosms, which are raised, rightly or wrongly, to the rank of models. These bubbles of knowledge provide us at least with near-certitudes, which, albeit isolated, are probably preferable to the uncontrollable generalities that merely skim over reality. For history is made with good questions, that is true, but it is made above all with sources. To paraphrase Aby Warburg, questions are gold but the Good Lord dwells in detail (der liebe Gott steckt im Detail).


Professor at the Collège de France
Liz Libbrecht (Traducteur)

© Collège de France, 2013

Conditions d’utilisation :