On Middle Grounds

European adventurers and conquerors, scientists and missionaries, approached the Americas with a battery of preconceptions and received wisdoms (Elliot, 1970; Pagden, 1993). These included habits of ritual, items of dogma, notions of divinity, understandings of priesthood and church, and much else that seemed to them general truths but to us are transparently local understandings transplanted into alien soil. Some flourished, others never took root, a few produced strange and fascinating fruit. All of this is well known and well explored. Now there is a reverse flow, one in which concepts and inventions developed to understand early modern and more recent encounters are being transplanted back to assist in the study of antiquity. This chapter is concerned with one such, the notion of the middle ground and its application to religious activity in the ancient Mediterranean.

The notion of a middle ground has been employed quite widely in discussions of antiquity, especially by archaeologists and historians looking for ways to describe conditions in the archaic Mediterranean.1 Part of the attraction is the capacity of the concept to describe “messy” and complicated patterns of interaction. Many of the same studies also employ entanglement theory or evoke hybridity and creolization or badge themselves as postcolonial. For those trained in the study of the ancient world, the middle ground evokes an alluring space in which many agencies intersected. It offers an attractive alternative to narratives of conquest, assimilation, imperialism, or colonization that approach the experience of encounters more from one direction than another. The middle ground is usefully complex compared to “Greeks and others” or “Romans and natives.” For ears sensitized to Francophone usage, there is perhaps also an echo of sur le terrain, “in the field,” the counterpoint to high theory, a commitment to recognizing the intricate details of local realities and their materiality.

The origin of the term is Richard White’s influential monograph of the same name, published in 1991. The middle ground described the situation in the Great Lakes Region in the period between 1650 and 1815, where different groups of native Americans and Europeans encountered each other across a broad swathe of territory known in French as the pays d’en haut, the “upper country.” The distinctiveness of this period and this encounter, in White’s view, was that local societies had been critically disrupted by the advance of European fur traders, settlers, and military expeditions but had not (yet) been brought under the control either of European empires or of the emerging colonial republics of North America. One of the book’s many strengths is that it evaded simple dichotomies between Indians and Europeans. The world into which French Canadians and then others moved was already traumatized by violent raids carried out by the Iroquois on various Algonquian-speaking peoples. The world they came from was equally convulsed by wars between the English and French and by the emergence of creole elites in the Americas. The story ends with local populations facing the expansion of the new American republic, fresh from its successful War of Independence.

With hindsight this seems a transitional period, an intermission between first contact and the incorporation of territories and peoples into imperial regimes. At the time, it was simply messy. White’s interest was in the kind of accommodations and negotiations, local understandings and misunderstandings, in the cross-cultural conventions involving many parties that emerged on the middle ground. Within a region divided by ancestry and language, technology and religion, connections were made and institutions of a sort emerged that lasted or evolved over nearly two hundred years, five or six human generations.

The book has been much discussed.2 White himself has commented on his surprise at the way a fine-grained study of a quite particular historic space has been appropriated to form a general model of cultural interaction, one that might be applied to quite different parts of the world and even to antiquity (White, 2011, xi–xxiv).3 Gallantly he goes on to say that the author of a book about the creative power of misunderstandings has no right to complain if others have made something unexpected out of his ideas. All the same, he makes clear what he considers the elements necessary for the creation of a space that is similar to his middle ground:

a rough balance of power, mutual need or a desire for what the other possesses, and an inability by either side to commandeer enough force to compel the other to change. Force and violence are hardly foreign to the process of creating and maintaining a middle ground, but the critical element is mediation. (White, 2006, p. 10)

White also insists on the spatial dimensions of the world within which mediation takes place and on the creation of new infrastructures in conditions where no one group dominated the other. I shall not discuss how far individual appropriations of his ideas by particular archaeologists and ancient historians satisfy White’s criteria (nor whether they should have to). Naturally, we have not all done the same things with his ideas.

Chris Gosden makes one generalizing appropriation of the concept of the middle ground in his tripartite taxonomy of colonialisms: the other two types he labels “terra nullius” and “colonialism within a shared cultural milieu” (Gosden, 2004, pp. 24–40). Terra nullius denotes a situation where colonizers arrive with overwhelming force and do not recognize the prior arrangements or rights of indigenous inhabitants. This is the territory of atrocities, genocides, and mass expropriations. The conquest of New Spain is explicitly cited as an example of this mode of colonization. Colonialism within a shared cultural milieu describes political or military expansion when there is little culture gap. Greeks dominating Greeks is one example. The middle ground is an intermediate category, meetings of strangers when there are some power differentials but without either side having overwhelming force. Accommodation, the generation of new cultural differences, and a destabilizing of values on both sides characterize situations of this kind. Relations between Greeks or Romans and the nonurban populations on their peripheries provide one of his examples. Crucially Gosden stresses that his taxonomy serves heuristic purposes only, that real-life situations were more complex and not fixed or stable, and that these types are designed as tools for examining individual cases, not as adequate descriptions on their own of any particular situation:

The last qualification I would make is that the typology should not be seen as a linear progression from one form to another: within one colonial formation all three types can exist simultaneously: there can be movement from one to another, or one form can be found alone. (Gosden, 2004, p. 25)

The notion of movement between forms of colonialism is an important one, especially if we stipulate that movement is not always in one direction. White’s middle ground referred to a particular period, one that had a place in a larger narrative of the history of the First Nations over the last five centuries. If the description is to be generalized it should not be teleological. Middle grounds should not always represent halfway stages toward complete subordination, even if that has often been the experience of modernity. Gosden’s insistence on the coexistence of different forms of colonialism is important too. Perhaps few encounters are ever quite as devastating as the term terra nullius suggests. Probably there is always some need for mediation.

Roman and American Middle Grounds Compared

There are enormous contrasts between Roman experiences in their provinces and those of the Spaniards in Mesoamerica. Gunpowder and iron working, horses and oceangoing vessels, gave the Spaniards immense advantages over indigenous populations, quite apart from the disease flows that came with conquest (Gosden, 2004).4 Yet despite this disparity in power, we can still find instances of accommodations similar to those described by White in The Middle Ground. Gosden presumably classified the conquest of New Spain under terra nullius because of the lack of respect paid to existing rights to land and power. This was certainly correct. Yet at the ragged fringes of Spanish power, spaces were opened up for productive misunderstandings.

Religious mediation provides many examples of this. White’s (1991) account of encounters between Jesuit missionaries and local peoples shows that the former were prepared, when it suited them, to take advantage of local misunderstandings about Christian religion. For instance, they did not challenge the locals when they at first interpreted the Jesuits and Christ as manitou or spirit forces (pp. 25–28). There are obvious Mexican parallels such as the contentious process through which cult offered to Tonantzin at Tepeyac was in some sense replaced by cult offered to Our Lady of Guadalupe. Misunderstanding is perhaps the wrong term. It seems that at least some of the participants had a very clear understanding of what was going on and made instrumental use of what they understood as their interlocutors’ misperceptions. These conform in general terms to White’s notion of actors seeking out “cultural congruencies, either perceived or actual” (p. 52) and his dictum that “any congruence, no matter how tenuous, can be put to work and take on a life of its own if it is accepted by both sides” (pp. 52–53). There is no doubting the scale of the trauma experienced by American societies in the centuries of colonial domination, the staggering death toll, the loss of indigenous culture, and the violence suffered by many individuals. Yet Gosden’s injunction to acknowledge the complexity of these histories encourages us to notice that even against the background of these events, new hybrid forms did emerge that included traces of local imagery and ritual.

If even imperial Spain relied in part on mediation, then the same must be true of the early Roman Empire (Dench, 2018, p. 46). Roman armies had none of the technological advantages later enjoyed by Europeans over the populations of the Americas and suffered the same numerical disadvantages as the conquistadors. Mediterranean states had some slight economic and organizational superiority over the inhabitants of continental Europe and Africa. Yet Roman expansion occurred within an interconnected world that had been formed by millennia of technology flows and other exchanges. Rome and those who would ultimately become her subjects used the same metals, the same domesticated animals, and most of the same domesticated crops, and also had in common coinage and alphabetic writing systems. Centuries of warfare and alliance, mercenary service and commerce, meant that Mediterranean and European populations knew each other quite well. Whatever the rhetoric employed in documents like the Res Gestae of Augustus, or on triumphal monuments from the first century BCE and the first century CE, Rome was not conquering terra incognita but extending dominion over a largely familiar ecumene.5

That familiarity had some advantages when it came to mediation. In the second chapter of The Middle Ground, White considers a range of media through which the French and the Algonquian-speaking peoples created pragmatic mutual understandings. Some involved repurposing existing concepts such as the French notion of le sauvage. There were also evolving diplomatic rituals such as gift exchange and the smoking of peace pipes. There were attempts to mythologize relations, some drawing on Christian elements as well as folktales and local motifs. Intermarriage was a means of building alliances at the microsocial level. Conventions arose to regulate trade and to limit feud. White described how ad hoc institutions were set up to adjudicate in the case of conflicts, creating what he terms “bizarre cultural hybrids” (p. 79). Ceremonial was invented or modified to represent and solidify new relationships. Rituals, mythology, and ceremonial all played central parts in creating the middle ground in le pays d’en haut.

When Roman power was decisively extended beyond the Mediterranean littoral—mostly during the first century BCE, in fact—much less needed to be invented. I have discussed elsewhere how existing conventions of mythography and science provided Romans, Greeks, and indigenous peoples with the means to create new relations of kinship, underpinned by new histories, in the Roman west (Woolf, 2011b).6 Others have pointed out how elements of a shared aristocratic culture had emerged from the archaic period on. Rituals of guest-friendship and of formal wine drinking are already visible in the Homeric poems. The spread of sympotic imagery, drinking equipment, and wine itself in the archaic and classical periods has been richly documented (Dietler, 1989; Murray, 1990; Murray & Tecusan, 1995). It occurred wherever Greeks traded or settled, in Etruria, in the situla art of the Veneto, and in temperate Europe, with local variations but on a recognizably cosmopolitan theme.

This familiarity did not mean there was no need for mediation, nor that further misunderstanding was impossible. It simply meant that, as Rome reduced neighboring peoples to subject populations, there was less of an initial gulf to cross and media were available through which new situations could be negotiated. Those media of cross-cultural communications contributed—along with other factors such as a shared disease pool—to one final contrast between the Roman Empire and New Spain. Roman expansion was not a catastrophe. It did not entail a demographic collapse, mass expropriations of land, or the demolition of cosmological certainties. For casualties of conquest and those enslaved after it, Roman conquest was indeed brutal. Recent work by both archaeologists and historians has made it clear not only that episodes of genocide did take place, erasing communal identities and obliterating local knowledge, but also that Romans regarded such tactics as legitimate and necessary in some circumstances.7 But Roman conquest did not bring about the cataclysmic end of a way of living, as the success of the conquistadors did in what they made into a New World.

Religious Mediation in the Roman Empire

Ritual is an elastic term. Smoking a pipe together, taking part in a symposium, intermarrying, and naming one’s children after one’s guest-friends are all appropriately described as rituals.8 Like routines, they are associated with social conventions, and their practice is entangled with particular objects and substances. What distinguishes these rituals from other routines is a shared understanding of their significance. Taking part in rituals of this kind effects changes in the relationship of the participants. Like a shared rite of passage—and most rites of passage are in some sense shared—these rituals built and modified relationships and identities. All this happened across the many middle grounds of the ancient Mediterranean world throughout the last millennium BCE and the first centuries CE.

For the remainder of this paper I will be concerned with a narrower set of rituals, those in which some of the participants were divine beings. My argument is that here too, mediation between Romans and others was made much easier by long-established modes of interaction, by conventions formed on pre-Roman middle grounds.

Once again there are some obvious differences with the situation in New Spain that need to be born in mind. Most obviously the ancient world knew nothing like the varieties of Christianity that were imported by early modern Europeans along with their guns, germs, and steel (Diamond, 1997). Historians of classical antiquity are now very wary of writing of “religion” at all (Nongbri, 2013).9 Even those who continue to use the term immediately point out the absence of a centralized authority, exclusive membership, or dogma from the ancient world. The world of the polytheists was far from tranquil, but it knew neither schism nor heresy. Ritual acts accompanied almost every kind of collective activity—political, social, festal, and sporting—and were ubiquitous in family life and in daily routines of work and leisure. Yet the absence of a secular sphere meant religion was barely separated from other activities.

Religious authority in particular presents a contrast. Across the ancient world few religious leaders were not also members of the educated classes that ran and owned most ancient states. The most often-cited exceptions are Druids and the priesthoods of Judaea-Palestine and of Egypt (Goodman, 1987; Gordon, 1990a).10 Even in these cases it is not clear how distinct these groups were from other elites in terms of interests, backgrounds, outlook, and behavior. The authority they exercised was over ritual action, which was mostly controlled very locally. The kinds of religious politics conducted in the early modern world between popes and kings, religious orders and colonists, bishops, military commanders, and civil governors are unimaginable in antiquity. Missionary activity was virtually unknown (Goodman, 1994).

The paradigm for religious mediation in the Roman case is usually taken to be ruler cult. There is now a broad consensus that this was not a unified religious program (and certainly not a religion) emanating from the center, but rather the cumulative project of dozens of accommodations through which the Roman Senate, people, Rome herself, and eventually the emperors and some of their relatives were incorporated into local systems of ritual.11 That incorporation took many forms. It ranged from adding the names of emperors to hymns and oaths, carrying their statues in processions or placing them in temples, and inserting imperial anniversaries into religious calendars, to creating altars, temples, and priesthoods dedicated to individual emperors. Emperors found a place in ritual but never in myth or cosmological thought. Ancient polytheists made many distinctions between different kinds of divine beings. Living and dead emperors and their kin were never confused with the ancient gods and in iconography, oaths and prayers to them were assigned subordinate places and status (Nock, 1930; Scheid, 1999).

When we can disentangle the agency through which these religious innovations were created, we find local aristocrats, councils, and assemblies taking the formal initiative, with Roman governors sometimes involved in a secondary role. Before and behind this we must assume there were negotiations over what would be acceptable and welcome, locally and in the center, and in some cases it is likely Romans took the lead. For example, the creation of parallel cult organizations for Greek cities and Roman citizens, at about the same time in the two neighboring provinces of Asia and Bithynia-Pontus in 29 BCE, is implausible without the involvement of Octavian, based at Pergamum at the time (Madsen, 2016). Equally, the forms of provincial cult created at Lyon, Cologne, Colchester, and some other western centers owed so little to local ritual traditions it is difficult not to see Roman initiative as predominant. Yet if we put origins aside, the new cult organizations and ritual performances established in the first century CE did succeed in engaging the participation of the wealthiest provincials, who competed at considerable personal expense to hold priesthoods. Imperial cult has been seen as a form of gift exchange, a device through which a new temporal order was naturalized and a divine mandate, a theodicy of good fortune, established for the status quo (Gordon, 1990a, 1990b; Price, 1984).

One of the longstanding obstacles to acceptance of this view of ruler cult was a sense that it was different in kind to the other forms of collective ritual practiced in Rome’s provinces. Imperial cult was homage, other cult was proper religion; one was political, the other more spiritual; one was a matter of displays of loyalty, the other more sincere. In fact, the boundary between what we label “imperial cult” and other forms of collective ritual was seamless, and the processes through which rituals were modified or devised to involve the emperor were the same as those more widely used when rituals were invented or adapted to new ends. Ancient polytheisms were never static. Councils and assemblies were frequently called on to consider whether a new festival should be added to the civic calendar, whether new gods should be invited or admitted into the body of those that received public worship, whether land should be allocated for the construction of new temples, and so on. In formal terms the decision whether or not to seek to hold the provincial temple of Claudius or Domitian was no different to debating whether or not to fund an annual festival in honor of Isis.12

If the processes through which ruler cult was established are obscure, we know almost nothing of the processes through which specific syncretisms such as Mercurius Dumias and Mars Lenus were agreed. As far as the western provinces are concerned, however, broad consensus has emerged on Latin epigraphic and some comparative evidence, based on what we can infer from colonial and municipal charters.13 It is broadly agreed that Mediterranean observers, at least, sought to recognize familiar gods under unfamiliar names and rituals (rather as Caesar did in the case of the Gauls), that some equivalences gained widespread acceptance by indigenous groups as well as visitors, and that at the moment when Roman-style polities were constituted in the west, some of these equivalences fed into the stipulations of the public cults. So, for example, at some point around the turn of the millennia, the Roman idea that the chief god of the Gauls was a version of Roman Mercury was accepted by some Gauls (perhaps under the form that Mercury was the name Romans gave to their chief god) and then when the civitas of the Arverni was formed and needed to define its public cults, the cult of Dumias became that of Mercury Dumias. Perhaps there were fierce debates over this. Could Dumias have become Mars Dumias? Were there voices opposed to making equivalences with alien gods? But the politics of ancient syncretism, a contentious issue today (Stewart & Shaw, 1994), are lost. At a larger scale, the effect was that hundreds or thousands of local male gods came to be represented as variants of just a few, most of them rendered into versions of Jupiter, Mars, Mercury, Apollo, Hercules, and Saturn. Mostly no trace of earlier names or images survived. That small part of the population that travelled beyond their own state of origin—most of them male traders, soldiers, and a few landowners—would not have encountered very much religious diversity.

Where detailed regional studies have been conducted, many nuances emerge (Cadotte, 2007; Derks, 1998; Spickermann, 2003, 2008). There were clear regional preferences when it came to which male Roman gods were associated with local deities: Mars was especially popular in northern Gaul and Germany, Saturn in Africa, Hercules on the lower Rhine. Local epithets were used in some areas to differentiate between Martes or Mercurii who were worshipped alongside each other. In other areas they are very rare indeed. The relative popularity of the main Roman gods varied from one area to another, although Jupiter was almost always associated with the chief deity. Most provincial gods are known only under their Roman names, a few sometimes have a local epithet attached, and others (mostly goddesses) appear only under local names such as Rosmerta or Sirona. During the first centuries CE new deities arrived, including Mater Magna–Cybele, Isis, Mithras, and various male deities from Syria already syncretized with Jupiter. Their take-up in the western provinces was not even, yet all the same they had some claim to be global deities. The situation in the east was more complex. In some areas, local deities had become associated with Greek ones, even before Alexander’s conquests, and the world Rome expanded into was full of complex syncretisms. Some could be the basis of further connection to Roman deities, so Syrian Baalim, already connected to Zeuses, might easily be reinterpreted as Jupiters. Once again, we are aware of no controversies over this. Syncretism undoubtedly resulted in a less diverse cosmos, one drawn together by myth and iconography rather than dogma and authority. But the process by which this happened seems neither to have been coordinated nor resisted.

The Bases of Ritual Mediation in the Roman Empire

White identified rituals as one means by which new relationships were formed on his original middle ground. This seems even truer in Roman antiquity. Several factors help explain how this came about.

The first is the fundamental similarity of religious systems across the eventual territory of the Roman Empire. Virtually all of Rome’s subjects were polytheists, virtually all practiced animal sacrifice, virtually all made images of the gods, and most placed them in temples which were in some sense transformations of the houses that humans, or at least powerful humans, inhabited. Ancient gods were often, perhaps mostly, thought of as part of the human communities that worshipped them. As those communities were joined up, so their gods too came into alignment with each other.

Second, when there were differences, Romans and many of their subjects were already equipped to deal with them. Romans of the first century BCE were prepared for variations in the ways the gods were portrayed, local peculiarities of ritual, unusual names, and so on, and this is because they had been living in a world marked by these differences for centuries. Romans were equipped with a range of ways of dealing with these differences, modes of understanding that were philosophical, ethical, ethnographic or even satirical. Many of these responses had been learned from Greeks (and perhaps others) who had been encountering alien religious forms throughout the last millennium BCE. It was widely understood around the Mediterranean that the same god might be called different names by different peoples, and that local images of familiar deities, and even the rituals paid to them in particular places, were often peculiar. These modes of understanding—we might almost say of translation—were inherited from encounters on earlier middle grounds.

All this is one aspect of the general contrast between events in New Spain and those in the Roman provinces. The former resulted from a sudden encounter, accompanied by overwhelming force; the latter were built on centuries of encounters and connections of different kinds between peoples who already had much in common.

Religious mediation in the ancient Mediterranean did have a history. The gods in Homer are everywhere the same. Achaeans, Trojans, Egyptians, Ethiopians, and Phaeacians all knew the same gods and knew them by the same names. When this fiction was created is not clear, but we can be sure that there was no period from the Bronze Age on in which some Greeks would not have been aware that other peoples worshipped other gods. Even in Homer’s time the unity of the gods was a mythic convention. By the sixth and fifth centuries BCE, there is a much greater awareness of religious differences in the theological speculations of Ionian philosophers and in Herodotean ethnography. Presumably, Greek adventurers in the Far West and mercenary soldiers in Anatolia, the Levant, and Egypt would have been aware of this from the seventh century on. Perhaps it is the absence of prose that conceals earlier knowledge of alien gods. The responses attested in fifth-century material include the construction of equivalences between deities (such as between Isis and Demeter), attempts to resolve apparent inconsistencies (as in the attempt by Herodotus to reconcile the myths and chronologies of Herakles), and the philosophical response of regarding all local knowledge of the divine as limited and incomplete. Alongside these intellectual responses are iconographic ones, such as the representation of Melqart as Herakles (or vice versa), and epigraphic ones, such as the bilingual gold tablets from Pyrgi in Etruria, which offer complementary views of the same cosmos (and ritual) in Etruscan and Punic.14 Bi- and trilingual texts are known from all around the Mediterranean and the Black Sea and in western Asia as well. Most mention gods. These too were artefacts created on middle grounds.

Roman expansion was a bloody business. Even if the human tragedy was not on the same scale as in New Spain, provincial societies underwent convulsive transformations. Yet religious conflict made almost no contribution to the process. Even synoptic studies of provincial revolt have found only a few cases of millenarian leaders, and studies of revolt narratives show that religious dissent was rarely a central theme.15 There was nothing to rival the entanglement of faith and violence in the Middle Ages or after. On the contrary, ritual offered powerful resources when the time came for mediating new relationships. A shared language of cult and image, of performance and myth, had already been formed on the middle grounds of the archaic Mediterranean. During and after the chaotic course of Roman conquest it had become a matter of habit to reach for the gods, and they did not disappoint.

Notes

References