What Is Religious About Founding A City?

How to found a city

This paper seeks to approach the issue of religion and urbanity through the lens of foundation rituals. Many ancient Mediterranean cities have a foundation discourse, an attempt to understand their starting point. Some date from that moment. Founders can be gods, heroes or men (rarely but sometimes women – Dido at Carthage is a good example) who receive honours later. Much of the discourse inevitably revolves around what we call colonies, that is, artificial foundations, cities created afresh in the historical period. It is possible that the discourse around these cities then impacts on more inventive mythological stories which are set in earlier periods.

This has been well studied by historians, archaeologists and philologists. Much of the attention has been on historicity – do the historical and archaeological data coincide? Other areas include the role of foundation stories in identity discourse, or the literary qualities of the accounts. However, it is worth emphasizing that for the most part, foundation is presented in terms of religion. It is ritualized, it may be repeatedly celebrated and it ties the city directly to the gods; yet it is also clear that many foundation stories, and where they existed, events, are fictional narratives or performative enactments, at some distance from the sometimes faltering and often long term process of settlement, conquest, and integration. Focusing on the essential fictionality of this discourse, and its relationship to urban religion, is my goal, and I come at this not through ancient texts, but through an unusual but fascinating reflection on the subject by Michel Serres.

From ancient history to modern structuralism

There is an evident and obvious tension between literary stories of foundations and archaeological evidence. One gives us founders and founding moments; the other gives us process. Repeatedly we find that the story of a beginning is belied by the evidence of continuity. Yet we continue to be fascinated by beginnings; just as the ancients loved stories of the first person to do something, so we look for the first human, the first drawing, the first city. ^[1]

It is an interesting question as to whether this was a general Mediterranean phenomenon from the outset, or whether, as Elias Bickerman argued, it was a particularly Greek trait and was then picked up by others as they elaborated national stories along the lines of an existing model (Bickerman 1952). This still requires some explanation of what drove the Greek interest in beginnings, and it is possible that the colonization moment, in which Greeks founded cities elsewhere, opened the question of how existing cities had begun. The search for legendary origins was also competitive. Moreover, autochthony is relatively rare; the city is often the outcome of movement.

In antiquity we sometimes engage in rather dispiriting fights over this, trying to prove or disprove Thucydides’ dates for colonies, or worrying about what happened in April 753 BC in Rome (Donnellan, Nizzo, and Burgers 2016; Morakis 2011 with a good review of the chronological issue; Carandini 2011). For the most part, it seems to me, stories of origins are interesting as stories rather than having any value as evidence for an event. But this is not, cannot be, the whole story. The notion of the foundation clearly existed, and once it did it became archetypal, so that the foundation ritual which almost certainly did not occur in Rome perhaps did in some Roman colonies.^[2]

One way of reading the focus on the foundation event is in terms of its functionality. Bluntly it is very useful to begin somewhere. One of the points of utility is in terms of time; and we are no exceptions, with our Christian Era start to the stopwatch. Whatever date is chosen for foundation makes the critical contribution of allowing time’s arrow to fly forwards; it is if nothing else a point from which the community is said to care about counting, and it perhaps has more salience in an urban context. It is interesting that at Rome this date was calculated three times, with the arbitrary Romulean moment, the Eratosthenic calculation of the Aeneas connection, and perhaps the earliest of them all, the emergence of what Purcell called Capitoline time, from the beginning of the Republic, which was itself almost certainly subject to significant constructive calculation. True sceptics will argue that year one is probably some time in the fourth century with the Licinian-Sextian legislation. It will be immediately obvious that we have slipped from foundation to invention of foundation. Counting was important, and competitive counting is also perhaps a feature of the ancient world, though as Herodotus noted, the Egyptians had everyone beaten in this game (Feeney 2007; Purcell 2003).

However, time is not neutral and foundation cannot be reduced to mere functionality. It mattered for ancient cities, and it continues to matter to us, and this paper tries to work through some ideas as to why that should be. I have chosen to work backwards, from a modern thinker on foundations, Michel Serres. I hope that by so doing I will be able to disclose both something of our way of thinking about foundations, as well as eventually speaking more directly to the ancient concern with how to begin.

Michel Serres was a renowned philosopher, with a particular interest in the history of science^[3]. He wrote his thesis on Leibniz under Gaston Bachelard, and wrote on Lucretius (Serres 2000, cf. Serres 2007; Holmes 2012; Holmes 2016) as well as more broadly on the Anthropocene, technology and climate. Serres’ work has been enthusiastically received in some areas of postmodern philosophy, where his almost incantatory, and quasi-poetic style of writing is prized. His writing covers a huge range of subjects from geography to space travel to history to biology, but rests largely on an allusive methodology which seeks to develop ideas through connection rather than privileging any single discourse, or admitting the claims of an objective science.

There are clear connections with another great theorist of religion and violence, René Girard. The issue of original sin and scapegoating is a shared interest with Girard, as will be clear, but Serres was interested in objects as well as people – he anticipated the wave of object-oriented archaeology. Serres’ challenge to the notion of modernity and science influenced the cultural theorist Bruno Latour, and indeed Latour has written of and with Serres.^[4]The shared ground with Latour is complicated. It arises partly from a belief that the discourse of science needs to be revised, that what has been characterized as science is both less than what is qualified to be included and yet should still be seen as part of a subjective discourse.

It is worth just spending a bit more time on this intellectual genealogy because it will help to locate Serres’ unusual project. Serres’ intellectual journey undoubtedly starts in the sciences. Up until 1968 Serres was very close to the philosopher of science Georges Canguilhem, who was also Foucault’s mentor.^[5]They then fell out very badly, and it seems that Serres’ critique of the epistemological underpinnings of science was even more acute and destructive than that of Canguilhem (he himself tells a story about a difference of intent, with Canguilhem’s circle being deeply attracted to the exercise of political power).^[6]Serres has a narrative of ‘expulsion’ from philosophy which was clearly painful but also perhaps somewhat liberating because what Serres writes is some distance from anything which looks like formal philosophy.^[7]So here we have a combination of a classically trained scientist who turned to philosophy and then moved on.

The critical background to the work of Canguilhem, who is such an important figure in French philosophy, is described in Stefano Geroulanos’ brilliant account of the development of antihumanist atheism in the decades before and after WWII. The replacement of a philosophy based on God with one which placed human beings at the centre was a process of long duration. Geroulanos notes Christian humanism, Renaissance humanism, Humboldtian education, the Enlightenment, nineteenth century liberal humanism with a strong emphasis on natural law and socialist humanisms as some of the relevant movements. He identifies specifically in France the notion the egalitarian human subject in post-revolutionary liberalism, and a utopian socialist tradition (developed by Marx) with an emphasis on ‘the socio-political goals of a “human nature” conceived as good and perfectible’ (Geroulanos 2010: 19).

This had little chance of surviving the miseries of the first half of the twentieth century, and indeed one can also trace the problem of intellectuals who were aware of, courted or shunned the dangers of their work being used within totalitarian discourses. With only limited opportunities for a revival of Christianity, and profound scepticism about the perfectibility of human nature, the tide flowed for some towards a different and more disillusioned philosophy. The better known figures in this move include Heidegger, Bataille, Levinas and Sartre, the immensely influential Hegelian Kojève, and then Blanchot and Hyppolite. But Geroulanos also introduces Alexandre Koyré, whose work with that of Bachelard on philosophy and quantum physics (and responding to Brunschvicg’s humanism) is one of the lines through to Canguilhem’s antihumanism.^[8]

As Geroulanos notes, and as was picked up in the conversation between Latour and Serres, if Canguilhem’s science was one outcome of antihumanism, another was structuralism (Latour, Serres, and Lapidus 2011: 10). Subordinating action to structure, history to paradigm, and to some extent humans to language, structuralism in the work of Lévi-Strauss for instance effectively displaced humanity.^[9] However the larger influence on Serres, by his own account, was Georges Dumézil (1898-1986), best known for his study of the Indo-European roots of classical and especially Roman culture and religion, and the theory of a tripartite structure of warrior, priest and farmer.^[10]

Dumézil

Dumézil is a strangely muted figure in many accounts of French thought either side of the Second World War, but he was both more influential and more interesting than he is sometimes given credit for. In saying this, I do not wish to undercut the sharp critique of Momigliano, Grotanelli and Lincoln on the fascist undertones of what he was doing. In fact that is central in a way to making sense of the complex skein of ideas which I am unravelling.^[11]

For a good idea of what was actually at stake, one may turn to a recent account of the argument between Dumézil and the Russian linguist Prince Nicolai S. Trubetskoi. Behind the mudslinging, their debate which was centred on the Caucasus, was about method and political intent. Trubetskoi rejected the Indo-German diffusionist model which Dumézil was imposing on the Caucasus, preferring to assume that language was more local, and that mechanisms of interaction were not to be turned into a trace of an original Heimat. For Dumézil the critical point was the capacity to make comparisons. The discovery of the tripartite Indo-European structure permitted the co-ordination of facts. As Geroulanos and Phillips argue, quoting from lectures by Dumézil in 1940:

“The object of my [Dumézil’s] work is to reconstitute, through comparison, the fragments of a prehistoric religion: that of the people who spoke the common IndoEuropean language.” Prehistory could no longer be sought out on the linguistic level, it had to be examined mythologically; it could not be elaborated at the structural or systematic level, only at the comparative one. The IndoEuropean universe – whose name Dumézil presented after the war as “purely symbolic” – still existed in 1940, specifically to facilitate a reconstruction of prehistory from the ruins it left behind. He could reach properly back to these ruins, to what he, as we have seen, called the conquering armies that “enslaved all of Northern, Western, Southern, and South-Eastern Europe,” and reconstruct their way of thinking (Geroulanos and Phillips 2018: 377).

The comparative element is absolutely critical. For Dumézil, it underpins the entire enterprise of his work. It is unsurprising that Dumézil was Levi-Strauss’ most important supporter in his advancement. And critically we have to see that the basis of comparison brings us back to a particular vision of the west; mythological rather than Christian, structural rather than humanist, conflictual rather than progressive.

Now if there is one sharp flaw in the Dumézilian model, looked at from an empirical point of view, it has always been the issue of transmission. How did the model work over long periods of time? How did an Indo-European model of thought diffused in the Neolithic survive to influence first millennium BCE Rome? What were the mechanics which transcended the evident rewritings and rethinkings of subsequent centuries? To my (incomplete) knowledge, Dumézil does not answer this convincingly, and I will simply say that I find models of cultural memory to be problematic. There is limited evidence that these structures form genuine parts of our cognitive inheritance either.

Michel Serres

It is therefore extremely interesting that the issue of time is something which Serres, influenced as we have seen by Dumézil, addressed head on, though in various places and ways (Assad 1999; Herzogenrath 2013). Two of his favourite metaphors are the folded or crumpled cloth and the wasp’s flight. The first is ‘topological’; points which are far from each other in one conception can be brought close in another. The second is an expression of the random, and drawn from a poem of Verlaine

L'espoir luit comme un brin de paille dans l'étable.

Que crains-tu de la guêpe ivre de son vol fou ?

Hope shines - as in a stable a wisp of straw.

Fear not the wasp drunk with his crazy flight! (Trans. Gertrude Hall)

Time moves backwards and forwards, and all time is present at any time. The fascination with time is another French legacy, from Bergson and Merleau-Ponty, but also increasingly a scientific concern. As often, Serres philosophized in places where science would subsequently be found.

Serres’ theory of history appears to allow for the co-presence of different times, and to allow circular or random passage across time, or, as Eileen Joy puts it, a notion by which ‘every present moment is inhabited by and also inhabits (consciously and unconsciously) multiple, heterogeneous temporalities – some at a distant remove and others more contiguous.’ The freedom to move across time in this way then permits the association and analogy of ideas which characterises Serres’ freewheeling associations, for instance between the explosion of the Space Shuttle Challenger and the rites of Baal at Carthage.

However, in the specific volume which I want to consider, for the next section of the paper, Rome, it is not clear that time is so open (Serres [1983] 2015a). Instead, Serres presents Rome as a darker and more difficult vision of an eternal return to a moment of bloody foundation. It is, as Assad says, a ‘discursive plotting of a conservative dynamical system that “describes the world in terms of deterministic and reversible trajectories.”’^[12]That word reversible is slightly problematic; it does not, I think, mean that Serres thought that the foundation of Rome could be undone, but rather that here there is a kind of unforgiving loop which takes one repeatedly back to the beginning, or even something worse. It is perhaps for this reason that Rome is a rather terrifying read. It is time to look a little harder at Serres’ work.

The book is divided into four sections and eight chapters, and is the first part of a trilogy.^[13] The structure of the book is taken very loosely from Livy; some sections begin with a reference to his historical work ab urbe condita.^[14] But as befits the work of an author who eschews linearity, there are frequent loops back to earlier sections. It may be easier to understand the book if we introduce its main themes, although it must be stressed that Serres’ highly elusive and allusive style evades any kind of easy summary.

The first section treats the act of foundation as a sort of concealment. Rome’s foundation story conceals aspects of its darker nature, and Serres places Rome in dialogue with Alba Longa, the other foundation story, which Rome’s story supplants. The second section elaborates on these negative aspects with a focus on death, the creation of empire and the problem of a community whose voice or vote can mean death. The third section looks at the individual leaders, their relationship with myth and gods, and the way their actions drive history and myth; whilst the last section looks at the crowd or multiplicity, and asks questions about its capacity to act and create, and the way this may reflect a more hopeful form of action. If Rome is a resolutely political history of knowledge, the second volume of the trilogy, Statues, looks at representation, and the third, Geometry, is a sort of account of the emergence of science.

On my reading Rome circles around two challenges, one epistemological or phenomenological, the other political or historical. The first relates to Serres’ difficulties with the emergence of the object and the extent of its knowability. What is there which is independent of me is an old question in philosophy and Serres claimed in his conversations with Bruno Latour to have given up on epistemology (Serres, Latour, and Lapidus 2011: 9, 17, 52, 113, 128), but at various stages in Rome he claims that what comes into existence at the foundation is the object, but that it is not directly knowable, it lies within his frequently used metaphor of the black box. Serres develops an alternative kind of story about knowledge.

The other challenge is the historico-political division between one and many. The role of the multiplicity is critical to the book as a whole. We shall see at the end Serres’ hope for the emancipation of multiplicity from a tragic cycle which keeps returning to a unity which must then be sacrificed again to be unified again.

There is a degree of isomorphism between the two arguments; the definitive object and the single entity are more graspable but less generous, less liberating, than openness of reference and a liberated multiplicity. However, Serres never allows his discourse to fall into dualism. Part of the frustrating richness of reference arises from the way Serres repeatedly undercuts dualities, either by stylistic phrases (‘a collective working at its sum’ is a good brief example (114), but they are ubiquitous), or through the introduction of the concept of an excluded middle which keeps returning. Throughout it is clear that Serres is no structuralist, or if he is, it is in the highly modified sense that Philippe Descola is, seeing repeated spectra which operate somewhat independently of each other, but which are all important in describing a society (Descola 2014). So Rome moves backwards and forwards across independent spectra; one, the extent to which reality is fixed, hidden and misleading, or variable, transparent and enabling, the other the degree to which the multiplicity is allowed to be, or is dragged back to unity.

Serres begins with an extended reflection on the story of Hercules and Cacus, which is read as a problem about knowledge and history.^[15] The story is of Hercules’ confusion, the disappearance of the animals, the slaughter of Cacus, the black box of the cave; Evander then turns the problem into history, and excludes the voice of Cacus. The multiplicity of voices, Cacus, the shepherds, become simply the divinization of the murdering god. ‘History hears everyone cry out: the historian, most often, only hears one voice’ (14). From the beginning, therefore, Serres problematizes our very ability to tell the story; later on, Serres convicts himself and history of being nothing more than a pastiche of detective work, seeking out the victim. So the work of the first chapter is to see if there is another, less schematic, less binary way of telling the story, and here Serres returns to his concept of La belle noiseuse, a name which features in a story by Balzac, Le chef d’oeuvre inconnu, which tells the story of the painter Frenhofer and two younger artists, Poussin and Porbus. Frenhofer is a massively talented artist, but is blocked for want of a model to complete a painting he has laboured over. Poussin offers his beautiful mistress; Frenhofer thinks he has completed the painting but Poussin and Porbus can see nothing in the chaos of the canvas:

In a corner of the canvas, as they came nearer, they distinguished a bare foot emerging from the chaos of color, half-tints and vague shadows that made up a dim, formless fog. Its living delicate beauty held them spellbound. This fragment that had escaped an incomprehensible, slow, and gradual destruction seemed to them like the Parian marble torso of some Venus emerging from the ashes of a ruined town.

For Serres, this offers a vision of noise and chaos which is liberating, and from which history is always trying to drag us towards a more binary, judgemental world, and one less full of promise and opportunity.^[16]

One alternative world is Alba Longa, and Serres plays with the notion of white as being not colourless but containing all colours. Similarly figured as white is the doomed Vestal Virgin, Tarpeia. Both Alba Longa and the Vestal Virgin are destroyed, hidden, in the black box that is Rome’s foundation. The first part of Rome is about knowledge, about how the object of knowledge is constructed, and what is hidden or buried in the process of making the indeterminate fixed.

The second part then begins to turn towards the historical problem of the multiplicity. The tearing apart of Romulus, of Mettius Fufetius, or the multiplicity of the crowd are unified around Serres’ notions of power, shared perception, a version of the social contract, but a guilt-ridden one which entails acts of violence and dispersal, shared secrets. I think Serres is hinting at the development of ethics as opposed to metaphysics here, though again nothing is so clear.

Part three of the book continues the more historical aspect of the account, with the notion of exclusion. The process of forming a group includes the potential to exclude someone, and thereby some way of behaving, some mode of thought. The third way is no longer possible, or rather struggles to be possible. To some extent here Serres traverses similar ground to Agamben’s later Homo Sacer, by finding in the notion of exclusion of one individual something radically diminishing for all individuals. Agamben speaks of the reduction to bare life, Serres of the attempt to push something to the margin, but for Serres, the attempt is not necessarily successful. What is excluded is never entirely excluded, and to return to Serres’ topological concerns, there is a multiplicity of spaces ‘finely imbricated into each other, inextricably.’ (153). In other words, finally setting boundaries turns out to be impossible. In another set of metaphors, Serres compares the army which united the multiplicity and the plague which disunites the citizen body.^[17]

This leads us to the concept of instauration, and another qualification: ‘Rome is never founded enough, solidly enough, deeply enough, in its rules, in its rites, in the right gesture and the right rubric’ (187). So one keeps returning to the beginning and trying again. The repetition of founding, which is absolutely central to Serres’ idea (see also Kraus 1994), entails the possibility of doing things differently, but there are obstacles. Serres deploys a particularly dense set of ideas – myth, history which is actually myth but conceals it and so is a lie, death – all of which seem ranged towards the repetition of the same grim story. On the other side, there is dance, silent and in the open.

The image of the sunlit theatre which closes part three gives way to the agitated multiplicity of part four. Here we seem to go back again to the darker circle: ‘Foundation happens when multiplicity becomes unity. The multiple hardens around the unity of the corpse, around the place of the dead man.’ However, part four is about the dissolution of the unified multiplicity – external shocks, plague, the Volscians, are somewhat like the clinamen which Lucretius describes, the jolt or swerve that breaks things up and changes direction (Serres 2000).

In the last chapter, Serres offers another an alternative foundation, and one that is entirely peaceful. The crowd offers grain to the Tiber, which was once the Albula river before its name was changed. The grain is poured, the grain becomes solid, and the Tiber Island is formed. This then becomes an immensely fertile metaphor – and the pun is intended, because just as the rite is connected to agriculture and nature, so the swirling inchoate mass of mixed material, grain and chaff, improbably reaches down beyond the historical murderous foundations to make a new kind of foundation, a bridge from nature to culture and back again. This multiplicity works beyond history, out of time, ‘without murder, without putting to death, beneath the motionless summer sun’ (234).

This far from comprehensive account of a rich, at times baffling book will hopefully suffice to give an idea of the kind of work we are dealing with. What value does the book have for classicists, and what does it have to say in the context of our present context?

Evidently Serres cannot be used as a commentary on Livy in any direct way. Although some classicists have found inspiration from his work, it is an allusive, and if we are honest inaccurate reading, which is highly impressionistic, and which conflates unhelpfully different aspects of the Livian account.

The text riffs interestingly on some well-worn epistemological and phenomenological problems, whilst having ambitions to position itself within a canon of work which included among others Foucault, Girard and Dumézil. And it is impossible to read Serres I think without a degree of hindsight. As we have seen, Agamben, Latour, Descola all contain elements of the thought of Serres. The most significant crossovers are with Deleuze and Guattari, who also developed a theory of multiplicity, and that takes us to Hardt and Negri.^[18]Serres’ aversion to footnotes means that he concealed to a large extent his own influences, and he has perhaps been overlooked as an influence on others. Although the role of the classics in the development of post-structural thought is beginning to be more systematically acknowledged, here we have a central text by one of the most significant thinkers in the poststructural canon which is an extended and to some degree undervalued reflection on a key classical text (Miller 1998).

Religious, historical and discursive contexts for foundation stories and rituals

One way of concluding this essay then might be to note how Serres develops the theme of rite and myth as a postmodern technique of explaining the occlusion of the underlying realities around Rome’s foundation. The deconstruction of this deeper truth does indeed link religion and the city, in the sense that here at least we have a concerted attempt to take one of what Serres regards as the three canonical cities, the others of course being Athens and Jerusalem, and interpreting the mythical roots of Rome. ‘Athens is mind, Jerusalem is sign, and Rome is object’; Rome is of ritual not of myth (49); or rather, the myth is there to do the work of concealing the object, preventing not disclosing the multiple possible truths.

This is of course disputable, on many levels. The trio of cities, and the sketchy characterization of each is a problem. So too is the jump to an aphorism about Rome’s non-mythical status which is elsewhere presented quite differently, and where the terms are undefined. The book stands in a line of characterizations of Rome which compare it unfavourably against the Greek world. The bloody start of the city, which we did not need Serres to explicate, is used to ground an extended meditation on violence which parallels, but is not necessarily more illuminating, than that of Livy. Finally, the way religion is treated explicitly is almost entirely unhelpful; it is largely functional, divorced from any genuinely historical understanding, and is used to stand for models and symbols of authoritarianism.

Two key aspects of what Serres is doing are clear. First, Girard’s theory of the scapegoat is critical to the argument. Also, Girard’s theory of an original violence, which is then concealed through the secondary violence performed by the instruments of civilization, ritual, religion, ethics, is very close to Serres’ argument (Girard 2016, Girard 2017; see also Dougherty 1998). Repeatedly we find that the sacrificial victim (dismembered, buried, murdered) is needed to sustain the complex argument of how to resist multiplicity. The key moment then is the bloodless sacrifice of grain which is a different kind of foundation, and one which respects the multiplicity. But whereas Girard saw the scapegoat as a universal, Serres has a more complex relationship with the comparative approach. First he clearly differentiates Rome as we have seen, thereby undermining the universalist schemes of the structuralists, and second he shows no interest in any sort of sequence – his deities are not Indo-European, as were those of Dumézil.

Even so, Serres plays with archetypes all the time, and in this way his work in some sense replaces the Dumézilian triadic structure with a different structure, more fluid and more indeterminate, but nonetheless intended at some level to be a key to unlock an understanding of the past.

It is from this point that I want to try to weave a different kind of argument about what Serres might help us to see which is genuinely valuable to thinking about urban religions and foundation rites. I shall divide this into three parts: first the recurrence of foundation, second the mendacity of foundation and third the politics of foundation.

Chris Kraus’ brilliant article No Second Troy is one of the few I know to engage seriously with the repetitive nature of foundation as a Roman trope. Serres’ insistence on the recurrent refoundations of the city reminds us of Harriet Flower’s excellent demonstration that we may need to drop the idea of a single res publica in favour of a series of redefinitions of the Republic (much as the French do). The ancient sources present both the Romulean and the 509 BCE moment as res publicae. The Republic, for the Romans, was not born with the consulship. To some extent our problematic obsession with that moment of transition, with all the ensuing arguments about the early consulship, is a product of a rather modern periodization (Kraus 1994; Flower 2011; Smith 2011).

It is also important to note that development of the notion of the foundation in 753 BCE was slow. Serres’s approach, which risks making Livy’s text a sort of timeless witness to events, shares something with the methodological flaw underlying Carandini’s work in ignoring the evolution of stories, but pushing the Romulus story back from front stage into the mix of other accounts was rather perceptive. Both Peter Wiseman’s critique and Mary Beard’s earlier article on the Parilia show the extent of late Republican redescription of this particular story, and the increasing gravitational attraction of the Romulean myth, something which Poucet called ‘romulisation’ (Wiseman 1995; Beard 1987; Poucet 1981). This is not to say that some version of a foundation myth was not present at an early stage, but it may have been unrecognisable to us, and was at least in part the product of the sharpened chronological awareness which Denis Feeney described so well (Feeney 2007).

In short, Rome was founded many times, and the outline of the story was changed many times. We might choose with Serres to see the sacrificial element as a constant, and then argue that around this story repeated narratives of sacrifices were invented. Let us leave that as an open possibility, and draw out a further consequence of the argument, and one which Serres is clear about. That is, the story is in one shape or form untrue. For Serres, the mendacity of foundation lies somewhere between its obfuscation of knowledge and its reinforcement of power structures. Again, we might not have arrived at this point in the same way, but the argument has some force.

It is not only the Roman account which is multiple. As Naoíse Mac Sweeney has shown, this is characteristic of foundation stories. Her suggestion that, faced with this multiplicity, we should look at ‘complete foundation discourses … the sum total of several different myths together and the various relationships between the stories and variants’ is Serresian in approach, and compelling. Her work seeks to find a way between positivist readings which seek to rescue some historical truth and constructivist readings which make each variant a political discourse in and of itself. As she writes, ‘A more pragmatic approach would recognize that strategic invention and instrumental manipulation must certainly have been crucial factors in shaping both the content and the expression of foundation myths, but these were necessarily constrained by popular expectation, generic conventions, and plausibility. Investigating the audiences of myth, as well as the popular reception of foundation stories, is important to contextualize the literary or artistic expressions of these myths.’^[19]

By placing foundation stories in a more discursive context, and in relation to each other, Mac Sweeney jolts us out of the Rome-centred focus of this paper and into the much wider and more complex world of Italic foundations. We think therefore and immediately of Cato the Elder, and his work Origines, which itself set Rome precisely in the sort of discursive multiplicity of foundation stories, which eventually we find ourselves recovering largely through later sources such as Dionysius of Halicarnassus, Strabo and indeed the learned allusions of Virgil.^[20]

We are here in a historiographical mode, replacing the elemental with the erudite. Serres, who as we have noted dislikes the footnote, has elided this completely. What ultimately were these foundation stories about? What was at stake in retelling them? Mac Sweeney emphasizes the construction of identity, and this must be to some degree right; especially as time progresses, the overlapping of stories allows for increasingly complex and multifaceted approaches to identity and plurality, which were of exceptional significance in the fraught period when Ionian cities were establishing their identities in the sixth century, especially in relation to Homeric stories, and again in the tortured politics of the Hellenistic period.^[21]

The Roman foundation story likewise has to be contextualised ethnographically, historiographically and politically. Much work has been done here, though there are areas where our evidence still will not permit certainty. The Hercules story may well have been old; Stesichorus was writing about the cattle of Geryon in the sixth century BCE, and the figure of Aeneas is well known in the figurative depictions in Greek pottery, largely to be found north of Rome in Etruria (Davies and Finglass 2014; Enea nel Lazio 1981). As we have seen, it was a chronological challenge in part which developed the story of Aeneas, Alba Longa and Rome, but that needs also to be understood against the development of complex traditions around the indigenous or Greek peoples of pre-Roman Italy. The specific development of Rome’s relationship to Troy, as Andrew Erskine showed, was part of the same Hellenistic power politics as occupied the Ionian states (Erskine 2001). In addition, Magna Graecia was fertile soil for foundation stories, and their elaboration, and this historiographical tradition, represented for instance by Timaeus of Tauromenium, was likely to have been more important for the emergence of Roman historiography than we can now easily see (Baron 2013).

The fact of Greek colonization is, in all this, one of the very few indisputably factual elements. Mac Sweeney works hard to argue that the question of whether the foundation stories are true is unhelpful unless the definition of truth is established with care. She argues for useful fictions; and this is consonant with Wiseman’s approach to the myth of Remus, which seeks to locate it in specific political contexts.

Serres pushes us to acknowledge that these ‘fictions’ are knowing, and challenges us to call them out. The notion of foundation is effectively a conscious mendacity, an artistic redescription of reality. The pervasiveness of the capacity of the Greeks and Italians to rewrite these stories draws attention to their malleability. They are all black boxes, which operate opaquely to conceal the object of their action. Myth is, as Serres suggests, a delicate machine, difficult to see, which it would suffice to tamper with a bit in order to change the direction of history (21).^[22] So we arrive finally at religion. In what sense, if any, is the foundation of a city religious? One possible answer would lead us back towards a civic religion model, in which religion is purely instrumental for the interests of dominant powers. Whatever pre-existing narrative(s) we might think existed, religion was subsequently used to solidify through ritual calendars, festivals, chronicles, and together with the evolution of myth legitimates the identity stories, and is then recursively reaffirmed in its utility and truth through repetition.

However, this story has been found wanting by modern scholars, and Serres surprisingly and intriguingly pushes us in a different direction. On his argument foundation conceals and works against multiplicity. But the multiplicity keeps escaping, and one might argue that the number of variant foundation stories is another part of this phenomenon. The black box is not inescapable, but it remains the case that foundation stories are fundamentally about epistemology and politics.

If we take the view that foundation stories exist to solve a problem, or to answer a question, then we are it seems to me constantly directed to the issue of what makes a community and what sustains a community. Serres is I think right to talk about the role of multiplicity. This is in my reading a step past identity as such, if we mean by that a sense of geopolitical identity. Rather it may distinctly and sharply address an aspiration of identity, even at a personal level.

The historical problem is about causes and beginnings. The idea of a moment at which someone acts in a way which means that from that point onwards there is a community which did not exist before does not, as we have seen, mean that there was literally nothing before – in order to arrive at that paradox we need some sort of catastrophe such as the flood. Rather we find an accumulation of stories which point to the process of building a community. The question then remains as to what sort of community that will be, and whilst divine favour is a sine qua non, it is by no means the only and perhaps not the major component. Founders are often human (even if subsequently heroised or with a divinity somewhere in their background), sometimes all too human, and are nothing without their followers. This combination of privilege, frailty and dependence is a parable about leadership at its most stark. Even in the most politically effective act of foundation, a leader will be seen in his mortality and his insufficiency. This is precisely what opens up the space for the community.

Clearly one can espouse a Durkheimian view which would make religion the product of society’s need for a functional system of symbolic coherence, so that the religious celebrations of founding moments become expressions of a society’s need for meaning. The complication which the urban context may give is that there is no single community, but rather a complex set of groups looking for stories which retrospectively explain their co-existence. It is for that reason that the problem of foundation, the question it seeks to explain is, as Serres suggests, in part about the object of knowledge and in part about the community. What it is that is founded is dependent on the integration of pre-existing parts, and the work of knitting this together does not cease at that point, but has to be repeated and can be subtly changed.

Critically and logically, almost everything we know about actual foundations, including foundation rituals, postdate the beginning of community and presuppose the existence of community. To break this down a little, we know of foundations of cities and foundations of buildings. Most city foundation stories are later than the actual foundation, though as indicated the existence of dating systems suggests that there is an agreed starting point in some instances, and there is no reason in the case of colonies for instance that this is not agreed early on. That is not the same as saying that this was actually year one, and the literature on precolonial Greek presence is substantial and needs no elaboration here.

Foundation can be by new creation or by synoecism, or we may suspect by the conscious decision to ‘rebrand’ an existing group or groups. Foundation may be a matter of clarifying the boundaries between ethnic groups which had mixed more promiscuously before. In this sense the narrative of foundation accompanies rather than precedes the evolution of identity. It follows that whatever ritual of foundation we may think we can see or find is already the subject of discourse; it cannot be separated out from discourse. We cannot therefore disinter an act from which the story flows; what we think of as foundation of a community is in fact another way of telling a story about that community.

A slightly different argument needs to be made for Roman colonization (Bradley and Wilson 2006; Stek and Pelgrom 2014). Here, it has been argued, there was a very clear moment when the settlers marched out under the flag and took over land.^[23] However even here the idea of a foundation ab initio seems practically improbable, and overlooks the rocky start many colonies had, especially in northern Italy in the middle Republic. As Emma Dench notes, colonial foundations imitated the original Roman foundation, and thereby may have contributed much to the construction of the Roman myth; but they were also about appropriation and reinterpretation, as with the Oscan law from Bantia (Dench 2018). It is also true that we have been too quick to assume identity between what happened at Rome and what happened in the colonies, a sort of fixed rule book, which almost certainly never existed (Bispham 2000; Bispham 2006). It is thus dangerous to overstate the typicality of the few accounts, including figurative ones, that we have.^[24] However we look at colonies, they are fundamentally political acts in their patronage, establishment and intent, in which the favour of the gods is delivered through augury, a quasi-technological process which marries some notion of the transcendent with the measurement and division of space.^[25]

Even the act of dedicating space to the gods, and thereby placing it out of human hands, presupposes a community and non-sacred space. Foundation is a process of recognition and definition, by which something is said to come into existence, so foundation rituals are not yet acts of consecration. Even in the microcosm of a temple, the foundation is not singular.^[26] And it is interesting that in an exhaustive study of temple foundation rituals properly speaking – the preparation of a site – Gloria Hunt found no evidence of a reference to a deity (Hunt 2006: 194). Although this may have happened in the presumed accompanying prayers and so forth, the overall sense is of a generalised purification (Hunt 2006: 204; see also Osborne 2004). She goes on to illustrate the importance of the role of the patron; the presence of the support of the gods is as it were an affirmation of the status of the dedicator. The process of dedication was likely accompanied by dance and procession, an experiential involvement of the community (Marconi 2013; Hölkeskamp 2015 for more general remarks).

In all the many discussions of the foundation of Rome, which is surprisingly often taken as somehow typical, we risk losing sight of these three key points: that the act of foundation was not singular, it was seldom real, and its subsequent performances were always political, not merely in support of individuals but as enactments of community. The intersection of this with urbanity and religion tells us something about both. The urban context sustains and demands recurrent but persistently reshaped stories of foundation precisely because the community is large, segmented, and evolving.^[27] As for the religious element, I venture to suggest that religion gives the kind of symbolic language which answers the epistemological problem of what the city is, what thing it is that was founded. Although the answer is political and historical, the language is religious. That is because, I would suggest, religious thought offered the natural and appropriate mechanism and methodology. It is the delicate machine which Serres calls myth (see above). It is at least arguable, in the context of a research project on religion and urbanity, that one product of the intensity of urban exchange is the complexity of this machine and its operations. Foundation myths can be highly ambivalent, and each change in the story is a rewriting of history. The density of the urban situation permits this polyvalence.

There may be a further element at play because foundations are potentially fragile. We know that some colonies failed, and cities could be destroyed. Rome’s own story includes within it the destruction of Alba Longa and of Veii, and her own near disaster at the hands of the Gauls. The obsessive thinking and rethinking of foundation myths, the delicate machine of myth which constantly re-presents the point of origin, is also perhaps a response to the notion of precariousness. In other words, this was a story that needed to be told again and again, as a reaffirmation of the city’s continuing existence and against the possibility of its extinction. Here again, Serres’ circular repetitive account allusively reflects a reality which we see in Roman celebrations of her foundation, for instance at the Parilia, which we referred to above. In this ancient festival, annually performed but also in its evocation of a pastoral past increasingly distant from the bustling metropolis, we see the operation of the delicate machine of myth, ritual and religious action responding to the passage of time and the transformation of Rome itself (Beard 1987).

As a final point I wanted to return to the intellectual genealogy of these ideas. Geroulanos’ account of the deposition of the individual human being from the centre of philosophical attention in the post-war period closes as the brief reign of structuralism begins. Serres lived through and to a degree reacted against Althusserian Marxism and Levi-Straussian anthropology, both of which looked at structures rather than individuals, but inevitably his thinking responds to and is shaped by those ideas, and perhaps more significantly by the notion of archetypes which operate as tropes within human history. Just as Dumézil populated his world with competing abstractions, and Girard uses his concept of ‘double concealment’ to construct a narrative of violence and mimetic repetition (and in the process claim to be revealing something hitherto unknown), so Serres invents a cast of characters and presents them as refracting truths which are otherwise hard to discern. The result is a good example of what Kant might have called dogmatic speculation.^[28] However, there is some reason to suppose that while this may be worthless as history, and dubious as philosophy, it might be analogous to ancient thinking. Once we recognise that for the most part foundation stories are either invented or else imitate inventions, and that they serve the purpose of legitimating views of community which are usually freighted towards privilege and wealth even within a citizen body (let alone the disenfranchised slaves and quasi-silent women), they too appear as dogmatic speculation. In this sense, even if in no other, Serres’ work has the advantage of disclosing the notion of urban foundation and its mythical entourage as an intellectual construction, and one perhaps intimately connected to the challenges of urban life.