Work in Progress: Cities and Civilizations

Carl Stephenson

Recently I received a copy of Methods in Social Science: A Case Book (edited by Stuart A. Rice) by interlibrary loan, which I had ordered for the paper by Henri Pirenne that it includes, “What are historians trying to do?” Another paper in this volume, “The Work of Henri Pirenne and Georg von Below with Respect to the Origin of the Medieval Town” by Carl Stephenson, has given me much to think about. The name of the author didn’t ring a bell with me, but I looked him up found that he is the author of the widely available Medieval Feudalism (I assume that it was once commonly used as a textbook). I had acquired a copy of this book years ago, read it, and was impressed by the clarity of the exposition.

Stephenson notes at the beginning of his paper that, “A hundred years ago historians still glibly asserted the persistence of Roman municipal institutions.” Despite the many footnotes to the paper (most references are to German papers, with a smattering of French and English works), this claim is not footnoted, and is presented as though it is common knowledge. So I wonder who the historians were in the early nineteenth century who regarded the persistence of Roman municipal institutions as common knowledge. But that isn’t what interested me about that paper.

Stephenson compares the views of five historians and their theories of the origins of medieval cities, and then he gives a list of properties (p. 369) which could be said to be the differentia between villages and cities. These properties include: 1) having a market, 2) being walled, 3) constituting a jurisdictional unit, and 4) enjoying special dispensation with regard to fiscal, military, and other political responsibilities. After naming these four differentia, Stephenson adds, “All four of these characteristics were matters of public law, for they came by virtue of state endowment.” I will come back to this.

After outlining the state of the debate over medieval cities in the late nineteenth century, Stephenson introduces Pirenne and his ideas, which focus on the city as a locus of trade and the emergence of a distinct mercantile class involved in trade. These two characteristics — trade and a trading class, which I guess could be thought of as one or two characteristics — can be added to the list given in the previous paragraph, so now we have five or six characteristics as the differentia of village from city. Having a market might be considered to be the same as being a center of commerce, and there is some discussion of this in the paper. Even in the early Middle Ages there were small markets, but these markets were insufficiently large to support and sustain a distinct mercantile class, so one might consider the emergence of such a class as more distinctive that the existence of trade per se.

Henri Pirenne

In Stephenson and the scholars he cites, the emergence of cities in medieval Europe is treated as something new in history. Stephen writes in terms of, “the beginnings of urban life in Europe” (p. 371), while Pirenne, in his famous work Medieval Cities, wrote, “The birth of cities marked the beginning of a new era in the internal history of Western Europe” (p. 153; this is the first sentence of the final chapter). But we know that cities are certainly not new to history, and Stephenson said it was formerly a commonplace that Roman municipal institutions survived from classical antiquity.

Cities were not new to history, and not new to Europe, but they were new to the civilization of medieval Europe that was then in the process of development. Europe had, essentially, lost its Roman cities (cities had ceased to function as cities, though they often remained settlements), and when the civilization of medieval Europe began to re-establish commerce, cities began to appear, some of them at the sites of former ancient cities, and some of them founded de novo.

Western civilization passed through a distinctive, and perhaps unique, stage of development in the early Middle Ages, which we formerly called the Dark Ages, except that historians don’t use this term anymore. During the early Middle Ages, cities that had been depopulated in the late Roman period declined further into mere ruins with an adjacent village. Rome itself declined from a peak of around a million to a few thousand residents — some put the number at 50,000, some put the number less than 30,000. But it is obvious that a once built-up area had largely returned to agricultural use, as statues like the Laocoön group were dug up in gardens and vineyards that had been planted long after the structures of the built environment had been forgotten.

During the Dark Ages (for lack of a better term), there were fortresses, administrative centers, churches, and monasteries. Many of these had adjacent villages, but according to Pirenne and Stephenson and most other scholars, a castle plus a village does not a city make. The royal court was essentially nomadic, moving from one castle to another, or from one military campaign to another, without settling down in one place for a period of time sufficient to attract the kind of population that would turn some center into a city. Above all, there were manorial estates, ruled by a feudal lord who had sworn his loyalty to the crown, but who, for all practical purposes, was the unquestioned ruler within his domain. I find it interesting that Stephenson had so little to say about manorial estates, as I would maintain that this was the fundamental unit of civilization in the early Middle Ages, and this is what made early medieval civilization so distinctive, so different from other civilizations and so different from its classical ancestor: instead of civilization being based on the city, it was based on a loose network of manorial estates.

By invoking Western Civilization above (i.e., the above paragraph but one) I am implicitly relying on the more abstract conception of civilization (civilization at a higher “level”) that I discussed in newsletter 173, i.e., civilization as it exists on the level of a cluster of civilizations or a series of civilizations; in a less abstract conception of civilization, the civilization of classical antiquity came to an end, and the civilization of medieval Europe began to form, the latter following the former, influenced by the former, but distinct from it. When we posit a construction like “Western Civilization” we are talking about a different kind of civilization than the kind of civilization that classical antiquity constituted on its own. The implied continuities that make Western Civilization into one great entity comprising ancient, medieval, and modern civilizations around the Mediterranean, Europe, and eventually the New World, are not continuities of administration, of political regimes or institutions, not even, according to Pirenne and Stephenson, continuities of cities (and the institutions based in cities).

I found myself so intrigued by Stephenson’s list of five or six characteristics that distinguish a village from a city because it reflects a later discussion of the origins of pristine cities at the very origins of civilization. The standard by which we naturally measure theories of urbanization is V. Gordon Childe’s 1950 paper “The Urban Revolution” published in Town Planning Review (V. Gordon Childe, “The Urban Revolution,” The Town Planning Review, Vol. 21, №1 (Apr., 1950), pp. 3–17); Childe’s paper dates from a couple of decades after the debates discussed by Stephenson. Childe, too, wanted to distinguish a village from a city, but he wanted to distinguish a Neolithic village from a city, while Stephenson wanted to distinguish a medieval village, which may well have be co-located with a now-defunct ancient city, and a medieval city.

V. Gordon Childe in the center.

Childe had also produced a list of characteristics that he said were diagnostic of an urban revolution in the ancient world. His list consisted of ten items, including 1) the extent and density of settlements, 2) division of labor, i.e., craft specialization, 3) surplus value transferred to social elites (which might also be called “capital accumulation”), 4) monumental architecture, 5) social stratification, 6) writing, 7) science, 8) art, 9) trade, and 10) prioritizing residence over kinship. Of Childe’s ten items, the only item shared by Stephenson’s list is trade.

Childe’s list of ten characteristics of a city has been enormously influential, and continues to be influential today. Moreover, it has expanded beyond diagnostic criteria for an urban revolution to commonly being used by archaeologists as diagnostic for civilization. In a sense, Childe’s list of urban characteristics has functioned as a framework for the discussion of the emergence of civilization much as the Drake equation has functioned as a framework for the discussion of ETI and its ability to community with us (or to us). What this means is that later scholars are continuously revising the list of characteristics, and there is no settled consensus, but there is at least a common conceptual framework within which shared problems can be discussed. But Childe was a prehistorian and an archaeologist; accordingly, his interest was in the foundation of pristine cities, i.e., cities that appeared where no city had previously existed, and this makes Childe’s conceptual framework for urbanization distinct from Stephenson’s conceptual framework for urbanization.

The first four characteristics introduced by Stephenson were, as he said, matters of public law that came by virtue of state endowment. That cities should arise in this environment immediately distinguishes medieval cities from the emergence of ancient cities. The ancient cities that arose in the earliest history of civilization, and indeed which cities constituted the earliest history of civilization, could not have arisen by virtue of state endowment, because there was no state as of yet. But while the medieval state was weak and often ineffectual, because many of the fictions of Roman law and classical antiquity were maintained well into the Middle Ages, this fiction could always be drawn upon to impart a legal basis for a city.

Another unavoidable reference in the scholarly discussion of the origins of cities is Fustel de Coulanges’ classic work The Ancient City, one of the great works of nineteenth century history, which takes a period in western history between that of Childe (prehistory) and Stephenson (the Middle Ages), viz. the city of classical antiquity, and, even more so, the city of early antiquity. Coulanges uncompromisingly states his thesis early in the book, which is that the city developed from the religious tradition of the ancient family.

Numa Denis Fustel de Coulanges (18 March 1830–12 September 1889)

Coulanges concludes the book on an equally uncompromising note:

“Our study must end at this limit, which separates ancient from modern polities. We have written the history of a belief. It was established, and human society was constituted. It was modified, and society underwent a series of revolutions. It disappeared, and society changed its character. Such was the law of ancient times.”

The final chapter, of which the above paragraph is the last, begins, “The victory of Christianity marks the end of ancient society.” What Coulanges is saying here is that, already in classical antiquity, the kind of city that constituted the earliest human societies had changed to the point of no longer being recognizable, and therefore, in a sense, ceased to exist. A new kind of city came into being, and, with it, a new society came into being. Thus the new cities of medieval Europe, which were distinct from the cities of classical antiquity, were distinct from cities that were themselves distinct from earlier cities, the cities that constituted the bulk of Coulanges’ inquiry, and perhaps these cities, too, were distinct from the cities that Childe had in mind when he wrote about the urban revolution.

I have already explained above part of how this interests me: if the cities of medieval Europe were new, then urban continuity is not the basis for the continuity of Western civilization, which basis, if there is any, must be located in institutions that transcend the city and its institutions. Regular readers will understand how this resonates for me, as I often express my formulations of civilization in terms of its institutional structure. But we can easily segue from an abstract discussion in terms of institutional structure to a much more concrete and pragmatic idiom. If the city and its institutions are the basis of civilization — and this is something that many historians and archaeologists can agree upon, even if they dispute the details of any agreement — then the kind of fundamental changes that we see in the conception of a city in Childe, Coulanges, Stephenson, and Pirenne should constitute fundamental changes in civilization.

Indeed, I would agree with this, which means that a corollary of my definition of civilization in terms of its institutional structure must be that the institutional structure of civilization remains invariant even as the city and its institutions undergo change. A further corollary is implied by the above observations about constructions of civilization that apparently involve multiple civilizations in series or in parallel (what I call series and clusters): the ascent to a more abstract and general conception of civilization leaves the institutional structure invariant, even as the particular institutions that jointly constitute the higher-level institutions may change, and may even be in flux.

Sometimes as an alternative to my definition of civilization in terms of institutional structure — a conceptual framework joined to an economic infrastructure by a central project — I employ a pragmatic definition of civilization as a network of cities in relationships of cooperation, competition, and conflict. This pragmatic definition of civilization constitutes what I called a “basal civilization” in newsletter 173. I believe it can be shown that the pragmatic definition of civilization is effectively equivalent to the institutional structure definition, and vice versa, but I will not belabor that at present.

This pragmatic definition of civilization is helpful as a rule of thumb, to identify civilizations in history, but it also has theoretical implications. The pragmatic definition of civilization gives us a lower bound for the extent of a civilization in space and time: there must be enough space and time, and population sufficient for multiple cities to form and to establish relationships with each other. A lower bound, in turn, is necessary to the most basic quantification of our understanding of civilization, which quantification has been sorely lacking.


The pragmatic definition of civilization is also much more particularistic that my definition in terms of institutional structure, and here we see the relevance of the various conceptions of the city as theorized by Childe, Coulanges, and Pirenne, inter alia. The city at the beginning of history — say, Nineveh or Babylon — the city of early antiquity — say, Athens or early Rome — the city of late antiquity — say, Antioch or Constantinople — and the city of medieval Europe — say, Paris or Regensburg — are each of them institutionally distinct from the others. They can’t even be defined by the same list of characteristics, as we saw in the difference between Childe’s litany and Stephenson’s litany of characteristics of cities. And yet, all were cities. All had urban institutions, though to put it in that way is obviously circular, and it cannot satisfy us except as a placeholder to later be formulated in a more analytical fashion.

All of these theoretical problems are, at the same time, theoretical opportunities to further clarify the concepts and distinctions I have been using. Just by writing this I have started toward a better clarification of how basal civilizations are distinct from a more abstract conception of civilization, even as institutional structures remain invariant. And I see that a subtle distinction needs to be made, such that civilization does not inhere in the city as such, but in a network of cities; that is to say, civilization is an emergent from multiple cities that cannot be reduced to any one city. The institutions of civilization, such as they are, are then distinct from particularistic municipal institutions; any municipal institutions can enter into the institutional structures of civilization, so that, while the institutions of the city and the institutions of civilization are closely related, they are not identical.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store