Monday, January 31, 2011

Do all cities have neighborhoods?

It's hard to imagine a modern city that does not have neighborhoods. What would residential areas in such a city look like? Is this even possible? Given the prominence of neighborhoods in social science research on life in cities today, I would guess that all modern cities do have neighborhoods. If a sociologist or planner, for example, identified a city that lacked neighborhoods, I'm sure they would study the situation and publicize it for being so strange.

For premodern cities whose housing and living conditions are described in historical documents, all or nearly all published examples have neighborhood organization (I haven't found a neighborhood-less city yet, and I haven't given up searching yet). As for cities only knowable through archaeology, my own specialty, neighborhoods are more difficult to identify but some progress is being made (Smith 2010). It seems that any time an archaeologist decides to look into housing and residential zones at an ancient city, the result is the identification of neighborhoods. My article on this is posted here.

What do I mean by neighborhood?  These are the working definitions I used in the article:

  • "A neighborhood is a residential zone that has considerable face to face interaction and is distinctive on the basis of physical and/or social characteristics" (Smith 2010:139).
  • "A district is a residential zone that has some kind of administrative or social identity within a city." (p. 140)

In the article I give some examples of premodern and nonwestern cities that have numerous small neighborhoods and a smaller number of (larger) administrative districts. The Hindu city of Bhaktapur in Nepal is an example (see Smith 2010 for details and citations). Although it may be difficult to distinguish neighborhoods and districts empirically, these concepts are important because they point to two of the major kinds of social dynamics that define and shape neighborhoods. On the one hand are bottom-up processes arising from social interaction among neighbors, and on the other are top-down processes of administration and control by city or state authorities. Much of what happens in urban neighborhoods is a result of the interaction of these bottom-up and top-down processes within a given built environment.

So far, we are batting 1,000. Whether one looks at modern cities, historically documented premodern cities, or archaeologically excavated ancient cities, all have neighborhood organization. But that's not all. Some large village settlements (e.g., prehistoric pueblo socieites in the U.S. Southwest) are divided into housing clusters or zones that resemble neighborhoods. And rapidly urbanizing sites, such as squatters settlements in the developing world, tend to have neighborhood organization. Even Black Rock City, the temporary city that is the site of the Burning Man festival each year, has neighborhood organization (generated by both bottom-up and top-down forces).

If neighborhoods are truly a universal aspect of urban organization, two questions are worth exploring: (1) why is this the case? and (2) what are the implications for modern cities and urban policy? Stay tuned, we don't have the answers yet. In the meantime, you can find out about a transdisciplinary research project on urban neighborhoods and open spaces.


Smith, M. (2010). The archaeological study of neighborhoods and districts in ancient cities Journal of Anthropological Archaeology, 29 (2), 137-154 DOI: 10.1016/j.jaa.2010.01.001

Thursday, January 27, 2011

Mark Twain on the urban villages of London

The newly published autobiography of Mark Twain contains this gem of a description of London in 1896:

One little wee bunch of houses in London, one little wee spot, is the centre of the globe, the heart of the globe, and the machinery that moves the world is located there. It is called the City, and it, with a patch of its borderland, is a city. But the rest of London is not a city. It is fifty villages massed solidly together over a vast stretch of territory. Each village has its own name and its own government. Its ways are village ways, and the great body of its inhabitants are just villagers, and have the simple, honest, untraveled, unworldly look of villagers. Its shops are village shops; little cramped places where you can buy an anvil or paper of pins, or anything between; but you can’t buy two anvils, nor five papers of pins, nor seven white cravats, nor two hats of the same breed, because they do not keep such gross masses in stock. The shopman will not offer to get the things and send them to you, but will tell you where he thinks you may possibly find them. And he is not brusque and fussy and unpleasant, like a city person, but takes the simple and kindly interest of a villager in the matter, and will discuss it as long as you please. They have no hateful city ways, and indeed no ways that suggest that they have ever lived in a city.
       -- Mark Twain  (2010)  Autobiography of Mark Twain, volume 1. University of California Press, Berkeley, page 108.

The metaphor of "urban villages" has been common in urban studies for some time, from the classic Boston ethnography of Herbert Gans (1962) through some of the recent new urbanist planning literature (Neal 2003) to contemporary research on Chinese urbanization (Hao et al. 2011).Writers often seem surprised to find that neighborhoods in big cities are like villages, but that surprise derives from the sterotypical western view of the urban way of life (Wirth 1938). In fact, neighborhoods are one of the few universal features of cities (Smith 2010), and the social dynamics of urban neighborhoods often parallel the social dynamics of villages. Life is lived at a social and spatial scale much smaller than the entire city; people often know their neighbors and cooperate in various ways. Urban villagers may have rural-like customs and practices, or rural-like cultural values (as noted by Twain).

One reason for the existence of urban villages is the role of migration in the formation and maintenance of neighborhoods. Premodern cities were demographic sinks, with high mortality rates. Neighborhoods often formed, and were maintained, by migration from a particular rural area, giving them both a rural complexion and a social or cultural distinctiveness. And migrants often maintained social relations with their relatives in the countryside.

Of course modern inner-city neighborhoods often depart from the "urban village" model, as research by Robert Sampson and other sociologists shows. But it turns out that there is a remarkable continuity in the presence and locations of the poorest neighborhoods from Twain's London until the present. Charles Booth's social maps of London in 1898 are surprisingly similar to such maps compiled today. This figure, first published in The Economist (May 4, 2006) is reproduced from Sampson (2009). Perhaps London's poorer neighborhoods today are less like "villages" than they were in the days of Twain and Booth, but some of the basic neighborhood structure has endured for more than a century. I wonder what Mark Twain would make of modern London (or Hannibal, Missouri, for that matter).

Thanks to Seven Tomek for bringing the Mark Twain quote to my attention.

Gans, Herbert J.
1962    The Urban Villagers: Group and Class in the Life of Italian-Americans. The Free Press, New York.

Hao, Pu, Richard Sliuzas, and Stan Geertman
2011    The Development and Redevelopment of Urban Villages in Shenzhen. Habitat International. In Press, corrected proof available online.

Neal, Peter (editor)
2003    Urban Villages and the Making of Communities. Spon Press, London.

Sampson, Robert J.
2009    Disparity and Diversity in the Contemporary City: Social (Dis)Order Revisited. British Journal of Sociology 60:1-31.
Smith, Michael E.
2010    The Archaeological Study of Neighborhoods and Districts in Ancient Cities. Journal of Anthropological Archaeology 29(2):137-154.

Wirth, Louis
1938    Urbanism as a Way of Life. American Journal of Sociology 44:1-24.

Monday, January 24, 2011

Myths of the urban revolution

The phrase “Urban Revolution” was coined in the 1930s by archaeologist V. Gordon Childe to describe the transformation of Neolithic village farming society into the urban states and empires of the Bronze Age. Prior to the Urban Revolution there were no kings, no cities, no writing systems, and no social classes. This was perhaps the most extensive and far-reaching social transformation in human history, with greater implications for life and society than either the Industrial Revolution or the Neolithic Revolution (yes, I know, historical sociologists will not believe that anything was more important than the Industrial Revolution).

Today, archaeologists have quite a bit of information about the Urban Revolution, which happened independently in several parts of the ancient world. For background and discussion, see Childe (1950), Smith (2009), Trigger (2003), or Johnson and Earle (2000). The Urban Revolution is not just about the origin of cities; instead it describes a broader set of social transformations, of which urbanization was only one component. (I realize that other writers and websites have used the phrase “urban revolution” to mean widely divergent things, but Childe’s usage dates to 1936, so archaeology wins out by precedent).

So, what are the myths of the urban revolution? I use this phrase to describe things that many people think are true, when in fact they are errors. Here is the big myth of the urban revolution:

  • ·        The urban revolution represented  progress for human life, with things improving for individuals and society.

It turns out that archaeology and anthropology have accumulated quite a bit of evidence showing that this was not a positive transformation for most people in most societies. Here is what happened to life and society after the Urban Revolution:

1.      People had to work harder to make a living.
2.      People had less freedom and self-determination.
3.      Human health went into a nose-dive: people had more diseases and lifespan was lowered.
4.      Violence and chaos increased in many cases.

Points 1-3 are strongly supported by considerable empirical evidence. Indeed, these are the kinds of things that get asked on exams in introductory anthropology classes. Point 4, as phrased here, is more controversial, but it includes an important insight often ignored by political scientists and economists—that social order was typically maintained in pre-state and non-state societies. Let’s review these points a bit:

1.      People had to work harder to make a living. Ethnographers have shown that modern hunter-gatherers spend LESS TIME making a living than farmers, and that tribal (not-state) farmers spend LESS TIME to produce the food they need than their peasant descendants after the Urban Revolution. This may seem counter-intuitive: weren’t hunters living at the edge of starvation until farming made life easy for everyone? Well, ethnographers have timed people as they live their lives, and hunter-gatherers put in less time to feed their families than do farmers. And after the Urban Revolution, people not only had to feed their families, they also had to produce a surplus to pay rent and taxes, which made them work longer hours. Archaeological research supports this model, showing, for example, that time-consuming agricultural features (e.g., irrigation canals, or hillside terraces) were generally built AFTER the Urban Revolution. They weren’t needed before then.

2.      People had less freedom. People in tribal societies had considerable freedom to live their lives as they saw fit.  In tribal society there are no leaders with coercive power, no laws or written regulations, no police force. If a family is dissatisfied with life in their group, they can leave and move elsewhere.

3.      Human health went into a nose-dive: people had more diseases and lifespan was lowered. Archaeologists can monitor ancient health and nutritional status by studies of the human skeleton from burials (this field is called bioarchaeology, or human osteology). Populations after the Urban Revolution had higher incidences of many diseases, they had more skeletal markers for childhood deficiencies of protein and other nutrients, and the average age at death went down. There are several reasons for this: diets were less varied, life in cities increased the level of communicable diseases (from dense and unsanitary conditions), and poverty increased. For discussion of points 1-3, see Johnson and Earle (2000), or a textbook such as Harris (1983), or Trigger (2003).

4.      Violence and chaos increased in many cases. This is the only one of my four points that is at all controversial. The extent, nature, and implications of ancient and non-western warfare are currently the subject of extensive debates. These debates draw on archaeological data, ethnographic observations, history, and theoretical models from ecology and economics. My own reading of the evidence is that warfare increased after the Urban Revolution (because states find more reasons to go to war, and their wars are larger scale and more deadly than tribal warfare).

But whether or not I am correct about levels of warfare, one aspect of point #4 is not controversial at all: non-state peoples maintained social order. Life was not chaotic or violent prior to states, and Thomas Hobbes was just plain wrong in calling life in the state of nature (i.e., outside of states) “solitary, poor, nasty, brutish and short.” Now I realize that some prominent modern scholars have claimed that pre-state life was chaotic and violent. Nobel Prize winner Dougas North and his colleagues, for example, have built a model for social change that posits chaos before the state (North et al. 2009a, b). Life was so chaotic and violent that nothing was accomplished (in their terms, there was little economic growth). So elites had to step in and establish laws and private property and formal government to reduce the endemic violence that Hobbes had described. I’m sorry, but this is just plain incorrect.

My suggestion for anyone who wants to believe in the violence of non-state society is to read some ethnography. This is pretty basic, low-level stuff in the field of anthropology; any student in an introductory cultural anthropology class can probably rattle off the ethnographic evidence for social order and control in tribal society, and the social mechanisms by which order is maintained. If you don’t want to take an intro anthropology class, then try a few of these works: (Harris 1983, 1989; Roberts 1979; Taylor 1982). Or check out Sillitoe and Kuwimb’s (2010) critique of one of Jared Diamond’s recent controversial papers.

So, if the social changes of the Urban Revolution were the opposite of progress (“De-evolution”?? Devo?), then why did it happen at all? Why aren’t we all still hunter-gatherers, or tribal farmers? Good questions. I’m out of space now, so I’ll duck this one for the moment. Let me just say that this is a big research question in modern archaeology.


Childe, V. Gordon
1950    The Urban Revolution. Town Planning Review 21:3-17.

Harris, Marvin
1983    Cultural Anthropology. Harper and Row, New York.

1989    Life Without Chiefs. New Age Journal Nov/Dec: 42-45, 205-209.

Johnson, Allen W. and Timothy K. Earle
2000    The Evolution of Human Societies: From Foraging Group to Agrarian State. 2nd ed. Stanford University Press, Stanford.

North, Douglass C., John J. Wallis, and Barry R. Weingast
2009a  Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Cambridge University Press, New York.

2009b  Violence, Natural States, and Open Access Orders. Journal of Democracy 20(1):55-68.

Roberts, Simon
1979    Order and Dispute: An Introduction to Legal Anthropology. St. Martin's Press, New York.

Sillitoe, Paul and Mako John Kuwimb
2010    Rebutting Jared Diamond's Savage Portrait: What Tribal Societies Can Tell us About Justice and Liberty. StinkyJournalism .Org website:published online.

Smith, Michael E.
2009    V. Gordon Childe and the Urban Revolution: An Historical Perspective on a Revolution in Urban Studies. Town Planning Review 80:3-29.

Taylor, Michael
1982    Community, Anarchy and Liberty. Cambridge University Press, Cambridge.

Trigger, Bruce G.
2003    Understanding Early Civilizations: A Comparative Study. Cambridge University Press, New York.

Sunday, January 23, 2011

Did ancient cities have urban sprawl?

In his controversial book, Sprawl, A Compact History, Robert Bruegmann (2005) claims that urban sprawl has always been with us. Low-density, unplanned residential areas have surrounded most cities, from the Urban Revolution to the present. Bruegmann makes this claim as part of an argument that sprawl is not as bad as modern urban critics have claimed. In his view, sprawl has always surrounded cities, and rather than complaining about it or trying to get rid of it, we should just learn to deal with it.

But I am not interested here in Bruegmann’s larger arguments or his debates with the new urbanists. Rather, my concern is with Bruegmann’s claim that sprawl existed “in almost every era in urban history.” Is that the case? The only evidence he marshals is the fact that wealthy Romans had villas outside of the capital, which is certainly not adequate support for his claim. But in fact I think he may be largely correct.

How should we define urban sprawl? Definitions are important, because they shape the way we look at the world, and they define the parameters of phenomena. Some definitions of sprawl rule out the possibility that sprawl existed before the Industrial Revolution. For example, Dolores Hayden’s definition of sprawl is: “a process of large-scale real estate development resulting in low-density, scattered, discontinuous car-dependent construction, usually on the periphery of declining, older suburbs and shrinking city centers” (Hayden 2004:7-8). For premodern cities, this simply is not a useful definition.

Bruegmann, on the other hand, provides a much broader definition of sprawl that is useful for comparison. Sprawl is defined as “low-density, scattered, urban development without systematic large-scale or regional public land-use planning” (Bruegmann 2005:18)

No one has yet carried out a systematic comparative analysis of sprawl (or sprawl-like settlement) around ancient cities. But I’ve looked at lots of city maps and read lots of archaeological and historical reports on premodern cities, and I am struck at how many of these have unplanned, low-density residential zones outside of the main city area.

Urban sprawl is especially easy to identify for walled cities. Whenever cities are walled (whether ancient Chinese imperial capitals or medieval European towns or Sumerican city-state capitals), they have settlement outside the walls. The map of medieval Ipswitch shows this pretty clearly (the city wall is indicated by the red line). Many ancient low-density tropical cities (such as those of the Classic Maya, or Angkor and other ancient Khmer cities) have declining density towards their edges. Two adjacent Maya cities in Belize, Xnaheb and Nim Li Punit, show this pattern: both have declining density as one moves away from the city centers (which have temples, pyramids, and other civic structures), but Xnaheb is more “sprawling” than its neighbor (this map was redrawn from Jamison 1993).
Based on many such examples, my subjective impression is that sprawl was quite common around ancient cities. But just how common was it? Did ALL walled cities have extra-mural housing, or only some of them? Do all Maya cities exhibit sprawl like the two shown above, or only some examples? Until someone makes the effort to select a scientific sample of ancient or premodern cities, we cannot answer these questions. Anthropologists know something about sampling non-western cultures (Ember and Ember 2009), but this knowledge has yet to be applied to premodern cities.

But even if we put together a group of ancient cities that might form a reasonable sample, can we analyze the dynamics of ancient urban sprawl? Obviously the cause of ancient Maya sprawl was not the automobile. It was probably not the real estate market and certainly not government subsidies for highway construction. But perhaps transport costs played a role. Was sprawl more prominent in societies with wheeled transport (e.g., Rome, medieval Euopre) than those that lacked the wheel (e.g, the Maya or Inka)? What about the size of the city, and/or the intensity of agriculture in the urban hinterlands? I’d love to know the answer to these and other questions about the origins and patterns and dynamics of ancient urban sprawl.

I think archaeologists and urban historians now have the data to make real headway here, but until my colleagues and I put forth the conceptual and empirical effort to address such questions, we will have to be content with impressions. Will information about ancient sprawl solve the problems of modern sprawl? No, of course not. But as scholars, planners, and officials search for solutions to today's urban problems, perhaps a broader base for comparison might help generate new ideas. And adding ancient cities to our knowledge of modern sprawl (and other issues) will certainly help scholars separate universal urban traits from those traits that are particular to individual cities or regions or time periods.

To read a lengthier and more technical discussion of this issue, see Smith (2010).


Bruegmann, Robert  (2005)  Sprawl: A Compact History. University of Chicago Press, Chicago.

Ember, Carol R. and Melvin Ember  (2009)  Cross-Cultural Research Methods. 2nd ed. AltaMira, Walnut Creek, CA.

Hayden, Dolores  (2004)  Field Guide to Sprawl. Pantheon, New York.

Jamison, Thomas R.  (1993)  Symbolic Affiliation, Architecture and Settlement Patterns in Southern Belize: Nim Li Punit and Xnaheb during the Late Classic. Ph.D. dissertation Dissertation, Department of Anthropology, University at Albany, SUNY.

Friday, January 21, 2011

The "Wide Urban World"

The phrase "wide urban world" encompasses the many manifestations of cities and urbanism on earth, from early Mesopotamia to modern Mumbai. It describes an approach to cities that is comparative and historical. A comparative perspective is necessary if one wants to draw conclusions about urban phenomena that go beyond a single city. Most comparative urban analysis in the social sciences is quite limited in scope, comparing, say Detroit and Pittsburgh, or New York and London, or perhaps Mumbai and Shanghai. But with a  historical perspective, comparisons can range back into deep history.

Does it make any sense to compare Phoenix and ancient Teotihuacan? Perhaps there are common dynamics of large cities in dry environments. How did the people (more than 100,000 of them) of Teotihuacan deal with shifting rainfall and agricultural productivity? Can this give us any ideas about the sustainability of modern Phoenix?

What about ancient Rome and modern Los Angeles? These are/were huge metropoli with great cultural and social diversity, from ethnicity to social class to occupation. Are there basic principles by which the leaders of large diverse cities have dealt with such diversity?

A broad perspective on comparison can be advantageous in several ways:
  1. Information about premodern cities may help scholars, planners, and officials better understand modern cities, and it will give them a larger inventory of cases to draw on in designing solutions for modern urban problems and issues. But this requires rigorous data on premodern cities, solid knowledge that is based on historical and archaeological research and not on speculation.
  2. Information about modern and many historical cities helps archaeologists understand ancient cities. Nearly all archaeological inferences about ancient society are based on analogy (inductive logic), which means comparisons with better-known cases. So a wide perspective helps archaeologists understand the ancient world and the first cities.
  3. The broad historical and comparative perspective of the wide urban world is necessary to address questions about cities and urbanism as general phenomena and processes. It is not scientifically defensible to look at modern U.S. cities and make conclusions about urbanism in general. One can talk about modern U.S. cities in general, but if we want to know about cities as a form of human settlement, then we simply cannot ignore the thousands of years of urban development before the Industrial Revolution.
These points give an idea of the perspective of this blog, the "Wide urban world." We will cover themes such as urban sprawl, sustainability, neighborhoods, housing, city planning, squatters, and the urban built environment.

For a more technical description of this broad perspective on urbanism, see this White Paper, submitted to NSF by members of our urban organization project:

Smith, Michael E., Christopher Boone, George L. Cowgill, Sharon L. Harlan, Alison Kohn, Barbara L. Stark and Abigail York  (2010)  An Expanded Social Scientific Perspective on Urbanism. White Paper, Future Research in the Social, Behavioral and Economic Sciences. National Science Foundation, Washington, DC.

This document is posted on the project website, and its now available on the NSF website as well.