Race, ethnicity, social class: Are most urban neighborhoods homogeneous or heterogeneous?

Many modern cities are segregated by race, class, or other parameters, and mixed neighborhoods seem rare. Most people believe that this is not a good situation, for many reasons. Much effort is devoted to trying to reduce the degree of segregation, and there is much research about how to do this. Some writers suggest that in the distant urban past, there were more mixed neighborhoods, and homogeneous neighborhoods are a modern phenomena. Others suggest that traditional cities always had neighborhoods organized by ethnicity or class or occupation. What is the truth here? What do we know about the extent of social clustering in premodern and nonwestern cities? When people who are alike cluster in neighborhoods, is this because they prefer this arrangement and make decisions to bring it about? Or are they forcibly clustered into ghettos, and then prevented from moving by laws and other top-down practices? Or perhaps such patterns arise as byproducts of other actions and decisions? If we study these things for premodern cities, can we derive any lessons for modern urbanism?

These are some of the questions that motivate a research project I am involved in called "Urban organization through the ages: Neighborhoods, open spaces, and urban life." It is part of a series of transdisciplinary research projects at Arizona State University called "Late Lessons from Early History." One of whose goals of this program is to make comparisons between modern and past societies and try to draw lessons for modern society. If you have followed this blog, you will know one of my main purposes here is to explore connections and comparisons between premodern and modern cities. Not only do I write about both modern and ancient cities, but I often compare them or use examples from both categories to make a point.

Our research project has six principle investigators, representing the disciplines of anthropological archaeology (yours truly), sociology, geography, and political science.Our first joint article was published over the summer:

York, Abigail, Michael E. Smith, Benjamin Stanley, Barbara L. Stark, Juliana Novic, Sharon L. Harlan, George L. Cowgill, and Christopher Boone  (2011)   Ethnic and Class-Based Clustering Through the Ages: A Transdisciplinary Approach to Urban Social Patterns. Urban Studies 48(11):2399-2415.

I'd like to be able to say that we have solved the historical puzzles of urban social clustering and segregation, but alas, we have only made a modest contribution. We describe our transdisciplinary and comparative approach to the problem; we discuss a number of drivers or forces that contribute to social clustering at the neighborhood level, and we give a few examples of how these have played out in different historical and geographical settings. We use the term "clustering" because the word "segregation" has considerable baggage in modern parlance, with implications that limit its application to premodern cities. There was ethnic clustering in many premodern cities, for example, but the dynamics were quite different from modern racial segregation. Clustering is a  more neutral term, better for comparative analysis.

One of our conclusions is that there was no single "traditional" form of social clustering. Many writers over the years have assumed that modern western cities developed out of a prior pattern of traditional cities (sometimes traditional means seems to mean medieval, sometimes early modern, sometimes Classical Greece or Rome). If we can understand the traditional situation and how it changed with modernization, this will help us understand modern cities. But there was never any single "traditional" pattern.

Another conclusion is that patterns of social clustering vary greatly, both within and between urban traditions. There is no such thing as a "typical" medieval European urban pattern; some medieval cities had homogeneous neighborhoods, some had mixed neighborhoods. There was no typical "Aztec" pattern or "Islamic" or "Chinese" pattern. Cities varied in their neighborhood organization within cultures or within urban traditions.

And a third conclusion of our paper is that there are many causes or drivers of clustering, and in any given city several of these are likely to play a role. We organize them into four broad categories, each of which has several individual drivers:

1. Macro-structural forces (capitalism, globalization, etc.)
2. The state (laws, policies, actions of governments)
3. Local regimes and institutions (real estate markets, zoning, local elites)
4. Bottom-up processes (individual choice, chain migration, neighborhood self-regulation).

Our next job is to refine our scheme and apply it to a greater number and range of case studies, with more systematic and in-depth analysis. Stay tuned.