Cities as social reactors

Fig 1. Inka storehouses

Some traits of cities seem universal, shared by all cities, large and small, past and present. Neighborhood organization is one of these urban universals, something I have talked about in previous posts (here, and here, for example).  See Smith et al. (2015). Other traits seem more limited to only some kinds of cities, or cities in only a few regions or periods. For example, Inka cities were full of standardized stone storage facilities, built and maintained by the state (Figure 1); such storehouses are rare in other cities. Contemporary cities have organized trash pick-up (whether government or privately run), while past cities lacked this institution. One of my reasons for writing this blog is to explore the nature of these similarities and differences among cities across the span of deep history and around the globe.

A number of lines of research now suggest that that the role of cities as “social reactors” may be another urban universal. This means that the presence of greater numbers of people—in larger settlements—leads to more social interactions, which in turn produces social and economic benefits far beyond what smaller settlements can produce. Cities are places where individuals interact socially in all kinds of ways. As the number of people in a city goes up, the number of potential interactions with other people also rises, but it rises much faster than the number of people. The increase is exponential (Figure 2). These interactions have been called "buzz" (Storper and Venables 2004).

Fig 2. Interactions & pop.

A long line of research in the social sciences focuses on the negative consequences of growing city size. More people means more crowding, and crowding in cities can cause stress, both psychological and social. Larger cities have more crime and more poverty. Social relations may be less personal in larger cities, with neighbors less willing to help one another. One intriguing theory, by archaeologist Roland Fletcher (1995), posits thresholds of population size and density that growing settlements cannot cross unless they develop new means of communication and social integration (Figure 3). Otherwise the stresses of living with too many people make life intolerable and settlements will disintegrate or break down.  When early village dwellers figured out how to communicate information with writing, and how to divide their settlements into increasingly specialized spaces and buildings, they could achieve the transition to the first cities, which had more people and denser populations.

 

Fig. 3. Fletcher's book

Growing city size also has its positive consequences. In early times, the most basic attraction of city life was protection. War and violence have been prominent in all eras of history. Once our ancestors adopted agriculture and settled down, the concept of property was born. Property and riches led to crime and warfare. People quickly figured out that there was strength in numbers. Kings and rulers in past times wanted their subjects to move into town. People were easier to tax and control in cities, and their economic activity created growth and prosperity that benefitted the rulers.

Now let's jump to the present. In contemporary cities, urban growth has what economists call “agglomeration effects” (Figure 4). This means that as firms and factories set up in cities, they attract workers and create greater specialization and increased productivity. These in turn generate growth, attracting even more people to cities and creating more output. The spatial concentration of people and firms in cities leads to economies of scale, which pushes urban productivity even higher. Knowledge and skills can be transferred among people, and from one industry to another, because of their concentration in a city.

 

Figure 4.  Agglomeration effects

This picture of agglomeration economies is a pretty standard view of contemporary cities, as seen by urban economists and economic urban geographers. It is easy to see how cities in this sense can be called “social reactors.” They are places where concentrations of people, firms, and institutions create wealth and growth, built on a foundation of the way the people interact with one another. This is definitely the way cities operate within modern capitalist economies. But what about cities before the modern era, before the birth of capitalism? They didn’t have wage labor; workers did not move among firms and industries as they do today; land markets were rudimentary or non-existent; and economies were far less dynamic than they are today.

Were cities in the past also “social reactors”? My answer to this question a few years ago would have been “no.” It seemed to me (and to many urban scholars) that the generative properties of cities—their capacity to grow quickly and expand economically, their ability to create wealth—arose from the capitalist economy. Ancient cities did not have agglomeration economies, and thus their roles as social reactors must have been far less important than today.

But then I started working on urban scaling with three colleagues. The results have been surprising and illuminating to me, and important for a general understanding of urbanism. I’ve blogged a bit about the scaling work previously (an early post, and later), and I will talk about it more in the future. For now, I will say that this research had convinced me that the concept of social reactors definitely applies to ancient cities, and it even applies to large villages and towns. Something fundamental happens when people live together in settlements, and these fundamental processes become stronger as a settlement gets bigger.

Here is a very brief description of three reasons why I changed my mind on ancient cities and settlements as social reactors.

First, the very regular patterns of how urban traits change with city population, as found for contemporary cities, also apply to past cities. These relationships are analyzed with scaling models, and our research group has found that the scaling patterns of modern cities are also present in cities before the modern era.

 

Fig. 5. Mandan village

Second, these scaling relationships are also found in village settlements among tribal peoples (Figure 5). To me, this is an astounding finding. Generative processes that are thought to be “urban” in nature also affect much smaller settlements in societies that lack cities, kings, laws, and social classes. Wow, this is nothing short of incredible.

Third, the explanation for the above reasons comes from Luis Bettencourt’s (2013) formal model of scaling. Rather than tracing the city dynamics that create scaling to agglomeration economies, this model traces them to face-to-face social interaction within a structured built environment. This means that they should be present in historical cities, in ancient cities, and in non-urban village settlements. And they are!

After a few years of working groups and meetings our project now has a name: “The Social Reactors Project: Human Settlement and Networks in History” and a website. We are physicist Luis Bettencourt, economist Jose Lobo, archaeologist Scott Ortman, and me. Our website has just gone live this week! Check it out for our work and papers, and for the names of postdocs and students who are helping.

The idea that face-to-face social interactions (buzz) are crucial to cities and to urban life is nothing new. Here are three quotes from diverse authors:

“The central theme of this book is that cities magnify humanity’s strengths. Our social species’ greatest talent is the ability to learn from each other, and we learn more deeply and thoroughly when we’re face-to-face.”

--Urban economist Edward Glaeser (2011:250).

“Again and again, we find that one key to creating social capital is to build in redundancy of contact. … Common spaces for commonplace encounters are prerequisites for common conversations and common debate.”

-- Political scientist Robert Putnam and Civic leader Lewis Feldstein (Putnam and Feldstein 2003:291).

“Stable physical settings” promote community. “The neighborhood accommodates and facilitates social interaction.”

--Urban planner Sidney Brower (Brower 2011:19)

I believe this notion was also crucial to past communities, a theme I develop in my new book, At Home with the Aztecs: An Archaeologist Uncovers their Daily Life.

The innovation of our scaling research group is to demonstrate, with solid quantitative data, that these face-to-face interactions were not only present and important in the past, but that they also generated change and output in the past, just as they do today. We have a bunch of papers in production and in press, and some published (see the website). The role of cities as social reactors is not just a contemporary phenomenon, an attribute of capitalist economies. Rather, it is a fundamental attribute of human settlements through deep history and across space.

References

Bettencourt, Luís M. A.

2013  The Origins of Scaling in Cities. Science 340: 1438-1441.

Brower, Sidney N.

2011  Neighbors and Neighborhoods: Elements of Successful Community Design. APA Planners Press, Chicago.

Fletcher, Roland

1995  The Limits of Settlement Growth: A Theoretical Outline. Cambridge University Press, New York.

Glaeser, Edward L.

2011  The Triumph of Cities: How our Greatest Invention Makes us Richer, Smarter, Greener, Healthier, and Happier. Penguin, New York.

Putnam, Robert D. and Lewis M. Feldstein

2003  Better Together: Restoring the American Community. Simon and Schuster, New York.

Smith, Michael E., Ashley Engquist, Cinthia Carvajal, Katrina Johnston, Amanda Young, Monica Algara, Yui Kuznetsov, and Bridgette Gilliland

2015  Neighborhood Formation in Semi-UrbanSettlements. Journal of Urbanism 8(2): 173-198.

Storper, Michael and Anthony J. Venables
2004    Buzz: Face-to-Face Contact and the Urban Economy. Journal of Economic Geography 4 (4): 351-370.

Why do all cities have neighborhoods?

I've been writing about urban neighborhoods for several years now. I have made the claim that all cities have neighborhoods. In fact, neighborhood organization is one of the very few urban universals. There are very few features shared by ALL cities, throughout history and around the world. Besides neighborhoods, other candidates for urban universals include the provision of urban services, and the fact that if a society has an elite class, then many or most of its members live in cities.  See: Do all cities have neighborhoods? (2011), or

I find that I always hesitate a bit when writing that "all" cities have neighborhoods. That is a tough claim to prove. We simply don't have information about all the cities that have ever existed, so a claim for the universality of something like neighborhoods must rest on indirect evidence. Here are the three lines of evidence that make sense to me.

First, every description of a city that is sufficiently detailed and focused to mention the existence of neighborhoods, does in fact mention neighborhoods. This is far from an air-tight argument. But I've been looking at city descriptions like this for a number of years now, and so far this claim has held up. These include ethnographic reports of cities around the world, historical accounts of cities before the modern era, and archaeological reports of ancient cities. Archaeologists started thinking seriously about neighborhoods about eight years ago, and guess what? Since then, many reports of neighborhood organization have popped up. Check some of the works in the bibliography below.

Second, many bin-depth studies of neighborhoods, in the past and the present, have found that neighborhoods are crucial social and spatial units within their city. They are important in many ways for urban residents, and they are important for the overall operation and functioning of the city. Some of my favorite such studies are Robert Sampson's analysis of Chicago neighborhoods today, Abraham Marcus's study of Aleppo in the 18th century, and Eva Lemonnier's identification of neighborhoods at the ancient Maya city of La Joyanca. See: Why are neighborhoods important? (2014).   Or, in Publishing Archaeology, see Archaeological concepts of community confront urban realities today (2015).

Third, I carried out a study, together with a bunch of undergraduates, of neighborhood organization at semi-urban settlements (Smith et al, 2015). The study was based on the assumption that if neighborhoods formed at these rapidly-formed, often chaotic, and sometimes specialized settlements, then they would form at any good-size human settlement. We found neighborhoods did indeed exist at Plains Indian aggregation sites, arts festivale, RV camps, protest camps, shantytowns, military camps and forts, internment camps, company towns (including ancient Egyptian workers villages), and refugee camps. The only kind of settlements where we could not confirm or discomfirm the presence of neighborhoods was disaster camps.  See : Neighborhoods in semi-urban settlements (2011).

So, if neighborhoods really are urban universals, why is that the case? In our 2015 article, we give two types of answers: ultimate causes, and proximate causes. These concepts, borrowed from evolutionary biology, refer to the deep underlying causes of social phenomena (the "ultimate" causes) and to the basic day-to-day reasons for their formation ("proximate" causes). The underlying, ultimate cause of neighborhood formation is that people in cities need, or want, to live their lives on a smaller scale than the entire city. Some studies suggest that this is caused by constraints on human memory; one can only recall so many people, and effective social networks cannot be too large. Other studies suggest that living in cities causes social stress, and neighborhood organization is a way of relieving that stress.

It is interesting to note that neighborhoods can form in two very different ways. The most common path throughout history was the bottom-up approach. People living in an area interact with those around them (their neighbors), and eventually clusters or people, or communities, develop on their own out of the day-to-day actions of people. But in some cases, city or government authorities create neighborhoods. They organize cities from the top down, and people move into ready-made neighborhoods.

In our paper we identify the following proximate causes of neighborhoods: For bottom-up neighborhoods, simple sociality--interacting with your neighbors-- is the primary cause of neighborhood formation. Group preservation and defense also contribute to neighborhood formation in some cases. For top-down neighborhoods, established by authorities, the most common proximate causes are administration (the need to administer the residents) and control/surveillance. Sociality is a secondary consideration; if not present from the start, it quickly develops once people start living in their pre-made neighborhoods.

REFERENCES

 Arnauld, Marie Charlotte, Linda R. Manzanilla, and Michael E. Smith (editors)
2012    The Neighborhood as a Social and Spatial Unit in Mesoamerican Cities. University of Arizona Press, Tucson.



Hakim, Besim S.
2007    Generative Processes for Revitalizing Historic Towns or Heritage Districts. Urban Design International 12: 87-99.

Lemonnier, Eva
2011    Des quartiers chez les Mayas à l'époque classique? Journal de la Sociéte des Américanistes 97 (1): 7-50.

2012    Neighborhoods in Classic Lowland Maya Societies: Identification and Definition from the La Joyanca Case Study (Northwestern Peten, Guatemala). In The Neighborhood as a Social and Spatial Unit in Mesoamerican Cities, edited by Marie Charlotte Arnauld, Linda R. Manzanilla, and Michael E. Smith, pp. 181-201. University of Arizona Press, Tucson.

Marcus, Abraham
1989    The Middle East on the Eve of Modernity: Aleppo in the Eighteenth Century. Columbia University Press, New York.

Sampson, Robert J.
2012    Great American City: Chicago and the Enduring Neighborhood Effect. University of Chicago Press, Chicago.

Smith, Michael E.
2010    The Archaeological Study of Neighborhoods and Districts in Ancient Cities. Journal of Anthropological Archaeology 29 (2): 137-154.

2011    Classic Maya Settlement Clusters as Urban Neighborhoods: A Comparative Perspective on Low-Density Urbanism. Journal de la Société des Américanistes 97 (1): 51-73.

Smith, Michael E., Ashley Engquist, Cinthia Carvajal, Katrina Johnston, Amanda Young, Monica Algara, Yui Kuznetsov, and Bridgette Gilliland
2015    Neighborhood Formation in Semi-Urban Settlements. Journal of Urbanism 8 (2): 173-198.

What are universal urban features?

1.Walled neighborhoods in Chang'an

What characteristics are shared by all cities, from the earliest to today, and around the world? Many of the features shared by all cities are not exclusive to cities or urban settlements. Things like housing, big buildings, wide streets, or social diversity are often found in villages and other non-urban settlements. Three features of cities seem to be true universals. By this I mean features that (1) are found in all known cities; (2) are often absent in non-urban settlements; and (3) have a major impact on life in cities. These three features are neighborhoods, urban services, and elites. There may be others that I haven't considered; let me know if you have ideas for non-trivial urban universals.

2. Neighborhoods (clusters) at La Joyanca

(1) Neighborhoods

For years I've been telling my classes that neighborhoods are one of the few urban universals. Figure 1 here shows the walled neighborhoods at the Chinese Tang city of Chang'an. Recent research of our urban group here at Arizona State University, has been targeting the neighborhood at cities through time. Archaeologists have woken up to the importance of urban neighborhoods, and this has become an active area of fieldwork and analysis; see the new book, The Neighborhood as a Social and Spatial unit in Mesoamerican Cities. The clusters of houses at Classic Maya sites were neighborhoods; figure 2 here shows one example, the city of La Joyanca (from the chapter by Eva Lemmonier) Even semi-urban settlements have neighborhoods - see my post on this.

3. Bhaktapur neighborhood plaza

4. Model of a Bkaktapur shrine

Neighborhoods are often focused on key features such as a plaza, a water-source, or a temple. In the Nepalese city of Bhaktapur, for example, neighborhoods are formed around open plazas, often with water sources (fig.3). In addition, each neighborhood has one or more shrines. Our urban project has a small exhibit in the Museum of Anthropology at ASU, and the museum folks created a nice model of one of the Bhaktapur neighborhood shrines (fig. 4). Central features like this give neighborhoods a central focus for people to gather and interact on a daily basis.

(2) Urban Services

The next research project of our urban group is a study of urban services in premodern cities. In background reading for this project it occurred to me that urban services are another urban universal. When people live together in cities, they cannot take care of all of their basic social needs in the same way that rural people in villages can, and they also take on new needs that must be met in the city. Someone has to provide basic services, which include infrastructure (water, roads), education, commercial outlets, ritual, and places to gather. In modern cities, governments provide most of the urban services, but in medieval cities many services came from guilds, church groups, and private citizens. How does this work in premodern cities? And how are services affected by elites and inequalities? These are the basic questions we will be studying over the next few years. I will talk about our project in a future post (click here for some preliminary information). For now, I will just mention some basic services and how they intersect with neighborhoods.

5. Neighborhood temple in Calixtlahuaca

We are studying urban services through their facilities -- the places where they are provided. For the comparative study we have singled out three services that occur in most cities: markets/shops; temples; and assembly spaces. The small plazas and shrines of Bhaktapur are examples of neighborhood-level services in that city. Neighborhood-level service facilities can be widely distributed in cities, and typically there are many of them. But cities also have higher-level service facilities - that is, facilities that are larger and serve more people, and there are fewer of these features. Thus in Mesoamerican cities, there are often many small temples, distributed around the city (fig. 5), while there are only a few large central temples (fig. 6).

6. High-order temple at Palenque

7. Central plaza at Copan

Also, many cities have numerous small neighborhood plazas (fig. 3), but only one or two large, central plazas (fig. 7). For more information on our upcoming study of premodern urban services, click here.

(3) Elites

8. Medieval noble and beggar

My third candidate for an urban universal is elites. My claim is that in any society that has both cities and elites (that is, most complex, state-level societies), some or all of the elites will live in the city. There may also be rural-based elites, or elites who maintain multiple residences, but some elites will live in the city, and they will exert an influence over the lives of the non-elites. This is important, because in ancient societies, typically 5% or less of the population were in the elite class, and the small number of elite families had a disproportionate influence on urban life in cities. This claim also applies to modern cities, although the system of inequality and elites is radically different in contemporary western societies compared to premodern societies.

9. Elite and commoner house at Cuexcomate

Exactly how did urban elites influence city life in the past? There is probably variation among cities and areas, and this is one question we will investigate in our project. One very preliminary finding, from a small sample of cities, suggests that elite residences had better access to service facilities than commoner houses (surprise, surprise). But commoners living in the same neighborhoods as those elite residences had no advantages (in the distance they had to walk to get services).

While elites played important roles in ancient cities and societies, there has been surprisingly little comparative research on elites around the globe. Archaeologists usually identify elites by the size of their houses (fig. 9), and as the excavation and analysis of houses moves forward around the globe, we will learn more about ancient elites and their roles in cities.

Are there other urban universals beyond neighborhoods, urban services, and elites? Let me know if you have any suggestions. There is still a lot to learn about cities throughout history in the wide, urban world.