For urban historians and urban historical geographers, the relevance and meaning of the city as a driver of human history is central to what we do, both theoretically and empirically.Footnote 1 For some, the question of how to define what a city does is a pressing one. For many of us, though, the question is rarely raised; it resides in that murky place behind our writing and thinking, and has little direct or conscious play over how we go about doing our daily work. Historical geographers, with their greater emphasis on theory and spatial relations, are more likely than historians, trained as they are to think through narrative, empirical evidence and temporality, to explore the city's role in explaining social change. Despite this difference, the fact remains that only a handful of urban historical scholars of whatever stripe are actively interested in thinking through the scope and significance of urban agency. The fact that few openly grapple with the question of urban agency, of course, does not mean that we do not work with some understanding of the city's ontological status. All of us do, for better or worse.
In this short essay, I would like to make a small contribution to our understanding of urban agency by focusing on the relational elements of space and how these are shaped by human and non-human action.Footnote 2 Building on the extensive work by geographers and others working on contemporary questions, my intention here is to suggest that a relational understanding of space can help us better understand the role of the city in determining social change. To this end, I start with a few prefatory remarks about how relational thinking can be a useful way of thinking through the character of urban agency. I then turn to think through these issues in the context of American and Canadian suburbs.
Prefatory remarks
Let me start with a simple comment. If by urban agency we mean that people are affected by the city itself or by the city's various components, then I think we can only answer one way, and that is yes. The city has discursive power. The draw of New York or Mumbai or Shanghai to millions if not billions of people from across the world is a clear sign that the city itself is a potent object, with the power to shape human lives. The city has material power. The impact of walls, from those encircling gated communities in Los Angeles and Beijing to those that have separated different populations in Berlin and Jerusalem, is well documented and is a testament to both the symbolic and material power of physical objects. In this view, the city produces a reaction. It causes people to undertake certain actions and to behave in certain ways. It has agency.
If by urban agency we mean that cities are active agents shaping human life, then the answer has to be no. If by cities we mean an assemblage of heterogeneous material and discursive spaces, then cities have no intentionality. ‘Space’ as a built environment cannot choose to do anything. Cities are socially constructed entities, beholden to the actions of individuals and social groups. But this does not mean that they have no agency; it does not mean that they are a mere backdrop to human action. They do cause change, but not under their own making. If this is the case, several questions have to be explored. To what extent does the ‘city’ as an extensive and complicated built environment create change? How can we differentiate what different urban agents do within the city defined as an assemblage of differentiated spaces? If the city is an agent, how do its effects differ from those of people? In other words, how do we go about understanding and differentiating the intentionality of human and non-human objects?
Most urban historians today would accept the post-structuralist insistence that space is not a simple backdrop for human action and that objects are not inert reflections of human agency. (This is probably not true for many other types of historians though.) Having said that, how many of us consciously deploy space as an agent in our work or think of the city as nothing more than a passive setting, a blank sheet to be written on by human actions? Not many, I think, but not all. In recent years, the idea that the city is a static entity has been questioned by scholars who argue that urban space is a relational site. Two elements are central to this thinking: first, boundaries are blurred, such that there is no clear separation between objects; and second, space is not absolute and does not exist separately from objects. What this suggests is that cities are flexible spaces that are ineluctably tied to non-proximate spaces. Cites are not passive vehicles that simply reflect human action, but are active and heterogeneous entities that are part of human action. Three recent approaches that have focused on the agency and attendant effects of human and non-human actions may be useful for investigating the relational links between material spaces, human agency and cities. They are the humanist spatial turn, Actor-Network Theory and political ecology.
The humanist spatial turn, with its origins in the work of Henri Lefebvre and Michel Foucault, emphasizes the complicated relationships between space and society.Footnote 3 It can be illustrated through a recent paper by Margaret Garb. In an analysis of a strike of restaurant waiters in 1890s Chicago, she argues that space is ‘a tool deployed in social struggles’ and that urban societies ‘produce and are produced by space’.Footnote 4 In this view, the built environment and its attendant spaces play a key part in thinking through the effects of the city on humans. The built environment is not a passive backdrop to human action. The built environment is the text through which we understand how space/cities have effects. Actions and the effects they have need to be teased out from the symbolic and material world of the city. For Garb and others, space and action are co-constitutive. We are not angels dancing on the head of a pin.
By emphasizing the effects of non-human entities, Actor-Network Theory provides a somewhat different picture from that of the humanists. This is best understood through the work of Bruno Latour, whose scholarly contribution has been, among other things, to show how material objects transform human life.Footnote 5 For Latour, the world is assembled through local associations around material objects, and the effects produced out of these associations. In this way, the spaces that events occur in are critical to the form, character and intensity of those events. Historians and sociologists, such as Timothy Mitchell and Michel Callon, have taken up this work to show how the bite of a mosquito and the building of economic texts shape human life.Footnote 6 In all of these works, the central point is that non-human agents trigger change. The world outside human action is not a background for that action. In this way, the city can be considered a non-human actor, transforming the world through the associations it forces on humans.
Urban political ecology also argues that non-human elements have a defining role in shaping the human world. In this case, the focus is not on spatial relations as in the humanist spatial turn or the germs and microbes so beloved by Latour, but on the political economy of nature–society relations. The aim is to move beyond binaries (nature/society, material/symbolic) and to show how space and action are intertwined. In this view, natural relations are not separate from human beings, but are produced inside social activity. From this perspective, and this is similar to Latour's argument, urbanization is a metabolic process ‘by which new and more complex relationships of society and nature are created’.Footnote 7 Space is not inert or static. Rather, the socio-natural metabolism of the city is the form through which the nature of both humans and non-humans is transformed.
These three theories are quite different from each other. Nonetheless, they raise important issues that can help us think through the question of urban agency. First, they all emphasize the ability of non-humans – from the mosquito to a wall – to shape human life. Cities, made up as they are of non-human forms, are not simple containers for human action. They are dynamic entities that both produce and are produced by their interaction with people. Second, they point to the difficulty of assessing both the content and the quality of the associations and processes that are important to understanding city life. It is one thing to theorize about the ways in which socio-natural metabolism transforms the city and the life of urban inhabitants; it is another to identify the specific mechanisms that underpin this relationship or to tease out the appropriate evidence from the archives and the built environment. Finally, all three perspectives have something to say about intentionality. Latour's Actor-Network Theory does not differentiate between human and non-humans in terms of agency – it is a level playing field. The humanists and the political ecologists, on the other hand, differentiate between the gradations of intentionality, arguing that social actors create unequal worlds through their interaction with space and natural systems. In this way, cities are both reflections and producers of environments that embody and reflect positions of social power.
These points have obvious implications for the question of urban agency. I want to explore these relational issues – the role of non-humans, the mechanisms of change and social differentiation – through a discussion of the literature on suburban history in the United States and Canada.
Suburbs and urban agency
The suburbs have been one of the most important areas of research in American and Canadian urban history over the past 30 years. The flurry of studies that followed the classic statements by Sam Bass Warner, Ken Jackson and Bob Fishman created the first batch of serious scholarly work on the suburbs.Footnote 8 They established the suburbs as the home of the white middle and upper classes, and a place associated with a particular spatial meaning – the middle-class dream. This work forced us to think of the suburbs as spaces created and worked on by property developers, white, middle-class households and governments. They forced us to take seriously the idea that the suburbs were different from the city. Suburbs had particular qualities that could not be replicated in the central city. This was most clearly articulated by Fishman when he suggested that the suburb ‘expresses values so deeply embedded in bourgeois culture that it might be called the bourgeois utopia’ and that it ‘reflects the alienation. . .from the urban-industrial world they themselves were creating’.Footnote 9 The key point here is that it reflects; it is a mirror. It does not contribute or shape; suburban space is a passive backdrop to social relations imposed on the urban fringe. The suburb is a spatial container not an object. It is absolute, not relational.
This work on white middle-class suburbanites established the importance of the suburb as an object of study and paved the way for a new generation to explore questions of class, race, gender and identity outside of the city centre. These new suburban historians created a more nuanced and wide-ranging understanding of the suburbs. The white middle-class suburbanite now had to share the suburbs with workers, minorities and industry.Footnote 10 This has allowed us to think differently about the suburbs. The more recent suburban historians, to paraphrase E.P. Thompson, have rescued the working-class and African-American suburbanite ‘from the enormous condescension of posterity’.
More importantly for the discussion here is that the suburb has been framed as an active space, one that places more emphasis on a relational view of urban space. The suburbs – whether they are the owner-built suburbs of Toronto, the African-American suburbs of Detroit, the blue-collar suburbs of Los Angeles or the industrial districts of Chicago – are no longer considered a passive backdrop to social relations. The suburbs are no longer a blank sheet on which class relations are played out. The suburbs are no longer a place created in reaction to the problems of the central city. The suburb's physical layout and homes have an appeal across class and racial lines. The locational assets of the suburbs attract a particular type of industry – capital-intensive factories linked to regional and national markets. The suburbs acquire a particular status that separates them and makes them different from the central city. Space matters in this perspective: it creates different responses and provides the opportunities for people to act in dissimilar ways.
Despite this, new suburban historians still consider the city and the suburbs to be ontologically different. For all the attention given to the worlds created by workers, minorities and industrial executives on the metropolitan fringe, the emphasis has been on the inherent differences between the spaces on either side of the city-suburban line. But is this appropriate? Are the suburbs different from the city centre? Do suburban spaces have different effects than central city ones? The new suburban historians would answer yes to both questions. In their view, the suburb is both a complex and contested entity, and a place with specific forms of agency; one quite different from the central city. But is this the case? Even though they frame the metropolitan area more relationally, they continue to emphasize human agency and to downplay the agency of non-human objects.
The blurring of the line between different spaces and the effects these spaces have on human action can be illustrated through the case of an area on the edge of metropolitan Milwaukee. This place had all of the elements of the typical working-class suburb. Its low land values were attractive to those seeking to own their home, especially those with young families wishing to escape the central city's high rents, congested conditions, dirty factories and rundown housing. Here were ‘new homes on quiet residential streets, with small front yards and an occasional tree. Neither tenement nor huge triple deckers were anywhere in sight.’ Here was the suburban dream for America's working class. As some readers may have realized, this suburban story is taken from Roger Simon's 1978 study of Ward 14, a neighbourhood within the City of Milwaukee.Footnote 11 Ward 14 had many of the functions typically assigned to the suburbs. It was new. It was socially homogeneous. It was leafy. It was residential. It was not the central city. But yet it was part of the central city; it was not the suburbs.
If Ward 14 is not a suburb, what then is suburban? How does Ward 14 differ from politically formal suburbs? Is a political boundary enough to create two worlds? And what does this mean for the presumed difference between the city and the suburbs and their attendant effects? I would argue that the idea of there being a city and the suburbs is a poorly constructed notion. Indeed, it is a ‘chaotic concept’. It is chaotic in that the objects under consideration do not have an internal set of relationships which are necessarily related. In this case, the notion of a suburb consists of too many contingencies and non-dependent relations, and, as such, it cannot hold the conceptual weight that is imposed upon it. Things that are spatially contiguous are not necessarily connected; and vice versa.
This brings us back to the perspectives outlined earlier. All three point to the importance of internally related relations and associations for understanding how space is produced and how space feeds back to and thereby changes social relations. Whether it is political ecology's ‘unruly nature’, or the spatial turn's politics of space, or actor-network's mosquito, they propose that the importance of urban agency is not so much to do with building boundaries around space as it is with thinking through the associations that connect people and places. This suggests that a relational perspective is a fruitful way to think through urban agency. Space is not ordered or absolute. It is both connected and unbounded; it is both interactive and articulated.
Suburbs can be framed by the social relations of place – i.e., the network of relationships bound up with a particular setting. In this perspective, interaction refers to the relationships that occur within a place – the associations that Garb, Latour and the political ecologists have pointed to. Whether it is the filling in of a stream or the building of a streetcar line, place-based relations give meaning to and create new meanings. The stream is now a sewer and the track a commuter line. In other words, suburbs are not just places in place, they are also unbounded objects – places out of place. They are articulated through a constellation of relations with other places. Suburbs are shaped by networks that span the world – information, capital, people and so on. In this sense, the suburb is the product both of itself and the world outside to which it is connected. In this view, space is not intentional – it cannot actively act upon another object or actively produce reactions. Rather, space is mobilized – it is something that is formed through what happens to it. It exists only through the relations between objects. But this mobilization means that it does have effects.
Conclusion
Let me conclude by drawing out two main arguments. First, if we can accept that ‘the social is necessarily spatially constituted’ then we can argue that the city and the suburb are as well.Footnote 12 All social and physical forms have a spatial referent. Cities and suburbs are given life through their spatial location and the relations this location has with other locations. The city is unavoidably spatial. But it is not necessarily coherent. The complexity of the city and the multitude of competing actors, however, ensure that cities do not have an internal spatial coherence in the same way as multinational corporations or other sophisticated organizations do. Even though they may not be coherent, they do have effects. Cities and suburbs are not a passive milieu for social relations as the historians of the suburbs would have us believe.
Second, in this brief consideration of urban agency, I have tried to make three points. First, metropolitan regions (cities and suburbs) are active in that they are unbounded. They are not absolute or contained; their unbounded form produces the potential for shaping themselves and the worlds they are connected to. Second, cities and suburbs are spaces of association; they are a world in which objects are connected; they are not discrete objects with their own essence. Cities and suburbs are mobilized spaces that exist only through the relations they have with other things. Finally, they are not intentional. We are not ‘in’ space – we act in and thereby produce space. This is different from giving cities agency. In other words, to paraphrase Marx's famous statement in ‘The 18th Brumaire of Napoleon Bonaparte’, ‘cities make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves’.Footnote 13
