Abstract
Territorial exclusion is a multi-scalar phenomenon. However, research has tended to focus on exclusion at separate scales. This paper develops a conversation between research on the territorial exclusion of international migrants at the national scale and the territorial exclusion of lower- and working-class residents at the urban scale. Both strands of research have encountered a common empirical puzzle: territorially exclusive practices rarely comport with official government policies. The paper argues that these apparent “policy gaps”—and efforts to overcome them—can be more fruitfully studied as outcomes of the scalar structuration of legitimate violence, which shapes the way that policy-makers seek to achieve exclusionary goals. The paper suggests that this approach may be used as the platform for richer inter-disciplinary conversations between Human Geography and International Relations (IR) about territorial exclusion and the historical scaling and rescaling of legitimate violence over time.
Territorial exclusion occurs at many different spatial scales, including the body, landed property, neighborhood, city, nation, and region. This paper develops a conversation between research on territorial exclusion at two specific scales. At the national scale, research focuses on the manner in which, under the aegis of immigration policy, passports, borders, detention centers, and deportations are used to identify, exclude, and remove those people attempting to enter and reside within state jurisdictions without authorization. At the urban scale, territorial exclusion is currently a focus of the literature on gentrification which explores how dynamics within the property market, policing, and culture have progressively pushed lower- and working-class populations out of urban centers. This paper brings together insights from Political Geography and International Relations (IR) to develop the case for a more holistic and scalar approach to the study of territorial exclusion.
This piece aims to open up new ways of interrogating these processes of territorial exclusion. The starting point for this endeavor is the observation that research on the political and institutional dimensions of these two types of territoriality has encountered a common empirical puzzle. Both forms of territorial exclusion are “global”—in terms of being observed across the world—but deeply contested phenomena. Immigration control is widely regarded as a fundamental attribute of statehood and a basic function of the modern state (Hollifield, 2004). Some have argued that gentrification has become a similarly ubiquitous phenomenon, albeit operating at a different scale (Lees, 2000; Smith, 2002). Yet, both types of territorial exclusion rarely comport with government policies. At the urban scale, governments that preach social balance nonetheless deploy policies that allow for the exclusion of working- and lower-class populations. At the national scale, governments that preach immigration control nonetheless accept informal migrants. In this paper, I argue that we can better understand these outcomes and their implications if we study territorial exclusion in a more multi-scalar, comparative and relational way.
More specifically, I first propose that both of these “policy gaps” ought to be studied as generic features of the social organization of violence. Second, I suggest that the means that political actors use to overcome these policy gaps differ between scales. Organizing the violent exclusion of people is a difficult social and political process which necessarily confronts powerful coordination dilemmas. Yet, over time, and across the globe, violent exclusion has become more legitimate at the scale of the nation and property, but less so at the urban. This process of scalar structuring shapes the way violent exclusion can be organized at these different scales. At the national scale, policy-makers create formal bureaucratic structures to compel state agents to locate, arrest, detain, and deport unauthorized residents from territorially defined jurisdictions. At the urban scale, only more indirect methods are used, usually involving the coordination of practices of violent exclusion at a lower scale: that of property. Crucially, it is only by comparing and contrasting exclusion at different scales, and considering the relations between them, that the unique properties of exclusion at the individual scales come to light.
By developing these points, this article aims to outline three starting points for an inter-disciplinary, and multi-scalar study of territorial exclusion. First, the article suggests that there are opportunities for
The paper develops the argument in three steps. The first section uses geographers’ concepts of territory and scale to engage critically with Weberian theories of the state and international relations. This results in a scalar portrayal of legitimate exclusionary violence. The second section explores how this portrayal may inform our inquiries into immigration and gentrification policy gaps. Finally, I consider how we might address these questions through the synthetic deployment of approaches developed in the fields of International Relations and Political Geography to the comparative study of territorial exclusion at separate and related scales.
Territorial exclusion, legitimate violence, and coordination problems
Organizing territorial violence is a challenge
This paper rests on the assumption that violent means play a unique role in processes of territorial exclusion (Jones, 2016). Territorial violence consists of any effort to physically prevent people from moving and/or expelling people from places where they are deemed unwanted or out of place. 1 In practice, this includes acts such as blocking entry, arrest, detention, and forced removal. This approach rests on a broadly Anglophone understanding of territoriality as a spatial strategy (Sack, 1986) rather than a social relation (Painter, 2010; Raffestin and Butler, 2012). As a spatial strategy, territorial violence both directly and indirectly impacts patterns of human movement and settlement. It not only has the effect of physically moving human bodies from here to there, but may also have deterrent effects. Since people commonly act in ways to avoid becoming the victim of a violent act, the mere threat of a violent response may change how people make decisions about where to move and where to reside (Carling, 2002).
While exclusionary violence is potentially very powerful, it is not necessarily easy to actively deploy, particularly when exercising control requires the effort of more than one person. As organizational theorists have long asserted, compelling people to work in concert is a challenge (Lipsky, 2010; Pressman and Wildavsky, 1984). To say this another way: “principals” experience difficulty compelling “agents” to do their bidding. Various actors may possess a range of ideas about the groups they would like to exclude and the places they would like to protect, or simply have interests in ensuring that processes of territorial exclusion continue. However, in order for such territorial ideas and ambitions to be realized, the people who are in direct command of violent means, and are directly responsible for deploying violence (whether they are police, army, security guards, militia, or ordinary civilians), must consistently work in concert toward a common end. Such an outcome cannot be assumed, but must be actively engineered.
These assessments have significant implications for how we understand the relationship between violence and territorial exclusion. If the principal–agent dilemma is assumed to be endemic to all forms of violent exclusion, then we have no
The question then is—how are these obstacles overcome and in what ways? The answer to this question varies across time and depending on the scale we are talking about. Actors seeking to territorially exclude possess different sets of tools for overcoming the principal–agent dilemma, depending at which scale they are operating.
Scale as a social construct
In order to understand this point it is important—without rehashing old debates—to clarify the manner in which I am using the concept of “scale.” The effort to describe and understand the similarities and differences between territoriality at different scales has been an enduring, albeit not always explicit feature of the study of human territoriality. Early behavioral studies understood large-scale social and political territoriality as driven by the same underlying animating causes as micro-territorial and individualistic forms (Ardrey, 1969; Malmberg, 1980). Later socio-historical forms of analysis—exemplified by the work of Robert Sack (1986)—emphasized the very different underlying origins and causes of territorial strategies at different scales, and instead sought to generate a conceptual vocabulary through which these contrasts and comparisons could be described and explained.
More recent work has taken a more critical and relational approach to scale, as less defined by abstract measures of extent, and as a more historically produced ordering of sets of spatial practices. Working from this perspective, scholars have sought to understand territorial practices at given scales in terms of their historically produced differentiation from similar sets of practices operating at other scales. Neil Brenner (2001) is more precise, describing the aim of this literature as exploring the “production, reconfiguration or contestation of particular differentiations, orderings and hierarchies among geographical scales…their embeddedness and positionalities in relation to a multitude of smaller or larger spatial units within a multi-tiered, hierarchically configured geographical scaffolding of territoriality” (600). I adopt this structurationist approach to the study of territorial scale, but depart from the pre-dominant structurationist preference for neo-Marxist lines of explanation and inquiry, where scaling and rescaling are consistently described as derivative of the changing nature and demands of the organization of production or changing forms of capitalism (cf. Ferdoush, 2018). Instead, and drawing upon more relational views of scale (MacKinnon, 2011; Marston, 2000), I emphasize the independent role of processes of legitimation in the way territorial violence has become relatively “fixed” at different scales, and the manner in which these scalar fixes both constrain and enable territorially exclusive behavior.
In deploying the concept of scale in this way, I recognize that strong cases have been made for either the return to a flat ontology in which scale is more or less eliminated from our conceptual vocabulary (Marston et al., 2005) or for a complete untethering of analysis from topographical perspectives so that the topological relations defining contemporary spaces and power relations can be more fully revealed (Allen, 2011; Allen and Cochrane, 2010). While recognizing the need for work that continues to critique the political construction of scale, in this paper I make the case for the continuing analytical value of scalar concepts, by demonstrating how the social and political projects aimed at achieving immigration control and gentrification have encountered scalar fixes as social facts. I specifically demonstrate the power of scalar fixing in the stark contrast between policy-makers’ (a) ability to hierarchically order the violent exclusion of unauthorized migrants at the national scale and (b) inability to order the violent exclusion of unwanted urbanites at the urban scale. At the same time, I note the importance of relational understandings of scale by demonstrating how the legitimacy of violence at lower scales may impact efforts to territorially exclude at higher scales. In developing these points, I defend the value of topographical concepts of scale for efforts to understand territorially exclusive projects and their outcomes.
Using scale to reconceptualize the levels of legitimate violence
The idea that there is a scalar structure to the organization of violence is not new. My version of this argument is based on a critical engagement with a quasi-scalar differentiation that has been core to the development of much IR theory. This has involved a differentiation between the nature of governance at separate “levels of analysis.” The primary basis for this claim is a Weberian distinction between the state and anarchy. Weber (2004) defined the “state” as a “human community that (successfully) lays claim to the monopoly of the legitimate physical violence within a particular territory” (33). For Weber, violent monopoly constituted the pre-condition for the development of a particular means of governing: hierarchy. Bureaucracy and the rule of law rely heavily on the expectation that violence will be consistently deployed to ensure civilian compliance with policy-maker’s decisions and rules. Weber (2004) clarified his depiction of the hierarchical, territorial state by differentiating it from a condition of anarchy: if there existed only societies in which violence was unknown as a means,
Before considering how the scalar dimensions of this argument might be clarified, it is worth considering the types of theorizing that the domestic/international distinction engendered. Those theorists who sought to engage with Waltz did not assume that a condition of anarchy ruled out efforts to govern at the international level would always fail, but that they would require
Geographers have rightly critiqued IR theorists’ tendency to suggest that political action tends to be structured into a singular, fixed, and timeless set of quasi-spatial “levels.” 2 We can use the concept of “scale” to rework Weber/Waltz: examining how legitimate violence is scaled and rescaled by ongoing social process. The foundation for this approach is my understanding of the nature of legitimate violence: not as a possession or attribute which determines who can act, but a construct that shapes agency. Theorists of violence and the state have tended to lean toward the first of these two formulations. At one extreme is Waltz, who argues that the assumption that the state possesses a monopoly on legitimate violence can be used to make categorical distinctions between the defining “rules” of domestic and international spheres of politics.
Charles Tilly (1985) offers a more complex portrait, concurring with the idea that the state’s territorial monopoly on violence divides politics into two discrete levels of action, but explaining this as an outcome of a contingent and contested scalar process. Yet, Tilly still provides an unduly narrow role to processes of legitimation in the scaling of violent behavior. More specifically, legitimation processes do not change the scales at which violence is legitimate because they have been fixed by long past and forgotten changes in military economies of scale.
Yet, “legitimacy” is not something that can be collected and monopolized like so much loose change, but the product of a complex amalgam of material and communicative processes. The same goes for legitimate violence. Of course, it would be foolish to neglect the banally material dimension to the way ideas about legitimate violence have been reproduced, whereby many institutions of rule have established the exclusivity of their claim to legitimate violence through the “monopolization” of the material
Crucially, if legitimate violence cannot be possessed, it does not make sense to think of it as prone to monopolization. Instead, and reformulating Weber, the relative legitimacy of a given form of violence determines the ways in which it can be organized. The key issue here is that legitimacy allows for formal hierarchies—and more specifically law-making authorities and bureaucratic agencies—to help ensure that legitimate violence is routinely and effectively deployed. By way of contrast, and following directly on from the assessments of IR theorists, illegitimate violence can only be promoted by more indirect means, perhaps by seeking to shape the normative frameworks or incentive structures that guide how violent actors use their discretion in their daily work. An absence of legitimacy, therefore, does not entail the absence of territorial violence, but rather a significant handicap for efforts to organize territorial violence at a given scale.
This broadly social constructivist understanding of the nature and power of legitimate violence frames my approach to the scaling of territorial exclusion. More specifically, I begin by asking how violent exclusion has been legitimized and delegitimized at different scales. A full account of the process whereby our current scalar structure of violent exclusion has emerged is beyond the scope of this paper. Instead, I will merely attempt to identify some apparent differences between developments at two specific scales, suggesting that over time, exclusionary violence has become more legitimate at the scale of the national jurisdiction while less legitimate at the urban.
The scaling of territorial violence
The critical foil for my ideas here is John Torpey’s (2000) classic work on the invention of the passport. This work began with the proposition that, prior to the 20th century, many types of actors—including slave owners, feudal lords, cities, states, and empires—operating at and across various scales, legitimately territorially circumscribed the movements of different categories of people across space. However, over the course of the latter 19th and early 20th century, national states gradually and collectively abolished many competing forms of movement and settlement control—territorial or otherwise, established a set of instruments for regulating the international movement of people, and claimed the prerogative to violently exclude outsiders from their territory. The relative legitimacy of this claim is most clearly evident in contemporary immigration law, which defines removal as a lawful response to the unauthorized presence of an individual on the territory of a state. For example, s. 10 of the United Kingdom’s Immigration and Nationality Act (Revised 2014) reads: “a person may be removed from the United Kingdom under the authority of the Secretary of State or an immigration officer if the person requires leave to enter or remain in the United Kingdom but does not have it.” A similar claim may be found in s. 1226 of the US code “On a warrant issued by the Attorney General, an alien may be arrested and detained pending a decision on whether the alien is to be removed.”
While the statutory coding of this principle pays testimony to the legitimacy of this form of territorial violence, it also should not blind us to the historicity and contingency of the idea. The British case is an excellent example of this. The legitimacy of efforts to locate, detain, and deport foreign nationals in the United Kingdom has been consistently contested by moves to promote free movement at different scales, including the sanctuary city movement at the urban scale, the European Union’s promotion of the principle of free movement at the regional scale, and the refugee convention at the global scale. As the Brexit movement—defined by its clarion call to “take back control of our borders”—suggests, the legitimacy of territorially exclusive practices is not something that is simply won and thereafter possessed in perpetuity by a given actor, but gradually and historically achieved and challenged by social groups and forces.
While my depiction of the history of immigration politics is largely in line with Torpey’s account, I part ways with his presentation of “internal” movement control. Torpey presents the growth of a state monopoly over internal movement as parallel to its control over international mobility. In contrast, I would argue that the increasing legitimacy of violent exclusion at the national scale was paralleled by a movement in the opposite direction at the urban scale. Spatial partitioning amongst social groups is an enduring feature of the urban form. In certain contexts, state laws have sanctioned these spatial partitions. To take a relatively extreme example, the Group Areas Act (n. 41 of 1950) divided the South African population into racial groups, afforded the Governor General power to declare any area suitable for the occupation by one of these groups, made unlawful occupation or ownership of land within these areas a punishable offence, and empowered an inspectorate to—in conjunction with the police—enforce these laws. Backed by these powers, South African governments created a powerful, hierarchically organized set of bureaucratic agencies designed to build and sustain the most racially segregated cities on earth (Christopher, 1990, 1994).
While similar—if less draconian—laws were evident in many cities prior to and during the early parts of the 20th century, over the course of the last century, efforts to create and enforce partitions within the city or at the city’s edge have been progressively delegitimized. This has not meant that territorial exclusion at the urban scale has vanished but that the means for organizing it have been circumscribed. Peter Marcuse (2002) describes these changes in the following way: “the state has been decisive in creating, maintaining or destroying partitions in all periods, but the state has been itself the decisive power in earlier periods, while today its actions largely reflect and reinforce the power of groups external to it and operating through the market” (31). We may attribute this change to a range of disconnected forces and processes, including the decolonization and desegregation of African and Asian countries, the civil rights movement and the desegregation of the US South, and the collapse of authoritarian regimes in Eastern Europe. The collective outcome of these processes is that—while many states possess a variety of “internal movement controls”—we rarely find laws explicitly condoning the use of violence to prevent people moving to and within the city. Examples like China’s
To summarize the argument to this point, violent exclusion constitutes a powerful means of controlling human mobility and settlement. However, exclusionary violence cannot be readily deployed in ways that match any actor’s preferences and plans. Based on this preliminary assessment, I expect to observe different sets of tools determining whether policy gaps narrow or widen at different scales. At both scales, more horizontal processes of coordinating violence may be observed. At the national scale, policy-makers have the option of hierarchically organized violence, while at the urban scale they do not. This has significant implications for the way we understand the relationship between, on the one hand, immigration and gentrification policy, and on the other, territorially exclusive practices. The following section will develop this argument by engaging with the contemporary literatures on territorial exclusion at the urban and national scale, and identifying how researchers have problematized these phenomena, and how they might be better reinterpreted through a scalar lens.
In focusing on the literatures on immigration control and gentrification, it is important to recognize that neither literature constitutes an effective summary of territorial exclusion at a given scale. At the national scale, we might want to talk about literatures on exclusion of unwanted travelers, political opponents, and other categories of people identified as risks or threats. At the urban scale, we might instead seek to discuss literatures on racial segregation, slum clearing, or household registration systems in authoritarian states. I have selected these two literatures less for their ability to represent what is happening at a given scale and more for the types of claims they develop about the forms of territorial exclusion they observe, which reveal the value of inter-disciplinary learning.
How should we study the politics of exclusion?
Do states “accept” unwanted migration?
The preceding thoughts help us to sharpen our inquiries into processes of territorial exclusion at different scales. Let us begin at the national scale, where scholars studying immigration have commonly seen the state’s failure to achieve control as an empirical “puzzle.” One of the principal concerns of the literature has been the “control gap” or why states accept “unwanted migration.” As Christian Joppke (1998) originally defined the problem: “[u]nwanted migration is not actively solicited by states… [r]ather it is accepted passively by states” (266). Despite the fact that immigration laws state that foreigners who do not possess the right to remain in a country may be removed, in most cases large numbers of undocumented migrants remain. This observation has been puzzling to those researchers who have focused on the tendency of states to legitimize the status of unwanted migrants through legalization, or through policies which afford officials the discretion to decide whether deportable persons ought to be sent home. Scholars have placed particular emphasis on the capacity of liberal lobby groups—both capitalist and human rights oriented—to compel states to make exceptions to the law (Freeman, 1995; Linklater, 1998; Soysal, 1994). They argue that states reluctantly “accept” unwanted migration, because they are liberal, meaning pliable to pro-migrant humanitarian and business lobbying. While subsequent research moved on to consider ways in which states might escape this dilemma (Guiraudon, 2000; Lahav, 1998; Lavenex, 2006), the interpretation of the failings of immigration enforcement as a “gap” has not been displaced (Bonjour, 2011; Czaika and De Haas, 2013).
An alternate starting point from which to make sense of the observation that unauthorized migrants are rarely removed from state territory is that hierarchical control over the violent expulsion of immigrants has only been historically achieved across a relatively small expanse. Whereas in the veritable “no-man’s land” of the international border line, airport security zones, and international air and maritime space, officials can be expected to screen the population and separate it out into wanted and unwanted groups in accordance with immigration law; away from these borderlines, consistent enforcement is rare. In the domestic setting, a variety of actors have the legal and physical capacity to assist in the process of locating, arresting, and deporting unauthorized migrants, but retain significant degrees of discretion as to whether they will do so, and if so, in what way. These include municipal governments that explicitly instruct their agencies not to police immigration laws and instead provide all-comers with access to services and in some cases official identification (Bilke, 2009; Lippert and Rehaag, 2012; Ridgley, 2008; Van der Leun and Bouter, 2015), sub-state agencies such as the police and department of labor that prioritize the policing of crime and labor law over immigration enforcement (Lee, 2011; author, 2008; Leerkes et al., 2012), businesses that employ the most qualified and cheapest labor regardless of their immigration status (Cornelius, 1994: 391–392; González-Enríquez, 2009: 155; Miller, 2001), and religious organizations that privilege their own understandings of legitimate asylum over those of the state (Rehaag, 2009). The result is not that immigration laws tend to be unenforced within state borders but that evidence of enforcement is geographically variegated across the domestic sphere (Walker and Leitner, 2011). This unevenness of domestic enforcement outweighs the significance of acts of state leniency in accounting for the continued presence of large numbers of informal migrants in most states. Government officials often actively formalize the status of those informal migrants whom they are able to locate and arrest. However, they far more routinely simply do not detect the presence of informal migrants, verify their status, and/or take them into custody.
If the above claims are accepted, then efforts to exclude unwanted migrants within state territory constitute a classic coordination problem: consistent enforcement across space requires the concerted action of multiple (relatively) autonomous actors. The fact that across the globe, millions of people are deported each year, is evidence of the fact that these are not insurmountable obstacles. Our efforts to study such outcomes ought to focus more closely on whether and how coordination problems are commonly overcome.
One way in which policy-makers’ coordination problems have been commonly overcome is through the establishment of hierarchies. However, efforts that fall short of eliminating the discretion of officials to decide whether and where they will search, how they will determine who to arrest and which detainees they will deport, will consistently encounter obstacles. Another approach involves the attempt to engender aggregate effects. Instead of working against the discretion of local officials, this involves affording greater discretionary power to a wider variety of officials under the assumption that the cumulative effect of their behavior will align with state ideals. The work of Matt Coleman and Austin Kocher on the devolution of immigration enforcement in the US helps us to identify these two methods of resolving the coordination problem (Coleman, 2007, 2009, 2012; Coleman and Kocher, 2011). On the one hand, they have focused our attention on a number of parallel policy processes initiated by the US Federal government that seek to expand and strengthen hierarchies by limiting the discretion of officials vis-à-vis immigration enforcement. These include efforts to (a) limit the discretion of judges in deportation cases and (b) expand the efforts of immigration officials to police the interior. On the other hand, they write of initiatives designed to expand the discretionary authority of officials by (a) insulating immigration enforcement from judicial review and (b) devolving powers of immigration enforcement to police officials. They speak of these processes as intertwined with one another, but I would suggest there is value in emphasizing their differences. One involves the attempt to use the law to mandate that investigations, arrests, and deportations shall occur. The other involves the attempt to liberate a hypothetical “enforcement market” from the constraints of the law, under the assumption that officials will use their discretionary power to an exclusionary end. In short, one resolves a coordination problem through hierarchy while the other does so by way of aggregate effects of behavior at an alternate (lower) scale.
This second approach to coordination problems in immigration enforcement has received limited attention in the literature, despite the fact that it may constitute a crucial difference between a highly exclusionary and less exclusionary state. In order to develop this point, I draw on my own research in South Africa, which reveals how inter-scalar relations can generate aggregate effects upon territorially exclusive outcomes at the national scale.
Unbeknownst to many, during the mid-2000s South Africa became one of the most prolific deporters of foreign nationals in the world, deporting more people per capita than many more powerfully equipped states. Yet, this outcome occurred in the near absence of a hierarchically ordered immigration bureaucracy (Vigneswaran, 2020). The immigration department was corrupt, poorly staffed, under-resourced, and incapable of compelling other departments to follow its instructions. In order to understand this outcome, my research looked beyond the conventional hierarchical relationships between immigration policy-makers and implementers, to see the alternate ways in which South Africa’s immigration enforcers had been “put to work.”
The key to understanding the South African case proved to be recognizing the complex relationship between forms of territorial exclusion and legitimation at different scales. More specifically, exclusionary immigration policy-makers in South Africa were the unwitting “beneficiaries” of processes of legitimation operating at the urban scale. Briefly, Apartheid era policies and laws had developed a powerful movement control apparatus designed to enforce segregation at the urban scale. In particular, the South African police force had taken on the policing of urban mobility—through arrests, detention, and forced removal of black South Africans—as their defining function and primary occupation, and policy-makers provided the police with extensive human and material resources to fulfill this set of tasks. In the twilight of the Apartheid era, the South African government delegitimized this form of movement control at the urban scale, when they abandoned the Group Areas Act in 1986. However, this change at the urban scale had dramatic effects for exclusionary processes occurring at the national scale, as police officials—facing an existential crisis—took up the task of internally policing unauthorized migration, targeting, and arresting immigrants from Zimbabwe, Mozambique, and Lesotho in the same way they had previously targeted South African Zulus, Xhosas, and “Coloureds” (Vigneswaran, 2011).
During the transition to Apartheid, various officials, including the police, attempted, through policy and law to move the police out of immigration policing, but failed. This “failure” was surprisingly “effective” as the aggregate effects of police officials’ use of their discretionary power to arrest and detain international migrants generated huge numbers of arrests leading to deportations. In the absence of hierarchy, acts which promoted the discretion of officials had generated powerful exclusionary outcomes at the national scale, primarily due to processes of legitimation playing out at the urban scale.
The South African case points to the need for research to further explore the potential power of both non-hierarchical forms of coordination and inter-scalar relations to shape immigration enforcement outcomes. If the above assessments were correct, the study of immigration enforcement would be better equipped to address its central puzzles if it adopted a revised set of Weberian assumptions. Instead of beginning from the assumption that states have monopolized legitimate movement across their jurisdiction and then seeking to understand why in many contexts they fail to use this power effectively, they might instead begin with the assumption that territorial exclusion necessarily confronts coordination problems as soon as we move away from the borderline, government office, and detention center. In part, this move encourages us to re-direct our attention away from efforts to exclude under the assumption of, or through hierarchy, and toward efforts to marshal the various forms of exclusionary power that reside at alternate scales. The South African case offers an example of how this sort of analysis might complement the study of hierarchical control at the border. At the same time, this move does not mean that we ought to neglect the possibility that states might respond to coordination problems through efforts to establish hierarchical control over the immigration enforcement process in their interior. As Coleman and Kocher argue, minor changes to the discretionary authority of officials can have major effects on exclusionary outcomes. As we shall see in the next section, the availability of this form of command structure constitutes a crucial difference between government efforts to exclude at the national and urban scales. The absence of a claim to legitimate exclusion from the city means that there is only one means by which coordination challenges may be resolved: through the aggregate effects of exclusion at a lower scale.
Why do “inclusive” policies generate gentrification?
If immigration researchers have been peculiarly puzzled by the failure of policy-makers to generate exclusion at the national scale, then we may say that gentrification research has been plagued by the opposite problem. They have perhaps made too little of the disjuncture between gentrification policy and practice. The term “gentrification” refers to “a change in the population of land-users such that the new land-users are of a higher socio-economic status than the previous users, together with an associated change in the built environment through a reinvestment in fixed capital” (Clark, 2010: 25). In 20th century studies of gentrification, policy-makers played a relatively peripheral role. Researchers tended to argue that such changes were primarily the product of economic, cultural, or social dynamics. 3 However, at the turn of the century, many scholars began to speculate whether the peripheralization of lower- and working-class groups was becoming a defining motive of urban policy. This proposition was popularized by Neil Smith’s (2002) observation that gentrification had become “thoroughly generalized as an urban strategy that takes over from liberal urban policy” (427). These sentiments were echoed in Elvin Wyly and Daniel Hammel’s(2004) observation that “more than ever before, gentrification is incorporated into public policy” (35). Other researchers took the point further, to suggest that public policy had become “the main driver of gentrification” (Cameron and Coaffee, 2005: 39). The principal exemplar of the notion that policy ought to be a focal point of efforts to understand the politics of gentrification is Loretta Lees’ (2000) idea that various governments have deployed a gentrification “blueprint,” a “one size fits all” policy (404) to transform the class structure of their cities’ residential core.
Given the growing conviction that gentrification has become a policy-driven process, it is somewhat surprising to note that urban policies rarely explicitly endorse the exclusionary outcomes of gentrification. Public policy-makers and documents rarely describe the process of in-migration and displacement that their policies have encouraged as “gentrification.” In particular, policy-makers have rarely clarified which members of society ought to move out and which should move in to a “gentrifying” area. Instead, they have tended to describe their aims in terms of ensuring that diverse groups may benefit from the processes of “regeneration,” “restructuring,” “renewal,” “improvement,” “balance” and “mixing,” etc. that they proscribe as redevelopment agendas for those zones that subsequently gentrify (Bridge et al., 2012; Herzfeld, 2010; Lees, 2008; Slater, 2006; Smith, 2002).
Critical scholars have commonly regarded policy-makers’ failure to explicitly label their plans “gentrification” as a form of obfuscation. So, for example, Neil Smith (2008: 196) noted that the term “regeneration” was merely a way of making class warfare more palatable. Some authors have gone further, to suggest that the new gentrification euphemisms are part of the means by which the state engages in class-driven ideological warfare (Loopmans et al., 2010: 182). So, for gentrification researchers the policy gap works in reverse: an exclusionary set of interests dominates despite a formally or rhetorically inclusive policy.
I would argue that this “policy gap” might be more fruitfully studied as a sign of the difficulties involved in violently excluding populations at the urban scale and therefore a crucial object of study. This begins with the recognition that, in the absence of a claim to legitimate exclusion, policy-makers lack the dedicated laws, policy instruments, and bureaucratic agencies—to exclude specific groups from parts of the city. There is a prerogative to exclude at the level of property but no monopoly over the ends to which this prerogative is deployed. Exclusion at the level of property may generate aggregate effects. However, ensuring that legitimate violence is deployed to such an end constitutes a significant coordination problem.
This point begins with the recognition that the primary way that violence has been deployed to promote gentrification is through the commodification of space (Clark, 2010: 27). In this respect, scholars have noted how forms of state violence, including the demolition of derelict property (Goetz, 2011a, 2011b; Weber et al., 2006), the privatization or relocation of public housing (Crump, 2002; Goetz, 2011a; Newman, 2004; Newman and Ashton, 2004), and the evictions of unauthorized occupiers (Atkinson, 2000a, 2000b; Newman and Wyly, 2006) have each contributed to processes of gentrification. However, the adoption of commodification as a policy does not, in and of itself, determine which group will have access to a given part of the city. Rather, it creates a context in which market forces
The question of whether commodification results in displacement of lower- and working-class groups depends on a variety of mitigating factors. In part this has to do with the fact that the spatial structure of the housing market can stall or slow gentrification processes in specific neighborhoods. The quality and character of the housing stock, home ownership rates, and the housing preferences of the resident population stall, slow, or prevent gentrification in areas that would otherwise appear ripe for such change (Ley and Dobson, 2008; Shaw, 2004; Shaw and Hagemans, 2015). However, the more pertinent issue for this paper is that exclusionary violence consistently stalls commodification processes or shapes them in ways that do not lead to the eviction of lower- and working-class groups. City planning mechanisms designed to fast-track demolitions may not strongly influence where demolitions occur (Weber et al., 2006). Public actors may formalize the status of informal dwellers and squatters instead of responding to private demands for eviction (Kim, 2010; Pruijt, 2003; Thörn, 2012). Public agencies may fund the efforts of marginal groups to buy or lease housing in commodified areas (Badcock, 1989; Newman, 2004).
Gentrification researchers have studied the coordination problem through painstaking empirical work, in individual cases, to demonstrate how various actors have combined to generate exclusionary outcomes from commodification processes. Much of this research takes place under the banner of the “revanchist city” in which “capital, embodied by middle-class gentrifiers, battles it out, block by block, house by house, to retake the city” (Lees, 2000: 399). This work could be further strengthened by efforts to explore the extent to which and ways in which gentrification has displaced more inclusionary urban norms.
In addition to the difficulties involved in pre-determining the outcomes of market forces, efforts to commodify space also often struggle because obstacles to residential segregation do not simply take the form of policy-maker aspirations or commitments, but are hard-wired into municipal institutions at various scales. Wouter van Gent (2013) has demonstrated the power of such constraints in his study of the limits of state-driven gentrification efforts in Amsterdam. He notes that Amsterdam’s policy-makers have been unashamedly calling for gentrification, by prioritizing rising prices of housing stock in the city center and recruiting middle-class residents to these areas. However, these broad schematic plans to pursue gentrification at the urban scale have been muted by protections against exclusionary practices grounded at various institutional scales. In particular, guarantees of minimum proportions of social housing stock in the city center and in new developments, protections for tenants against eviction, and the autonomous functioning of working-class housing associations have significantly muted the exclusionary impacts of commodification policies. As Van Gent argues, it is not necessarily the original intent behind social protections that matter, but the question of whether social movements and political coalitions actively made use of these protections that decides whether market forces will succeed in excluding poor and working-class populations from the city.
While the Amsterdam case reveals the manner in which enduring taboos against territorial exclusion form a bulwark against gentrification policies, in Mumbai we see similar contestation in the absence of enduring protections for working- and lower-class residents. Instead, in Mumbai working-class and middle-class activists have secured their place in the city through the creative reconstruction and redeployment of municipal policies and laws.
Over the past two decades, Mumbai’s cotton industry has largely collapsed and the mills, millworker residences, and adjacent slums in these areas have become the target for property developers who have primarily sought to build sky-rise apartments for upper class Maharashtrians and other internal migrants from the North of India. Property developers have been powerfully resourced and politically connected, and have largely won this gentrification battle. However, both working class and slum dweller movements were able to creatively use their existing rights of individual and collective occupancy, along with colonial era municipal laws to resist processes of expulsion from the inner-city. The key institutions here were municipal laws limiting the height and density of development in Mumbai, which means that almost every skyscraper in Mumbai has been built as an exception to municipal codes. Slum dweller and cotton mill worker movements were able to use these constraints—in conjunction with municipal officials—to exert leverage in their negotiations with developers over land. They have compelled developers to build social housing—sometimes on new-build and non-inner-city sites, but often in areas directly adjacent to the new developments—for the previous occupants, in order to gain government approval for exceptional building permits. As a result, rather than the process of mass lower- and working-class expulsion that one might expect in a city like Mumbai, which has few protections for the housing needs of these groups, activists have managed to creatively utilize a range of
The Amsterdam and Mumbai cases reveal that territorial exclusion still suffers from a legitimacy deficit at the urban scale. Anti-gentrification norms have been and continue to be institutionalized in municipal policy, law, and practice. Even when powerful political actors are explicitly committed to territorially exclusive gentrification processes, there are a range of custom-built and readily manufactured measures which can be utilized to subvert and subdue gentrification plans. The challenge for researchers would be to move beyond local case studies, to ask to what extent the legitimacy of territorial exclusion at the urban scale has changed in recent years. In this respect, the IR/Political Science literature could be of significant utility to the geographers working on gentrification in the city. More specifically, scholarship on norm origin and diffusion could provide a framework within which we could begin to explore both the institutional forms that taboos against urban exclusion have taken in different contexts, and the extent to which these taboos have been superseded or overcome by new gentrification norms (e.g. Finnemore and Sikkink, 1998). Instead of dismissing these obstacles as mere rhetoric and simply heralding the arrival of a new gentrification policy consensus, this sort of research would seek to actively question the extent to which policy-makers might be increasingly able to not simply enable gentrification markets but forcefully implement such policies.
Concluding remarks
In summary, this piece provides reasons to suggest that the study of immigration control and gentrification would profit from a revised set of assumptions about the state’s exclusionary power. If we begin from the assumption that exclusionary violence can be equated with the mere enforcement of policy and law, our research will tend to focus on explaining enforcement “failures” and “biases.” Liberal, Marxist, and Foucauldian streams of research have profited significantly from the process of explaining why states tend to act in ways that depart from their stated exclusionary or anti-exclusionary aims. However, I have suggested that these interpretations of state behavior may rest on a flawed assumption. Within state borders, hierarchy over state practices of violent exclusion cannot be assumed.
Beginning with a Weberian set of assumptions does not necessitate a statist theory of territorial exclusion. Rather, it opens the door to a debate about government: how violence is coordinated to exclusionary ends. Liberal, Marxist, and Foucauldian ideas can compete or work synthetically with Weberian accounts to explain how and under what conditions such problems are overcome. The general gist of this article is that the proto-typical Weberian answer to the problem of how territorial exclusion is governed has its limits. Hierarchy is a relatively enduring answer to the problem of anarchy but one that does not stretch very far. Some of my readers will want to go further, to annihilate the “myth” of the hierarchical state. This piece provides some solid grounds for not going that far. On the one hand, this is because I stand by the notion that hierarchy constitutes a valid assumption at many borders and across a variety of places within the domestic realm. On the other hand, this is because the scalar geography of legitimate violence enables some forms of hierarchical control, but not all. At a minimum, the discussion suggests that it would be premature to ignore the difference between the options available to states at the national scale—hierarchy or aggregate effects—and the urban scale—aggregate effects alone.
If these points hold, then studying the coordination of violent exclusion in the domestic sphere ought to hold opportunities for both political geographers and international relations scholars. It would appear that there is ample opportunity for synthesis and debate on this front. IR journals offer an array of ideas regarding how coordination problems have been resolved. While Weber tends to sit in the background of contemporary geographical journals, I would expect that many of IR’s critical traditions of research into the problem of anarchy would map neatly onto the Marxist, Feminist, and post-Structuralist frameworks that are more commonly discussed within this discipline. In this dialogue, IR’s traditional skepticism of the power of global norms could usefully temper geographers’ more bullish reflections on an emergent gentrification policy norm. As this study suggests, while policy-makers might be more openly declaring their gentrification ambitions, gentrification still lacks the deep institutional legitimacy to allow for more hierarchically ordered exclusion of poor and working-class groups from the city.
The challenge that lies ahead is to develop a truly comparative research project on the politics of exclusion across scales. This would not only include a more rigorous conceptualization and comparison of exclusion at the scales of the city and national territory. It should also extend—at a minimum—to exclusionary projects at the “international” or “global” scale. In this respect, geographers are already helping to shift the IR debates. Mainstream IR scholars have been largely inattentive to the governance of violent exclusion at the global scale because there has been no treaty or policy which explicitly states this goal (Betts, 2010; Hollifield, 1992; Koslowski, 2011). Geographers have formed a major part of an inter-disciplinary conversation, which has filled this gap, discussing the emergence of a “global mobility regime” (Glick Schiller and Salazar, 2013; Shamir, 2005). This research points to “sedentarization”—efforts to prevent people from the developing world from reaching the developed world—as an emerging and common exclusionary agenda aim that shapes a variety of practices of violent exclusion, including border control, detention, deportation, and encampment (Hindess, 2000; Hyndman and Giles, 2011; Malkki, 1992). However, to this point the literature has primarily focused its attention on the critical interpretation of gaps between state policy and practice. It has yet to take on the ancillary task of exploring to what extent violence has been coordinated to sedentarization effects at the global scale. It would seem that geographers’ experience in studying the exclusionary violence within state borders could be readily transposed to help IR identify whether anarchy is being overcome at its preferred analytical “scale.”
Footnotes
Acknowledgements
The author would like to thank Marieke de Goede, Marlies Glasius and two anonymous reviewers for their comments on previous drafts.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.