This page intentionally left blank
Policing Transversal Borders
Sharon Pickering and Leanne Weber
In this chapter we discuss how traditional policing agencies are being shaped by the political emphasis on border control, consider whether immigration authorities are becoming more police-like, and identify how other government and non-government agencies are being brought into a migration policing role. We do so by focusing on Australia, a nation which has expended significant political, material, and ideological efforts on crafting an expansive border policing regime over the past 15 years.
Border Policing and Transversal Borders
Border control can be regarded as the heart of the regulatory effort to sustain national sovereignty (Sassen 1996). Historically, states have sought to construct their sovereignty as fixed, spatially and temporally. These fixed borders have often been unquestioned in view of the ‘inescapable fact’ that international state borders ‘all have territory, they all are territory’ (Donnan and Wilson 1999: 44, original emphasis). However, border policing increasingly requires activities that are both beyond territory and operate within physical borders. This decoupling from territory fundamentally changes the nature of that policing, and in turn the nature of the border. While we might argue about which is cause and which effect, it is reasonable to conclude that they are mutually constitutive. As Pratt (2005: 185) reminds us: ‘The border is an ongoing accomplishment, yet the processes by which it is continually produced are erased by its apparent self-evidence.’ It is through acts of policing that the juridical border is ‘performed’ by both state and non-state actors (Wonders 2006).
Conceptualizing borders in terms of statecraft transforms a criminological analysis of border policing from an understanding of the border as fixed to understanding it as dispersed to multiple ‘sites of enforcement’ (Weber and Bowling 2004; Weber 2006), a process in which policing plays a critical role (Pickering 2004). In considering the policing of multiple borders we seek to apply the idea of transversal- ity. In so doing we take up the work of Soguk and Whitehall (1999) who argue that prior to identifying the multiple boundaries, frontiers, and borders that overlay one another there is a condition of transversality which does not occupy a space easily locatable within this complex, geographical environment. Rather, a state of transversality exists prior to conventional sovereign boundaries ‘that enable[s] political inclusions, exclusions and cultural separations across peoples and places’ (Soguk and Whitehall 1999: 675). Soguk and Whitehall argue that the transversal is defined both for and by migrants and their movements. Their analysis does not start with the state in order to explore the transversal, but with transversal subjects—the migrants and their crossings. Consequently, studies in this tradition have used the voices of migrants to disrupt official accounts of sovereignty and borders. Applying these ideas about transversality to the internal border shifts the focus away from physical border crossings and towards ethnographic accounts of lives lived in states of illegality (Nunez and Heyman 2007; Fan 2008); the role of migrants as active self-definers, seeking to cross fluid legal boundaries that demarcate zones of differential entitlement (Coutin 2005; Schuster 2005); and the often exclusionary responses of extant populations, as described by Khosravi (2010: 75): ‘After crossing many physical, national borders, I found myself facing other kinds of border ... those in the minds of people.’
We argue here that conditions of transversality are not only produced by migrants and their movements, but also by those seeking to police borders. Transversal borders, in other words, are performed by a range of people—openly and officially, or tacitly and unofficially—who may seek to reinforce, transgress, or transcend the juridical border. Inasmuch as the transversal histories of migrants challenge the notion of a sovereign essence it is possible to imagine a policing function both central to, but floating apart from, the state, able to invest in a crossborder, de-territorialized existence which is unbounded in many respects by temporal or geographic constraint. One of us has made this argument elsewhere in relation to the Australian Federal Police (Pickering 2004). The intersection of the many practices of border policing brings into being what we call transversal policing. The Latin definition of transversus, which translates as ‘lying across’, has been applied in medical and other settings to connote a more active process of ‘cutting across’ an axis, so that the resulting incision or structure is situated at right angles to the original frame of reference. Transversal border policing practices which run, conceptually, at right angles across established borders open up spaces of governance which are ripe for colonization.
While processes of geographical transversality influence where border control functions are performed, changes in governance characteristic of globalizing, neoliberal states are bringing about a dispersal of authority across networks or chains of actors (Garland 1996) which affects who performs these state-defining functions (Weber and Bowling 2004). This shifts the boundaries of existing government agencies, creates spaces for new private and commercial actors, and redefines the state through border policing processes that are both embodied (through performances by state and non-state actors) and informated (through information exchange that creates an invisible, virtual border). In this chapter we consider how transver- sality brings new actors into the border policing realm as novel spaces ofgovernance are made available for exploitation by entrepreneurs. At the same time, other agencies resist imposed changes to their established roles. We conclude that transversal policing is an active and entrepreneurial practice which generates deep contradictions and paradoxes for the agencies involved as it cuts across established norms and practices.
In the remainder of the chapter we will consider how Australian immigration authorities have become more police-like under pressure to secure borders; argue that Australian police are increasingly occupied in migration policing roles; and demonstrate that other government and non-government agencies are being drawn into wider migration policing networks. In each context we consider border policing developments at both the external and internal borders.
Proposition 1: Immigration Authorities are Becoming More Police-like The external border: scientifically managing migration—policing, prediction, pre-emption
The convergence of policing and immigration functions can increasingly be viewed in the use of pre-emptive border control measures by immigration authorities— that is, their efforts to police out undesired persons prior to their arrival in Australia. Pre-emption is concerned with predicting and preventing acts before they occur. Primarily this is occurring through the desire to achieve certainty in an increasingly ‘scientific’ approach to managing migration through the use of predictive analytics that share common features with crime science, in particular the use of crime mapping by police agencies, models of intelligence-led policing, and the use of biometrics (Wilson 2006). The politics of border control can be neutralized and made objective, on the one hand, by translating border politics into problems to be scientifically known and solved and, on the other hand, are rendered banal as calculated risks to be managed. ‘Scientific’ management—the massing of data and its analysis—are the functions that animate both the transversal border policing space and the frontiers of information-driven policing more generally. It has been argued that the emergence of intelligence-led policing is ‘the most significant and profound paradigm change in modern policing’ (Ratcliffe 2008: 88). Biopolitics and what Bigo (2002) calls the ‘governmentality of unease’ (characterized by practices of exceptionalism and acts of profiling and containing foreigners) in the form of immigration-as-policing activities occur prior to and apart from the border. This governmentality of unease produces a border function on/in the individual bodies/lives of those perceived as having a threatening intent. Governing mobilities, Amoore has argued in the US context, means that ‘the management of the border. ... is more appropriately understood as a matter of biopolitics, as a mobile regulatory site through which people’s everyday lives can be made amenable to intervention and management’ (2006: 337).
Border policing is a core function of the Department of Immigration and Citizenship (DIAC) in Australia. Currently, the DIAC describes its external policing role as ‘promoting visa integrity’. Australia operates a total visa system. It is the prevention aspect of ‘promoting visa integrity’ that primarily occurs at the external border—that is at the many points prior to arrival in Australia where Australian authorities seek to deter and deflect unwanted travellers. Similar to the pre-emptive and intelligence functions of security and policing agencies (see Zedner 2007; McCulloch and Pickering 2009), the DIAC maintains an offshore network of border intelligence officers, migration intelligence officers, and airport liaison officers to undertake these tasks. The DIAC approaches prevention as ‘intent management’ which is defined as ‘a high level use of information management and predictive analysis [in which] DIAC aims to, in a virtual sense, “extend the borders” by identifying and managing high risk travellers before they get to Australia’ (McCairns 2011).
This seemingly Orwellian desire to manage intent requires a range of elaborate and systematized processes to manage large populations of potential and actual travellers, only some of whom may seek ‘migration outcomes’ within Australia. It is to assess and respond to undesired travellers to pre-empt and prevent any potential claim on Australia’s protection obligations or other contravention of migration arrangements (including anticipated post-arrival compliance breaches). Pre-emptive border control has much in common with pre-emptive policing endeavours (Weber 2006; Wilson and Weber 2008) and pre-crime models (Zedner 2007) which seek to identify and respond to crime before it is committed, particularly crimes related to national security concerns.
The DIAC describes its ‘layered’ approach to intent management as follows: ‘The process starts even before the person books a ticket or lodges an application for a visa to enter Australia’ (McCairns 2011). This places intent management as an outermost rim in border policing—one that occurs temporally well before the usual preparatory actions associated with travel (for example buying a ticket or applying for a visa) and physically well before any border crossing. Those who are ‘intent managed’ are policed out of the licit market in the movement of people. Those who are not intent managed out of the system are then streamed into policing categories by immigration, described by the DIAC as ‘risk tiering’:
The client’s details are assessed against a set of risk ratings. Key objective of risk tiering is to be able to predict high and low risk clients based on historical patterns. Aligning processing effort to client risk levels. Depending on the risk rating, visa applications will be streamed to a particular processing queue: streamlined, standard, high rigour. (McCairns 2011)
Even when a person successfully satisfies the conditions for issuing a visa, a second layer of border policing is undertaken by immigration authorities using predictive modelling to identify risky travellers. These analytics (see Table 1) are routinely changed and updated; however, they always work to filter the most risky aspects of a traveller which include suspect personal attributes (country of birth, citizenship) and suspect authorities (airline, post-granting the visa, point of arrival).
Predictive modelling is operationalized through the ‘risk scoring’ of all air arrivals. This has two key phases. The first is entirely automated risk scoring based on predictive modelling, detailed above. This identifies persons who will be subject to a second stage of checking by an analyst at the airport of arrival 30 minutes after a traveller has checked in at the airport of embarkation. This analysis,
Table 1 ‘Risk Profile’—Attributes of interest
Source: McCairns (2011).
witnessed during a period of ethnographic research by one of the authors, is undertaken at great speed and seeks to identify and discount any data anomalies— such as data from different systems containing misspellings and being incorrectly matched. Just as risk profiles are used to direct police resources to crime hot spots and repeat offenders (Ratcliffe 2008), risk tiering is used to reduce the list of persons of interest for immigration authorities so that they can be subject to more intensive scrutiny upon arrival, or in some circumstances be prevented from boarding the plane.