Content uploaded by Austin Kocher
Author content
All content in this area was uploaded by Austin Kocher on Aug 02, 2020
Content may be subject to copyright.
Detention, deportation, devolution and
immigrant incapacitation in the US, post 9/11geoj_424228..237
MATHEW COLEMAN AND AUSTIN KOCHER
Department of Geography, Ohio State University, Columbus, OH 43210-1361, USA
E-mail: coleman.373@osu.edu; kocher.51@osu.edu
This paper was accepted for publication on May 2011
In this paper we examine perhaps the most significant shift in US immigration enforcement since the
militarisation of the US–Mexico border in the late 1980s and early 1990s – the now decade-long
transformation of immigration enforcement from an outwards-looking power, located at the terri-
torial margins of the state, into also an inwards-looking power focused on resident immigrant
everydays. In large measure this shift in the geography of immigration policing is due to an
unprecedented devolution of a once exclusively federal power to regulate immigration to non-
federal law enforcement agencies operating in non-border spaces in the post-9/11 environment. We
argue that the result of this shift in the ‘where’ of immigration enforcement amounts to a spatialised
tactic of immigrant ‘incapacitation’.
KEY WORDS: criminalisation, immigration control, detention and deportation, post 9/11
Two sisters attending high school in Wake County, North
Carolina were detained by the school’s discipline officer
in September 2009 for fighting. A Raleigh Police Depart-
ment officer summoned to the school determined that the
sisters were born in Mexico, and arrested both on charges
of simple assault and disorderly conduct. A third girl
involved in the fight, who told the officer that she was
US-born, was not arrested. The sisters were sent to the
Wake County jail. Both were interviewed during the
booking process by a sheriff’s deputy trained by federal
immigration authorities to investigate immigration status.
The sisters were identified as undocumented. The older
sister was deported; the younger sister was allowed to
finish high school before being deported.
In September 2009, a Mexican-born male in his 30s left
his house following an argument with his US citizen
partner and sat in his car listening to music. The man was
arrested by a Raleigh Police Department officer for a
noise violation, following a complaint from a neighbour.
During the booking process at the Wake County jail, he
was identified as a deportable non-citizen and placed
into deportation proceedings.
In May 2009, a single mother on her way to a hospital in
Raleigh to care for a friend who was in labour, was pulled
over at a red light. The officer asked for a valid driver’s
license, and took the driver into custody under a ‘no
operator’s license’ charge. During the booking process at
the county jail, the driver was determined to be undocu-
mented. The driver was released on her own recognisance
with a set deportation hearing date before an immigration
judge.
In August 2010, a 19-year-old was arrested in Raleigh for
throwing a soft drink can out of a car window and hitting
a pedestrian. The teenager is currently in deportation
proceedings as a result of the arrest. His mother cannot
file a cancellation of deportation request because she
would likely face deportation if she came before the
court.
Immigration control in the US context is often
thought of as primarily a matter of border enforce-
ment. We can trace this equation between immi-
gration control and border control in part to the
exceptional place that the US–Mexico border occu-
pies in popular and legislative imaginations in the US.
Exceptional, that is, in terms of the large volume of
unauthorised traffic that crosses it, as well as the threat
to law and order that such transgressions are said to
pose (Andreas 2010; Dunn 2010; Maril 2004; Nevins
2010). Most recently the US–Mexico border has
emerged as exceptional in relation to homeland secu-
rity and the national security threats posed by unau-
thorised entry. Political pundit and erstwhile Fox news
anchor Michelle Malkin (2002, 9), for instance,
expresses a widespread post-9/11 fear in her claim
that ‘bin Laden-funded operatives . . . ride the rails
undetected from Mexico along with hundreds of thou-
sands of other undocumented workers’. She adds:
‘We don’t have borders. We have the world’s longest
backdoor welcome mat’.
Malkin’s description of the border as the ‘world’s
longest backdoor welcome mat’ – and, crucially, the
way she traces an identity between terrorists and
The Geographical Journal, Vol. 177, No. 3, September 2011, pp. 228–237, doi: 10.1111/j.1475-4959.2011.00424.x
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011 © 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
‘other’ undocumented labourers who freely infiltrate
the body politic via the US Southwest – is far from
original. On the one hand, immigration control at the
border throughout the Cold War period was repeat-
edly the product of densely interwoven national secu-
rity and domestic policy issues (Tichenor 2002;
Zolberg 2006). On the other hand, Malkin’s post-9/11
reading of the border as under potential siege repeats
a well worn (and contradicted) reading of the border
as a simultaneously pedagogical and performative site
(Chapin 1998; Doty 2003). Indeed, nearly continu-
ously throughout the period of modern immigration
enforcement, which can be dated from the late nine-
teenth century Chinese Exclusion period, the
US–Mexico border has been represented as a given
and self-evident marker dividing national populations
at the same time as its apparent lack of solidity has
authorised numerous technologies and practices
which have sought precisely to divide and secure
populations and space territorially, for the most part
by surveilling possible entrants in order to impede the
physical entry of political, economic and social
‘undesirables’. In other words, despite its givenness in
the sense of marking out a space of security on the
inside from a space of insecurity without, the border
has been an ongoing birthing ground for the territorial
organisation of identity and difference, suggesting a
new reading of the ways in which the words ‘undocu-
mented’ and ‘labour’ are put together there. In this
sense, Malkin’s call for tougher border enforcement,
which draws attention to America’s ongoing self-
making in the borderlands rather than to its concrete-
ness, is business as usual.
As a result of this longstanding attempt to control
the movement of populations at and across the border,
the US Southwest has been transformed, particularly
over the last 40 years, into a densely militarised space
– literally littered with old and new interdiction tech-
nologies. These include fences, walls, vehicle barriers,
vehicle checkpoints, ground sensors, and most
recently, electronic surveillance towers. The militari-
sation of the US Southwest, of course, contradicts
Malkin’s portrayal of the border as a space without the
state and its infrastructures (more generally, see
Purcell and Nevins 2005). Indeed, claims that the
US–Mexico border is an unpoliced landscape and
that non-citizens can enter freely between official
ports of entry stand starkly in contrast to federal appre-
hension data for the region. For example, of the global
number of 24 million individuals taken into custody
by federal immigration authorities between 1991 and
2009, 22 million (or 90%) were arrested by the Border
Patrol in the US–Mexico border region (DHS 2010).
We should also note that the US–Mexico border is a
decidedly lethal landscape. Due to the growing use of
border interdiction technologies such as those noted
above, many thousands of would-be labourers have
died from dehydration and exposure while trying to
cross the border on foot (Eschbach et al. 2003). It is for
this reason that Rosas (2007) refers to the border
region as the ‘killing deserts’.
However, that immigration law and enforcement
has historically played out as border enforcement in
the US context, and lethally, should not force an iden-
tity between immigration control and what has been
called the ‘gatekeeper’ state (Nevins 2010). Rethink-
ing the equation between US immigration control and
border control, particularly in the post-9/11 environ-
ment, is our central interest in this paper. Although US
immigration control is well described as a ‘manage-
ment of territory’ (Bigo 2007), in the sense that it is
principally about border enforcement, the externally
oriented territoriality of US immigration enforcement
practice has – within the past decade in particular –
been supplemented with a ‘management of popula-
tions’ strategy focused not on stopping territorial entry
per se but on policing immigrants who have already
settled in the US. To refer quickly to Michel Foucault’s
work on territory and security, immigration control
implicates both a sovereign territorial power, which
divides and encloses social space via state bordering
projects, as well as a superpanoptic mode of govern-
ing (on the superpanoptic, see Puar 2007, 114–65),
which is no longer that of fixing and demarcating the
territory, but of allowing circulations to take place, of
controlling them, sifting the good and the bad, ensuring
thatthingsarealwaysinmovement... butinsuchaway
that the inherent dangers of this circulation are canceled
out.
Foucault (2007, 65)
This emerging ‘management of populations’ approach
to immigration enforcement is not historically unprec-
edented. Nevins and Aizeki (2008; see also Balder-
rama and Rodríguez 2006) argue persuasively that the
active disenfranchisement of undocumented immi-
grant communities in the US interior – a form of legal,
political, and economic apartheid – has always been
a strategic element of US immigration enforcement.
Nonetheless, the immigrant pacification strategy that
Nevins and Aizeki articulate in relation to the mass
deportations during the 1930s as well as the 1954
Operation Wetback are now, unlike at any other point
in US history, part of the everyday fabric of immigrant
life in the US interior. In other words, post-9/11 deten-
tion and deportation have become routinised as a
potential aspect of everyday life for millions of
undocumented immigrant residents in the US (De
Genova 2002).
Before we continue, we want to highlight our agree-
ment with Elden (2009) about the oftentimes too-easy
watershed use of ‘9/11’ which ultimately rewrites
world historical time in terms of an American-centric
calendar of pain. However, in the case of US immi-
gration enforcement, 9/11 needs to be recognised as
an important event. Although as noted above US
immigration policy has long been conditioned by
Detention, deportation, devolution and immigrant incapacitation 229
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011© 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
foreign policy considerations, and as such is best
thought of as constitutively ‘between’ the realms of
foreign and domestic policy in the sense that it follows
from parallel ‘internal’ and ‘external’ inscriptions of
the dangers posed by immigrant bodies, and results in
the securitisation of a range of policy areas with no
obvious direct linkages to national security issues
(Coleman 2008). 9/11 did enable a significant esca-
lation in interior enforcement. This was largely a result
of the new premium placed on homeland security
after the attacks, as well as the way in which immi-
gration policy was immediately sutured to the latter.
Specifically, as we argue below, the attacks intensified
an already operational merger between the civil
immigration and criminal law enforcement appara-
tuses in the US, such that mundane, everyday policing
with no direct relevance to national security by non-
federal authorities can now lead to detention and
eventually deportation. This is indeed how we intend
the arrest stories that we opened with, above, to be
read. These stories, collected during fieldwork in
Raleigh-Durham, North Carolina on the mundane
mechanics and outcomes of local immigration
enforcement, show that contact with non-federal
police for a range of reasons – for example, a driver’s
license-based traffic stop, a minor infraction such as a
noise violation, or a misdemeanor offense, which his-
torically have not implicated an immigration status
check – is now ground zero for US detention and
deportation strategy in the post-9/11 era. Indeed, data
suggest that over the past 10 years hundreds of thou-
sands of individuals have been deported as a result of
these sorts of encounters with local and state police,
in sites far removed from US territorial borders
(Coleman forthcoming). In what follows we propose a
tentative explanation of how this post-9/11 ‘manage-
ment of populations’ works and to what effect.
Non-federal detention and deportation
The growth of immigration enforcement qua popula-
tion management in technically non-border spaces
can be traced to a suite of federal laws passed in the
early to mid 1990s. These expanded the criminal
grounds for removal from the US and created new
detention and deportation procedures with limited
court oversight. We can think of these changes as
sanctioning, first, the criminalisation of immigration
law, and second, the dejuridicalisation of immigration
enforcement (Coleman 2007).
The criminalisation of immigration law – that is, the
treatment of immigration as a criminal problem, as
well as the infiltration of criminal law statute into
immigration law (Miller 2005; Stumpf 2006) – dates
most recently to the aggravated felony charge in the
1988 Anti-Drug Abuse Act. The charge was initially
designed to ensure the detention and deportation of
drug kingpins under various felony charges, but was
expanded repeatedly during the 1990s to encompass
a broad assortment of crimes that would count as
neither felonious nor aggravated for citizens – indeed,
any crime with a sentence of a year or more. The
charge was also made retroactive, meaning that past
convictions could, if brought to the attention of immi-
gration authorities, serve as the basis for deportation.
These developments were accompanied by a parallel
move to sequester immigration policing from the
courts. One example of this dejuridicalisation of
immigration enforcement was the creation of expe-
dited removal in 1996. This procedure permitted
immigration authorities to summarily deport individu-
als apprehended within 100 miles of the US–Mexico
and US–Canadian borders. The most important
example, however, is the way that, in a steady stream
of laws passed after the 1990 Immigration Act,law-
makers hacked away at the possibility for court scru-
tiny of deportation hearings. By the end of the 1990s,
a significant number of deportation cases were by
definition exempted from thorough judicial review
beyond a basic hearing before an immigration judge
within the Department of Justice. These include cases
involving a criminal deportation order, any expedited
removal case, as well as any case in which discretion-
ary relief – such as a waiver or stay of deportation – is
denied by immigration authorities. In summary, as a
double-pronged strategy to merge criminal law
enforcement with immigration law enforcement, and
yet deny non-citizens court protections associated
with the former, the criminalisation and dejuridicali-
sation of immigration enforcement literally meant an
expansion of the criminal infractions that could be
counted as grounds for deportation. There was also a
drastic reduction in the ability of individuals detained
as a result of criminal enforcement to petition their
deportation.
Arguably much more important in terms of the shift
to population management is the recent, that is post-
9/11, direct involvement of non-federal law enforce-
ment officers in immigration control. For much of the
twentieth century, only federal immigration authori-
ties had the power to police immigration, as recogn-
ised by legislators as well as the courts; accordingly,
the recent devolution of immigration enforcement to
non-federal proxies is an historical about-face. There
are a number of examples of non-federal immigration
policing, including ad hoc local–federal policing
related to anti-gang enforcement, workplace raids,
fugitive operations (targeting non-citizens who have
not complied with formal deportation orders), drug
enforcement operations, as well as certain local–
federal information sharing initiatives such as the
recent Secure Communities program, which allows
for local and state police to determine the immigration
status of individuals arrested for a crime and booked
into a non-federal holding facility. In order to make
the discussion more manageable, however, we will
focus here on a specific devolutionary program
known as 287(g), named after the section of the
230 Detention, deportation, devolution and immigrant incapacitation
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011 © 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
Immigration and Nationality Act amended to allow for
the delegation of federal immigration enforcement
responsibilities to non-federal police (Coleman 2009).
The 287(g) program allows state and local police to
investigate immigration cases and ultimately make
immigration arrests on behalf of federal authorities.
There are two basic types of 287(g) authority, both
negotiated between the Department of Homeland
Security’s Immigration and Customs Enforcement
(ICE) office and the local law enforcement agency in
question: the jail enforcement model and the task
force model (for an overview, see Seghetti et al. 2006).
Under the jail enforcement model, federally trained
local law enforcement agents and/or locally based
federal immigration officers check for immigration
status as part of the general booking process. A similar
program known as Secure Communities functions in
much the same way as a the jail enforcement 287(g),
with the exception that the former does not guarantee
federal custody of an individual in the event that
immigration status is found to be in question. The
second form of 287(g) is the task force program.
Unlike the jail model, the task force 287(g) does not
require a criminal arrest per se in order to determine
immigration status. At least as originally conceived,
the task force model awards non-federal officers
broad discretionary authority to make arrests based on
civil immigration status alone during the course of
routine policing. Regardless of the type of 287(g)
agreement active in a jurisdiction, once an individual
has been identified as deportable, local and state
authorities may hold the individual for as long as it
takes to process their deportation, as if they were
formally in federal custody.
A quick mention should be made on the impor-
tance of 9/11 to 287(g) adoption. The 287(g) program
was available to non-federal police well before 9/11
but taken up only as of 2002. Prior to 9/11 there was
little interest in the program, in large measure because
it seemed to go against a then prevalent community
policing model, which required close ties between
immigrant communities and municipal and county
police. After 287(g) was made a possibility in 1996,
Mayor Giuliani of New York sued the federal govern-
ment to overturn aspects of the law on the basis that it
would create immigrant communities aloof from city
police, employers, educators and health workers. The
dramatic shift away from community policing and in
favour of 287(g) in the aftermath of the 9/11 attacks
arguably followed from the new emphasis thereafter
on local–federal law enforcement cooperation as well
as from the way in which immigration enforcement
specifically was made central to the homefront war on
terror from almost the minute that the first airliner hit
the World Trade Center towers. Moreover, the turn to
287(g) was no doubt heavily influenced by the federal
experimentation with the criminalisation and deju-
ridicalisation of immigration enforcement in the
1990s, as above. By the end of the 1990s the popu-
lation of non-citizens subject to deportation had vastly
expanded by virtue of the increase in crimes (many
non-serious) warranting deportation, and by virtue of
the accelerated, extra-judicial channels through
which certain criminally classified non-citizens could
be removed (Coleman 2007). In the wake of 9/11, the
deputisation of non-federal police to enforce immigra-
tion laws in the course of routine criminal policing
was an obvious way in which to further increase this
selective crossing of the civil and criminal law
enforcement systems, such that the discretionary
authority to detain individuals under the criminal law
system and the exemption from significant court over-
sight from the civil immigration system could be maxi-
mised simultaneously, as well as extended from the
post-conviction environment to the everyday policing
context (see below).
More than 70 registered jail model and task force
model 287(g)s are currently active in the US, and at
least as many are thought to be awaiting ratification by
ICE. Many of the currently active 287(g) agreements
are located in the so-called ‘nuevo New South’ (Fink
2003; Mohl 2003; Smith and Furuseth 2006; Suro and
Singer 2002; Winders 2005), or in municipalities and
counties in the US South with historically marginal
Latino/a settlement and with unprecedentedly high
rates of growth in Latino/a populations over the past
decade [for a map of current 287(g) agreements which
draws attention to the importance of the US South, see
Coleman (2009); see Winders (2007) for an account of
interior enforcement targeting Latino/a populations in
the US South, post 9/11]. In the aggregate, over the
January 2006–October 2010 time period for which
data are available, the 287(g) program has led to
the identification of more than 185 000 deportable
non-citizens and the removal of more than 125 000
individuals from the US [ICE 2010; cf Coleman forth-
coming on the difficulty of using federally available
statistics to measure the extent and impact of pro-
grams like 287(g) on undocumented communities].
The most important sites for 287(g) enforcement, mea-
sured in terms of total individuals placed into depor-
tation, or otherwise in detention and awaiting
deportation, are Arizona (50 000 cases), California
(48 000 cases), North Carolina (18 500 cases), Texas
(15 000 cases), Georgia (15 000 cases), and Tennes-
see (8500 cases) (ICE 2010a). To put these numbers in
historical perspective, total 287(g) detentions and
deportations since 2006 exceed all decadal counts of
federal deportations registered before 1970, and are
roughly on par with total federal deportations effected
in the 1970s (231 000 cases) and 1980s (212 000
cases) by the then federal Immigration and Natural-
ization Service (data from INS 1997, Table 58). We
should also note that the 287(g) program is but one of
a number of devolved immigration policing practices
which comprise the post-9/11 immigration enforce-
ment landscape; the Secure Communities program
noted above, for example, has produced an additional
Detention, deportation, devolution and immigrant incapacitation 231
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011© 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
180 000 administrative immigration arrests and/or
local–federal custody transfers during the October
2008–February 2011 period (ICE 2011). The overall
point here is that non-federal custody cases resulting
in detention and deportation are now a cornerstone of
overall US immigration strategy.
Our fieldwork in central North Carolina suggests
several important qualifications about how 287(g)
works (discussed in Coleman forthcoming). First,
287(g) is far from a uniform delegation of power.
Although the memorandum of understanding that
local and state law enforcement agencies sign as part
of the 287(g) delegation process is in theory standar-
dised such that all signatories possess the same
authority to determine immigration status and detain
individuals in the event that status is found to be in
question, what happens on the ground in terms of
how the power to police immigration is taken up is
very much conditioned by site-specific legal and
political contexts (more generally on immigration
enforcement as conditioned locally; see Coleman
2007; Hiemstra 2010; Varsanyi 2008; Wells 2004). In
some localities, 287(g) authority occasions wide-
spread immigration status checks for literally any indi-
vidual who comes into contact with enrolled law
enforcement agencies. In other localities, in contrast,
287(g) is used very selectively to prosecute serious
felony offenders. In this sense, 287(g) authority can
produce radically different categories of deportees as
well as volumes of deportations across sites (in the
North Carolina context, see also Weissman and
Headen 2009; Gill and Nguyen 2010). Second, and
despite this possibility for local-level variation, a
majority of 287(g) agencies appear to be using their
new found authority to focus on minor infraction and
misdemeanor offenders than on serious felony offend-
ers. Indeed, in the bulk of our case studies we found
that it was mostly individuals brought into routine
contact with local and state police that found them-
selves in detention and eventually deported from the
country. For example, aggregate data from the North
Carolina Sheriffs Association regarding its member
287(g) agencies show that regularly more than 80% of
immigration actions result from misdemeanor arrests
(data are discussed in Coleman forthcoming). Third,
the growth in 287(g) deportations is not related to an
increase in worksite enforcement. Rather, 287(g)
implicates immigrant mobility, and specifically auto-
mobility. Regularly 50% of the deportations originat-
ing from 287(g) agencies across our central North
Carolina sites started with a minor driving infraction
(i.e. driving without a license, taillight out, improper
lane change, etc.) or with a non-felonious driving
while impaired charge. In addition, the high number
of non-citizens deported as a result of traffic violations
in the region correlates with the fact that Latinos spe-
cifically are overrepresented in almost all traffic
enforcement stops by all law enforcement agencies in
the region (data available from authors on request).
Moreover, our research shows that the chances of
running into a traffic checkpoint where one would
have to provide a valid driver’s license, and risk arrest
and a status check without a license (which is not
available without a valid social security number), is
greater in higher-concentration Latino and lower
income neighbourhoods than it is in majority non-
Latino and higher income neighbourhoods. In
general, this points to perhaps the most important
aspect of 287(g) authority, in addition to its site speci-
ficity: its generation of insecurity for immigrant popu-
lations on the move between spaces of labour, social
reproduction, and leisure (see also Nuñez and
Heyman 2007; Stuesse 2010). Finally, 287(g) is in
many ways the tip of the iceberg in terms of which law
enforcement agencies are involved in immigration
enforcement. If the 287(g) program is housed at a
common county jail, then any law enforcement
agency that books individuals through the jail – which
in the US South is usually all agencies making an
arrest in a county – could in theory be engaged in
immigration enforcement by virtue of knowing that an
individual’s immigration status will be checked at
some point during the booking process. From this
perspective whether or not local and/or state police
are formally enrolled in a 287(g) is not necessarily a
good indication of the extent of local immigration
enforcement in that area (see also Gladstein et al.
2005). Methodologically this means that studying offi-
cially enrolled 287(g) agencies may not be the best
way to study how immigration enforcement is being
devolved to non-federal law enforcement agencies
and to what effect for undocumented communities.
The newly localised powers of detention and depor-
tation under the 287(g) program have significantly
restructured the way that the criminal law and civil
immigration law systems interact, post-9/11. Three
major differences from the 1990s deserve to be under-
lined. First, the criminalisation and dejuridicalisation
of immigration enforcement charted in the wake of the
1988 Anti-Drug Abuse Act required some form of
criminal prosecution on the part of non-federal
authorities. In other words, there had to be successful
follow through on a criminal charge – be it a minor
infraction, a misdemeanor, or a felony – in order for
deportation to be effected. Prior to the 2000s, the
criminalisation of immigration enforcement was in
large measure a product of the so-called ‘institutional
hearing program’, which required that ‘criminal
aliens’ convicted and/or serving time in state and
federal jails be deported either while in custody or
immediately following the end of their sentence. The
more recent delegatory revolution in immigration
enforcement through programs like 287(g) is altering
this matrix. What we see now is a much looser rela-
tionship between the criminal law and immigration
law systems. While certainly individuals convicted of
crimes are being deported under programs like 287(g),
it is also the case that there is no formal requirement
232 Detention, deportation, devolution and immigrant incapacitation
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011 © 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
under 287(g) for successful criminal prosecution in
order for the deportation process to kick in (see over-
view of 287(g) agreements in GAO 2009a 2009b).
This is well signalled by the fact that immigration
detainers identifying individuals as deportable are
assigned during the booking process and before any
charges have been adjudicated in the courts (Capps
et al. 2011). Given the elasticity of the criminal law
enforcement system, as well as the significant amount
of control that local 287(g) agencies can exercise over
the detention and deportation process, the fact that
individuals are identified as deportable before they
are convicted greatly amplifies the possibility that the
immigration and criminal law systems will come
unglued. During our fieldwork in North Carolina, for
example, we encountered a number of cases (and
heard of many more during interviews with lawyers
and immigrant rights advocates) in which individuals
held on minor infractions such as driving charges
were not selected for prosecution by local authorities
before they were deported, or alternatively were
deported after they had posted bail and/or before they
had a chance to defend themselves in court for the
arresting offense. In other words, a major difference
from a past era of deporting ‘criminal aliens’ is that
they may now be deported as ‘criminals’ without
having been tried or convicted in the criminal law
system. This absence of a required conviction to
deport raises the disquieting possibility that the
general investigatory powers that police possess to
stop and question individuals on criminal law
enforcement grounds can be used for immigration
status inquiries under the guise of public safety – a
possibility that we suggest requires much more in the
way of in-depth research. Importantly, this means
moving beyond an analysis of federally available data
regarding deportation cases and criminal convic-
tions1. In order to examine the relative priority of the
criminal and immigration law systems, in line with
our conclusions above about the site specificity of
programs like 287(g), researchers will have to scruti-
nise the administration of due process in situ and on a
case-by-case basis relative to localities’ specific
deployments of 287(g) authority.
Second, the crossing of criminal law enforcement
with immigration law enforcement in the form of
287(g) authority does not produce a uniform field of
policing in terms of who gets asked for immigration
status and who does not. Putting aside the question of
which jurisdictions are subject to 287(g) authority and
which are not, an important question in terms of
coming to grips with the spatial unevenness of immi-
gration enforcement, within 287(g) jurisdictions immi-
gration law will be brought to bear unevenly. This
selective (rather than systematic) enforcement of
287(g) is, of course, a product of the fact that criminal
law enforcement is itself a spatially fractured and
incomplete logic of control (Herbert 2006; see also
early legal pluralism research, i.e. de Sousa Santos
1987, on how similar ‘social objects’ are translated
into different ‘legal objects’ depending on context).
Indeed, 287(g) policing, made possible by the wide-
spread use of identification and database technologies
at the scale of routine policing, is a form of legal
assessment that is contingent and shifting rather than
guaranteed. In this sense, 287(g) is not a legal power
in crude opposition to ‘illegality’ qua undocumented
residence. Rather, 287(g) authority is about an admin-
istration of immigrant ‘illegalisms’ applied selectively
across space and incompletely in reference to its
target immigrant populations.
Third, even in the case where select law enforcement
agencies in the mid 1990s (in Arizona, California,
Florida, Illinois, New York and Texas) could access
federal immigration databases to determine the immi-
gration status of arrested individuals prior to sentenc-
ing, this very rarely guaranteed federal custody. Getting
individuals from local to federal custody was in general
a problem because relatively few federal resources
were dedicated to interior enforcement (relative to
border enforcement), and because immigration status
checks at the federal scale depended upon the use of
surnames and dates of birth data and oftentimes pro-
duced an unmanageable number of hits. In contrast,
the recent delegation of immigration authority to non-
federal agencies has in many instances created a more
certain custody chain between local, state and federal
authorities. The now greater chance of transferring
individuals from local to federal custody is due to the
use of biometric data (such as fingerprints) to determine
identity during the booking process, as well as federal
authorities’ now methodical use of local and state jail
facilities to hold individuals continuously until they
can be deported. The last point is of crucial signifi-
cance. Programs like 287(g) allow local law enforce-
ment agencies to continue to hold individuals even
when the criminal grounds for detention have been
resolved by virtue of their power to detain on civil
immigration grounds without a criminal charge. This
does not mean that individuals cannot either be
released on their own recognisance or bond out of
local custody pending the outcome of a criminal
charge, particularly if it is a lesser minor infraction.
However, on the whole the delegatory trend in immi-
gration enforcement in the post-9/11 environment
means that non-federal detention is now much more
likely to result in uninterrupted transfer to federal
custody and deportation than in years past.
Immigration enforcement as immigrant
incapacitation
How might we conceptualise the growth of non-
federal immigration enforcement away from US state
borders, for example in the form of a delegatory
power like 287(g)? One possible explanation would
be to refuse any easy distinction between the ‘man-
agement of territories’ and the ‘management of popu-
Detention, deportation, devolution and immigrant incapacitation 233
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011© 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
lations’ by virtue of the way that detention and
deportation practice ‘on the inside’ by non-federal
officials nonetheless works to regulate and (re)distrib-
ute populations on a specifically territorial basis.
Walters (2002, 277) provides an insightful account of
deportation from this perspective. Walters argues that
deportation follows from the ‘sovereign right of states
to control their territories and the discretion they have
regarding the admittance and residence of aliens’.
That deportation is inescapably about state authority
and territory, however, does not mean that deportation
is a neatly sovereign-juridical form of authority.
Rather, Walters sees deportation, by virtue of its his-
torically consistent deployment against shifting state-
based categories of political and social enemies, as a
territorialising security practice which works through
an exhaustive, detail-oriented policing of populations.
The term police is not used here in the sense of crime
prevention, as above, but in reference to Polizeiwis-
senschaft, or the ‘science of police’ of mercantilist
and/or cameralist states (for more on police, see
Gordon 1991). Walters argues that deportation is the
‘routine administration [of the population] with expul-
sion as an instrument to perpetuate sovereign power’
(2002, 282). Moreover, Walters argues that deporta-
tion is an international police power insofar as it oper-
ates according to a ‘logic of dividing and allocating
populations to the territorial authorities deemed prop-
erly responsible for them’ (2002, 283).
Walters’ key point, borrowing from Foucault’s The
history of sexuality (1979, 136), is that deportation is
at its heart an appropriative ‘right of seizure’ levied
over specific non-citizen populations as well as a
territorialising ‘mechanism of subtraction’ (literally, a
power to exclude) which finds its authority in a base-
line spatial assumption that all peoples across the
globe can be identified with specific national state-
territorial spaces. This is a provocative take on depor-
tation by virtue of its undoing of the identity between
border control and immigration control (see also
Walters 2006). Walters’ conceptualisation of deporta-
tion is moreover a significant challenge to the often
too-easy characterisation of the present as biopoliti-
cal. Indeed, with respect to this last point, Walters’
analysis reminds us that the link between security and
population is not always about the more liberal
problem of biopower, as Foucault discussed in The
history of sexuality. Lastly, Walters’ argument draws
much needed attention to immigration enforcement
as the proving grounds for states’ territorial integrity,
that is, as a form of population management intended
to signal to domestic as well as other audiences that the
state is in fact an ongoing territorial concern (see also
Dauvergne 2008). Nonetheless, a possible critique of
the ‘regulation mania’ of police power approach to
deportation is that, at least in the US context, it argu-
ably dovetails more with what the state says ‘it’ is doing
in the realm of detention and deportation rather than
what is actually happening on the ground.
An interesting example of this is a leaked 2003
Department of Homeland Security strategic plan titled
Endgame.Endgame reads very much like a police
document: it outlines a comprehensive, multi-agency
and multi-jurisdictional program to deport all deport-
able non-citizens – at the time the document was
written the total was roughly 11 million individuals –
as well as eliminate the then several hundred thou-
sand ‘final orders of deportation’ backlog before the
courts, all within a 10-year window. Endgame does
help to make sense of the significant increase in
deportations over the past several years. However, it is
also important that Endgame’s ‘100% rate of removal’
strategy – and the array of federal through local law
enforcement practices the report details – is a ‘regu-
lation mania’ only on paper. Despite the significant
increase in deportations over the past decade, there
has been less than a double digit decrease in the
number of undocumented residents in the US; current
estimates put the undocumented populations at 11.1
million people, down from a high of 12 million in
2007. Moreover, certain specific populations of
undocumented residents have actually increased sub-
stantially over the past decade. For example, estimates
suggest that there are 2.1 million more undocumented
Mexican nationals resident in the US in 2009 than in
2000 (data from Passel and Cohn 2010). This trend
can be explained via the militarisation of border
enforcement – individuals are less likely to chance
additional border crossings and more likely to stay put
– but is nonetheless at odds with the detention and
deportation strategy put forth in Endgame.
If we consider Endgame performative of the US
state as a potentiality rather than an actuality per se,
the question thus arises as to how to best understand
current US detention and deportation policies strate-
gically. If the ramping up of detention and deporta-
tion, especially via non-federal law enforcement
officials historically precluded from administering
federal immigration law, is not appreciably altering
the size of the undocumented population as an aggre-
gate, what is the goal of such a policy? What we have
in mind here is detention and deportation practice as
a ‘management of populations’ that is not tethered
explicitly to a ‘management of territory’ in terms of
physically removing non-citizen individuals to their
countries of origin. Obviously the US–Mexico border
is a primary site where the ‘management of popula-
tions’ is still very much coincident with a ‘manage-
ment of territory’, as noted above. Away from the
territorial margins of the state, however, we see the
post-9/11 amplification of interior detention and
deportation as a policy of migrant ‘incapacitation’.
We are indebted to Gilbert (2009) for this term; she
uses it in her analysis of the localisation of immigra-
tion enforcement in the US (and Canada) to move
beyond debates about assimilation as the basis of
immigration policy (see also Nagel 2002 2009 for
a geopolitical account of assimilation which
234 Detention, deportation, devolution and immigrant incapacitation
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011 © 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
approximates Gilbert’s concept of incapacitation).
From an incapacitation standpoint, the localisation of
immigration enforcement can be understood as an
exclusionary moment of ‘civic stratification’ rather
than a corrective, levelling politics of incorporation or
sameness (Kofman 2002; McGregor 2008; Morris
2001; Varsanyi 2008). But incapacitation, we would
argue, is more than the deliberate sorting of immigrant
populations into various strata of precarious legal,
social, political, and economic exceptions relative to
the rights and freedoms enjoyed – at least theoretically
– by full citizens. Incapacitation suggests something
more inherently geographical. As Gilmore (2007, 14)
explains in her own discussion of the term, incapaci-
tation is a spatial fix which ‘doesn’t pretend to change
anything about people except where they are’.
Gilmore’s discussion of incapacitation is in reference
to the growth in racialised prison populations in the
post-industrial US city, and as such is not exactly
analogous to the question of interior immigration
enforcement we are discussing here. However, we
feel that Gilmore’s emphasis on incapacitation as
‘extensively and repeatedly removing people from
disordered, deindustrialised milieus and depositing
them somewhere else’ is worthy of consideration in
the immigration context (Gilmore 2007).
Understanding incapacitation as a spatial fix to
undocumented migration entails thinking through the
difference between territorial presence and legal pres-
ence, and specifically how immigration enforcement
is not necessarily about policing territorial presence in
an absolute, inside/outside sense. Although all
undocumented immigrants are de jure subject to
deportation on account of their extra-legal territorial
presence or ‘undocumentedness’, in practice immi-
gration enforcement plays out very differently. By
virtue of the enormously difficult task of locating and
deporting all undocumented migrants, de facto immi-
gration enforcement is about selective policing and
the subsequent production of a population of ‘territo-
rially present’, yet not legal, residents who live and
labour under the ongoing threat of deportation. We
propose, then, that interior, everyday immigration
enforcement works through the production of an
exemplary migrant precarity: socio-economic and
legal insecurities are visited upon select immigrant
bodies (as with Nevins and Aizeki 2008 above) but
reverberate more widely through familial and other
social networks (Martin forthcoming), and as such
contribute to the indirect regulation of a broader
population of unauthorised immigrant subjects.
Now we are in a better position to see de facto
immigration enforcement as a two-step spatial strat-
egy of migrant incapacitation, borrowing from
Gilmore, as above. On the one hand, via a very public
process of physical detention and removal to their
countries of origin, sensationalised immigrant bodies
(even if the conditions of their arrest are less than
sensational, as noted in the opening vignettes) are
singled out by various local arms of the state in a move
which both performs state power as well as the
alleged dangers and disturbances posed by undocu-
mented aliens (see Bigo 2002 for deportation policy as
a ‘governmentality of unease’ directed at citizen
populations). On the other hand, this is not simply a
performance of state potentiality to a domestic audi-
ence of citizens; it is not simply a tactic designed to
‘worry’ citizens about the dangers posed by migrant
others and then ‘reassure’ them about state compe-
tencies. Such acts of literally inter-national physical
relocation, exercised unevenly across space as well as
populations, are the conditions of possibility for a
differently spatialised but no less tangible social
control project which encourages the disappearance
of immigrant bodies from the public sphere. Specifi-
cally, we propose that programs like 287(g), which
bring the threat of detention and deportation into the
most intimate recesses of immigrant life by virtue of
the possibility of contact with non-federal police,
threaten to scatter immigrant subjects into territorially
present, yet increasingly invisibilised, ‘shadow popu-
lations’. We take this phrase from US legislative
debates in the late 1970 and early 1980s leading up to
the 1986 legalisation bill, in order to draw attention to
immigrant communities who are increasingly invis-
ible on the social landscape, and who as such are
forced to engage in increasingly tenuous underground
means of social reproduction in a society in which
they nonetheless labour openly. We also use this
phrase to signal the contradictions of state immigra-
tion policy. In its initial usage, for example, ‘shadow
populations’ was used by lawmakers to underscore
how it was that ham-fisted state policies designed to
deport undocumented migrants had created large
communities of individuals who were literally unde-
portable by virtue of their location ‘outside the law’
(see critical commentary in Coleman 2008). Now we
would point out that despite being driven in important
ways by the post-9/11 devolution of immigration
authority to non-federal police, the production of
‘shadow populations’ contradicts the state’s interest in
making immigrant populations legible to practitioners
of statecraft, which in many ways is the overarching
aim of the homeland security state.
We conclude by referencing briefly Butler and Spi-
vak’s Who sings the nation state (2007), a provocative
examination of the politics of nation and state in the
post-9/11 period. In the book Butler specifically sug-
gests that deportation is no longer adequately concep-
tualised only as a mode of expulsion qua removal
from one state space to another. As she explains, the
nation-state as a ‘political formation . . . requires peri-
odic expulsion and dispossession of its national
minorities in order to gain a legitimating ground for
itself’ but that the expulsion per se does not require
extra-territorial removal (2007, 33). As Butler explains
it, expulsion is a highly differential process of informal
territorial inclusion coupled with formal legal exclu-
Detention, deportation, devolution and immigrant incapacitation 235
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011© 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
sion, or a mode of social and juridical containment
‘within the polis as its interiorised outside’ (2007, 16).
This process of juridical and social deportation
without formal territorial expulsion is, we argue,
exactly the strategic spatial logic behind the post-9/11
amplification of detention and deportation policy in
the US via delegatory programs like 287(g).
Note
1 General federal data on deportations do show that officially
convicted individuals, although increasing in share, make up a
minority of total so-called ‘enforcement and removal opera-
tions’ cases. For example, over the October 2007–September
2010 period, in excess of 60% of removal cases were classified
as non-criminal (919 780 of 1 497 320 cases). Moreover, of the
total number of criminal conviction cases, a slim percentage of
individuals were deported following serious so-called ‘level 1’
offenses. Using disaggregated 2010 data, only 16% of total
removal cases followed from successful conviction for ‘aggra-
vated felony’ charges (cf Coleman 2007 on aggravated felonies
as including less serious, non-felonious, as well as non-violent
offenses); an additional 22% and 12% respectively were for
‘level 2’ charges (misdemeanors) and ‘level 3’ charges (minor
infractions) (data from ICE 2010b).
References
Andreas P 2010 Border games: policing the US–Mexico divide
Cornell University Press, Ithaca
Balderrama F E and Rodríguez R 2006 Decade of betrayal:
Mexican repatriation in the 1930s University of New Mexico
Press, Albuquerque
Bigo D 2002 Security and immigration: toward a critique of the
governmentality of unease Alternatives 27 63–92
Bigo D 2007 Detention of foreigners, states of exception and the
social practices of control of the Banopticon in Rajaram P K
and Grundy-Warr C eds Borderscapes: hidden geographies
and politics at territory’s edge University of Minnesota Press,
Minneapolis
Butler J and Spivak G C 2007 Who sings the nation state? Seagull
Books, Calcutta
Capps R, Rosenblum M R, Rodríguez C and Chishti M A 2011
Delegation and divergence: a study of 287(g) state and local
immigration enforcement Migration Policy Institute, Washing-
ton DC
Chapin J 1998 Closing America’s ’back door’ GLQ: A Journal of
Lesbian and Gay Studies 4 403–22
Coleman M 2007 Immigration geopolitics beyond the
Mexico–US border Antipode 38 54–76
Coleman M 2008 Between public and foreign policy: US immi-
gration law reform and the undocumented migrant Urban
Geography 29 4–28
Coleman M 2009 What counts as the politics and practice of
security, and where? Devolution and immigrant insecurity after
9/11 Annals of the Association of American Geographers 99
904–13
Coleman M forthcoming The ‘local’ migration state: the site-
specific devolution of immigration enforcement in the US
South Law & Policy
Dauvergne C 2008 Making people illegal: what globalization
means for migration and law Cambridge University Press,
Cambridge
De Genova N P 2002 Migrant ‘illegality’ and deportability in
everyday life Annual Review of Anthropology 31 419–47
de Sousa Santos B 1987 Law: a map of misreading, toward a
postmodern conception of law Journal of Law and Society 14
279–302
DHS 2010 2009 Yearbook of immigration statistics Department
of Homeland Security, Washington DC
Doty R L 2003 Anti-immigrantism in western democracies: state-
craft, desire, and the politics of exclusion Routledge, London
Dunn T J 2010 Blockading the border University of Texas Press,
Austin
Elden S 2009 Terror and territory University of Minnesota Press,
Minneapolis
Eschbach K, Hagan J and Rodríguez N 2003 Deaths during
undocumented migration: trends and policy implications in
the new era of homeland security In Defense of the Alien 26
37–52
Fink L 2003 The Maya of Morganton: work and community in
the Nuevo New South University of North Carolina Press,
Durham
Foucault M 1979 The history of sexuality: an introduction
Vintage Books, New York
Foucault M 2007 Security, territory, population: lectures at the
Collège de France 1977–78 Palgrave Macmillan, New York
GAO 2009a Immigration enforcement: better controls needed
over program authorizing state and local enforcement of
federal immigration laws US Government Accountability
Office, Washington DC
GAO 2009b Immigration enforcement: controls over program
authorizing state and local enforcement of federal laws should
be strengthened US Government Accountability Office, Wash-
ington DC
Gilbert L 2009 Immigration as local politics: re-bordering immi-
gration and multiculturalism through deterrence and incapaci-
tation International Journal of Urban and Regional Research 33
26–42
Gill H and Nguyen M T 2010 The 287(g) program: the costs and
consequences of local immigration enforcement in North
Carolina communities University of North Carolina Latino
Migration Project, Chapel Hill
Gilmore R W 2007 Golden Gulag: prisons, surplus, and opposi-
tion in globalizing California University of California Press,
Berkeley
Gladstein H, Lai A, Wagner J and Wishnie M J 2005 Blurring
the lines: a profile of state and local police enforcement of
immigration law using the National Crime Information Center
Database, 2002–2004 Migration Policy Institute, Washington
DC
Gordon C 1991 Governmental rationality in Burchell G, Gordon
C and Miller P eds The Foucault effect: studies in governmen-
tality University of Chicago Press, Chicago
Herbert S 2006 Citizens, cops and power University of Chicago
Press, Chicago
236 Detention, deportation, devolution and immigrant incapacitation
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011 © 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
Hiemstra N 2010 Immigrant ‘illegality’ as neoliberal governmen-
tality in Leadville, Colorado Antipode 42 74–102
ICE 2010a 287(g)-Identified aliens for removal Department of
Homeland Security, Washington DC (www.ice.gov/doclib/
foia/reports/287g-masterstats2010oct31.pdf) Accessed 7 May
2011
ICE 2010b ICE total removals as of December 7 2010 Depart-
ment of Homeland Security, Washington DC (www.ice.gov/
doclib/about/offices/ero/pdf/ero-removals.pdf) Accessed 7
May 2011
ICE 2011 Secure communities: IDENT/IAFIS interoperability
monthly statistics through February 28 2011 Department of
Homeland Security, Washington DC (http://ndlon.org/pdf/
scommfeb/nationwidestats20112.pdf) Accessed 7 May 2011
INS 1997 1996 Yearbook of immigration statistics Immigration
and Naturalization Service, Washington DC
Kofman E 2002 Contemporary European migrations, civic strati-
fication and citizenship Political Geography 21 1035–54
Malkin M 2002 Invasion: how America still welcomes terrorists,
criminals, and other foreign menaces to our shores Regency,
Washington DC
Maril R L 2004 Patrolling chaos: the U.S. border patrol in deep
South Texas Texas Tech University Press, Lubbock, TX
Martin L forthcoming The geopolitics of vulnerability: children’s
legal subjectivity, immigrant family detention, and US immigra-
tion law and enforcement policy Gender Place and Culture 18
McGregor J 2008 Abject spaces, transnational calculations: Zim-
babweans in Britain navigating work, class and the law Trans-
actions of the Institute of British Geographers 33 466–82
Miller T 2005 Blurring the boundaries between immigration and
crime control after September 11th Boston College Third
World Law Journal 25 81–124
Mohl R A 2003 Globalization, Latinization, and the nuevo New
South Journal of American Ethnic History 22 31–66
Morris L 2001 The ambiguous terrain of rights: civic stratification
in Italy’s emergent immigration regime International Journal of
Urban and Regional Research 25 497–516
Nagel C R 2002 Geopolitics by another name: immigration and
the politics of assimilation Political Geography 21 971–87
Nagel C R 2009 Rethinking geographies of assimilation Profes-
sional Geographer 61 400–7
Nevins J 2010 Operation gatekeeper and beyond Routledge,
New York
Nevins J and Aizeki M 2008 Dying to live: a story of US immi-
gration in an age of global apartheid Open Media/City Lights
Books, San Francisco
Nuñez G and Heyman J M 2007 Entrapment processes and
immigrant communities in a time of heightened border vigi-
lance Human Organization 66 354–65
Passel J A and Cohn D V 2010 US unauthorized immigration
flows are down sharply since mid-decade Pew Hispanic
Center, Washington DC
Puar J K 2007 Terrorist assemblages: homonationalism in queer
times Duke University Press, Durham
Purcell M and Nevins J 2005 Pushing the boundary: state restruc-
turing, state theory, and the case of US–Mexico border
enforcement in the 1990s Political Geography 24 211–35
Rosas G 2007 The fragile ends of war: forging the United States–
Mexico border and borderlands consciousness Social Text 25
81–102
Seghetti L M, Viña S R and Ester K 2006 Enforcing immigration
law: the role of the state and local enforcement CRS Report for
Congress Congressional Research Service, Library of Congress,
Washington DC
Smith H A and Furuseth O J 2006 Latinos in the new South:
transformations of place Ashgate, Burlington
Stuesse A C 2010 Challenging the border patrol, human rights
and persistent inequalities: an ethnography of struggle in South
Texas Latino Studies 8 23–47
Stumpf J 2006 The crimmigration crisis: immigrants, crime
and sovereign power American University Law Review 56
367–419
Suro R and Singer A 2002 Latino growth in metropolitan America:
changing patterns, new locations Center on Urban and Metro-
politan Policy and the Pew Hispanic Center, Washington DC
Tichenor D 2002 Dividing lines: the politics of immigration
control in America Princeton University Press, Princeton
Varsanyi M W 2008 Rescaling the ‘alien,’ rescaling personhood:
neoliberalism, immigration, and the state Annals of the Asso-
ciation of American Geographers 98 877–96
Walters W 2002 Deportation, expulsion, and the international
police of aliens Citizenship Studies 6 265–92
Walters W 2006 Border/control European Journal of Social
Theory 9 187–203
Weissman D and Headen R C 2009 The policies and politics of
local immigration enforcement laws: 287(g) program in North
Carolina ACLU and the UNC Immigration and Human Rights
Policy Clinic, Raleigh
Wells M J 2004 The grassroots reconfiguration of US immigration
policy International Migration Review 38 1308–47
Winders J 2005 Changing politics of race and region: Latino
migration to the US South Progress in Human Geography 29
683–99
Winders J 2007 Bringing back the (B)order: post-9/11 politics of
immigration, borders, and belonging in the contemporary US
South Antipode 39 920–42
Zolberg A 2006 A nation by design Harvard University Press,
Cambridge
Detention, deportation, devolution and immigrant incapacitation 237
The Geographical Journal Vol. 177 No. 3, pp. 228–237, 2011© 2011 The Authors. The Geographical Journal © 2011 Royal Geographical Society
(with the Institute of British Geographers)
