Special Section: Theorizing the Liberty-Security Relation:

Sovereignty, Liberalism and Exceptionalism

Foucault in Guantánamo: Towards an

Archaeology of the Exception

ANDREW W. NEAL*
School of Politics, International Relations and Philosophy,
Keele University, UK

This article offers a critique of the discourse that has emerged around
the problem of ‘the exception’. ‘The exception’ is shorthand for
the problem of certain events and situations, such as 9/11, being
designated as ‘exceptional’ in order to legitimate exceptional policies,
practices, executive measures and laws. The article terms this
discourse and practice ‘exceptionalism’. It begins by identifying
problems in the treatment of the exception and exceptionalism in the
work of Carl Schmitt, Giorgio Agamben and securitization theorist
Ole Wæver. A different theoretical approach to the problem of the
exception is then offered, drawing upon Michel Foucault’s early
work, The Archaeology of Knowledge. The narrative focuses on the detail
of Foucault’s ‘archaeological’ methodology, relating it to specific
problems in the political-theoretical discourse of exceptionalism. The
reasons for an emphasis on ‘archaeology’ rather than Foucault’s later
‘genealogical’ slant are explained. The article concludes by arguing
that ‘archaeology’ – conceived as a neo-Kantian mode of critique that
is discursive and historicist – is a more appropriate and less problem-
atic method for engaging with the problem of the exception.

Keywords Exception • Foucault • Schmitt • Agamben •

securitization

T
HE AIM OF THIS ARTICLE is to sketch a different response to the
problem of exceptionalism. Exceptionalism encompasses an array of
illiberal policies and practices that are legitimated through claims
about necessary exceptions to the norm. The Guantánamo Bay detention
camp is the most exemplary empirical site of exceptionalism. As such, it has
come to serve as a figure for a far-reaching and highly dispersed array of
exceptional practices, ranging from the securitization of immigration and
asylum1 to the ‘extraordinary rendition’ of ‘terrorist suspects’ to third coun-
tries widely believed to practice torture.2 In this article, I am particularly

© 2006 PRIO, www.prio.no

SAGE Publications, http://sdi.sagepub.com
Vol. 37(1): 31–46, DOI: 10.1177/0967010606064135
32 Security Dialogue vol. 37, no. 1, March 2006

concerned with the legitimating and critical possibilities of the exceptional-

ism discourse. The legitimation of exceptional practices may not always
succeed, and indeed the concept of exception can be deployed critically to
de-legitimate such policies and practices. While discourses of exceptional-
ism, both official and critical, have re-emerged as sharply relevant in the last
few years, I believe their political status is far from settled. The origins of the
concept of the exception are themselves a source of disquiet: the contest over
the resurrection of the sometime Nazi jurist Carl Schmitt rages on. Schmitt’s
exception, as I will explain, is pointed but suspect. Similarly, the work of
Giorgio Agamben and securitization theory are key contemporary theoreti-
cal approaches that bear a strong relationship to Schmitt and exceptionalism,
neither of which escape certain problematic implications. My suspicion
of this triumvirate of Schmitt, Agamben and securitization theory is the
starting point for this paper. After outlining the key stakes involved in this
encounter, I will sketch a critique and a different approach to the problem of
the exception through the early ‘archaeological’ work of Michel Foucault.

Schmitt, Agamben and Securitization Theory

I am profoundly suspicious of Schmitt’s invocation of an ‘exceptional event’

that brings forth its own political and existential imperatives. I think
Schmitt’s use of this idea in his 1922 Political Theology is actually a discursive
strategy to present a particular politics of naming, interpreting and repre-
senting ‘exceptional events’ as logical or necessary rather than as a specific
political ethic. Schmitt uses the hypothetical and somewhat metaphysical
problem of ‘exceptional’ contingency in order to eulogize the primacy of
exceptional sovereign power and construct an extreme nation-statist ethic
(see Schmitt, 1985). Giorgio Agamben proffers a less political and more
philosophical totalization of exceptional sovereign decision by erecting it as
a central pillar of the whole paradigm of Western politics, projecting the
problem backwards and transcendentalizing it in a rather vain way (see
Agamben, 1998, 2005).
In light of the politico-philosophical slipperiness of Schmitt and the totaliz-
ing projections of Agamben, there is much of merit in the ‘securitization
theory’ of Ole Wæver et al. Instead of treating the exception as a ‘real possi-
bility’ (Schmitt, 1996: 45) or a ‘great historico-transcendental destiny’,3 theo-

1
See, for example, the work of an ESRC-funded project by Jef Huysmans, Thomas Diez and Pat Noxolo on
‘Migration, Democracy and Security (Midas)’ at http://www.midas.bham.ac.uk.
2
See, for example, discussion of the report of Swiss Senator Dick Marty into CIA activities in Europe in The
Guardian (2005).
3
A term I appropriate from Michel Foucault (2002a: 231).
Andrew W. Neal Foucault in Guantánamo 33

rists of securitization seek to treat security as a process. In this process, issues

are ‘securitized’ by security elites and state agents through ‘speech acts’,
whereby those elites and agents attempt to convince their audience that a
particular issue is in fact a security problem, and thus bring about certain
forms of political and social mobilization. The advantage of this innovation
is that it treats ‘exceptional’ security situations as a nominalist rather than an
essentialist problem – that is, the politics of exceptional situations have more
to do with discourses and sociopolitical processes than with any funda-
mental metaphysical problem. To some extent, securitization theory defuses
Schmitt’s attempt to seamlessly link the idea of exceptional events with
exceptional nation-statist politics by stressing that security is a discourse that
can be deployed and manipulated through strategies of securitization. As
such, there are no events that in themselves dictate particular political
responses; rather, any events or issues can be turned into security issues by
particular discursive strategies.
Nevertheless, in terms of responding adequately to the problem of the
politics of the exception, securitization theory fails to live up to its initial
promise. Despite being a nominalist or constructivist approach, securitiza-
tion theory ultimately reifies the state in which ‘securitization’ occurs: it is
usually (but not exclusively) the state that securitizes. In itself, this is not a
major concern, perhaps merely a symptom of the ceaseless debate about the
coming or going of the state in the disciplines of politics and international
relations. What is concerning is that securitization theory waters down its
constructivism by treating the discourse, field or tradition of security as a
structural, sociological and linguistic unity with discernable limits and
thresholds that are crossed according to certain formal rules. Despite its
constructed nature, this image of security still entails a characteristic and
very Schmittian account of urgency, extraordinary circumstances and excep-
tional measures. Thus, it is possible to recognize when these conditions are
moved into and out of. In this way, securitization theory still treats security
as a special category: security is still distinguished from politics, and the
exception is still distinguished from the norm. Despite seeking to move away
from practices of securitization and exceptionalism, securitization theory in
fact perpetuates their ‘exceptional’ structure. Consequently, the residual
statism of securitization theory is more than a simple empirical position in
the ongoing ‘whither the state?’ debate. For Wæver, whatever the discursive
aspects of security problems, the field is still characterized by an urgency
that brings forth certain imperatives, tied to exceptional state prerogatives
and a mass polity. Wæver (1995: 51) writes that the security field is charac-
terized by:
urgency; state power claiming the legitimate use of extraordinary means; a threat
seen as potentially undercutting sovereignty, thereby preventing the political ‘we’ from
dealing with any other questions.
34 Security Dialogue vol. 37, no. 1, March 2006

Wæver’s use of discourse is too ‘thin’; it is still tied to formal qualities, con-
crete structures and a certain politics. It still betrays the dubious aspects of
Schmitt.
Securitization theory argues that security issues are constructed according
to a dialectic between securitizing state agents and an accepting or rejecting
public audience, mediated by a stock discourse of security characteristics
such as urgency and exceptionality. This does not in itself reify a single
political model of authoritarian nation-statist sovereignty, because both the
securitizing agent and the political ‘we’ may be constituted and constitute
each other in different ways. Indeed, a sophisticated reading of Schmitt
should be able to detect these dialectical and even discursive elements in his
political theory too. The problem is that by ceding imperative and mobiliz-
ing qualities to what Wæver considers the basic figurative elements of
security problems – urgency, extraordinary circumstances and exceptional
measures – certain structural, institutional and ethico-political implications
are already implicit. The security discourse, as a both a figurative and a
historical tradition, cannot be divorced from the figurative and historical
tradition of the strong nation-state. Wæver knows this already, of course, but
argues that by understanding the discourse of security and the processes
of securitization, one can then adopt that knowledge for de-securitizing
strategies and ends. The problem here is that there is no way of guarantee-
ing that knowledge of securitization will be used for ‘good’ political pur-
poses; that it is even possible to ascertain or agree on what ‘good’ political
purposes are; and, indeed, that better knowledge of securitization will
not simply be recolonized by the state to improve its own securitization
strategies. Of course, this criticism is simply an old-fashioned Weberian
argument. What I would add is that by reifying the basic figurative elements
of the security discourse, certain political prerogatives are reified as well,
and this is to fall straight into Schmitt’s trap. The exceptional event or situa-
tion dictates an exceptional response. It is not simply a dubious ethic, but a
dubious metaphysic dictating a dubious ethic. It is no good being construc-
tivist about the state, constructivist about ‘the people’, and constructivist
about ‘security’, if one is not sufficiently constructivist about such conditions
as ‘urgency’, ‘necessity’ and ‘exception’. Having said that, more construc-
tivism is not what is needed. Rather, the question of discourse needs to be
taken more seriously.
These discourses of exceptionalism in Schmitt, Agamben and securitization
theory privilege a sovereign centre. The problem of exceptionalism is suc-
cessively reduced to, borrowing terms from Foucault (2002a: 226), a ‘single
system of differences’ and to ‘absolute axes of reference’: the friend–enemy
distinction, the norm and the exception, bare life and political life, politics
and security. Each approach remains dualistic to a lesser or greater degree,
pursuing clear boundary lines that delimit unities, identities, categories,
Andrew W. Neal Foucault in Guantánamo 35

jurisdictions and limits. If these limits do not attempt to mark off the domain
of sovereign power, then at least the decision over the shape and location of
these limits is accorded to sovereign power. Although each approach has
sought to deal with the political implications of the underlying philosophi-
cal and metaphysical problem of the exception, the problem of exception-
alism has been reinforced rather than overcome. While exceptionalism
characterizes an array of contemporary events and practices that seem some-
how new and disturbing, each successive politico-theoretical approach has
sought to reduce the newness and particularity of these events and practices
to an underlying theoretical continuity or return. Each approach explains
exceptionalism as the expression of a continuous underlying philosophical
and political structure, prerogative, imperative or necessity.

Foucault’s Archaeological Method

I began this article by invoking an array of contemporary claims and prac-

tices that seem somehow new. I called these ‘exceptionalism’. These claims
and practices not only seem new, but they themselves invoke the new. It is
claimed that an exceptional event has brought about exceptional times, and
that exceptional times require exceptional measures. On 11 September 2001,
‘everything changed’. With the emergence of this constellation of exceptional
new circumstances and exceptional new responses, we seem to be faced with
a great break in history, the beginning of something new, a sudden irruption.
Except that the politics of the exception seems depressingly familiar and pre-
dictable. Did we ever doubt, on that fateful day, that the horrors unfolding
before our eyes would not be met with an American reign/rain of fire? The
terrifying uncertainty of the exceptional event and its interpretation has
proved to be a chimera. The meaning and interpretation of the event are
now thoroughly incorporated into a regime of legitimation for exceptional
sovereign practices. Perhaps the processes and prerogatives that named and
interpreted the event had a hold on it before it even happened, awaiting its
capture with well-established discourses of threat, urgency, emergency and
exception. The ‘new’ appears to have only reaffirmed the ‘same’: the perma-
nence of the prerogatives of exceptional sovereign power.
In my analysis so far, these two figures of the new and the same have
supported each other. The new, the rupture, the event, the exception, has
been used to reveal the continuing sameness of sovereign power. As such,
Schmitt uses the contingency of the event, the limit inherent in every schema,
the ‘real possibility’ (Schmitt, 1996: 45) of the exception, to reaffirm a ‘single
signified’ (Foucault, 2002a: 133): the metaphysical and political necessity of
exceptional nation-statist sovereign power. Similarly, to invoke Foucault
36 Security Dialogue vol. 37, no. 1, March 2006

again, Agamben finds in the problem of sovereign exceptionalism ‘the great

historico-transcendental destiny of the Occident’ (Foucault, 2002a: 231). And
securitization theory affirms, with its entrenched security discourse, the
continued existence of the nominalist prerogatives and practices of sovereign
power under constructed but ultimately formal conditions of ‘exceptional’
urgency and threat.
Foucault’s Archaeology of Knowledge inspires a riposte to these theorists: it is
the theoretical approaches themselves that perpetuate the ‘sovereign’ struc-
tures of discourses. Discourses of exceptionalism reify a certain vision of
sovereignty: accounts of decision under conditions of contingency; naming,
interpretation and representation through authoritative processes; the author-
ization of authority according to certain conditions of necessity; resolution at
the point of contradiction; sovereign judgement at the limit/threshold/border
of the normal and the exceptional. It is as if the discourse of exceptionalism
contains a ‘hidden discourse’, a return of the ‘same’, a ‘transcendental act that
gives them origin’ (Foucault, 2002a: 226). Foucault directly opposes all of this.
Instead, he offers the method of archaeology, which
is trying to deploy a dispersion that can never be reduced to a single system of differ-
ences, a scattering that is not related to absolute axes of reference; it is trying to operate
a decentring that leaves no privilege to any centre. (Foucault, 2002a: 226)

Foucault rejects the dualistic or dialectical opposition of rupture and conti-

nuity, the new and the same, the norm and the exception. Archaeology is not
concerned with the opposition of the original and the banal, the regular and
the irregular.
This is an invitation to adopt an archaeological approach in which all state-
ments (e.g. claims about exceptional events, exceptional times and excep-
tional measures) are considered not in terms of their sudden irruption, but in
terms of their historical conditions of possibility. These conditions, which are
discursive and not formal, apply to the exceptional statement and the normal
statement alike.
One must seek to describe both what appears new and what appears the
same according to dispersed formations of statement, event and archive, the
historical and spatial limits of which can only be questioned, not formally
deduced. Both newness and sameness need to be described, in Foucault’s
terms, at the level of positivity, of exteriority. As such, transformations and
recurrences should be described at the level of appearance rather than with
recourse to the idea of transcendental rupture, metaphysical contingency or
objective necessity. To avoid the pitfalls of other approaches to exceptional-
ism, the contemporary political problem should be described neither as a
full-spectrum break in history in which ‘everything changed’, nor as a
confirmation of the formal sameness of sovereignty, security and modern
politics. As such,
Andrew W. Neal Foucault in Guantánamo 37

the idea of a single break suddenly, at a given moment, dividing all discursive forma-
tions, interrupting them in a single moment and reconstituting them in accordance with
the same rules – such an idea cannot be sustained. (Foucault, 2002a: 193)

But, in the same vein,

[archaeology] considers that the same, the repetitive, and the uninterrupted are no less
problematic than the ruptures; for archaeology, the identical and the continuous are not
what must be found at the end of the analysis. (Foucault, 2002a: 192)

What Foucault offers is a form of analysis that combines a description of

historically situated discursive formations and assemblages of power with
an understanding of how those dispersed configurations are transformed in
their contemporary expression and exercise. One is not looking for a simple
succession of statements and practices, for that would be to reproduce a
metaphor of progress or development, again reducing the new to a form of
the continuous that is more an Occidental philosophy than a description of
transformations in their particularity. Statements and practices add and
relate to each other in a variety of not necessarily coherent or predictable
ways. As Foucault makes clear,
[statements] each have their own way of merging together, annulling one another,
excluding one another, complementing one another, forming groups that are in varying
degrees indissociable and endowed with unique properties. (Foucault, 2002a: 140)

With regard to the question of recurrence, therefore, Foucault explains that

every statement involves a field of antecedent elements in relation to which it is situ-
ated, but which it is able to reorganize and redistribute according to new relations. It
constitutes its own past, defines, in what precedes it, its own filiation, redefines what
makes it possible or necessary, excludes what cannot be compatible with it. (Foucault,
2002a: 140)

The relation of the novel new synthesis to the dispersed archive of the
already-said is not one of direct historical determination. To chart resem-
blances and recurrences archaeologically is not to invoke paradigm, epoch,
progress or historical spirit. Rather, it is to describe the new in terms of a
complex and specific relation of temporally and spatially dispersed regu-
larities with plural, not singular, horizons. The archive is not a fixed deter-
minant, but an amorphous and fungible body of historical conditions of
possibility for contemporary statements, events and transformations.
If one considers that all statements belong to discursive formations, and all
discursive formations consist of a dispersed and problematic array of objects,
subject positions, statements, concepts and strategies (Foucault, 2002a: 35–
39), then the figures of both the ‘new’ and the ‘same’ lose their transcendental
qualities. Instead of these overdetermined figures, archaeology describes
discursive formations in terms of appearances alone. It seeks to describe
the historical, rather than formal or ideal, conditions of possibility for the
38 Security Dialogue vol. 37, no. 1, March 2006

appearance of discursive practices and statements. It seeks to describe the

specific relations and regularities of object, statement, concept and strategy
that constitute the archive: ‘the law of what can be said, the system that
governs the appearance of statements as unique events. . . . It is the general
system of the formation and transformation of statements’ (Foucault, 2002a:
145–146).
This approach is therefore quite the opposite of those I have discussed so
far. Archaeology would not seek to describe what I have called ‘exceptional-
ism’ in terms of either uniqueness or sameness. It would not seek to find
beneath the appearance of exceptional statements and practices the return of
the formal conditions of sovereignty, security or the metaphysical possibility
of ‘the exception’. It would not describe exceptionalism in terms of trans-
cendental problem, nor in terms of transcendental act. Nor would it describe
the present appearance of particular statements and practices in terms of
historical epoch, ideal form, eternal truths or heavy traditions. It would not
give exceptionalism ‘a special temporal status’ (Foucault, 2002a: 23). Rather
than a single analytic horizon of norm and exception, politics and law, or
politics and security, archaeology calls for a pluralization of horizons:
The horizon of archaeology, therefore, is not a science, a rationality, a mentality, a
culture; it is a tangle of interpositivities whose limits and points of intersection cannot
be fixed in a single operation. Archaeology is a comparative analysis that is not intended
to reduce the diversity of discourses, and to outline the unity that must totalize
them, but is intended to divide up their diversity into different figures. Archaeological
comparison does not have a unifying, but a diversifying effect. (Foucault, 2002a: 177)

Archaeology and Genealogy

In the work of Foucault, a distinction can be made between the method of

‘archaeology’, which came first, and ‘genealogy’, which came later, but it is
important to note that the difference is only one of emphasis. As Dreyfus &
Rabinow (1982: 104) stress, ‘There is no pre- and post-archaeology or genealogy in
Foucault. However, the weighting and conception of these approaches has
changed during the development of his work.’ Archaeology is more con-
cerned with discourse and structure, while the two key departures of gene-
alogy are a greater emphasis on, first, historical power relations and, second,
technologies and practices of power that are focused on the body in particular.
These genealogical departures are exemplified by the essay ‘Nietzsche,
Genealogy, History’ (Foucault, 1999) and the book Discipline and Punish
(Foucault, 1991) respectively. In this article, I place the emphasis on archae-
ology for four reasons.
First, the terminology of archaeology is much closer to that which is used
Andrew W. Neal Foucault in Guantánamo 39

by the theorists of concern here. Schmitt, Agamben and the proponents of

securitization theory all construct their arguments around a structural rela-
tionship between objects, subjects and concepts. Agamben reads Foucault
structurally rather than genealogically. Similarly, the terminology of securi-
tization theory is very close to the structural and discursive language used in
The Archaeology of Knowledge. For all three, the ‘exception’ is constituted as an
‘object’ of special discourses, imperatives and practices. Principles, dialectics,
processes and practices of sovereign subjectivity play a decisive constitutive
role in the naming and designation of ‘exceptions’. The concepts of threat,
danger, necessity and security are placed at the centre of the discourse of
exceptionalism, invoking a legitimacy that is supposedly deeper and more
profound than that of the law and the ‘norm’. The language of archaeology
closely correlates with the categories that are in play in the discourse of
exceptionalism, but it problematizes them, critiques them, expands them and
disperses them. Archaeology is thus well placed to tackle the dialectical play
of categories that permeate the discourse of exceptionalism in all its forms.
Second, the emphasis of genealogy in understanding the emergence of
historical developments as ‘the hazardous play of dominations’, a ‘stage of
forces’ and ‘the struggle these forces wage against each other’ (Foucault,
1991: 83) does not play well as a critique of exceptionalism, because excep-
tionalism is already understood through the modality of violence and force.
For example, in the popular debate about the balance between liberty and
security under exceptional conditions, ‘liberty’ is subject to a struggle
between those who would defend it from the state and those who would
defend it from ‘terror’. To offer the classic Nietzschean genealogy of liberty
as a dominating ‘invention of the ruling classes’ (Foucault, 1991: 78, citing
Nietzsche’s The Wanderer and His Shadow) is not helpful for understanding
how liberty has come to play a polyvalent and ambiguous role in the dis-
course of exceptionalism, contradictorily standing for individual freedom,
on the one hand, and illiberal security practices in the name of ‘freedom’, on
the other. It is too easy to describe exceptionalism as the domination of the
‘ruling classes’, manifested as the sovereign decision on the exception
(Schmitt), the sovereign decision on bare life and political life (Agamben), or
the elites declaring issues to be security problems (securitization theory). It
would be simple to describe exceptionalism as a special field constructed,
bounded and continually reinforced by violent practices of inclusion and
exclusion, but this is what is already offered by Schmitt, Agamben and
securitization theory. Rather than simply understanding the politics of dis-
courses as ‘the violent or surreptitious appropriation of a system of rules’
(Foucault, 1991: 86) or ‘the hazardous play of dominations’ (Foucault, 1991:
83), archaeology places more emphasis on relations between objects, sub-
jects, concepts and strategies, the conditions under which each of these
categories is constituted, and the way they interact and supply authority to
40 Security Dialogue vol. 37, no. 1, March 2006

each other. In this sense, in contrast to the Nietzschean slant of genealogy,

archaeology is much more of a ‘critical’ project in the traditional Kantian
sense of ‘seeking the conditions of possibility and limitations of rational
analysis’ (Dreyfus & Rabinow, 1982: 122). The key departure from Kant is the
fact that ‘Foucault accepts this project but rejects the attempts to find a uni-
versal grounding in either thought or Being’ (Dreyfus & Rabinow, 1982: 122);
that is, Foucault seeks to describe limits and conditions of possibility, but of
the historical and discursive, rather than rational or ontological, kind.
Third, the genealogical idea of history as the imposition of violence and
domination is a deeply ambiguous argument that, I want to argue, Foucault
ultimately steps back from. In ‘Nietzsche, Genealogy, History’, Foucault
posits the idea of history as a relation of war and battle. As he writes, ‘the
successes of history belong to those who are capable of seizing these rules’
(Foucault, 1991: 86). The aim of ‘seizing the rules’ begs the very question of
exceptionalism. Is the way to tackle the problem merely to seize the means
of declaring the exception?
In an interview conducted in the midst of his ‘genealogical phase’ (if one
can speak of such a thing), Foucault is asked ‘Does the military model seem
to you . . . to be the best one for describing power: is war here simply
metaphorical model, or is it the literal, regular, everyday mode of operation
of power?’ In response, Foucault posits a number of difficult questions about
the ‘war model’, commenting that, ‘a whole range of problems emerge
here’ (Foucault, 1980: 123). In Society Must Be Defended, a series of lectures
delivered around the same time, Foucault thoroughly explores and tests the
hypothesis of this ‘war model’, asking ‘if we have to think of power in terms
of relations of force, do we therefore have to interpret it in terms of the
general form of war? Can war serve as an analyzer of power relations?’
(Foucault, 2002b: 266). It is my argument that in these lectures Foucault ulti-
mately steps back from the ‘war model’ of power relations to revert to a more
critical, ‘archaeological’ position.4 Instead of straightforwardly pursuing
‘war’ as an ‘analyzer of power relations’, Foucault critically examines the rise
and fall of the ‘war model’ of power relations in historical and political dis-
course, describing its conditions of emergence and the multiple inversions,
transpositions and syntheses it undergoes. Thus, Society Must Be Defended is
not genealogical in the sense described in ‘Nietzsche, Genealogy, History’,
but is more reflective and ‘archaeological’ in the way it critically describes a
discursive terrain upon which multiple historical and political claims are
possible. The possibility of these multiple claims and inversions stems
not simply from ‘the struggle of forces’, but through discursive and political
synthesis, inversion, co-optation and disciplinarization.

4
I make this argument in much greater detail in Neal (forthcoming); see also Neal (2004).
Andrew W. Neal Foucault in Guantánamo 41

Fourth, exceptionalism is a much wider problem than can be explained

through a genealogical analysis of technologies of power alone, as found in
Discipline and Punish, for example. Beyond the technologies of exceptional
practices themselves, such as torture and detention without trial, exception-
alism raises profound questions about the workings of political principles
and discourses. How are key political ideas such as liberty and security being
used in a discourse of exceptionalism that works to legitimate exceptional
practices? How is the problematique of liberty implicated in exceptionalism?
How do claims about imperatives and necessities that are more authentic
than law find critical purchase? The discourse and practice of exceptionalism
cannot and should not be separated into ‘levels of analysis’, but my particu-
lar focus on the widespread failure to find critical purchase on the problem
of the exception does call for a slant in emphasis towards archaeology. It
must be made clear, however, that archaeology is not concerned exclusively
with statements, and genealogy is not exclusively concerned with practices.
Emphasising archaeology does not exclude genealogy; they are comple-
mentary approaches that largely overlap. As Foucault (2002b: 10–11) puts it,
‘archaeology is the method specific to the analysis of local discursivities, and
genealogy is the tactic which, once it has described these local discursivities,
brings into play the desubjugated knowledges that have been released from
them’. The critical focus of this project calls for the analysis of discursivities
more than it calls for genealogical tactics.
As a final cautionary point, a focus on discourse does not mean an exclu-
sive focus on language. The argument presented here is that the problem of
the exception, and the way it has been treated critically, raises questions
about the role and constitution of subject positions, objects, statements, con-
cepts and strategies (although not exclusively in the military sense), and as
such the problem is immediately wider than that of discourse understood
merely as language. Discourse, as Foucault understands it, is a problem of
formations of language and concepts, but also of extra-discursive structures,
practices and positions of authority. As he makes clear:
Of course, discourses are composed of signs; but what they do is more than use these
signs to designate things. It is this more that renders them irreducible to the language
and to speech. It is this ‘more’ that we must reveal and describe. (Foucault, 2002a: 54)

Archaeology and Exceptionalism

Employing an archaeological mode of critique, how would one begin to

describe the dispersed ‘tangle of interpositivities’ that I have called the prob-
lem of the politics of the exception? What does Foucault’s methodology look
like? Foucault begins The Archaeology of Knowledge with an undertaking to
42 Security Dialogue vol. 37, no. 1, March 2006

‘describe the relations between statements’ in terms of what he calls ‘discur-

sive formations’ (Foucault, 2002a: 34). He seeks to describe these discursive
formations in terms of dispersed objects, statements, concepts and strategies.
Rather than seeking unity, hidden origins or the expressions of a sovereign
consciousness in these formations, he seeks to describe the diverse relations
between objects, statements, concepts and strategies. He seeks to describe
transformations in these relations. He seeks to describe correlations within
them. He seeks to describe these things in their particularity, their specificity,
their historical positivity. He does not seek an open terrain of freedom, nor
the determinations of history. It is not a question of seeking the deeper
historical meaning or expression in what was said or done. Nor is it, contrary
to many contemporary readings of Foucault, a question of how things could
have been otherwise. Rather, the task is to describe statements and events in
terms of their rarity, their historical conditions of possibility, ‘the limit that
separates them from what is not said, in the occurrence that allows them to
emerge to the exclusion of all others’ (Foucault, 2002a: 134). It is clear, then,
in Foucault’s terms, and contrary to what has been found in the theorists
being criticized here, that ‘discourse is not an ideal, timeless form’ (Foucault,
2002a: 131). Rather,
it is, from beginning to end, historical – a fragment of history, a unity and discontinuity
in history itself, posing the problem of its own limits, its divisions, its transformations,
the specific modes of its temporality rather then its sudden irruption in the midst of the
complicities of time. (Foucault, 2002a: 131)

All these methodological precautions of Foucault speak directly to the

problematique of this article. With these in mind, it must immediately be
questioned whether the terms ‘exception’ and ‘exceptionalism’ can be used
critically, analytically or even descriptively without participating in the
reinforcement and functioning of the discourse and practice of sovereign
exceptionalism. Indeed, a paradox Foucault tackled now presents itself here.
The unities of discourse have been questioned. The idea of an identifiable,
unified category of the exceptional has been thrown into suspicion. The
philosophical deduction of the formal necessity or recognizability of such a
category and the subsequent deduction of political necessities and sovereign
prerogatives have been met with mistrust. The return of the same, the reduc-
tion of the particular to the continuous, and the consistent discovery of the
transcendental under the contradiction or limit have all been problematized.
Foucault has instead argued for an understanding of discourse as dispersal
and difference. As he writes: ‘At the outset I questioned those pre-established
unities according to which one has traditionally divided up the indefinite,
repetitive, prolific domain of discourse’ (Foucault, 2002a: 79). How, then, is
it possible to continue to identify, describe and question a grouping of prac-
tices and statements that I have called ‘exceptionalism’? Or, as Foucault
Andrew W. Neal Foucault in Guantánamo 43

(2002a: 79) asks: ‘Why, then, proceed to such dubious regroupings at the very
moment when one is challenging those that once seemed the most obvious?’
Despite the need for dispersal, how and why can one persist with the
‘dubious regrouping’ of practices and statements of ‘the exception’ and
‘exceptionalism’? How can one continue to pursue these investigative con-
cerns without succumbing to the reification of special categories that charac-
terizes the discourse of exceptionalism as I have discussed it so far? In
response to these problems, Foucault (2002a: 85) spells out the way to dis-
tinguish these investigations from those that are symptomatic of the very
problem of exceptionalism in the following passage:
One is not seeking, therefore, to pass from the text to thought, from talk to silence, from
the exterior to the interior, from spatial dispersion to the pure recollection of the
moment, from superficial multiplicity to profound unity. One remains within the
dimension of discourse.

In their treatment of the empirical problem of exceptionalism, the approaches

I have looked at have played out just such a movement from ‘superficial
multiplicity to profound unity’. With particular reference to the analysis of
discourse, and thus with great relevance for the approach of securitization
theory in particular, Foucault (2002a: 133) levels the criticism that
generally speaking, the analysis of discourse . . . substitutes for the diversity of things
said a sort of great, uniform text, which has never before been articulated, and which
reveals for the first time what men ‘really meant’ not only in their words and texts, their
discourses and their writings, but also in the institutions, practices, techniques, and
objects that they produced. In relation to this implicit, sovereign, communal ‘meaning’,
statements appear in superabundant proliferation, since it is to that meaning alone that
they all refer and to it alone that they owe their truth: a plethora of signifying elements
in relation to this single ‘signified’.

Foucault urges a move in the other direction, to resist the move from the
particular to the formal, from the dispersed to the unified, from the different
to the same. Instead of moving to this ‘implicit, sovereign, communal “mean-
ing”’, Foucault insists on remaining ‘within the dimension of discourse’.
‘The enunciative domain is identical with its own surface’, he writes. ‘There
is no subtext’ (Foucault, 2002a: 135). One can still describe a certain regular-
ity of statements and practices as a discourse, but only one that is dispersed
and whose limits are thrown into question and problematized, not one that
is unified and formal. ‘Exceptionalism’ is a broken surface, a dispersed array
of appearances drawing upon an archive of the already-said. The discourse
and the archive can be described, but to search for hidden meanings, tran-
scendental destinies and metaphysical imperatives is to reify. One must
move away from the aim of establishing the critical ‘legitimacy’ or ‘right’ of
the discourse of exceptionalism. One must cease to look for the underlying
‘truth’ or sovereign continuity of the problem of exceptionalism. To critique
44 Security Dialogue vol. 37, no. 1, March 2006

is not simply to deduce the immanent coherence of a theory or discourse, but

to describe its conditions of possibility, and to pursue a Foucaultian archaeo-
logical critique is to describe historical and discursive conditions of possibility.
These rather abstract formulations of Foucault can be distilled into an alter-
native methodology for tackling the problem of exceptionalism. The aim
should be to refuse the metaphysical dialectic between norm and exception,
to resist a move towards an implicit sovereign structure, a hidden meaning
or a ‘single signified’. Instead of moving from empirical instances of excep-
tionalism to a political theory of exceptional sovereign prerogatives and a
mobilized mass polity, the task is to assert the dispersal and historicity of
the conditions of possibility of exceptionalism; to stress that successful and
mobilizing declarations of exceptions are only possible because of an already
existing discursive formation of objects, subject positions, enunciative
modalities, concepts and strategies. Foucault’s ‘discursive formation’ does
not mean an underlying tradition, a return to the same affirmed by the irrup-
tion of the new, but a dispersed archive of historical statements and practices
that make contemporary invocations and enactments of exceptionalism pos-
sible. When one describes ‘exceptionalism’, the aim should not be to describe
a special category, but to describe a dispersed regularity, an assemblage of
practices, an already existing archive of statements, an array of competing
subject positions, a body of tactics and strategies, a formation of historical
conditions of possibility, the limits of which can never be distilled and
formalized, only problematized.

Foucault in Guantánamo
It is almost as though Guantánamo Bay were the public face of exceptionalism.
We have access to detailed accounts about the camp and its practices from
former detainees. We have been fed official pictures of the camp through the
media. Although the happenings inside Guantánamo are carefully guarded,
the very existence of the camp appears to be for domestic and global public
consumption. However, we also know that the USA, for example, holds
thousands more detainees in less publicized locations across the globe.
Amnesty International (2005: 4) estimates that approximately 70,000 detainees
are held outside the USA. These include 520 in Guantánamo Bay naval base
(plus 234 releases or transfers); 550 in the Bagram and Kandahar air bases in
Afghanistan; an unknown number of detainees, estimated at scores, in other
US forward operating bases; in Iraq, 6,300 in Camp Bucca, 3,500 in Abu Ghraib
prison, 110 in Camp Cropper, and 1,300 in other US facilities; 40 estimated to
be held at undisclosed CIA locations worldwide; several thousand held by
foreign governments at the request of the USA; and 100 to 150 estimated secret
transfers (‘renditions’) of detainees to third countries.
Andrew W. Neal Foucault in Guantánamo 45

With this global archipelago of exceptionalism in mind, Foucault’s com-

mitment to describing discursive formations in their dispersal is productive
from the outset. The issue of contemporary exceptionalism exploded into
political discourse in the aftermath of 9/11. One should be wary of identify-
ing exceptionalism as the ‘sudden irruption’ in time of an identifiable unity
of practices. But nor should exceptionalism be described as a continuity. It is
not possible to identify, without a violent degree of reduction or overdeter-
mination, a single, unified discursive practice of exceptionalism in territorial
or political space.
If one is to describe a discursive formation of exceptionalism, then it must be
considered that such a formation is neither a smooth surface nor a unity. One
must describe the positivity of this discursive formation in terms of the con-
ditions of possibility specified by the historical, not philosophical, a priori of
the archive. If one is to talk about a discursive formation of exceptionalism,
consisting of objects, statements, enunciative modalities, subject positions,
concepts and strategies, it can only be to pose the problem, not the unity, of its
delineation in space and time. The discursive formation of exceptionalism is
neither formally identifiable nor philosophically coherent. It is a terrain of
contradictions, oppositions, correlations, additions, recurrences, remanences,
accumulations and complementarities. What is needed is to describe the
spatio-temporal dispersal of the discursive formation of contemporary excep-
tionalism. This means considering the distribution of objects, statements, con-
cepts and strategies across the broken surface of territory, history, politics,
material practice, techniques of power, theoretical discourses, and public
debate, to name but a few locations. This analysis can never be comprehen-
sive or totalizing, only partial.
It is possible, to a certain extent, to sketch what Foucault calls the ‘positive
limits’ of such practices – the agencies, the policies, the transactions, the
memos and orders, the reports and accounts, the places and sites, the subject
positions, the power structures, the positivity of objects, statements, concepts
and strategies that make certain practices possible – but not the ‘critical’ or
‘negative limits’ that oppose ‘knowledge to ignorance, reasoning to imagina-
tion, armed experience to fidelity to appearances, and fantasy to inferences
and deductions’ (Foucault, 2002a: 211–212). One should not identify a formal,
unified and underlying exceptionalism as distinguished from the norm, as do
Schmitt, Agamben and securitization theory, but only an array of dispersed
appearances. Following Foucault’s questions about the describable ‘positivi-
ty’ rather than ‘critical legitimacy’ of the appearance of the ‘sciences’, one
should ask after the ‘positivity’ of exceptionalism rather than question the
‘critical legitimacy’ of its legality/extra-legality, normality/necessity, coher-
ence/contradiction, or continuity/irruption (Foucault, 2002a: 212). Rather
than identify the formal, sovereign-centred qualities of exceptionalism, as
Schmitt, Agamben and securitization theory have done, the task is to sketch
46 Security Dialogue vol. 37, no. 1, March 2006

its broken surface in order to ask what it is for ‘exceptionalism’ to be ‘excep-

tionalism’ in the describable fact of its existence.
* Andrew W. Neal is a Research Associate with the European Commission FP6 project
CHALLENGE (The Changing Landscape of European Liberty and Security) at the
School of Politics, International Relations and Philosophy, Keele University, and Lecturer
in Security Studies at the University of Birmingham. Email: a.w.neal@keele.ac.uk. For
helpful and thoughtful comments on this and previous drafts, he would like to thank
Didier Bigo, Debbie Lisle, Rebecca Palmer, Angharad Closs Stephens, Halit Mustafa
Tagma, Rob Walker, the participants in the panel ‘Articulations of Politics: Violence,
Security, Society’ at the BISA Conference, 20–22 December 2004, and the two anonymous
reviewers for Security Dialogue.

References
Agamben, Giorgio, 1998. Homo Sacer: Sovereign Power and Bare Life. Stanford, CA: Stanford
University Press.
Agamben, Giorgio, 2005. State of Exception. Chicago, IL & London: University of Chicago
Press.
Amnesty International, 2005. ‘Guantánamo and Beyond: The Continuing Pursuit of
Unchecked Executive Power’, AMR 51/063/2005, 13 May; available at http://web.
amnesty.org/library/Index/ENGAMR510632005.
Dreyfus, Hubert L. & Paul Rabinow, 1982. Michel Foucault: Beyond Structuralism and
Hermeneutics. Brighton: Harvester.
Foucault, Michel, 1980. ‘Truth and Power’, in Colin Gordon, ed., Power/Knowledge: Selected
Interviews and Other Writings 1972–1977. New York: Harvester Wheatsheaf (109–133).
Foucault, Michel, 1991. ‘Nietzsche, Genealogy, History’, in Paul Rabinow, ed., The
Foucault Reader. London: Penguin (76–100).
Foucault, Michel, 1999. Discipline and Punish: The Birth of the Prison, trans. Alan Sheridan.
London: Penguin.
Foucault, Michel, 2002a. The Archaeology of Knowledge, trans. A. M. Sheridan Smith.
London: Routledge.
Foucault, Michel, 2002b. Society Must Be Defended, trans. David Macey. New York:
Picador.
The Guardian, (London), 2005. ‘Investigator Links Europe’s Spy Agencies to CIA Flights’, 14
December; available at http://www.guardian.co.uk/usa/story/0,12271,1666824,00.html.
Neal, Andrew W., 2004. ‘Cutting Off the King’s Head: Foucault’s Society Must Be Defended
and the Problem of Sovereignty’, Alternatives 29: 373–398.
Neal, Andrew W., forthcoming. ‘War on Terror or Law on Terror: “Society Must Be
Defended” as an “Archaeology” of the Exception’, in Andrew W. Neal & Michael
Dillon, eds, Foucault: Politics, Society, War. London: Palgrave Macmillan.
Schmitt, Carl, 1985. Political Theology: Four Chapters on the Concept of Sovereignty, trans.
George Schwab. Cambridge, MA & London: MIT Press.
Schmitt, Carl, 1996. The Concept of the Political, trans. George Schwab. Chicago, IL:
University of Chicago Press.
Wæver, Ole, 1995. ‘Securitization and Desecuritization’, in Ronnie D. Lipschutz, ed., On
Security. New York & Chichester: Columbia University Press (46–86).