









































isparchitecture.com

Koolhaas’ revision of foucault’s 
panopticon; or, 
how architecture and philosophy 
just met

ANDRÉ PATRÃO

1. To ask about the interactions between philosophy and 
architecture

 
       Architecture and philosophy have engaged 
with one other, directly, marginally, or just simply 
implicitly, in the works and discourse of academics, 
practitioners, and critics, most evidently in 
architectural modernism and postmodernism, but 
most intensively in the reactions to both. Philosophy 
has been avidly sought by architects to help question, 
disclose reveal, systematize, express, and expand the 
understanding of architectural works, the world 
upon which they intervene, the discipline itself, 
and the role of its practitioners and theoreticians. 
Architecture, in turn, has grown within philosophy 
from a passing example or an illustrative metaphor 
of some other matter, or a generally misfit artform 
within an aesthetic theory, into a topic in its own 
right—as testified by the recent development of a 
‘philosophy of architecture.’ 
      The particularly acute multiplication and 
radicalization of interactions between philosophy 
and architecture throughout the 20th century 
produced a number of well-known cases in which an 
author, work, school, movement, or approach from 
one discipline had a direct decisive effect the other. 
Famously, Norberg-Schulz relied heavily on Martin 
Heidegger’s philosophical writings to elaborate his 
own distinctive architectural phenomenological 



AP . vol 5 . No 1 . 2020

60

PA
TR

ÃO
 

theories. Kenneth Frampton, also well-read in the German philosopher’s 
writings, was never shy about the effect Hannah Arendt exerted on him. 
At times philosophers and architects collaborated, such as in 1985 when 
Bernard Tschumi invited Jacques Derrida and Peter Eisenman to design 
a garden pavilion in Parc de la Villette (1982-1998), or when discussions 
between Jean Baudrillard and Jean Nouvel were published as the book The 
Singular Objects of Architecture (2002). Preceding them all is of course the 
philosopher who also briefly became an architect, Ludwig Wittgenstein in 
the design and construction of House Wittgenstein (1928).
       These celebrated examples represent the epitomes of a still widespread 
tendency which, in diverse ways and to different degrees, shapes both 
philosophical and architectural works. And yet, seldom does either 
discipline take a step back to reflect upon the motives and methods of 
these interactions as a research topic in itself. How does architecture make 
use of philosophy? How does philosophy speak of architecture? Why does 
one turn towards the other? What comes about in their doing so? 
      We shall engage these questions by analyzing one particular case-
study of such interactions: the potential influence of philosopher Michel 
Foucault’s Discipline and Punish (1975) on architect Rem Koolhaas’ Koepel 
Panopticon Prison renovation project (1981). This connection is far less 
well-known than the previously mentioned examples, as it is far more 
low-key. In fact, it is difficult to point out or even discern its existence. 
However, as shall be seen, this discreteness does not mean that there is no 
case-study, but rather that discreteness is one of its principal traits, and 
one which distinguishes it uniquely from the cannon of architecture and 
philosophy’s interactions. 
       We shall begin by introducing the two protagonists of each domain 
and their respective contributions to the case-study: first, Foucault and 
the popular section of Discipline and Punish’s chapter “Panopticism;” 
second, Koolhaas’ renovation proposal for the Koepel Panopticon Prison 
and the explanatory text Revision that came with it. Third, we will dwell 
on how the former’s philosophical thinking may have exerted direct yet 
nearly imperceptible consequential influence upon the latter’s architectural 
project. Fourth and finally, the distinctive characteristics of this case-study 
shall be brought before the broader questions of how philosophy and 
architecture can engage one another.



61

isparchitecture.com

2. Michel Foucault, and Panopticism

       Michel Foucault has long been a recurring 
reference in architectural discourse. His writings 
on the socially marginalized, the power structures 
that set them apart from the mainstream, and 
the role of a certain conception of knowledge 
in enforcing this system have informed and 
influenced architectural concerns, generally—
but not only—through the social sciences.  His 
concept of heterotopia, for example, continues to 
sprout a rich variety of contradictory readings and 
reflections, while delighting the imagination for 
architecture researchers, practitioners, researchers, 
and even students, very much due to his original 
social critiques and the inspiring images with 
which he conveyed them. The force of Foucault’s 
impact on architecture is rather surprising, though, 
considering that architecture is not a major  
Foucauldian topic, nor even a minor one, nor did he 
ever produce any literature whatsoever exclusively 
or even primarily dedicated to architecture. For 
example, the famous lecture about heterotopias 
“Of Other Spaces” (Des Espaces Autres), delivered to 
the Circle of Architectural Studies (Cercle d’Études 
Architecturales) in 1967, is in fact little more than a 
copy of “Heterotopias” (Des Hétérotopies), the 1966 
radio broadcast on France Culture about utopia and 
literature, and with absolutely no intent to engage 
architecture at all—much like the concept’s first 
appearance in The Order of Things (Les Mots et Les 
Choses, 1966) which also focused on literature. The 
closest Foucault came to addressing architecture 
directly, thematically and extensively was in “Space, 
Knowledge, and Power,” a 33-question interview 
conducted by American cultural anthropologist 
Paul Rabinow, and published in the architecture 
and design journal Skyline in on March 1982. 
However, even there, architecture is not so much 
discussed directly as a topic in itself but rather as 

seldom does 
either discipline 

take a step 
back and 

reflect upon 
the motives 

and methods 
of these 

interactions 
as a research 

topic in itself. 
How does 

architecture 
make use of 
philosophy? 

How does 
philosophy 

speak of 
architecture?

“

”



AP . vol 5 . No 1 . 2020

62

PA
TR

ÃO
 

a gateway into topics that underlie it (e.g. knowledge and power) which 
actually interested him. It so happened that they interested architects too.
       Indeed, in Foucault’s writings architecture and architectural works only 
tend to appear as metaphors or as case-studies to illustrate and develop 
some issue other than architecture itself. And perhaps the most effective 
deployment of both strategies is found in the second section of the chapter 
“Panopticism” of his book Discipline and Punish.
       Discipline and Punish: the Birth of the Prison (Surveiller et Punir – 
Naissance de la Prison), published in 1975, is a genealogical inquiry into 
the emergence of the modern prison system, particularly in France, and a 
study of its disciplinary mechanisms’ widespread presence in many other 
areas of society. The emphasis falls on the reformist disciplinary penal 
system developed since the late 18th / early 19th century which, though 
far more benign than the punitive system it supplanted, nevertheless 
constituted an intensive form of generalized social control. At the center 
of this kind of disciplinarity are surveillance strategies—as hinted at in the 
discrepancy between the book’s original title (Surveiller et Punir) and its 
English translation (Discipline and Punish), where the French surveiller (“to 
monitor”, in the sense of “surveillance”) is replaced by “discipline.”
    To illustrate these disciplinary mechanisms, Foucault resorts to what 
he considers their most paradigmatic example: the panopticon, a building 
typology devised, curiously enough, not by an architect but by renowned the 
English philosopher Jeremy Bentham (1748-1832), inspired by his brother 
Samuel’s organization of industrial labor compounds. In its simplest and 
most characteristic form, the panoptical typology consists of a cylindrical 
multi-story arrangement of partitions around a central observation house. 
The observers at the center have a constant unhindered view into every 
partitioned space; hence the typology’s name from the combination of the 
Ancient Greek pan (all) and optikos (pertaining to sight), meaning “all-
seen” or “all-under-sight.” From within the partitioned spaces, however, it 
is impossible to perceive what happens in the observation house.1 Thus, 
in a panoptical system, a small number of observers can monitor a large 
number of people at any given time; the prominent feature, though, is that 
those monitored, unable to verify when they are or are not in fact being 
watched act as if they always are. The prisoners thus become their own 
guards.
       Foucault, like Bentham, regards the panopticon as an architectural 
formulation of certain Benthanian philosophical and social principles—
in a way making it too a case-study of architectural and philosophical 
interaction—which Bentham celebrates optimistically in Panopticon; or, 



63

isparchitecture.com

the Inspection-House2 (Figure 1) 
and Foucault denounces bleakly 
in Discipline and Punish. For 
both, however, architecture is 
only employed as a means to 
make a point, but is not the 
point itself. Foucault does not 
invoke the panopticon because 
it is architecture. In Discipline 
and Punish the panopticon 
is both a metaphor—the 
disciplinary power relations 
described throughout the 
book symbolized by the 
dramatic image of the 
archetypal surveillance 
techniques employed to enforce it—and a case-
study: not just a representation but a paradigmatic 
operational mechanism that enables and enforces 
the power structures of which Foucault speaks. 
Throughout the “Panopticism” chapter, Foucault’s 
interpretation and depiction of the panopticon 
ends up producing a philosophical description of an 
architectural typology, by expounding its historical 
origin, purpose, possible functions, modes of 
operation, formal characteristics, and societal role. 
Inadvertently, though, in doing so his reading also 
becomes an interpretation of the panopticon qua 
architectural object, albeit one grounded on his own 
understanding of architecture seen later in “Space, 
Knowledge, and Power,” as an mere instrument of 
far greater underlying forces. Hence this is why, in 
his words:

the Panopticon must not be understood 
as a dream building: it is the diagram of a 
mechanism of power reduced to its ideal 
form; its functioning, abstracted from any 
obstacle, resistance or friction, [may very well] 
be represented as a pure architectural and 
optical system: it is in fact a figure of political 
technology that may and must be detached 

 architectural 
works only 

tend to appear 
in Foucault’s 

writings as 
metaphors, 

to illustrate 
a non-

architectural 
point, or as 

case-studies, to 
illustrate and 

develop some 
other non-

architectural 
issue

“

”

figure 1:

Bentham’s panopticon 



AP . vol 5 . No 1 . 2020

64

PA
TR

ÃO
 

from any specific use.3   
Discipline and Punish thus unwittingly offers a philosophical critique of an 
architectural object and, with it, an accessible way from architecture into 
Foucault’s philosophical thinking. 
       Between 1979 and 1981, Koolhaas made use of both. 

3. Rem Koolhaas, and the Koepel Panopticon Prison

       The second protagonist of this story is a young Rem Koolhaas, 
renowned in the architecture world after Delirious New York (1978) but 
still seeking to establish himself as a practitioner. In 1978, his Office for 
Metropolitan Architecture produced what would become the first entry in 
their official list of projects: the Dutch Parliament Extension, in The Hague, 
Netherlands. Despite an ex aequo first place in the competition, alongside 
the Dutch architect Leo Heijdenrijk , the proposal was ultimately rejected 
along with all the others. However, OMA was granted what Koolhaas called 
a “consolation prize:” in 1979, the Dutch Government commissioned 
OMA to assess the possibility of prolonging the lifetime, by another 50 
years, of a prison, built in Arnhem between 1882 and 1886. (Figure 2) 
The original building, designed by Dutch architect and engineer Johan 
Frederik Metzelaar, was known as the Koepelgevangenis—a conjunction of 
the Dutch words “jail” (gevangenis) and “cupola” (koepel), alluding to its 
distinctive large domed roof. Along the inside of its cylindrical walls were 
four floors of holding cells, around a guardhouse at the center from which 
guards could originally see into every cell without being seen. In other 
words, the Arnhem Koepel was a panoptical prison.
       OMA’s proposal adds a series of new functions and structures to the 
prison complex: outside of the original building to the east there is the 
entrance, lobby, reception area, porter’s room, and meeting room; from 
here a visitor heads south, to the visitor’s garden and visitor’s room; to 
the north is the pavilion for difficult prisoners, a kitchen, and a patio in 
between, as well as a storage area; to the west  is the guard’s’ canteen, 
guard’s’ cloak room, instruction department, shops, and more storage 
space, as well as the sports area, which includes a gymnasium, judo room, 
multi-purpose room, running track, sports field, and a pool. Inside the old 
Koepel dome is a library, barber, meeting room, shop, and an infirmary. 
      The original building undergoes modifications as well: two new satellite 
structures house living quarters, dining rooms, and bathrooms, enabling 
the possibility of dividing the inmates of each floor into two groups (of 



65

isparchitecture.com

sensibly 48 people each), or even more if the satellite 
is divided too; and the cells, no longer constantly 
translucent to the guards, offer privacy. The most 
significant change, though, is the replacement of 
the central guardhouse by two lower-level streets 
intercepting in a cruciform manner, containing 
several of the aforementioned functions and opening 
access to the rest of the complex. In S, M, L, XL 
(1995), Koolhaas 
alludes to this 
s i m u l t a n e o u s l y 
metaphorical and 
practical gesture 
with two gruesome 
frames from 
Luis Buñuel and 
Salvador Dali’s film 
Un Chien Andalou 
(1929), displaying 
an eye being cut: 
the crossing streets 
cut the all-seeing eye, the pan-optikos. In other 
words, Koolhaas removes panopticism from the 
panopticon.
      Both generalist and specialized literature 
commonly assume that OMA’s redesign of the 
Koepel Panopticon Prison was strongly influenced 
by Discipline and Punish, published just four four 
years before the commission, and by Foucauldian 
thinking in general. One of the earliest and most 
prominent examples of these texts was Anthony 
Vidler’s article “The Ironies of Metropolis: Notes on 
the Work of OMA,” published in Skyline in March 
1982—by no coincidence in the same Skyline edition 
issue that included Foucault’s interview “Space, 
Knowledge, and Power.” According to Vidler, 
“[in the Arnhem prison project] we find echoes 
of a reading of Michel Foucault, whose studies of 
discipline and power have strongly influenced the 
politics and strategies of the generation of OMA.”, 
and it is “in the space marked by Foucault after 

the crossing 
streets cut the 

all-seeing eye, 
the pan-optikos. 
In other words, 

Koolhaas 
removes 

panopticism 
from the 

panopticon.

“

”

figure 2:

arnhem prison



AP . vol 5 . No 1 . 2020

66

PA
TR

ÃO
 

Nietzsche, that the project has been conceived.” 4 
       Indeed, the easily accepted assumption of Foucauldian influence on 
the Koepel renovation does not appear to raise any red flags that would 
lead us to question it. Until we do question it, that is: Exactly where is 
this influence? We ought to remember that the current inquiry does not 
merely search for a possible Foucauldian interpretation of the project, one 
amongst many that a critic may adopt to understand Koolhaas’ project 
once concluded, but rather the actual Foucauldian influence during the 
design process that produced the project. Thus, instead of assuming the 
existence of a case-study, we must first ask if there really is a case to study.
       The task presents difficulties though. As it turns out, Koolhaas and 
Foucault actually met in what we could well imagine to be a pivotal 
event: Around seven years before the Arnhem commission, architecture 
student Rem Koolhaas had just arrived in Ithaca (New York, USA), 
where he would attend Cornell University from 1972 to 1973. In  the 
fall semester of 1972, Foucault spent three weeks at Cornell lecturing at 
Cornell’s Romance Studies Department. The celebrity French thinker 
and the promising Dutch pupil became acquainted during this time. The 
encounter itself, though, was not especially life-changing. Recalling the 
occasion in an interview with The Cornell Journal of Architecture 8: RE 
(2004), Koolhaas affirms that he “cannot claim any kind of significant 
intellectual influence”5 from Foucault. Admittedly, the answer is 
disappointing, but not surprising, coming from someone who readily 
makes use of his formidably rich knowledge of his and other various fields 
in his works, but resolutely resists sacrificing his authorial originality to 
them. In 2008, in an interview for Radical Philosophy, the interviewers Jon 
Goodbun and David Cunningham noted how other architects of the same 
generation—Eisenman, Tschumi, and Libeskind—explicitly associated 
themselves with various philosophical trends while Koolhaas did not, to 
which he answered, after listing multiple philosophers he was acquainted 
with: “I don’t think these influences or relationships necessarily need to be 
flagged up. But it’s not an indifference to these thinkers.”6 
       Indeed, any claim of an ‘influence’ upon Koolhaas must be cautious in 
its use of the word. Rather than a powerful force that invades and pervades 
his work, an ‘influence’ for Koolhaas seems more like an ingredient which 
he takes hold of, tinkers with, processes and mixes with others to produce 
something of his own. The original sources of these external contributions 
become almost untraceable. It is in this sense that we must understand an 
exceptional confession, made at an event hosted by ETH Zürich in 2011, 
regarding Delirious New York, published in the same year as the Arnhem 



67

isparchitecture.com

commission:
[There is something which is generally not 
recognized,] that actually this book was a 
kind of French book, in the sense that it was 
profoundly influenced not only by Barthes, 
but also to some extent by Foucault, and 
particularly by somebody who had become 
my personal friend, Hubert Damisch.7

       Could the same be said specifically for the 
Arnhem project, though? In 1981 OMA handed 
in the Arnhem renovation project to the Dutch 
Ministry of Justice with an essay called Revision 
containing an overall historical analysis of the 
Arnhem Koepel prison and its typology; the 
theoretical and conceptual framework in which 
OMA operated; and an objective programmatic 
and formal description of the project itself. The text 
was later published in S, M, L, XL (1995), a book 
traversed through and through by a cornucopia of 
words simulating a dictionary, and which includes 
three entries quoted from Foucault: ‘exclusion’, 
quoted from Foucault’s 1970 address to the 
Collège de France under the entitled “L’ordre du 
discours,” translated by Alan Sheridan in 1972 as 
“The Archaeology of Knowledge & The Discourse 
on Language;” and, from Discipline and Punish, 
‘visibility’ and ‘power.’8 We may, however, argue 
that these dictionary entries came after the fact, as a 
later recognition of an influence. 
       Nevertheless, Revision also contains other little 
(but more than coincidental) connections. For one 
thing, the text’s description of the panopticon uses 
a series of words that appear in Alan Sherridan’s 
English translation of Discipline and Punish, but 
not, for example, in Bentham’s Panopticon; or, the 
Inspection House (e.g. referring to the panopticon’s 
floors as ‘rings’ instead of ‘circumferences’). 
        More importantly, though, Revision echoes 
Foucault’s bleak critical interpretation of the 
panopticon, resorting to similar reasons and 

Indeed, the 
easily accepted 

assumption of 
Foucauldian 
influence on 

the Koepel 
renovation does 

not appear to 
raise any red 

flags that 
would lead 

us to question 
it. Until we do 

question it, that 
is: exactly 

where is this 
influence?

“

”



AP . vol 5 . No 1 . 2020

68

PA
TR

ÃO
 

description, so much so that we can even pinpoint Revision’s similarities 
to the first two paragraphs, and perhaps part of the third, of the second 
section of “Panopticism.” Once again, though, we may counter by noting 
that perhaps these similarities only suggest a shared  intellectual context 
between Foucault and Koolhaas, with crossing issues, discussions, and 
approaches, like Vidler had noted so well. This was already the case at 
Cornell, for the same academic milieu that brought Foucault to campus, 
and that even pioneered the introduction of continental philosophy in 
the USA through the journal Diacritic, surrounded Koolhaas during his 
studies. It is no coincidence that Koolhaas had also read Roland Barthes, 
another renowned structuralist and later post-structuralist from the same 
Parisian intellectual scene as Foucault’s before producing his “French 
book.”
       Indeed, skepticism over the existence of a case study at all  seems 
reinforced by the fact that throught the entire text of Revision Foucault is 
not once quoted, mentioned, or even alluded to even though he explicitly 
references the evaluation by the governmental Jacobs Committee, the 
accusations launched by the 19th century Dutch Parliament member 
Wintgens, and prison typologies experimented with in Amsterdam and 
Maastricht. Other unnamed decisive forces may also have come into 
play. Four years before Discipline and Punish, Robin Evans published 
the article “Bentham’s Panopticon: An Incident in the Social History 
of Architecture” (1971) in the Architecture Association Quarterly, while 
Koolhaas, an eager reader of these issues, was still a student at the school.9 

Evans, curiously enough, quoted one of Foucault’s earlier publications 
in the essay, suggesting that not only was he acquainted with the French 
philosopher’s work, but found him pertinent for the topic too. Their 
readings differed though: for example, Evans’ focuses less on the role of 
authority—distinguishing his reading from Koolhaas’ and Foucault’s—
but more on the panopticon’s solitary confinement—like Koolhaas’, but 
unlike Foucault’s. This may explain certain particularities of Revision’s 
interpretation of the panoptical typology: on the one hand its two main 
pillars are the all-seeing surveillance system and solitary confinement, not 
so emphatically highlighted by Foucault; on the other hand, it often omits 
critical components of Discipline and Punish, such as the panopticon’s 
imposition of self-surveillance upon the surveyed, a fundamental point for 
both Bentham and Foucault.
       An alternative possible explanation, though, comes from the same 
event in which Koolhaas admitted some degree of Foucauldian influence:

[S]omehow I have a kind of sense of reticence, or chastity, or perhaps 



69

isparchitecture.com

it is that I write in my own kind of mind more 
literature than theory that prevents me from 
mentioning all these sources. And also because 
I am deeply aware that I am the kind of result 
of very respectable influences but also very 
unrespectable influences and that everything 
is a kind of amalgamation of trash and high 
culture […].10

       In other words, we should not expect to find 
a smoking gun of direct influence as it may be 
dissolved with other elements, or even not exist at 
all. The difficulty in confirming the assumption of 
a case-study is both due to the fact that because the 
links between the architect’s and the philosopher’s 
works are not explicitly demarcated, and also 
because these are irreducible to a single simple and 
uncorrupted instance, act, or event.  The cause lies 
partially with Foucault’s less than marginal interest in 
architecture, but mostly with Koolhaas’ own attitude 
towards external influences. It is nearly impossible 
to pinpoint an irrefutable, straightforwardly 
Foucauldian effect upon the Arnhem design, or to 
determine exactly how much can indeed be directly 
attributed to Foucault, or how many links were 
consciously forged by Koolhaas. Notwithstanding, 
a reading of Revision shall come to show that, to a 
significant degree, the Koepel design does indeed 
deal with a Foucauldian approach to a Foucauldian 
representation and discussion of a Benthanian 
object, regardless of whether this was or was not 
deliberate. It is in this sense that we can claim the 
existence of a case-study. 

4. The Foucaldian ‘influence’ (and lack of) in Koolhaas’ 
designRevision: Koolhaas from, through, beyond, besides 
Foucault

        How does Foucault’s and particularly Discipline 
and Punish’s influence appear in OMA’s project 
for the Arnhem panopticon renovation? To better 

    the Koepel 
design does 

indeed deal with 
a Foucauldian 

approach to 
a Foucauldian 

representation 
and discussion 

of a Benthanian 
object, 

regardless of 
whether this 
was or was 

not deliberate.

“

”



AP . vol 5 . No 1 . 2020

70

PA
TR

ÃO
 

understand the design process we return to the explanations given in 
Revision, whose seven small chapters may be split between a critical analysis 
of the old Koepel—from historical overviews to the broader architectural 
debates they provoke, and of the new proposal that follows—from its 
conceptual aim to its formulation as an architectural project.
       The first few paragraphs contain OMA’s interpretation of the 
panopticon qua typology, and of the Arnhem panopticon in particular. 
More than simply an introduction for the reader, this description reveals the 
architects’ grounding interpretations of the site’s pre-existing conditions, 
thus constituting the object upon which to intervene. For example, we 
could conceive of the Arnhem prison in terms of its urban situation, or 
of its distinct cylindrical form, or its colossal dome, the interior lighting, 
materiality, running costs, or environmental sustainability. Each different 
reading of the site’s pre-existing conditions derives from different 
approaches, theoretical frameworks, priorities, interests, and many other 
factors which reveal and emphasize different aspects of the site and call for 
different responses
          Koolhaas construes the Koepel according to “the Panopticon Principle, 
with its mechanistic ideal,” as an application of “a universal principle of 
organization for situations in which a small group of supervisors monitors 
a much larger group of supervised.”  For Koolhaas, “[t]he Arnhem Koepel 
represents the principle in its purest form: a single, all-seeing “eye” is 
placed dead center in a circle of the observed.” However, this principle, 
“[o]ne hundred years later, has become intolerable.”11 In these words we 
recognize a synthetic and superficial semblance to Foucault’s reading of 
Bentham’s description of the panopticon: the stress on its mechanistic 
character and the purity of an ideal put into an architectural form is even 
reminiscent of  the previously quoted passage of Discipline and Punish 
concerning the panopticon’s not being a ‘dream building’. In other words, 
OMA’s interpretation of the pre-existing object, upon which they, as 
architects, shall intervene upon, broadly coincides with a Foucauldian 
critical philosophical description of the panoptical typology.
         Turning Revision’s pages and moving towards the contents of OMA’s 
own proposal, implicit reference to the philosopher, no longer limited to 
Discipline and Punish, persists. Consider, for example, Koolhaas’ argument 
for programmatic flexibility, rather than ideal pre-determined architectural 
formal typologies (Koolhaas’ real interest), as a manner of accounting for 
the inevitably incessant historically shifting ideologies and their social 
formulation (Foucault’s real interest): 

If prison architecture today can no longer pretend to embody 



71

isparchitecture.com

an “ideal”, it could regain credibility by 
introducing the theme of revision as raison 
d’être. A “modern” prison architecture would 
consist of a prospective archaeology, constantly 
projecting new layers of “civilization” on old 
systems of supervision.12  

        A reader familiar with Foucault cannot help 
hearing echoes of the philosopher’s own concept of 
“archaeology”, as conveyed for example in The Birth 
of the Clinic: an Archaeology of Medical Perception 
(Naissance de la Clinique – une Archéologie du 
Regard Médical, 1963) and Archaeology of Knowledge 
(L’Archéologie du Savoir, 1969). Koolhaas’ 
prospective archaeology both presupposes and 
builds from Foucault’s archaeological method: on 
the one hand, it rejects trans-historical and trans-
cultural absolute social and (consequentially) 
architectural models, embracing instead the 
constant transformation of what societies, or 
“civilizations” as the architect put it, regard as the 
best model in each period; on the other hand, it 
then turns this Foucauldian analytical tool deployed 
on the past for an understanding of the present into 
a generative design principle projected towards the 
future. Hence prospective archaeology, which does 
not simply create a new archaeological layer with 
the next prison typology, but rather builds on the 
fact of their inevitable demise and replacement. 
The Arnhem renovation project is founded on the 
thematization of prospective archaeology. 

“Arnhem could be an experiment with a form 
of renovation that articulates programmatic 
and ideological change without destroying the 
building itself.”13

     Foucault’s implicit presence as a reference 
understandably disappears as the design progresses 
deeper into architectural and technical territory. 
OMA’s actual design gravitates around the surrealist-
inspired metaphor of cutting the panopticon’s all-
seeing eye, executed by replacing the guardhouse 



AP . vol 5 . No 1 . 2020

72

PA
TR

ÃO
 

with the two sunken streets from which the rest of the project spreads out. 
While this gesture may act upon Foucault’s reading of the panopticon, it 
certainly does not act according to him. In fairness, Discipline and Punish 
is mostly a descriptive endeavor, and thus does not offer prescriptive 
orientations that Koolhaas could follow. But would he, who so vehemently 
rejects external dominance over his work, ever agree to follow them even 
if he could? 
     In the end, the full weight and extent of Foucault’s influence upon 
OMA’s design process is measured not only by philosophy’s presence, 
but also by philosophy’s absence. While Discipline and Punish, Foucault, 
and philosophy in general do introduce decisive inputs throughout the 
design process, particularly in the initial interpretation of the pre-existing 
and in preparing an approach towards it, architecture is never sidelined 
as the dominating force. The Koepel renovation project was not meant 
as a mere literal architectural formulation, expression, or translation of 
philosophical principles, discussions, and ideas. Koolhaas meant to design, 
first and foremost, an architectural project—a seemingly banal yet radically 
simple approach within the larger architecture world’s interactions with 
philosophy.

5. Philosophy for architecture,, architecture for philosophy

     Despite two years of discussions with the Dutch authorities and, 
eventually, an approval for construction, OMA’s proposal for the Arnhem 
prison was never built. Its indefinite suspension was not due to any 
particular conceptual or programmatic problems, but to an all-too-familiar 
reason for architects; lack of funding. The Koepel would eventually be 
renovated by Martin van Dort, of Archivolt Architecten BV, in a process 
that started in 1994- and only ended in 2005.14 But the declining number 
of prisoners in the Netherlands led the classified national monument to be 
closed in 2016, and sold in 2018.
       Nevertheless, OMA’s Arnhem panopticon renovation project remains 
an exceptional, unique, and—design-wise—highly successful instance of 
a consequential interaction between philosophy and architecture. This 
outcome may surprise considering the two protagonists involved, the 
way they work within their own disciplines, and how they interact with 
the other’s: Foucault’s disinterest in architecture as a research topic and 
Koolhaas’ almost indiscriminate  appropriating, and critiquing of the vast 



73

isparchitecture.com

multiple sources he knows. 
       The result is a case-study as much about 
architecture and philosophy’s interested interaction 
as concepts and topics become shared across 
disciplinary boundaries as it is about their 
disjunction. We do not find an authoritative 
philosophical imposition of ill-fitting, inadequate, 
and even just unrealistic principles, norms, and 
judgements upon architecture. Instead, Foucault 
works within his own concerns, but develops 
them in such a way that they offer themselves to 
appropriation. In an interview with the geographers 
of the journal Herodote, in 1976, Foucault, 
insistently confronted with the fact that he had not 
said much about geography in his writings, offers a 
reply that can just as well be applied to architecture:

If I do the analyses I do, it’s not because of 
some polemic I want to arbitrate but because 
I have been involved in certain conflicts 
regarding medicine, psychiatry and the penal 
system. […] It’s up to you, who are directly 
involved with what goes on in geography 
[or architecture], faced with all the conflicts 
of power which traverse it, to confront them 
and construct the instruments which will 
enable you to fight on that terrain. And what 
you should basically be saying to me is, ‘You 
haven’t occupied yourself with this matter 
which isn’t particularly your affair anyway 
and which you don’t know much about’. And 
I would say in reply, ‘If one or two of these 
“gadgets” of approach or method that I’ve 
tried to employ with psychiatry, the penal 
system or natural history can be of service to 
you, then I shall be delighted. If you find the 
need to transform my tools or use others then 
show me what they are, because it may be of 
benefit to me.15 

     This is where Koolhaas comes in. Not, however, 
as the architect who obsessively attempts to 

       Koolhaas 
did not 

simply copy 
Foucault into 
architecture, 

but rather 
borrowed his 

conceptual and 
terminological 

clarity to 
develop and 

better express 
his own ongoing 
architectural 

discussions

“

”



AP . vol 5 . No 1 . 2020

74

PA
TR

ÃO
 

translate philosophical concepts into architectural form as if there lay an 
unassailable absolute foundation for architectural designs, and much less to 
superficially legitimize his work with empty or misunderstood quotations 
of a celebrated philosopher. Koolhaas, both a brilliant theoretician and 
practitioner, respectfully instrumentalizes philosophy and, along with 
many other tools, takes what he needs as and when he needs it for the 
sake of the production of an architectural project first and foremost. His 
sources, almost impossible to isolate in a single moment of contact, may 
also very well arise from the sheer shared discourse between disciplines, 
which often includes influential literature like Discipline and Punish and, 
it would seem, seminal works of architecture too.
      The Arnhem project thus shows architects how their designs can 
gain from resorting to philosophy without sacrificing architecture in the 
process. However, there are relevant insights for philosophers as well. While 
Foucault’s study of the panopticon created a philosophical critique of an 
architecture object, Koolhaas’ design in turn created a philosophically-
charged architectural reaction to a Foucauldian interpretation of Bentham’s 
panopticon—the conception of the pre-existing object upon which OMA 
intervened. The Koepel renovation project contains an architectural 
critique of the Benthanian utilitarian vision that originated the panoptical 
typology, and a critical conversation with Foucault’s own critique of it. As 
such, it can be read in philosophical terms, even if the architect’s action 
itself is not philosophy nor primarily philosophically motivated, for it does 
engage in a dialogue with philosophy even if through the distinct means 
of architecture. The Koepel Panopticon Prison renovation is therefore not 
just a case-study for architects, but for philosophers too.

Endnotes

1. Jeremy Bentham, “Panopticon; or, the Inspection-House.” In The Works 
of Jeremy Bentham, 4:36–142. (Edinburgh: William Tait) 1843, pp. 40-41.
2. Bentham, 1843, p. 39.
3. Michel Foucault, Discipline and Punish: The Birth of the Prison, 1975, 
p. 205.
4. Anthony Vidler, “The Ironies of Metropolis: Notes on the Work of 
OMA,” Skyline, March, 1982 p. 21.
5. “OMA re: OMU: Interview with Rem Koolhaas on Oswarl Mathias 
Ungers.” Interview by Jeremy Alain Siegel, Melissa Constantine, Matt 
Eshleman, and Steven Zambrano Cascante. The Cornell Journal of 
Architecture 8: RE, March 6, 2004. http://cornelljournalofarchitecture.



75

isparchitecture.com

cornell.edu/read.html?id=18. 
6. Jon Goodbun and David Cunningham, 
“Propaganda architecture: interview with Rem 
Koolhaas and Reinier de Graaf.” Radical Philosophy, 
October 30, 2008. https://www.radicalphilosophy.
com/interview/rem-koolhaas-and-reinier-de-graaf. 
7. Rem Koolhaas and Bernard Tschumi. “A 
Conversation,” Interview by Trüby Stephan. GTA 
Events - ETH Zürich official website, May 18, 
2011. https://video.ethz.ch/events/2011/d-arch_
gta/75db5f75-5ff9-41af-a221-c48ab14f8aa5.html.
8. “Revision.” [1981] In S, M, L, XL, by Rem 
Koolhaas, Bruce Mau, and OMA, 233–53, 2nd ed. 
New York: Monacelli Press, 1995.
9. I am greatly indebted to Mary McLeod for 
pointing out the existence of this article and 
its potential pertinence for Koolhaas’ project at 
Arnhem.
10. Rem Koolhaas and Bernard Tschumi. “A 
Conversation.”
11. Revision p. 237.
12. Revision  p. 241.
13. ibid.
14. P.I. De Berg, “Arnhem - renovatie, restauratie en 
uitbreiding koepelgevangenis” (Rijskmonument)” 
Archivolt Architecten BV. Accessed August 31, 
2018. https://archivolt-bna.nl/web/wp-content/
uploads/2015/03/PI-De-Berg-A4.pdf.
15. Michel Foucault, “Questions on Geography.” 
In Power/Knowledge: Selected Interviews and Other 
Writings, 1972-1977, translated by Colin Gordon, 
1st American ed., 64–77. New York: Pantheon 
Books, 1980. p. 65.

Foucault’s works cited in the text:
“Des Hétérotopies.” France Culture, July 12, 1966.
Discipline and Punish: The Birth of the Prison. 2nd 
Vintage Books ed. New York: Vintage Books, 1975.

     it does 
engage in a 

dialogue with 
philosophy and 

its subjects, 
even if through 

the distinct 
means of 

architecture.

“

”



AP . vol 5 . No 1 . 2020

Power/Knowledge: Selected Interviews and Other Writings, 1972-1977. 
Edited by Colin Gordon. 1st American ed. New York: Pantheon Books, 
1980.
“Questions on Geography.” In Power/Knowledge: Selected Interviews and 
Other Writings, 1972-1977, translated by Colin Gordon, 1st American ed., 
64–77. New York: Pantheon Books, 1980.
“Space, Knowledge, Power.” In The Foucault Reader, edited by Paul 
Rabinow, 1st ed., 239–56. New York: Pantheon Books, 1982.
Surveiller et Punir: Naissance de la Prison. Collection TEL. Paris: Gallimard, 
1975.
The Archaeology of Knowledge & The Discourse on Language. Translated by 
A.M. Sheridan Smith. New York, NY: Pantheon Books, 1982.
The Order of Things: An Archaeology of the Human Sciences. Vintage books 
edition. New York, NY: Vintage Books, 1966.
Michel Foucault and John K. Simon. “Michel Foucault on Attica: An 
Interview,” 1991 1974.

m
a

ha
n

 &
 k

lu
y

te
n

a
a

r

76


