Peter Eisenman

House I 1967
Cardboard Architecture: House I John Summerson represented this position The thesis presented in House I, the Baren-
Peter D. Eisenman quite explicitly when he attempted to make a holtz Pavilion, is as follows: one way of
case for a theory of architecture with such a producing an environment which can accept
These two articles by Peter D. Eisenman, "House I" programmatic basis. In essence, Summerson or give a more precise and richer meaning
and "House II" were first drafted in November of 1969 said the source of unity in modern architec- than at present, is to understand the nature
and April of 1970, respectively. In both cases they ture is in the social sphere, in other words, of the structure of form itself, as opposed to
were redrafted and necessarily condensed for publica- the relationship of form to function or of
in the architect's program. But it would seem
tion in the first edition of this book.
that the situation is more complicated than form to meaning.
In this edition the substance of the ideas remain the
Summerson allowed. For if the program is to
same as in the first publication. The only intention in sustain such an emphasis, it should be able House I posits one alternative to existing
the changes which have been made here has been to to specify and distinguish what the facts of conceptions of spatial organization. Here
clarify their content. a particular situation are, and except for there was an attempt, first, to find ways in
certain physical laws, facts in a programma- which form and space could be structured so
At present most buildings are burdened by their very tic sense are in reality a series of value that they would produce a set of formal rela-
description as "museums" or "country houses" with judgements. Much of the oeuvre of modern tionships which is the result of the inherent
a weight of cultural meaning which is here meant to architectural theory is involved in a basic logic in the forms themselves, and, second,
be neutralized by the opposition of an equally loaded dilemma precisely because it has refused to to control precisely the logical relationships
term. "Cardboard," usually a derogatory term in archi- distinguish between problems of fact and of forms.
tectural discussion (as Baroque and Gothic were when problems of value. And more specifically,
first used), is used here deliberately as an ironic and
because it has refused to recognize prob- There were three steps in this process in
pre-emptory symbol for my argument.
lems of form as predicated by anything ex- House I. First, an attempt was made to make
Cardboard is used to Question the nature of our per-
cept ideas of social and technological a distinction between those aspects of form
ception of reality and thus the meanings ascribed to change or as a matter for stylistic and which respond to programmatic and tech-
reality. Thus it is not so much a metaphor describing aesthetic speculation. nological requirements and those aspects of
the forms of the building but rather its intention. For form which relate to a logical structure. In
example, models are often made of cardboard, so the A museum as a program offers very little in order to make this distinction, an attempt
term raises the Question of the form in relation to the the way of specific functional requirements was made to reduce or unload the eXisting
process of design: is this a building or is it a model? which can act as either a suggestion for or meaning of the forms. Second, a formal
limitation to a formal development. This structure was made from these marks in the
Cardboard is used to shift the focus from our eXisting might account for the fact that many of the actual environment. Third, this formal struc-
conception of form in an aesthetic and functional con- best museums are ones which have been ture of marks was related to another formal
text to a consideration of form as a marking or nota-
created in buildings originally designed for structure of a more abstract and funda-
tional system. The use of cardboard attempts to dis-
other purposes. Equally, since it is difficult mental nature. The purpose of this proce-
tinguish an aspect of these forms which are designed
to act as a signal or a message and at the same ti me
to define a precise form from the functional dure was to provide an awareness of formal
the representation of them as a message. requirements, the form of a museum is often information latent in any environment which
realized as a very idealized shape. Since previously was unavailable to the individual.
Cardboard is used to signify the result of the particular very little is imposed on the form of a mu-
way of generating and transforming a series of primi- seum by its function, its form may be used to One aspect of the first step was an attempt
tive integar relationships into a more complex set of help clarify part of the problem outlined to reduce or unload the existing meaning of
specific relationships which become the actual build- above. the forms dictated by function so that the
ing. In this sense cardboard is used to denote the par- forms could be seen as a series of primitive
ticular deployment of columns, walls, and beams as The making of form can, for instance, be marks. This was attempted through a manip-
they define space in a series of thin planar, vertical considered as a problem of logical consis- ulation of the relationship of the color, tex-
layers. It is not so much a literal recognition of the
tency; as a consequence of the logical struc- ture, and shape of the built forms. White
actual surfaces as cardboardlike and thus insubstan-
tial but rather is meant to signify the virtual or implied
ture inherent in any formal relationship. The forms are used in House I to shift our visual
layering which is produced by the particular configura- making of form in this sense is more than perception and conception of such forms;
tion. the satisfaction of functional requirements from the perception of a real, tangible, white
and more than the creation of aesthetically volumetric architecture to the conception of
In this context House I and House " are experiments pleasing objects, but rather the exposition an abstract, colored planar space; from the
which attempt to translate these concepts into a pos- of a set of formal relationships. polemic of the "white" of the 1920's to the
sible working method and into a physical environment. neutrality of "cardboard." The white color
House I was an attempt to conceive of and and the flat texture are closer to an abstract
There is often an attempt made to rationalize understand the physical environment in a plane than say a natural wood or a cut stone
architecture in terms of its program. In a logically consistent manner, potentially in- wall. Also the very fact that the white planes
paper given at the R.I.B.A. in 1957, Sir dependent of its function and its meaning. carry a specific meaning related to a known

15
 style (the International Style), makes them ture was a primary consideration in the de- single coherent formal system directly from
less likely to take on new meaning. It should sign of the whole building. To focus on this, the physical fact. Rather they reinforce the
even be easier to reduce their existing mean- required a further shift in the primary con- intention that these marks in order to be
ing, as will be seen below, when they are ception of an environment; this time from understood first require disengagement of
placed in a different context. To this end, a concern merely for marking elements and the two systems from one another, an ac-
color and material will be seen in the argu- their meaning to a concern for their rela- tivity which takes place in the mind.
ment below to be used in House I as "mark- tionship in a formal structure. To force this
ing" devices. Traditionally, when white was shift in House I, the formal structure was in Such a marking of formal relationships, in
used, window mullions and hand rails were a sense over-stressed or over-articulated so the actual environment, has usually been the
painted black, and planes of primary or pastel that it would become a dominant aspect of extent of the architect's concern with formal
colors were introduced for aesthetic effect. the building. One means to over-stress such systems. But the present work takes one
In House I, white or black planes are used a structure was to suggest two simultaneous further step. If we analyze the nature of
simply as opposites in a formal structure structures which overlay and interact. These meaning in any specific context we realize
while grey or clear glass are considered as were based on a simple combination of two it has two aspects. The first is meaning
neutral. pairs of formal references: planes and vol- which is iconographic and symbolic and de-
umes, on the one hand; frontal and oblique rives from the relation of the form to some
A second aspect of the initial marking pro- relationships, on the other. reference which is external to it. For ex-
cess involved the structural elements - the ample, the particular juxtaposition of solids,
columns and beams. They appear initially to The two formal structures are marked by the columns, windows, and railings in Le Cor-
be rather conventional parts of a structural columns and beams. These are not deployed busier's Villa Savoye is intended as a direct
system. However, upon closer inspection in a regular pattern such as a columnar grid, recall of the super-structure of the modern
this is found not to be the case. It is actually which in such a condition could be seen as a ocean liners, and with it all the implications
not possible to determine how the structure neutral referent, nor are they to be seen as of the sea: discovery, newness, and ulti-
functions from looking at the columns and the residue of such a grid, but rather they mately man's conquest of nature. But un-
beams. All of the apparent structural appa- are intentionally placed in an apparently derlying that level of meaning there is
ratus - the exposed beams, the free stand- random order. This intention can be ex- another aspect, itself a potential source of
ing columns - are in fact non-structural. plained in the following way. In the first in- information, which conditions any icono-
When this is understood, a first step has stance, the space is conceived of as a layer- graphic interpretation; it is derived from,
been taken to unload, albeit in a very primi- ing or pi aiding (cross layering) of planes. and is in a sense inherent in the structure of
tive way, their structural meaning. While the The rectilinear columns and beams are the form. For example, the same juxtaposi-
apparent physical fact is the same whether placed so that they will read as a residue of tion of solids, voids, and columns at Poissy
they are load-bearing or not, their meaning these planes. Conversely, the round columns gives us cues to entry, sequence of move-
has changed because they are in fact not are used to mark the intersections of two ment, the relationship of open to closed
load-bearing, and thus the intention implied planes, which might possibly be read as space, of the center to the perimeter, and so
in their use in a particular location must now joined at this intersection, thus forming vol- forth. This information can be said to be the
be considered in a different way. Once one umes if the columns were square. The round product of the internal structure of form it-
has understood that they are not structural column prevents the possible interpretation self. While formal relationships can exist in
one must ask what are they? Why are they of columns as residual "corners" of vol- an environment at a real, actual level, where
where they are? Take them away, or change umes. In the second instance, the three col- an individual is aware of them through his
their shape, and what have you got? umns (a fourth is marked in the floor), be- senses - perception, hearing, touching-
cause of their particular disposition, also they can also exist at another level in which
It can also be asked, why go to all this trou- mark a diagonal system. They can be inter- though not seen, they can be known. This
ble? If the columns are supposed to be non- preted in the following way. If both pairs of second level is inherent in any envi ronment
structural, why not just cut them off at the round columns and beams were seen to span and is used by an individual whether or not
top so that we know immediately by the fact the entire space (Fig. 5) they would read, he is aware of it. This second level condi-
that they do not continue to the ceiling that despite the roundness of the columns, as tions the way we perceive the first level
they are not columns but merely a notation part of the frontal layering. By taking away by providing a structure for the visual cues
for some other purpose? But cutting the two columns, a round one in the space and which exist in the first level. And since it
columns short of the ceiling would in fact do one attached to the wall (Fig. 6) as well as has the capacity to be known, we must be
the opposite of what is intended. It would the portions of the beams connecting to concerned with how this happens. If we mark
give the column a further meaning by ob- these columns,. an implied diagonal is both these levels in the environment they
viously calling attention to itself as a non- created. can be explicitly perceived and understood.
supporting column, whereas it is supposed This is the third aspect of the work - a shift
to be merely one mark or a primitive element Thus the intention was to use the columns in focus from an actual structure to an im-
in a formal scheme. and beams to mark two systems without giv- plied structure and to the relationship be-
ing preference to either. Together the count- tween the two.
The second intention of this work called for erpoint of these two formal systems, the
taking these marks and deploying them in frontal planar layering and the diagonal vol- This second level may be thought of as a
such a way so as to make a complete umetric shift, overlaid and interacting with range of abstract and more universal formal
formal structure and to show that this struc- one another make it more difficult to read a regUlarities which exist in any conception of

16
physical space. These formal regularities primitive. It depends on an initial shift along posed to experience this intention does not
are universal in the sense that such formal a diagonal to create two implied square vol- depend entirely on the observer's particular
concepts as solid and void, centroidal and umes (Figs. 1 and 2). One square may be cultural background, his subjective percep-
linear, planar and volumetric are primitive seen as shifted out of the other or vice versa tions, or his particular mood at any given
notions which cannot be reduced and which so that the notations both for the plaid time, all of which condition his usual ex-
exist in a state of opposition in any spatial frontal layering and for the diagonal volumes perience of an actual environment, but rather
conception. This second level includes in can be seen as deriving from one, more basic it depends on his innate capacity to under-
addition to a set of irreducible formal regu- system. The diagonal is read as a resolution stand formal structures.
larities, the transformations of these reg- of the two directions in the plaid, or the plaid
ularities necessary to produce a specific is read as the result of the diagonal shift. Such a position introduces, as a primary
environment. Transformations may be de- Thus the deep structure is revealed only concern of architecture, the use of physical
scribed by such formal actions as shear, through an embedded relationship between form as a marking to produce, as it were, a
compression, and rotation, to produce a new two formal structures in the actual environ- new mental image of an environment differ-
level of formal information in any specific ment. Although one may perceive these two ent from that which we are actually seeing.
physical environment. Again the marking is structures in the actual environment, one is The deep structure, when it is combined with
used to signal the interaction between these unable to perceive the deep structure be- the perceptible physical reality, has the po-
two levels. The physical environment can cause of its existence in the envi ronment as tential, if it is structured in a precise fashion,
then be seen not only in its functional and an irregular gestalt. These actual structures to make available a new level of information.
iconographic dimensions but also in its thus have a common relationship in a deep The more this structure approximates a
formal one - as being generated from a structure which is not perceptible but which purely formal environment, the less tradi-
series of abstract formal regularities which can be understood after both structures tional the meaning it possesses, and thus the
may be described as a deep structure. These have been perceived. closer it is to an environment that might be
transformations and regularities have no a vehicle for such new information.
substantial existence but are merely a de- Any physical environment has this second or
scription of this second level of formal re- deep structural level, which not only has the To do this, form must be first considered to
lationships, in other words, a possible model capacity to convey information but does so be potentially separable from its existing
for an architectural deep structure. continually at a less-than-conscious level. It perception and conception, and second, it
exists without being consciously designed, must be considered as capable of changing
One means of making the deep structure in and there is a conceptual capacity within or raising the level of consciousness by pro-
a particular environment explicit is to force each individual to receive this information. posing a critique of the existing situation in
an individual to experience the environment Marking the deep structure in the actual en- architectu reo
as a notational system which has a recogniz- vironment may bring it to a more conscious
able relationship to a deep structure. This is level. As was said above, there is no reason
attempted in House I in the following man- or meaning intended in the use of this par-
ner. First, the series of formal relationships ticular formal strategy. The two overlaid
which are marked in the actual space (the systems are neither good nor bad in them-
parallel layers and diagonal volumes) create selves. They are intended merely to exem-
a contrast between actual space and implied plify the logic inherent in any formal struc-
space. This contrast makes one initially ture, and the potential capacity of that logic
aware of the presence of another level of to provide an area of new meaning.
formal structure. Second, the two sets of
formal notations which are discernible (one In summary, three shifts were attempted in
read as incomplete, the other asymmetrical) House I. Each concerned an attempt to sep-
because one can conceive of a symmetrical arate the actual physical environment from
and complete structure of formal regulari- its traditional relationship to function and
ties, are super-imposed. These notations meaning, to neutralize the influence of these
which are variations of the formula ABABA on the viewer. The first concerned the mark-
appear in the actual environment in the fol- ing of the elements of the actual environ-
lowing way. The first of these corresponds to ment; the second concerned the marking of
the formula A,BIAIA I (Fig. 3) and the second the formal structure in the actual environ-
to the formula A2B1AIB2A2 (Fig. 4); the mid- ment; the third concerned the marking of the
dle terms BIA I being common to both. When relationship of this formal structure to a deep
they are overlaid on one another, the under- structure.
lying structure is seen as compressed, but
when they are slipped apart in the mind, it Such a conception of design attempts to
reveals itself to be a simple symmetrical change the primary intention of architectural
structure. form from the perception of space to under-
standing the relationship of marks in that
The basis for creating this relationship of space to what is called here a deep struc-
actual structure to deep structure is quite ture. The capacity to understand, as op-

17
 ~:::::::::::::::~: _ : : :.' .' -. '. .: : :'-':.-.:..-_. : :..-._. . _.. =
~~.m_= . . . .:. . !mmtBffi I

IU !)-J o -

II .::::...... ':::.::::::" ....

I I I'
I
I I I I
,
:
I,
,, , :

I: iI I I I I: I I I I I
, I I' , I I
" "', , ,
!:llll il ! 11111111 .
I
,

:
I
I I
I I
,
I"
I :1

I , ' "
I

: ,
,
I I 1

I I I
I
: I I
H- ,
:I
::I I: III I ; : I I , I I

U-.LL.Ll..LJLL.Ll..L.L.Ll~1 ~:._.l~ ~D
: I ,
: Lt- I:,
' I

1 1

1III11111 I LLLiI..LLLl1..l.LL
1 I.L..L..L
ht-'-' ii,
!r...L..L..''nt+l-'--r-t-'
011 I1
1Tll1llr 0, 1'1,'1'" ;::::.

"
III I : !I 1-'-',--,-I'.L...l..
11.L...l..
11.L...l..\~.L.LL.L.L.L.L.L~+--t-T"t-'r'i"""r++--'-:,.n-
U1 ,+""";-r=r;!-t<---'i,c-=--=::
+;:J '--
It-t'I--+-L+-Y+-t-I,-:-,--+-'--+-'-+r--'-..........."
f-tL
I IT 1 !...~ i h--.:.-H+ll !-j:
t-H '----' ' '+-
I rl--'-r-t+---t-l: ~
I

/-r-+'-l'_+:7~~'-+:::-:'~=r=-~i='tt~=t=-t=tt"t=t=t=-~i='t't=t::t~~:t'U'9~~'~
,"
~
I l=hl~';:+ Itll=l~t+=
Ilj'fti
I I t! '+11
I:ljlftp
I I
H-+-I-+-,-.,.\-+- '-'
, I
I I II ITT I I I:
,-1
H-+-+--+-t-+-+--+-I-i-+-+--+-I-+,++
,
1-+-'+-+1-+,4:I -+r+
I
,-+-,-+-,+,-'-,-+"-'
1-i-1 --1 ,:++
I I:'
,-'I-'---,+-H
I I I
I
I : I

Lower Level Plan

o 5
I I I

West Elevation

111111111111111 ill

11111111

18
 [ 0
o I I III 1III I1II I I
I I

o I~
I II I I I : :
.....,...,..T"T"'l""...,........,..-r-T"'ii-,..,..!~,..,..I'...-i ~H ~ ~ lnmTlTnTmf=P=t+m=H"'=I1t -- :I I
I i H __ ~L!lft-;t-+-t-++-+-t-c:H-i1l_ _ H 11-:-'1+--++Hl1
o C

J_

! ,I I 'L'r-!
rg4-T
!

i I : I
I I lLL.L~LL.L.j
I I I 1

I I I- - I 1

i , ! I-i--
I

: I-i-;- I
I I I-f-r
I I I 1- 1'-+
I
o I o o

I I
:l=++ll1lr ::::::::::1-.L..L.Y- +} ~i~~ .. ~~ >t~t ;~Iu_;_m;_~_~_+-Hfc-
I I I
I I

rtt=I+++ _-LL-4I u -1 __
i 1

I
f-+-t-i+~H-H-f-+-t++-++1---++1~H-+++ 1-1 D f-t--L<t -j-HI

H-t-++++-H,-+-1-++++-H-'++-'++--H-++++-t+i'--+1-++_-+ -

Upper Level Plan -

Section

1=

I I
I I I -

I I I
'-

~
I I l-
I I '-

I I c I
19
 I IIL-__ ,----,

D
.--------===_11
" - - I_

Roof Plan

Section

. I

I I

20
I

21
 /~..
..."" : \

....
, .. .., ,
,.,'
., . "
i \'\\\
... : \

/,;;:;\~:::\ \\
'

,-
..,'

1\\\
j.//:\/ / ./"

:\ '" \ , .. /'
/"

, .. '

,,,,., .",,,,,
,,,'
..
'

.."'.... / ~
%' .,:'

\, ..... ;
..
/
", ,

,.,'
.,.,
\.,1
'. \ \:::::;:-1"/
\
i\L..
"
\ ,: ..
\L." 2

22