1957

In an address to the Royal Institute of British Architects upon being awarded its
annual Gold Medal, Sir John Summerson offered a compact formulation of
the theoretical basis of modern architecture. What distinguished the latter from
all its predecessors, according to Summerson , was its privileging of the program
as a source of unity, whereas the preceding periods of arch itecture had relied on
formal imitation. Summerson defined program as "a description of the spatia l
dimensions, spatial relationships, and other physical cond itions required for the
convenient performance of specific conditions."
As Summerson himse lf was to note upon repu blishing the paper more than
thirty years later, his argument represen ted a kind of last gasp of prewar English
modernism. The latter was a period dom inated by the MARS (Modern
Architectural Research) Group, the English wing of CIAM founded in 1933 by the
Canadian emigre architect and engineer Wells Coates. The group inc luded
Maxwell Fry, J. M. Richards, F. R. S. Yorke, Berthold Lubetkin (briefly), Denys
Lasdun , Leslie Martin , the three principals of the firm of Connell , Ward and
Lucas , the historian -critic P. Morton Shand, and other left-oriented London
practitioners. They succeeded in bringing the continental modern movement to
an eclectic and anemic English architectural scene, advancing a rational
approach to aesthetics and construction and promoting the use of materials
close to the industrial process. Although outside of several innovative houses
and schools they realized only a small volume of built work in the 1930s, they
effectively transformed the outlook of a generation of young British architects to
the ethical ideal of (in Fry's words) "a comprehensive architecture for the use of a
highly technical and reasonably integrated society."
During the war years a linear plan for Londo n was published by a German
emigre member of the MARS Group , Arthur Korn. Visionary in scope, it was
decisively rejected in 1943 in favor of a trad itional radiocentric plan by Patrick
Abercrombie and John Henry Forshaw. After the war ended the group continued
to exert a strong influence in the architecture offices of the county councils where
towns, schools, and housing for the new welfare state were then being designed .
At the same time, the objective aesthetic which Yorke had celebrated in The
Modern House in England in 1937 began to look increasing ly sterile. Richards , a
founding member of the group, advocated a return to a less rigid form of
architecture in his book Castles on the Ground ( 1946) and together with his
fellow editors in the pages of the Architectural Review. The Festival of Britain,
114-19 which was stB.ged in 1951 on the centenary of the Crystal Palace and coi ncided
135-36 with the MARS Group 's hosting of CIAM's eighth meeting in Hoddesdon , further
relaxed the prewar rigorism . It remained for the group's postwar inheritors to find
a new banner. Peter and Alison Smithson, who joined the MARS Group in 1951
and presented their work under its name at CIAM 9 in Aix-en-Provence , finished
181-83 by disbanding the group five years later, having recons tituted as Team 10.
The late 1950s were "the moment," as Summerson would write in hindsight,
"when the thought of my generat ion-the MARS Group generation-lost touch
with the real world." As such, a theory giving credence to architecture's ability to
carry out social aims through rational means "was not the conclusio n whic h
history requ ired ." Despite its opacity to the ferments in English architecture by
this date , though, Summerson's theory remains a register of the asp irations of a
generation of architects and an insightful analysis of the vicissitudes of
rationalism in the history of arch itectural ideas from Perrault through Le Corbusier.

From R.I.B.A. Journa l, June 1957, pp. 307-10. Republ ished in John Summerson,
The Unromantic Castle and Other Essays (London: Thames and Hudson , 1990),
pp . 25 7- 66. Courtesy of the author .

226
The Case for a Theory of Modern Architecture
John Summerson

Ever since the modern movement got onto its feet , questions have been asked about
what it stands on. An association of some kind between what is vaguely called "theory "
and what is vaguely called "modern architecture " continues , I believe , to be a topic
frequently debated , and I am told that teachers in some of the schools feel a practica l
need for some sort of theoretical formula as a means of introduc ing students to the
principles of modern design. Hence this paper , which offers nothing new but is simply
an investigation - an attempt to discover whether there does exist any basis of
principle applicable to modern architecture , different from the bases applicable to any
other architecture , or alternatively whether such a basis can be abstracted out of
prevailing practice and ideas .
I should like to take this alternative first because it offers an obvious pr ima facie
case. I think it is a bad case but it is necessary to put it up in order to put it down. Modern
architecture exists to the extent that there are plenty of buildings which everyone in this
room would immediately classify as products of the modern movement on the basis
of certain recurrent formal arrangements and relationsh ips. Embarrassed as we are by
the use of such expressions as "the modern style ," "manner ," or "idiom " there is
positively no denying the consensus of characterization. Modern architecture is there
all right. Furthermore , closely associated with this architecture are a number of ideas-
ideas expressing modernity in one sense or another , nearly always either by analogy
with the past or by analogy with some other activity than architecture . The architects
who design the buildings tend to quote and promote these ideas and it would be very
difficult to show that this complex of architecture and ideas is anything short of valid
in relation to present-day conditions . There is indeed no other complex of forms and
ideas which seriously rivals it. Now, in a situation like this , it may be argued , it should
be possible to put together a theory of architectu re without very much difficulty . It is
simply a question of two rather prolonged exercises in analys is and synthesis. First, of
assembling the ideas , examining their common trends of meaning , and reaching a
series of general concepts. Second , of abstracting formal characteristics from a select
repertory of modern buildings, eliminating merely mod ish elements , and providing a
grammar of form. It would then only remain to illustrate how the forms embody the ideas .
The whole exercise would, it may be supposed , add up to something like a Palladio
of modern architecture , a pedagogical reference book not in any way restricting further
development but consolidating the achievements of modern architecture , clarify ing
them and providing a departure platform for new experiments.
Such is the prima facie case for a specific theory of modern architecture . I have
tried to make it sound plausible but of course it is hopelessly gimcrack . Only imagine
for a moment the task of isolating characteristically modern forms from whole buildings.
Only imagine the horror of stirring around in the rag-bag of aphorisms , platitudes , and
fancy jargon and trying to determine their common trend and resultant meaning . The
imagination boggles , and when it does that is a sure sign that something stupid is be ing
attempted. So let us leave this whole enterprise and look for firmer ground on which to
start our inquiry.
We had better consider first what is in our minds when we think about a "theory "
of architecture. The elementary meaning is a conspectus of knowledge in any
particular field. A theory of architecture may be , like many of the treatises of the

227
eighteenth century , purely encyclopedic , without any explicit philosophical orientation
at all. It may be, like Julien Guadet's famous work ,1 a series of discursive studie s of
types and elements , in lecture form , within a closed tradition whose validity is taken for
granted. Or it may be of that curious kind represented by John Belcher 's well-known
book 2 of half a century ago in which a list of interesting words is compiled (scale , vitality ,
restraint, refinement , etc .), each providing the title for a short essay which gives it a glow
of meaning , without ever reaching down to fundamental concepts at all.
But I suspect that what is in our minds when we talk about architectural theory now
is something both less extensive and more profound than these-a statement of
related ideas resting on a philosophical conception of the nature of architectur e- in
short , principia . Since Alberti wrote his Oe Re Aedificatoria in the middle of the fifteenth
century there have been a certain number of statements of this kind , though not (when
all derivatives are written off) quite as many as you might think and few, merc ifully , as
difficult to understand as Alberti. They are usually to be found lodged in some section
of an encyclopedic work (e.g ., Alberti , lib. ix) or forming introductions to a course of
lectures (e.g ., Durand) or, more rarely , as independent polemical essays (e.g .,
Laugier). It is worth emphasizing that to state the principles of architecture does not at
any time take very many words . It is the demonstration by historical instance and the
exposition of grammar which fill up the tomes. This evening my quest is for statements
of root principle .
If we review the statements of principle which have attracted attention in the
course of the last five hundred years we may be struck by the fact that they are much
more easily related to each other than they are to the architecture prevailing at the time
they are written ; which suggests that just as architectural style has evolved from
generation to generation, each changing the favored accentuation of the last, so
architectural thought has developed phase by phase with its own dialectic. There has
been, in fact , an evolving process in theory just as there has been in style , and the two
processes have not made anything like the same pattern. Each has been and is in fact
autonomous , to the extent that it would be possible to write a history or architectural
theory without reference to a single actual building and even a history of architectural
style without a single reference to architectural theory - though I am not suggesting
that anybody should try.
The actual relationship of architectural theory to architectural production at any
given time is problematic . It is perfectly possible for a new idea to be announced ,
cherished by one generation , turned upside down by the next , and only in a third to be
validated in architectural designs. Something of the sort happened with the eighteenth-
century idea of rational architecture , to which I shall refer later on . On the other hand
it is possible for architectural style to be revolutionized without so much as one corollary
gesture on the plane of theory. Who has ever had a more powerful effect on architecture
than Michelangelo? Yet his effect on the theory of architecture was nil. So we must bear
in mind about theory that it is a historical process with a life of its own in its own medium
or words and that there is no question either of principles being abstracted wholly from
practice or of practice being necessarily a reflection of theory. This makes a pretty big
hole in the proposition called "a theory of modern architecture." But it brings us nearer
to a realistic view of what we are discussing.
In the present century a fairly large number of books - I make it about 120-- have
been written about the nature and principles of architecture. Up to 1925 there was a
modest issue of one book a year but in 1926 at least seven books (English , American ,

228
and French) appeared , though oddly enough not one of these recognized that any
fundamental changes were taking place in architectural thought. The general tendency
before 1927 was to rewrite the principles then stagnating in the Beaux-Arts tradition
and to comment on them in essay style, but I do not know of a single book which
investigated those principles historically or attempted to evaluate them philosophically
(there is one outstanding exception which I will mention in a moment). After 1927 books
stating the modern point of view began to appear . Between that year and the present
there have been statements from Behrendt , Lurc;at, Taut , Cheney , Platz, Hitchcock ,
Duncan , Gropius , Moholy-Nagy , Teague , Giedion , Fry, Saarinen , and Zevi , to mention
only some of those who have produced books ; to collect the statements appearing in
the form of papers , articles , and catalogue introductions would be a mighty exercise
in bibliography. The general character of all this writing is enthusiastic and propagandist.
The authors tend to start with a belief in the new architecture and to write around their
beliefs supporting them by picturesque and forceful analogies . Only rarely does one
detect a realization that architectural thought is a continuing activity sui generis in which
what is new must be distinguished by criticism of the past. But there are a few books
of great penetration and to some of these we must now pay attention.
I suppose nobody will doubt that Le Corbusier 's Vers une architectur e3has been
the most consequential book on architecture written in this century. Published thirty-
four years ago , it is still widely quoted and quite frequently read . It is not and does not
claim to be a theory of architecture. It is a series of critical essays , reprinted in the order
in which they first appeared in L 'Esprit Nouveau , starting in October 1920. In the whole
course of these essays nothing absolutely new is proposed in the way of architectural
principle , but a great deal that had been forgotten is brought into the light of the present
and exhibited with a quite uncommon flair for paradox. I think it would not be an unfair
generalization to describe Vers une architecture as a critique of the French rational
tradition - a critique marking a new phase in that always vigorous and controversial
zigzag of thought. This French rational tradition is not, of course , the Beaux-Arts
tradition personified in Guadet , for which Le Corbusier expresses a good deal of
contempt. It is, on the contrary , the tradition first of Jesuit intellectuals in the early
eighteenth century , later of rebels and academy haters , and indeed "tradition ,,,which
suggests a handing down of embalmed principle, is not at all the right word. It is a
historical process advancing by a series of contradictions and reassessments , of
which latter Vers une architecture is the most recent. As I am going to suggest that this
rational process is still a vital element in the contemporary theoretical situation perhaps
I may briefly explain what I understand it to be .
4
It all hinges on the ancient body of Mediterranean beliefs , restated by Alberti ;
and the hinge occurs in the age of Descartes. One could date its origin rather
pedantically from Perrault's critique of Vitruvius . It is picked up in the eighteenth
5

century by the Abbe Laugier 6 whose two essays were the standard statements for half
7
a century. But in 1802 Laugier was attacked as a muddler by Durand who presented
his students at the Polytechnique with an altogether tougher and more materialistic
case . So far , the argument had proceeded against a background of belief in classical
antiquity , but then , fifty years later , Viollet-le-Duc took up a new position , still rationalist
8

but transposing the background from classical to medieval antiquity and purporting to
show that the thirteenth century was the sole repository of rationalist principle . Viollet-
le-Duc was , directly or indirectly , the inspirat ion of many of the pioneers of the modern
movement: Berlage , Horta, and Perret among them .

229
 This is, of course, a grotesque simplification indicating only some of the more
obvious peaks in a great range of argument. Many more names should go in, not all
of them French: there is Cordemoy ;9 there is the mysterious Venetian rigorist Lodoli 10
whose influence is hard to estimate because he never wrote anything down ; there is
Frezier, 11 the engineer; there is half-French Pugin .12 Again in rough caricature, one
could sketch the process like this. Perrault said antiquity is the thing and look how
rational; Lodoli seems to have said rationalism is the thing, down with antiquity ; Laugier
said up with primitive antiquity, only source of the rational ; Durand said down with
Laugier , rationalization means economics ; Pugin said down with antiquity , up with
Gothic and look how rational; Viollet-le-Duc said up with Gothic , prototype of the
rational. Eventually a voice is heard saying down with all the styles and if it's rationalism
you want , up with grain elevators and look, how beautiful!
Well , now , in this process, which I take to be the main heritage of the modern
theorist , there are certain essentials which hold their own throughout. At the bottom of
it all is the axiom that architecture is an affair of simple geometric form-regular solids 13
and their elementary divisions . This is inherited from Italian tradition and has a peculiar
history of its own , passing from the quasimedieval numerology of Alberti to the visual
objectivity of the Cartesian world and on to the empathetic apprehensions of the
revolutionary school of Boullee and Ledoux. In some form or another it is always there .
Then there is the rational issue whose course through the eighteenth and
nineteenth centuries I have already sketched.
But there is also the question of antiquity and the measure of its authority and one
very important thing about the whole rational process is that it tends to exclude antiquity
as an absolute authority. However, antiquity was obstinately there all the time . Only the
theorists who never designed anything, like Lodoli and Laugier, could be really tough
about antiquity. Those who designed had, in one way or another, to admit it for the
important reason that the forms of classical antiquity or (in the nineteenth century)
medieval antiquity , provided something which is essential to the creative designer-
a bulwark of certainty, of unarguable authority on which his understanding leans while
his conception of the building as a whole , as a unity, takes shape. The most interesting,
indeed the dominating question, in a search for the modern principia is: where, if not
in antique forms, or some equivalent substitute, is the source of unity?
Le Corbusier provides no answer to this in Vers une architecture . There is no reason
why he should. The book is really nothing but a lightly etched reminder ("Trois rappels "
is the title of the first chapter) of the main content of the rational process and it contains
few ideas which could not be traced back into the line from Perrault to Viollet-le-Duc.
Le Corbusier 's designs, let me say in parenthesis, are a different thing altogether.
I have already said that architectural theory and architectural style are things apart-
each with its autonomous life, and this is nowhere more obvious than in the case of the
author of Vers une architecture. His conception of theory is simply the solid intellectual
platform, with foundations deep in the past , on which he stands to do something which
has nothing to do with the past whatever. Le Corbusier has not reasoned himself into
those architectural conceptions which have so profoundly influenced the expression
of modern building. Nor is there any mystery about how they have come about, for it
is by now an accepted fact of contemporary art history that Le Corbusier's vision in the
early days was that of the modern painters-the school of Picasso, Braque, and Leger ;
that after they had discovered the power of converting the commonplace into pure
conceptual painting, Le Corbusier discovered the power of composing the

230
common places and crude ingenuities of industrial building into equivalent architectural
realities . But there is nothing in Vers une architecture about that ; and if the pictures of
the author's own works were eliminated from the book it might eas ily be construed as
foreshadowing some frozen neoclassicism not far removed from that of Auguste
Perret.
Obviously , the only thing about Vers une architecture which helps us to envisage
a case for~ specifically new theory of architecture is the reillumination of principles
already established. If we were to argue from the example of Le Corbusier alone we
might well conclude that the theoretical process stemming from antiquity and the age
of reason was , in one form or another, the theory appropriate to the modern movement
in architecture. That may indeed be the case. But we cannot leave the matter there for
in another quarter altogether there have been theoretical inquiries of considerable
importance and entirely different character. I am thinking of the sphere of thought
represented by the Bauhaus .
Bauhaus thought has been pretty copiously manifested : in Gropius 's own writings
about Gropius and the Bauhaus and in the Bauhausbucher of the twenties. But for
anything like a systematic exposition of the Bauhaus theory the most significant book
is Moholy-Nagy 's The New Vision: From Material to Architecture , based on lectures
given at the Bauhaus in 1923-28. These lectures were given after Vers une architecture
had been published but they owe nothing to it, nor to the Esprit Nouveau circle from
which it emerged . Moholy, of course , was a totally different kind of person from Le
Corbusier-he represents in a fundamental sense that phenomenon of our time , the
displaced person. Le Corbusier's Swiss background was happy and stable . Moholy 's
Hungarian background was far otherwise and when Le Corbusier was bui lding a luxury
villa on Lake Geneva, Moholy was pitched into a hideous and incomprehensible war
without even the consolation of being on the winning side. 14 It is not surprising that
whereas Le Corbusier turns naturally to a reassessment of the past , Moholy turns back ·
on it altogether. I do not know how conscious he was of turning his back on Le Corbusier
but his book is in some respects a negation of Vers une architecture. Admittedly he
states what he calls the "basic law" of design as the obligation "to build up each piece
of work solely from the elements which are required for its funct ion ,"15 a statement which
is the genuine old-style rationalist article (it could well be a quotation from Laugier) , but
he then instantly declares that the basic law has limitations and he proceeds to search
for an ultimate authority .
This ultimate authority is of course likely to be the source of unity of which I have
already spoken. It is the something occupying the place wh ich used to be filled by
"antiquity ." What is it? Moholy says it is "biological. " The artist's freedom , he says , is "in
the last analysis determined biologically. " The words "biological ," "biologically " crop
up again and again throughout the book. "Architecture ," he says , "will be understood
. . . as a governable creation for mastery of life, as an organic component in living. " "The
standard for architects . . . will revolve around the general basis , that of the biologically
evolved manner of living which man requires ." And , finally , "architecture will be brought
to its fullest realization only when the deepest knowledge of human •life as a total
phenomenon in the biological whole is available ." 16
This preoccupation with biology and with the organic is obviously a very important
issue in our investigation . The word "organic " especially has had an almost magical
significance for architectural writers ever since Louis Sullivan wrote of it fifty odd years
ago as "a word I love because I love the sense of life it stands for , the ten-fingered grasp

231
of things it implies. " 17 That is not a very scientific statement but I have not yet found,
among the many writings about organic architecture, any statement that is. Yet it is
constantly used as an ultimate , as if organic values (whatever they may be) were
absolute values.
Moholy's treatment of the biological idea is more interesting than most since he
presses it harder and, in doing so , shows , in one direction, its perilous inefficiency.
When he declares that the artist's freedom is "in the last analysis determined
biologically " he leads us surely to a determinism which begs the whole question.
Moholy would like to construct a theory which is a perfect description of practice-
which coincides with practice. He cuts himself off from inherited theory and postulates
a new theory which would fit the biological (let us say pyschophysical) needs of man
like a glove. I suppose , if the most far-reaching implications of cybernetics were
realized , if the artist's functions were at last to be explicable in mechanistic terms , some
such theory might be arrived at. But that is such an awfully long way off that it is hard ly
worth considering in relation to the modern movement now in course of evolution; and
in any case I doubt if anybody yet sees the determination of the artistic needs of soc iety
as even a remotely possible point on the scientific horizon . Notwithstanding the fine
perceptions and immensely valuable practical suggestions contained in Moholy 's
book, it seems to me that his insistence on the biological is a premature and pure ly
verbal closure of the subject of modern architectural theory . It gives nothing to hold
onto but this elusive myth of "biological " finality .
Those who have written about "organic " architecture have usually gone in a rather
different direction from Moholy's . Frank Lloyd Wright's use of the expression "organic
architecture" is generally considered to be his own emotional tag for all fine, free , and
humane architecture , but especially for that of Frank Lloyd Wright. Behrendt, Steinmetz ,
Saarinen , and others have speculated on the "organic" in desultory philosophizings.
Bruno Zevi has investigated various recent uses of the word and in his book, Towards
an Organic Architecture , devotes a whole chapter to "the meaning and scope of the
term organic in reference to architecture." He does not discover any evidence of
strikingly profound thought on the subject nor does he commit himself to any precise
meaning. But he does write off various spurious or outmoded interpretations and , at
the end of his study , he does , in a single, rather casual remark , hit what I conceive to
be the nail exactly on the head. He says that the organic conception of architecture is
based "on a social idea and not on a figurative (I take it he means formal) idea ."18 That
rather wide interpretation would , I suspect, command almost universal agreement.
Zevi throws out his comment as if its truth was pretty obvious and I suppose it is,
but I want to underline the proposition and see how it relates to the picture of the
developing theoretical process which I have outlined. I suggested a few moments ago
that although the rationalist writers of the eighteenth and nineteenth centuries tended
to exclude antiquity as the ultimate authority, antiquity remained insistently there as the
source of unity, the focus at which the architectural design was realized . Where, I
asked , if not in antique forms , can the source of unity lie? Zevi's remark points to the
answer. The source of unity in modern architecture is in the social sphere, in other
words in the architect's program .
From the antique (a world of form) to the program (a local fragment of social
pattern): this suggests a swing in the architect's psychological orientation almost too
violent to be credible. Yet, in theory at least, it has come about; and how it has come
about could very well be demonstrated historically. First the rationalist attack on the

232
authority of the antique ; then the displacement of the classical antique by the medieval ;
then the introduction into medievalist authority of purely social factors (Ruskin); then the
evaluation of purely vernacular architectures because of their social realism (Morris) ;
and finally the concentration of interest on the social factors themselves and the
conception of the architect 's program as the source of unity- the source not precisely
of forms but of adumbrations of forms of undeniable validity. The program as the source
of unity is, so far as I can see , the one new principle involved in modern architecture.
It seems to be the principle which can be discerned through the cloud of half-truths ,
aperc;us, and analogies which is the theoretical effluent- not a very nice word , I'm
afraid-o f the modern movement.
Whether you accept this statement as a basic principle and a specif ically modern
principle depends upon a number of things. Mainly, there is the question , what a
"program " is. A program is a description of the spatial dimensions , spatial relationships ,
and other physical conditions required for the convenient performance of specific
functions. It is probably impossible to write out a satisfactory program withou t a certain
number of architectural relationships being suggested on the way and the character
of these relationships may well be something different from the relationships in a
predetermined stylistic discipline . The chief difference is that they involve a process
in time . It is difficult to imagine any program in which there is not some rhythmically
repetitive pattern - whether it is a manufacturing process , the curriculum of a schoo l,
the domestic routine of a house, or simply the sense of repeated movement in a
circulation system . Of course , this pattern does not dictate a corre sponding pattern in
the architect's plan or anything crude like that , but it does sanction relationsh ips which
are different from those sanctioned by the static , axially grouped dominants and
subordinates of the classical tradition-different , but carrying an equivalent authority.
The resultant unity can , I think , quite reasonably be described as a biological or organic
19
unity , because it is the unity of a process. Moholy-Nagy and after him Giedion 20 would
see it as a space-time unity , and you will recall Giedion 's brilliant analogies between
modern architecture and the concepts of modern physics on the one hand and the
Picasso revolution in modern painting (involving the concept of simultaneity) on the
other . Not that such analogies prove anything ; and there is always the danger that they
may seem to prove far too much; they are phantasms of the Zeitgeist. The actual reason
why the principle embodied here is new is this. It is only in the past half-century or so
that the program has ceased to be evaluated merely quantitatively and has come to
be evaluated qualitatively . This has to do with the fact that programs have become
more complex , more challenging , and therefore more susceptible to qualitative
generalization and evaluation . It has also to do with very much wider issues involved
in the social revolutions and reorientations of our time .
If we accept this principle - unity deriving from the program - as truly a basi c
principle of modern architecture , how does it look when lined up with the inherited
principles which we found that Le Corbusier had reilluminated in Vers unearchitecture?
Here comes the crux of the whole matter. The conceptions whic h arise from a
preoccupation with program have got , at some point , to crystallize into a final form and
by the time the architect reaches that point he has to bring to his conception a we ight
of judgment , a sense of authority and conviction which clinches the whole design ,
causes the impending relationships to close into a visually comprehens ible whole . He
may have extracted from the program a set of interdependent relationships adding up
to a unity of the biological kind , but he still has to face up to the ordering of a vast number

233
of variables , and how he does this is a question. There is no common theoretical
agreement as to what happens or should happen at that point. There is a hiatus. One
may even be justified in speaking of a "missing architectural language. " Gropius 2 1 has
stated the difficulty as the lack of an "optical 'key' ... as an objective common
denominator of design "-something which would provide "the impersonal basis as a
prerequisite for general understanding," which would serve "as the controlling agent
within the creative act." That is a precise description of the functions served by antiquity
in the classical centuries! The dilemma is really an enlargement of the flaw already
apparent in mid-eighteenth-century theory-the flaw that while antiquity was eliminated
as an absolute , nothing was introduced which took its place as a universally accredited
language of architectural form .
The flaw seems now to have widened into a veritable dilemma. Can it be resolved?
Well , I can think of two possible approaches to its resolution. The first involves an
extension of the rationalist principle into the sphere of engineering, and the second
involves a reconsideration of the geometrical basis and limitations of architecture.
Let us take the engineering question . The engineer is the heir to the basic tenet
of the old rationalism--economy of means in construction . So long as traditional
methods prevailed the architect could keep his eye on the ball, or at least persuade
himself that he was doing so ; but with the development of the science of the strength
of materials and the application of mathematics to design he was rapidly overpassed
by the engineer . The engineer ran away with the rationalist ball. It is no use pretending
that we can lop off this issue as a stray limb of the rationalist process which has got
outside the scope of architecture , because if we let the rationalist principle go modern
theory collapses in a heap. No. It is necessary to declare that no theory of modern
architecture can be logically complete which does not postulate the collaboration ,
immediate or remote, of architect and engineer ; and here collaboration must stand for
the design of components in factories as well as the personal achievements of a Nervi
or a Candela .
But let us be clear about what the engineer's role really is and how different it is
from that of the architect. For the architect, the source of unity for his design is, I have
suggested , the program. The engineer seeks unity in another way and another
direction altogether . He seeks it within one component-even if it is a very complex
component comprising the whole sectional trace of a large building. And it is a unity
of interdependent calculable issues adding up to a total whose criterion is performance.
His search for finality and the architect's are as wide apart as they can be. It would be
altogether too facile to suggest that they are even complementary . Nevertheless , a
whole view of architecture must necessarily extend to this latest metamorphosis of the
rationalist process in the hands of the engineer.
The idea can be and sometimes is upheld that the engineer, as a result of his
enforcement of the rationalist principle , invents forms and formal arrangements which
the architect then absorbs into his vocabulary of expression and uses , sometimes in
a strictly engineering way- and sometimes not.22 This certainly happens. But the
engineer is concerned strictly with components and although he may contribute
significant inventions he cannot contribute a continuously related system of inventions
-i.e. , a language.
Thus the engineering issue does not wholly resolve the dilemma of modern
architectural theory, and so we turn to the ancient axiom that architecture is fundamentally
concerned with the regular solids and simple ratios. It is getting to have an old-

234
fashioned look, this axiom, especially in an age which has discovered geometries
other than Euclidean. Moholy-Nagy was eager to go behind the axiom to "biological
assumptions. " Mr. Banham , in a recent article ,23 has offered us the attractive red
herring (I think it's a herring) of topology. In the field of practice, unfamiliar and complex
forms are cropping up. Candela has built a concrete church in which all the surfaces
are hyperbolic paraboloids. But surely the axiom stands as an overall absolute
necessity. Even if plans wriggle in the wildest of "free" curves , even if engineering
science introduces forms of great precision but visually unreadable complexity , we
shall always seek to read through the complex to the simple , to seek the assurance of
those simplicities which must be implied even when they are not stated . Very well. On
this principle of geometrical absolutes it is possible to erect systems or disciplines to
guide the architect toward that final ordering of form which he must achieve. Of these
systems the most celebrated is Le Corbusier's Modular. But the Modular , like any other
apparatus of the kind , is a system of control , not of expression (Le Corbus ier says this
as clearly as it could be said). It is not a language. And if I say that in my opinion the
erection of proportional disciplines - purely intellectual contrivances - does bring the
principia of modern theory into satisfactory relationship to each other and to actuality ,
it may well be objected that this theory excludes almost everything that has been most
valued in the art of architecture as a means of expression in the past three thousand
years . In answer to that , I have two things to say. The first is that if you accept the
principle that the program is the source of unity, the crucible of the architect's creative
endeavor , you cannot postulate another principle , another crucible, at the other end
of the designing process to satisfy the architect's craving for conspicuous self-
expression. You cannot have it both ways. You certainly cannot have two sources of
unity. Either the program is or it is not the source. It is part of my case for a theory of
modern architecture that it is the source. If you do not accept this case , I think you must
consider whether, after all, architectural theory does not stand very much where it
stood in 1920, or 1800, or even 1750, and whether the position of an architect who is
concerned about expression or style is not that of a man feeling his way back to
classicism or neoclassicism, or, to put the finest possible point on it, crypto-
neoclassicism .
The second thing that I would say is that it is quite possible that the missing
language will remain missing, and that in fact the slightly uncomfortable feeling which
some of us have that it ought to exist is nothing but the scar left in the mind by the violent
swing which has taken place in the lifetime of one generation from an old order of
principles to a new.
I have tried to demonstrate that in the light of all that has been written on
architecture in the past thirty years a specifically modern theory of architecture does
exist , and that it exists not as an arbitrary invention of our time but as a new stage in
the long evolution of theory since those forgotten men whom even Vitruvius knew as
the Ancients. Modern theory is part of the history of ideas . It is, I believe , only as the
history of ideas that it can be taught. The main thing is to get that history right and to
get it clear. It would be an outrageous assumption on my part that I have done either
this evening, and I have certainly been more speculative than historical. I have
presented what I feel may prove to be an exceedingly vulnerable thesis. I thank you
for listening to me with such patience.

235
Notes
1. J. Guadet, Elements et theorie de /'architecture (5th ed. , 1909).
2. J . Belcher, Essentials in Architecture ( 1907).
3. Le Corbus ier, Vers une architecture (1st ed. , 1923) . Trans. as Towards a New Architecture (1927
and 1946).
4. The De Re Aedificatoria was wr itten about 1450. The standard English translation is that of G .
Leoni ( 1726 and 1739); it is available in a reprint (ed . J. Rykwert, 1955).
5. C. Perrault , Les Dix Livres d 'architecture de Vitruve(1673), bk . V, chap . i (note 1) and bk . VI,
chap. v (note 8). These references are quoted by F. Algarotti, Saggio sopre J'architettura (vol. 3 in the
Opere , 1791) as predict ions of the rationalist attitude.
6. M . A. Laugier, Essai sur /'architecture ( 1753) and Observations sur /'architecture (1765) .
7. J. N. L. Durand , Precis des ler;ons donnees a /'Ecole Polytechnique (1802).
8. E. Viollet-le-Duc, Entretiens sur /'architecture ( 1863- 72).
9. De Cordemoy, Nouveau traite de toute / 'architecture ( 1714) .
10. For Lodo li see E. Kaufmann, Architecture in the Age of Reason ( 1955), pp. 95- 99.
11. Frezier, Dissertation sur Jes ordres d'architecture ( 1738).
12. The True Principles of Pointed or Christian Architecture ( 1841) is a plea for Goth ic as a rational
style.
13. Dr. J . Bronowski warns me that "regular solids" in a strict sense includes figures never regarded
as basic to architecture and excludes others which are. Time forbids reconsideration but, with this
warning, my meaning will not, I think, be misunderstood.
14. For biographies of Le Corbusier and Moholy-Nagy, see M. Gauthier, Le Corbusier, ou
/'architecture au service de l'homme (1944), and S. Moholy-Nagy , Moholy-Nagy: Experiment in Totality
( 1950).
15. The New Vision, p. 54.
16. Ibid., pp . 159-60 .
17. Kindergarten Chats (1947). Elsewhere (p. 47) Sullivan uses "organic" to mean "the part must have
the same quality as the whole," an idea which goes back to Alberti .
18. B . Zevi , Towards an Organic Architecture(1950) , p . 76.
19. The New Vision, p. 163.
20 . Space , Time and Architecture (3rd ed ., 1954), p. 432.
21. Scope of Total Architecture (1956) , p . 49.
22. For a discussion of this point by an engineer, see Ove Arup , "Modern Architecture: The Structural
Fallacy ," in The Listener , 7 July 1955.
23 . "The New Bruta lism ," in Architectural Review, December 1955 .

236