Affiles

Annals of the Architectural Association School of Architecture

12 O b i b l iot h e e k
BOUWKUNDE
Contents

3
TRANSLATIONS FROM DRAWING TO BUILDING
Robin Evans
19
BRANSON COATES: THE METROPOLE RESTAURANT AND CURRENT WORK
28

ZAHA HADID: OFFICE BUILDING I N BERLIN

THE CRYSTAL PALACE

REVIEWS
60
B U I L D I N G IMAGES Andrew Holmes: Addition
Thomas C r o w
65
POINT DE FOLIE - M A I N T E N A N T L ' A R C H I T E C T U R E Bemard Tschumi: La Case Vide - La Villette, 1985
Jacques Derrida

76
SITE U N S C E N E Peter Eisenman: Moving Arrows, Eros and Other Errors
John Whiteman
85
F E A R F U L SYMMETRY Figurative Architecture: Five Dublin Architects
David Gray
89
T H E METAMORPHOSI S OF T H E C O U R T Y A R D Houses of Sinasos, Cappadocia
Kalliope Kontozoglou

93
T H E K I N G OF C O N C R E T E Sir Owen Williams
Alan Harris
98
P A O L O Z Z I N O W Eduardo Paolozzi: Three London Exhibitions, 1986
Royston Landau
102
P R O D U K T I F O R M E T A M O R P O L I S NATO: Gamma-City
Brian Hatton
107
BOOK REVIEWS
Mask of Medusa: Works 1947-1983 by fohn Hejduk
Authentic Decor: The Domestic Interior 1620-1920 by Peter Thornton
C. R. Ashbee by Alan Crawford
The Bauhaus Reassessed by Gillian Naylor
The Secret Life of Buildings by Gavin Macrae-Gibson
112
CONTRIBUTORS

1
Tcirew Saint

ssociation

ides postage).

TRANSLATIONS
FROM
DRAWING TO BUILDING
Robin Evans

K I translate is to convey. It is to move something v/ithout come to pass in which, while on the one hand the drawing might be
altering it.' This is its original meaning and this is what vastly overvalued, on the other the properties of drawing — its peculiar
happens in translatory motion. Such too, by analogy with powers in relation to its putative subject, the building — are hardly
translatory motion, the translation of languages. Yet the substratum recognized at all. Recognition of the drawing's power as a medium
across which the sense of words is translated from language to turns out, unexpectedly, to be recognition of the drawing's distinct-
language does not appear to have the requisite evenness and continu- ness f r o m and unlikeness to the thing that is represented, rather than
ity; things can get bent, broken or lost on the way. The assumption its likeness to it, which is neither as paradoxical nor as dissociative as it
that there is a uniform space through which meaning may ghde with- may seem.
out modulation is more than just a naïve delusion, however. Only by Before embarking on the investigation of drawing's role in architec-
assuming its pure and unconditional existence in the first place can ture, a few more words might be spent on language; more particu-
any precise knowledge of the pattern of deviations from this imagin- larly, on the common antilogy that would have architecture be like
ary condition be gained. language but also independent of it. A l l things with conceptual
I would like to suggest that something similar occurs in architecture dimension are like language, as all grey things are like elephants.
between the drawing and the building, and that a similar suspension A great deal in architecture may be language-like without being
of critical disbelief is necessary in order to enable architects to per- language. Some might say that the recent insistence that architecture
ind Blue is a language is only the last wave of a persistent verbal tide eroding
nvas,
form their task at all. I would like to suggest also that, while such an
enabling fiction may be made explicit, this has not been done in archi- vision, bedevilling our ability to see without language to guide our
tecture, and that because of this inexplicitness a curious situation has eyes (Fig. i). In the words of the poet Paul Valéry, used as the title of a
recent biography of an American artist, 'seeing is forgetting the name
i5-9. /. T h e C r i t i c Sees, by JasperJohns, 1961. Sculpmetal on plaster with glass. of the thing one sees'.- Can we really be certain? Might not this purism
(Seep, 93.) 3'Ax6'/4x2'/ii inches. be in danger of becoming a ridiculous piety? Having recognized that

A A F I L E S 12 3
words effect vision, we are under no moral obligation to expel tliem responsible, smaller and less predictable, worth less but better, as the
f r o m it, even if tlie expulsion could be achieved. It is understandable hope would be, would it not, that in giving up grandiose pretensions
that, in the interest of the integrity of our art, we should imagine it to represent and define the social world in both its imaginative and
contaminated by other forms of communication, just as it is under- active aspects (a project the unlikelihood of which is comparable to
standable that, in the interest of its aggrandizement, we should the unlikelihood of compiling a legal code that is also a good novel —
imagine it comparable to language. But this is only to offer excuses for an ambition that can only be confounded in practice) architecture
the possession of incompatible ideas. may, by contraction and concentration, consthute itself anew? WeU,
Fastidiousness about the purity of vision arises fro m a fear that all this consohdation through withdrawal is already under way, and the
distinction w i l l be lost as one category forces itself into another. We problem is that it has become exactly this: a consolidation, a restor-
protect it because we think it in danger of being overwhelmed by a ation, a simple relocation of investment within the region staked out
more powerful agency. With our minds fixed on the predominance of long ago as belonging to architecture.
language we might even risk enclosing architecture within its own What might have occurred in architecture, but did not, occurred
compound, denying it communication with anything else to pre- outside h, and indeed outside painting and sculpture, in so far as these
serve its integrity. This would be possible, yet it seems very unlikely are categorically defined.' To insist on direct access to the work may
to occur because, for architecture, even in the solitude of pretended only be to designate the drawing as the real repository of archhectural
autonomy, there is one unfailing communicant, and that is the drawing. art. It may also be to reject drawing out of hand.
Some English art historians have been directing attention to the Of the works beyond the pale of architecture — earth art, perform-
transactions between language and the visual arts: Michael Baxandah ance, installations, constructions — which nevertheless deal with
w i t h the early Itahan humanists,' T. J. Clark with French nineteenth- recognizably archhectural themes, several are remarkable not just
century painting,'' and Norman Bryson with seventeenth- and for the fact that they make httle or no use of drawing, but for the
eighteenth-century French painting.^ Their studies, which have impossibility of their development through this medium.
advanced art history into an area never properly investigated, show The work of the Los Angeles artist James Turrell may be used as an
painters and commentators trying to extricate painting from language instance."* The mainstay of Turrell's work through the late sixties and
or trying to accommodate to it, in what was not so much a war seventies was the artificially lit room (Fig. 2). Most architectural of
between the verbal and the visible as an economy between them, full these were a series of empty spaces which, if drawn up within current
of friction though the deals back and forth may have been. I have architectural conventions, could only be construed as indicative of
found their work invaluable and stimulating. It seems to me, how- witless simplicity. Their effect as installations can none the less be
ever, that this economy dominated by the trade between two powers completely overwhelming. Such directly apprehensible quahties as
cannot be transferred to the study of architecture without adaption, they possess have nothing to do with the presence of the artist's
for the architectural drawing constitutes a third force that may well hands, feelings or personality. Fabricated as they are with tremendous
equal those of the art work and its commentaries. precision and parsimony, there is no more trace of Turrell in these
M y own suspicion of the enormous generative part played by rooms than of Mies in the most sparse of Miesian interiors. Evoking
architectural drawing stems from a brief period of teaching in an art gushes of transcendental mystification from some critics,'^ Turrell's
college.'" Bringing with me the conviction that architecture and the work is, ah the same, quite easy to understand and appreciate since it
visual arts were closely allied, I was soon struck by what seemed at the has to do with observers not being able to believe their own eyes. You
time the pecuhar disadvantage under which archhects labour, never look into something which you know is another rectangular room
working directly with the object of their thought, always working at with batteries of fluorescent tubes on the back of the partition
it through some intervening medium, almost always the drawing, through which you peer. You can see how it works. You can put your
whhe painters and sculptors, who might spend some time on prelim- hands into it. You can even see, standing out against the haze of ihum-
inary sketches and maquettes, ah ended up working on the thing itself ination that moves from mauve through to pink, evidence of some
which, naturally, absorbed most of their attention and effort. I still earlier investigator who took it into his head to climb into the illusion,
cannot understand, in retrospect, why the implications of this simple leaving his footprints in the otherwise spotless, spaceless interior.
observation had never been brought home to me before. The sketch Even then, only by deduction can you maintain ehher the depth of
and maquette are much closer to painting and sculpture than a draw- the room or the emptiness of it, for the light looks, if not sohd, then
ing is to a building, and the process of development — the formulation incredibly dense, as if its luminosity would not so much reveal the
— is rarely brought to a conclusion within these preliminary studies. image of anything thrust into it as devour it. Take a few steps back and
Nearly always the most intense activity is the construction and man- it is impossible to envisage its depth even by an act of conscious will, a
ipulation of the final artefact, the purpose of preliminary studies few more and the screen-like aperture through which you looked
being to give sufficient definition for final work to begin, not to seems to be standing out as a block of light in blatant contravention
provide a complete determination in advance, as in architectural of what you know to be true.'° The most remarkable properties of
drawing. The resulting displacement of effort and indirectness of Turrell's installations are local and not transportable. The result of
access stih seem to me to be distinguishing features of conventional direct observation of the play of electric hght on white-painted sur-
architecture considered as a visual art, but whether always and neces- faces and countless experiments tn situ, they cannot be adequately
sarily disadvantageous is another question. illustrated or photographed after their construction, and there is no
Two divergent definitions of the possibilities for architecture way that even the vaguest hint of their effect could have originated
follow from the recognition of this displacement. We may choose to through drawing. In this respect Turrell's ihuminated spaces of the
join architecture to the other visual arts more securely by insisting seventies and eighties — Orca, Raemar, the Wedgework series, etc. —
that only that which the architect manipulates with his own hands is were further removed f r om drawing and the drawable than the earlier
his work. It is all too clear that this new intimacy would first require works i n which shapes of light were projected onto walls through cut
a divorce because, as we gained more direct access to the work, we templates. Turrell made and published (and sold) prehminary draw-
would be relinquishing claim to the architecture that now flourishes ings for some of these (Fig. 3). One cannot imagine such drawings
within the political, economic and social order. If architecture were making any sense in the later works. By continuing in the same medium
redefined in this way, it might become more scrupulous and less while ehminating the projector, Turrell was effectively taking his

A A F I L E S 12
4
Jer no moral obligation to expel them responsible, smaller and less predictable, worth less but better, as the work outside the range of the drawing, for it was their projected shape
:ould be achieved. It is understandable hope would be, would it not, that in giving up grandiose pretensions that made works like Afrum drawable.
grity of our art, we should imagine it to represent and define the social world in both its imaginative and The drawing has intrinsic limitations of reference. N o t ah things
of communication, just as it is under- active aspects (a project the unlikelihood of which is comparable to architectural (and Turrell's rooms are surely architectural) can be
it of its aggrandizement, we should the unlikelihood of compiling a legal code that is also a good novel — arrived at through drawing. There must also be a-penumbra of quak
age. But this is only to offer excuses for an ambition that can only be confounded in practice) architecture ities that might only be seen darkly and with great difficulty through
ideas. may, by contraction and concentration, constitute itself anew? Well, it. If judgement is that these qualities in and around the shadow hne
ity of vision arises fro m a fear that all this consolidation through withdrawal is already under way, and the are more interesting than those laid forth clearly in drawing, then
:ategory forces itself into another. We problem is that it has become exactly this: a consolidation, a restor- such drawing should be abandoned, and another way of working
in danger of being overwhelmed by a ation, a simple relocation of investment within the region staked out instituted.
ar minds fixed on the predominance of long ago as belonging to architecture. Returning momentarily to the recently vaunted status of architec-
mclosing architecture within its own What might have occurred in architecture, but did not, occurred tural drawing within the schools: to regard a drawing as a work of art
unication w i t h anything else to pre- outside it, and indeed outside painting and sculpture, in so far as these as we usually understand it is to regard it as something to be consumed
be possible, yet it seems very unlikely are categorically defined.'' To insist on direct access to the work may by the viewer, so that his rapacious appetite for formulated experi-
ure, even in the solitude of pretended only be to designate the drawing as the real repository of architectural ence may be assuaged. A n y further use attributable to it is incidental
; communicant, and that is the drawing. art. It may also be to reject drawing out of hand. and detrimental in so far as it may reduce its value as food for
have been directing attention to the Of the works beyond the pale of architecture — earth art, perform- consciousness. We have witnessed, over the past fifteen years, what
and the visual arts: Michael Baxandah ance, instahations, constructions — which nevertheless deal with we think of as a rediscovery of the architectural drawing. This redis-
s? T. J. Clark with French nineteenth- recognizably architectural themes, several are remarkable not just covery has made drawings more consumable, but this consumability
nan Bryson with seventeenth- and for the fact that they make little or no use of drawing, but for the has most often been achieved by redefining their representational role
unting.5 Their studies, which have impossibility of their development through this medium. as similar to that of early twentieth-century paintings, in the sense of
-ea never properly investigated, show The work of the Los Angeles artist James Turrell may be used as an being less concerned with their relation to what they represent than
ing to extricate painting from language instance.* The mainstay of Turrell's work through the late sixties and with their own constitution. And so the drawings themselves have
it, in what was not so much a war seventies was the artificially lit room (Fig. 2). Most architectural of become the repositories of effects and the focus of attention, while the
bie as an economy between them, fuh these were a series of empty spaces which, if drawn up within current transmutation that occurs between drawing and buhding remains to a
jck and forth may have been. I have architectural conventions, could only be construed as indicative of large extent an enigma.
nd stimulating. It seems to me, how- witless simplicity. Their effect as installations can none the less be The second possibility flows directly from this. If one way of alter-
ated by the trade between two powers completely overwhelming. Such directly apprehensible quahties as ing the definition of architecture is to insist on the architect's direct
ady of architecture without adaption, they possess have nothing to do with the presence of the artist's involvement, either cahing the drawing 'art' or pushing it aside in
onstitutes a third force that may well hands, feelings or personahty. Fabricated as they are with tremendous favour of unmediated construction, the other would be to use the
its commentaries. precision and parsimony, there is no more trace of Turrell in these transmissive, commutative properties of the drawing to better effect.
;normous generative part played by rooms than of Mies in the most sparse of Miesian interiors. Evoking This latter option — which I call the unpopular option — I wish to
im a brief period of teaching in an art gushes of transcendental mystification from some critics,'' Turrell's discuss in this article.
conviction that architecture and the work is, all the same, quite easy to understand and appreciate since it The two options, one emphasizing the corporeal properties of
was soon struck by what seemed at the has to do with observers not being able to beheve their own eyes. You things made, the other concentrating on the disembodied properties
under which architects labour, never look into something which you know is another rectangular room in the drawing, are diametrically opposed: in the one corner, involve-
:t of their thought, always working at with batteries of fluorescent tubes on the back of the partition ment, substantiality, tangibility, presence, immediacy, direct action;
nedium, almost always the drawing, through which you peer. You can see how it works. You can put your in the other, disengagement, obliqueness, abstraction, mediation and
ho might spend some time on prelim- hands into it. You can even see, standing out against the haze of ihum- action at a distance. They are opposed but not necessarily incompat-
11 ended up working on the thing itself ination that moves from mauve through to pink, evidence of some ible. It may be that, just as some fifteenth-century painters (Masaccio,
ist of their attention and effort. I still earher investigator who took it into his head to climb into the illusion, Piero, Mantegna, Pinturicchio, Leonardo) combined the pithy irregu-
;t, why the implications of this simple leaving his footprints in the otherwise spotless, spaceless interior. larities of naturalism with the compositional regularities of perspec-
Jught home to me before. The sketch Even then, only by deduction can you maintain either the depth of tive construction, so architects might conceivably combine, in such a
o painting and sculpture than a draw- the room_or the emptiness of it, for the light looks, if not solid, then way as to enhance both, the abstract and the corporeal aspects of their
ess of development — the formulation incredibly dense, as if its luminosity would not so much reveal the work. Instead, they stand next to each other, in an unpropitious sort
sion within these prehminary studies. image of anything thrust into it as devour it. Take a few steps back and of way, as alternative candidates. Argumentative opposition is usually
; activity is the construction and man- It is impossible to envisage its depth even by an act of conscious will, a stifling. A tug of war works better between rugby teams than between
the purpose of preliminary studies few more and the screen-like aperture through which you looked opposed concepts or practices, yet this is the way we insist on playing
ion for final work to begin, not to seems to be standing out as a block of hght in blatant contravention games. I would like to avoid this partisanship, so much more effective
ition in advance, as in architectural of what you know to be true.'° The most remarkable properties of in drowning out sense than articulating it, but it should be said that in
cement of effort and indirectness of Turrell's installations are local and not transportable. The result of the present chmate the tendency is generally to place the abstract and
stinguishing features of conventional direct observation of the play of electric light on white-painted sur- the instrumental within the orbit of a suspect, culpable professional-
Jal art, but whether always and neces- faces and countless experiments in situ, they cannot be adequately ism, ahowing the direct and experiential presence only within a
r question. illustrated or photographed after their construction, and there is no covert architecture which can never be revealed fully in the former,
of the possibilities for architecture way that even the vaguest hint of their effect could have originated and which shows up as so many sporadic episodes of resistance. In
this displacement. We may choose to through drawing. In this respect Turrell's illuminated spaces of the consequence the direct and experiential appears far more ethical and
'isual arts more securely by insisting seventies and eighties — Orca, Raemar, the Wedgework series, etc. — far more interesting, far more at risk and far more real than the
-ct manipulates with his own hands is were further removed from drawing and the drawable than the earlier indirect and abstract approach. This can only be acknowledged as
this new intimacy would first require works in which shapes of light were projected onto walls through cut true to the degree that the varieties of indirectness, abstraction and
J more direct access to the work, we templates. Turreh made and published (and sold) preliminary draw- instrumentality found in practice are puerile, obstructive and dull,
5 the architecture that now flourishes ings for some of these (Fig. 3). One cannot imagine such drawings which on the whole they are, as also are the artistic pretensions ofthe
and social order. If architecture were making any sense in the later works. By continuing in the same medium schools. A contest between two kinds of dullness cannot be expected
't become more scrupulous and less while ehminating the projector, Turrell was effectively taking his to come to much, even if it does ensure fairness.

AA F I L E S 12
 A distinction might be made between the object of drawing as
practised i n architecture and drawing as practised tradition-
ally in Western art. A story of the origin of drawing, derived
from Pliny the Elder" and recycled into the visual arts as subject
matter i n the eighteenth century (like all stories of origins, far more
revealing of the time of its tehing than of the time of which it tells),
shows this up nicely. The story is of Diboutades tracing the shadow of
her departing lover. If we compare versions by two neo-classical art-
ists, one exclusively a painter, the other better known as an architect,
some indicative differences become apparent.
David Ahan's The Origin of Painting of 1773 (Fig. 4 ) shows the
couple in an interior, the dressed stone wall of which provides a plane
surface upon which Diboutades traces the shadow made by an oh
lamp, placed at the same level as the sitter's head, on a ledge close at
hand. Karl F. Schinkel's unusual variation on this theme was painted
in 1830 (Fig. 5). Significantly, and in contrast to most other treatments
(as well as departing from Pliny), the architect chose, not an architec-
tural interior for his reconstruction of the event, but a pastoral scene
whh shepherds and shepherdesses.'' In place of the worked surface of
stone, a naturahy exposed face of rock. In place of the lamp, the light
of the sun. Both paintings, true to the original story, show drawing as
a function of projection, and both show quite clearly the combination
of elements required: a source of light, a subject upon which it plays, a
surface behind the subject, and something to trace with. Schinkel, how-
ever, shows the minimum of material artifice needed to accomphsh
this. To judge from his painting, the first human mark put on nature
might well have been the line of charcoal on the rock, while in Allan's
the accoutrements of civhization were already i n place to provide the
necessary circumstances for this late, charming and reflective acces-
sory. So it is perhaps equally pertinent that, while Diboutades herself
4. The Origin of Painting, h)i David Allan, 1773. performs the task in Allan's painting, it has been delegated to a
muscular shepherd in Schinkel's.
The artifice shown by Schinkel is that of an already organized social
structure of deference in which is expressed also the distinction
between thought and labour, a distinction absent in the more inti-
mate surroundings of Allan's painting. In Schinkel's version drawing
precedes building, in Allan's it follows from it. Of the two, it was the
architect who was obliged to show the first drawing in a pre-architec-
tural setting, because without drawing there could be no architecture,
at least no classical architecture constructed on the lines of geometri-
cal definition. In Schinkel's work, drawing is, from the beginning, a
divided activity, resolvable into a prior act of thought and a con-
sequent manual undertaking which the arrival of architecture would
duplicate, on a much larger scale, as the difference between design and
construction. In this instance the man is servant to the woman: she
conceives; he does.
At least as important in this symptomatology is the manner of
lighting. Allan uses a lamp, that is, a local, point source of illumin-
ation from which issue divergent rays. Schinkel uses the sun, that is, a
source so remote that its rays have to be regarded as travelling parallel
to one another past the earth. The two kinds of hght correspond to
the two types of projection: central projection, based on divergent
projectors, which played a crucial part in painting through the de-
velopment of perspective; and parallel projection, based on parallel
projectors, which has played an equally crucial, though far less well
5. The Origin of Painting, byK. F. Schinkel, 1830. recognized role in architecture through the development of ortho-
graphic projection (Fig. 6). The painter's version less remote, more
intimate, less differentiated; the architect's more remote, public,
insistent on differentiation. Just as we would expect, perhaps, but the
specific expression of these tendencies in Schinkel betrays a profes-
sional proclivity, giving drawing a priority, potency and generality
not evident in Allan's rendition.
The most notable difference of all, however, is registered only in an

6 A A F I L E S 12
A
distinction might be made between the object of drawing as oblique way in the two paintings. This has to do with the subject mat-
practised in architecture and drawing as practised tradition- ter of the artist's work. In painting, until well into the twentieth cen-
ally in Western art. A story of the origin of drawing, derived tury, the subject was always, as in the story of Diboutades, taken from
from Pliny the Elder" and recycled into the visual arts as subject nature. It may have suffered vast idealization, distortion or trans-
matter in the eighteenth century (like all stories of origins, far more mogrification, but the subject, or something like it, is held to exist
revealing of the time of its telling than of the time of which it tells), prior to its representation. This is not true of architecture, which is
shows this up nicely. The story is of Diboutades tracing the shadow of brought into existence through drawing. The subject matter (the
her departing lover. If we compare versions by two neo-classical art- building or space) will exist after the drawing, not before it. I could list
ists, one exclusively a painter, the other better known as an architect, various riders and qualifications to this principle, which may be called
some indicative differences become apparent. the principle of reversed directionality in drawing, to show that it may
David Allan's The Origin of Painting of 1773 (Fig. 4 ) shows the occasionally be complicated, but these would not alter the fact that,
couple in an interior, the dressed stone wall of which provides a plane statistically speaking, if I may put it that way, it gives a good account.
surface upon which Diboutades traces the shadow made by an oil We might surmise, then, that the absence of an architectural setting in
lamp, placed at the same level as the sitter's head, on a ledge close at Schinkel's painting is a recognition of this reversal, by which the
hand. Karl F. Schinkel's unusual variation on this theme was painted drawing must come before the building, of so little consequence to
in 1830 (Fig. 5). Significantly, and in contrast to most other treatments Allan the painter, who follows Pliny, innocently imagining that
(as well as departing from Pliny), the architect chose, not an architec- architecture developed to classical maturity without its aid.
tural interior for his reconstruction of the event, but a pastoral scene Drawing in architecture is not done after nature, but prior to con-
with shepherds and shepherdesses.'' In place of the worked surface of struction; it is not so much produced by reflection on the reality
stone, a naturally exposed face of rock. In place of the lamp, the light outside the drawing, as productive of a reality that w i l l end up outside
of the sun. Both pamtings, true to the original story, show drawing as the drawing. The logic of classical realism is stood on its head, and it is
a function of projection, and both show quite clearly the combination through this inversion that architectural drawing has obtained an
of elements required: a source of light, a subject upon which it plays, a enormous and largely unacknowledged generative power: by stealth.
surface behind the subject, and something to trace with. Schinkel, how- For, when I say unacknowledged, I mean unacknowledged in prin-
ever, shows the minimum of material artifice needed to accomphsh ciples and theory. Drawing's hegemony over the architectural object
this. To judge from his painting, the first human mark put on nature has never really been challenged. A l l that has been understood is its
might well have been the line of charcoal on the rock, while in Allan's distance from what it represents, hence its periodic renunciation ever
the accoutrements of civilization were already in place to provide the since Philip Webb rejected the whims of paper architecture — while
necessary circumstances for this late, charming and reflective acces- continuing to draw prodigiously.'-' There are ah sorts of curious re-
sory. So it is perhaps equally pertinent that, while Diboutades herself minders as to the subliminal acceptance, beneath the level of words,
performs the task in Allan's painting, it has been delegated to a of its singular priority within the art of architecture, if art it be, such
muscular shepherd in Schinkel's. as in architectural portraits, where, as a rule with but few exceptions,
The artifice shown by Schinkel is that of an already organized social and as in Wilhson's portrait of Robert Adam (Fig. 7), they are por-
structure of deference in which is expressed also the distinction trayed with their drawings, as are sculptors with their sculptures and
between thought and labour, a distinction absent in the more inti- painters with their canvases, estranged, for posterity, from the results
mate surroundings of Allan's painting. In Schinkel's version drawing of their labour, the clients more usually retaining the privilege of
precedes building, in Ahan's it follows from it. Of the two, it was the being portrayed with the building. '••
architect who was obliged to show the first drawing in a pre-architec- It would take much more than an article to reveal the full extent of
tural setting, because without drawing there could be no architecture, drawing's intrusive role in the development of architectural forms, or
at least no classical architecture constructed on the lines of geometrk to investigate the way in which it creates a translatory medium of this
cal definition. In Schinkel's work, drawing is, from the beginning, a or that consistency. Three instances must suffice to give some idea of
divided activity, resolvable into a prior act of thought and a con- what we are dealing with.
sequent manual undertaking which the arrival of architecture would The importance of orthographic projection has already been
duplicate, on a much larger scale, as the difference between design and alluded to. Although the geometric principle of parallel projection
construction. In this instance the man is servant to the woman: she was understood in late antiquity, Claudius Ptolemy having described
conceives; he does. it in a work on sundials around AD 300,'-'' evidence of its use i n
At least as important in this symptomatology is the manner of architectural drawing is not found until the fourteenth century. The
lighting. Ahan uses a lamp, that is, a local, point source of illumin- earliest more or less consistent orthographic projection of a building
ation from which issue divergent rays. Schinkel uses the sun, that is, a to have survived is a large, detailed elevation of the Campanile of
source so remote that its rays have to be regarded as travelling parallel S.Maria del Fiore, preserved in the Opera del Duomo in Siena and
to one another past the earth. The two kinds of light correspond to thought to be a copy of an original by Giotto, produced after 1334
the two types of projection: central projection, based on divergent (Fig. 8)."' To say that this was the first instance is not to deny the
projectors, which played a crucial part in painting through the de- existence of many drawings of a similar sort — plans, elevations and
velopment of perspective; and parallel projection, based on parallel sections — going back to the second millenium BC. But the Campanile
projectors, which has played an equally crucial, though far less well drawing required two imaginative steps never before taken together,
recognized role in architecture through the development of ortho- as far as I know. Firstly, a completely abstract conception of projector
graphic projection (Fig. 6). The painter's version less remote, more lines;'' secondly, an abihty to conceive of the thing being represented
mtnnate, less differentiated; the architect's more remote, public, (the surface of the building) as not equivalent to the surface of rep-
insistent on differemiation. Just as we would expect, perhaps, but the resentation — not quite. The corner bastions of the tower, with their
spec, ic expression of these tendencies in Schinkel betrays a profes- chamfered sides, and the diagonal Gothic windows above are drawn
sional prochvity giving drawing a priority, potency and generahty obliquely but with no indication of perspectival recession. In other
not evident in Allan's rendition ^ ^ ^ ^ & ^ detahed medieval drawings that have survived, orthographic relations
The most notable difference of ah, helowever, is registered only in an are held for all the parts of a facade that are frontal and close to being

7
 coplanar, but not in surfaces receding at an angle from the picture
plane.'* In other words, the orthography apphed only if the building
itself was identified by the draughtsman as sufficiently sheet-like and
frontal. To maintain effectively the relations between an array of
invisible parallel projectors, a plane onto which they are projected at
right angles, and other surfaces at various angles to the plane of
projection, as did the author of the campanhe drawing, required, at
that time, insight of a different order; a good reason, perhaps, for
accepting, in the absence of firmer documentary evidence, Gioseffi's
contested ascription," since it is acknowledged that as a painter
Giotto gave to the presentation of pictorial space far greater
coherence than his predecessors.
A comparison of the Campanile drawing with the highly developed
proto-orthography of ancient Egypt, so weU preserved on a drawing
board of around 1400 BC now in the British Museum (Fig. 9), reveals
not only greater rehance on outline in the Egyptian example, and the
compensatory flattening of the figure across the shoulders, preparing
it for re-embodiment in the fosshized compressed form of a bas-relief,
but also rehance on a manual activity — the sculptor's chisel cutting
straight into the face of the cubic stone on which the profile was to be
inscribed — to make the projectors tangible. Prior to the abstractions
of orthographic projection, projectors could be kept in mind through
the thoroughly physical reahzation given them in the fabrication of
reliefs and sculptures.'"'
Another choice presents itself: two quite different possibilities
attendant on the use of architectural drawing are discernible in the
Campanile drawing. It could rest on the simple and primitive expedi-
ent of assuming near equivalence between the surface of the drawing
and the mural surface it represents. Through the miracle of the
flat plane, lines transfer with alacrity from paper to stone and the
wall becomes a petrified drawing, inscribed or embossed to lesser or
greater degree. Much of this ancient identity remains with us to this
day, carried, through classicism, into the professional pastime we call
implying depth. To imply depth within a sohd three-dimensional
body is to conceive of it as being made up of flat surfaces modulated
within a thin layer yet giving the impression of being much deeper. It
is to attempt to make virtual space and real space at one and the same
time and in the same place — a sophisticated idea utilizing simple tech-
nical means. In Pahadio's sketch of the S. Petronio facade (Fig. 10) the
close alignment (but not quite identity) between drawing and build-
ing is at once obvious. This is the kind of architecture that so much
fascinated Alberti: a massive, monumental architecture engendered
from the etiolated, reduced, bodyless elements of 'lines and angles
which comprise and form the face of the building';^' an architecture
made through drawing and made of the same species of ihusion as is to
be found in drawing. For into its patterns of lines stopping and start-
ing we project, by a weh understood reflex of overdetermination,^^ a
deeper space. A n d in just the same way we project into the solid build-
ings of Alberti, Bramante, Raphael, Giulio Romano and Palladio,
borne along by the same absorbing reflex of overdetermination, the
illusions of drawing.
I feel as uncomfortable discussing implied depth, which has become
one of architecture's most hallowed shibboleths, as I do when wear-
ing someone else's suit. It is nevertheless necessary to do so, if only to
point out how the pursuit of this particular illusion has retarded
architectural vision by keeping it restricted within the confines of
particular conventions. Yet to assert that these conventions were his-
torically uninteresting or fruitless would be to adopt an easy and false
posture of disdain. In fact they were responsible for establishing the
drawing as a viable medium, ahowing the architect to spill his imagin-
ation onto it, sure in the knowledge that much of the effect would
8. Elevation of project for the Campanile ofS. Maria del Fiore, Florence, Giotto travel.
copyist, after 1334. Only with this reassurance of sufficient affinity between paper and

8 A A F I L E S 12
 coplanar, but not in surfaces receding at an angle from the picture
plane."* In other words, the orthography applied only if the building
itself was identified by the draughtsman as sufficiently sheet-like and
frontal. To maintain effectively the relations between an array of
invisible parallel projectors, a plane onto which they are projected at
right angles, and other surfaces at various angles to the plane of
projection, as did the author of the campanile drawing, required, at
that time, insight of a different order; a good reason, perhaps, for
accepting, in the absence of firmer documentary evidence, Gioseffi's
contested ascription," since it is acknowledged that as a painter
Giotto gave to the presentation of pictorial space far greater
coherence than his predecessors.
A comparison of the Campanhe drawing with the highly developed
proto-orthography of ancient Egypt, so well preserved on a drawing
board of around 1400 BC now in the British Museum (Fig. 9), reveals
not only greater reliance on outline in the Egyptian example, and the
compensatory flattening of the figure across the shoulders, preparing 1

1
it for re-embodiment in the fosshized compressed form of a bas-rehef, ^ ^ ^ ^ ^ •
but also reliance on a manual activity — the sculptor's chisel cutting
straight into the face of the cubic stone on which the profile was to be
inscribed — to make the projectors tangible. Prior to the abstractions
of orthographic projection, projectors could be kept in mind through
the thoroughly physical reahzation given them in the fabrication of
rehefs and sculptures.^"
Another choice presents itself: two quite different possibilities
attendant on the use of architectural drawing are discernible in the
Campanhe drawing. It could rest on the simple and primitive expedh
ent of assuming near equivalence between the surface of the drawing
and the mural surface it represents. Through the miracle of the
flat plane, lines transfer with alacrity fro m paper to stone and the
wah becomes a petrified drawing, inscribed or embossed to lesser or
greater degree. Much of this ancient identity remains with us to this
day, carried, through classicism, into the professional pastime we call
implying depth. To imply depth within a solid three-dimensional 9. Egyptian drawing board. Left-hand area inscribed around 1400 BC.
body is to conceive of it as being made up of flat surfaces modulated •

within a thin layer yet giving the impression of being much deeper. It
is to attempt to make virtual space and real space at one and the same
time and in the same place — a sophisticated idea utilizing simple tech-
nical means. In Pahadio's sketch of the S. Petronio facade (Fig. 10) the
close alignment (but not quite identity) between drawing and build-
ing is at once obvious. This is the kind of architecture that so much
fascinated Alberti: a massive, monumental architecture engendered
from the etiolated, reduced, bodyless elements of Tines and angles
which comprise and form the face of the building';^' an architecture
made through drawing and made of the same species of ihusion as is to
be found in drawing. For into its patterns of lines stopping and start-
ing we project, by a weh understood reflex of overdetermination,^^ a
deeper space. And in just the same way we project into the sohd build-
ings of Alberti, Bramante, Raphael, Giuho Romano and Palladio,
borne along by the same absorbing reflex of overdetermination, the
illusions of drawing.
I feel as uncomfortable discussing implied depth, which has become
one of architecture's most hallowed shibboleths, as I do when wear-
ing someone else's suit. It is nevertheless necessary to do so, if only to
point out how the pursuit of this particular hlusion has retarded
architectural vision by keeping it restricted within the confines of
particular conventions. Yet to assert that these conventions were his-
torically uninteresting or fruitless would be to adopt an easy and false
posture of disdain. In fact they were responsible for establishing the
drawing as a viable medium, allowing the architect to spih his imagin-
ation onto it, sure in the knowledge that much of the effect would
Piore, Florence, Giotto travel. 10. Facade drawingfor S. Petronio, Bologna, by
Only with this reassurance of sufficient affinity between paper and Andrea Palladio, 1579.

A A F I L E S 12
 wall could the drawing have become the locus of the architect's activ-
t W T O U N A f ,Vr-MJl\Cl3<TIVM-K.Kllcv^v/uMA M i U ^ Z M I ^ ^
ity, capable of absorbing all his attentions and then transporting his Mlc;ll.M,l|,s A N M I I bONMlOTr.MU MITH rVIlMaiNOÜIlM.|llA
UOMAI- ANNO -IN Ul._v\-||
ideas into buildings without undue disfigurement. Still, if its advan-
tage was the ease of translation, its disadvantage stemmed from the
same source: too close a likeness, too cautious a liaison, too much lifeL...l|
bound up in the elaboration of frontalities.
It may seem obvious that only when fighting this tendency, seeing
outside the drawing technique, his imagination soaring above the
confines of the medium, can an architect create fully embodied three-
dimensional forms. Obvious it most certainly seems, because every-
one believes it to be true. It is also demonstrably false. I come now to
the second possibility attendant on the use of parallel projection. The
assurance and relative precision with which the splayed surfaces were
projectively determined in the Campanile drawing indicates that the
draughtsman did not need to imprison forms within orthography.
Although the correspondence of frontal surface and sheet was still
dominant, there is at least a hint that through the rigour of the
technique, not despite it, the represented surfaces might prise them-
selves from the surface of representation, floating free fro m their
captivity in paper — no — attempts at vivid phrasing can do so much
damage. Rigorous projection does not free anything, not in the sense
of emancipation. Things are just made more manipulable within the
scope of the drawing. For any material object to obtain freedom is for
its handler to lose control of it, and that does not happen.
Think of a sheet of paper sprouting thousands of imaginary orthog-
onals from its surface. In conventional architectural drawing, con-
servative and fearful of losing conformity, they would not need to be
very long before meeting up with the edges of the imaginary scaled-
down surface behind the paper to which the lines of the drawing
correspond, and, as in the Egyptian sculptor's elevational drawing, 12. Pavement of the Campidoglio, Rome, hy Michelangelo, c.1538. Engraving of
they are often identified with the direction of incision into the stone 1567.
or, more recently, with the direction of multiplied layering in service
of phenomenal transparency; in either event, they act as guide rails later), which de 1'Orme could have known about, there is one crucial
into the blindness of an as yet unrealized dimension — short ones difference. While all these others were determined metrically, de
securely attached at both ends. What if they were longer and more rOrme's was determined projectively. We know this because he
abstract? Would it strain the architect's power of visualization? tells us so:
Would it endanger his control? Would it jeopardize translation?
Ceux qui voudront prendre la peine, cognoistront ce que ie dy par k voute
r ^he next example I would like to consider involves one detail spherique, laquelle i'ay faict faire en la ChappeUe du chasteau d'Annet,
avecques plusieurs sortes de branches rempantes au contraire l'un de l'autre,
of a small building by Phhibert de 1'Orme. De 1'Orme, a truly & faisant par mesme moyen leurs compartiments qui sont a plomb & per-
fascinating subject, did more to wrest orthographic projec- pendicule, dessus le plan & pavé de ladicte Chappelle.^'
tion from the predominantly painterly usage of earlier practitioners
(Piero, Raphael, perhaps Giotto) than anyone, and his work deserves This statement suggests that the pattern in the paving is similar to that
better elucidation than I am able to give it in this article. For the sake of in the dome, and since this is exactly what we find on the floor of Anet
brevity and the argument, however, this one incident wih have to do. (Fig. 13) and since de 1'Orme has specified perpendicular lines that
In the dome of the Royal Chapel at Anet, a chateau west of Paris project into each other, we might let the case rest, as indeed his
enlarged for Diane de Poitiers by de 1'Orme after 1547, can be seen a commentators have.-''
net of lines, not exactly ribs and not exactly coffers, neither spiral nor Words are such powerful things, and when they correspond to
radial. They are nevertheless laid out and carved with unusual prech visual impressions — the floor looks like the dome — they may
sion (Fig. 11). Moreover, their properties, hard at first to describe in reasonably stand as prook Strange to say, this was all an elaborate
stylistic, geometric or structural terms, are directly accessible to hoax by de 1'Orme, or at least I cannot think of any other explanation
vision. Most noticeable of all is the continuous expansion of lozenges, for why he should have gone to such lengths to cover his tracks.^'
rib thicknesses and angles of intersection as they extend down f r o m More interesting than whether it was a hoax or not is why no one
the oculus towards the base of the dome. The effect is of a coherent noticed the difference. And far more interesting than either is the
diffusion and enlargement or, conversely, of concentration, remote- method he did use to derive the criss-crossing curves under the dome.
ness and rotary acceleration towards the lantern. There has never One reason it was not recognized is that all the drawings made of
been anything quite like it and, although there are similarly patterned the chapel from the sixteenth to the late nineteenth century are mani-
apse heads (as in the portico of Peruzzi's Palazzo Massimo, Rome), festly incorrect, unable always to transfer the tracery of the dome, or
Roman coffers (as in the Temple of Venus and Rome outside the even the pattern of the floor, without gross bungling (Fig. 14), though
Domus Aurea) and pavements (as in Michelangelo's Campidogho the rest of each of the drawings is quite competent.^' Yet a look at the
pavement (Fig. 12), possibly designed in 1538, though not laid till much patterns in the dome and on the floor of the actual buhding would be
enough to convince anyone of the impossibility of de 1'Orme's claim.
/1. Royal Chapel, Anet, hy Philibert de 1'Orme, 1347-52. View mto the dome. Simply count the number of intersections along one of the eighteen
 longitudinal lines of the dome, and then count the number of inter-
sections along a corresponding radius on the floor. In the dome there
are eight, on the floor six. This alone is conclusive proof that no
parallel projection could map the one into the other. De 1'Orme's
deception WAS of a pecuhar and uncharacteristic sort, because he vv'as
doing far more than he owned to, not less.
Another aspect of the difficulty of seeing through the claim is the
fugitive character of our third term, the drawing, and its virtual
absence from our account of the making of architecture. Invoking it, I
shah now try to reconstruct the procedure adopted by de 1'Orme for
making the tracery of curves.^-'
Put out of mind, for the moment, the floor, and look only at the
dome. First, notice how the curved ribs approach the oculus ring,
glance across it, and return, making eighteen continuous loops
around it (in fact the stone wreath around the oculus overlaps the
hnes of tangency — see postscript at the end of this article). Then
notice how the returning ends of the same loops, as they descend,
meet at points along the equator of the dome (obscured by the cornice
in the photograph). We may then think of the network as made up of
eighteen identical tear-drop shaped rings eccentrically placed on the ...

surface of the hemisphere as if spun round its vertical axis. These are iHHiiiii
obviously complex, three-dimensional curves, neither circular nor 15. Diagonal division of lines of equal longitude and equal latitude on a sphere,
elliptical. The most useful clue so far is in the fact that they make from Perspectiva C o r p o r u m R e g u l a r i u m , by WenzelJamnitzer, 1568.
closed loops. H o w could these complex curves be defined on the
spherical surface with such precision? It was certainly not through the
expedient of dividing the hemisphere hself into rings of lathude and
lines of longitude and then interpolating diagonal curves — the pro-
cedure adopted, as far as I can tell, by every other architect faced with
a similar problem (Fig. 15)^* — as no handy gradation of latitudes could
have procured tangency around the oculus. On the other hand,
de 1'Orme possessed an unusual and, within architecture, perhaps
uniquely vivid comprehension of projective relations, as can be seen
from the Premier Tome de l'Architecture that he published in 1567,^'
packed with abstruse stereotomic diagrams involving projections of
nameless exotic curvatures. One of the remarkable features of these is
that every last one has its origin i n a circle. But, as the circles are
collapsed, elongated, ramped, then projected onto cones, cylinders or
spheres at glancing angles, they metamorphose into thoroughly
plastic, volatile shapes, commensurable only through the procedure
of projection itseh. This is the other significant clue.
Is there, then, a format of circles on a plane surface that would,
through parallel projection onto a hemisphere, transform into a nest
of tear-drops with the requisite nurnber of intersections? The answer
is yes, and it turns out to be the simplest possible arrangement: an an-
nular envelope of circles (Fig. 16). This annular envelope, I suggest, is
the real plan of the dome. Each one of the circles within the envelope
would produce, under projection, another closed curve, but of quite
different shape. The easiest way to envisage this is to think of the
circle as the base of a cylinder (the sides of the cylinder being the
projector lines) which cuts through the hemisphere whilst touching
its rim.
The resulting closed curve on the hemisphere, half of what is cahed
a hippopede,'° looks nothing like the circle from which it has arisen
(Fig. 17) and, ahhough the number of intersections stays the same,
neither do the original ensemble of circles resemble their projected
translation on the dome. The envelope of circles on the plane can be
seen to have an unfortunate appearance, the middle lozenges of the
annular ring hmply slumped in a distribution that has neither the
dynamic suggestiveness nor the quaskstructural appearance of the
dome, and it fails conspicuously to register the accelerating contrac-
tion towards the inner ring so pronounced above. So, rather than
dutifully deposit a piece of didactic evidence on the floor, de 1'Orme 16. Suggested plan for the traceryi of the dome ofthe Royal Chapel, Anet, drawn by
tinkered with it, expanded it and then clipped off hs outer rim unth h the author.

A A F I L E S 12
13
 Accordingly, to fabricate would be to make thought possible, not to
delimit it by making things represent their own origin (as tiresome a
restriction in art as in social hfe).
The pattern of the dome ribs at Anet does have a provenance. A
somewhat dubious iconography may even be sketched out. Acker-
man, in his study of Michelangelo's Campidoglio pavement, found a
medieval astronomical chart in a similar pattern of twelve rings, indi-
cating lunar revolutions during the course of a year. It is possible,
though by no means certain, that an investigation of sources would
reveal links between the solar and lunar charts, other diagrams of this
form, the Campidoglio pavement and Anet." It may even divulge an
informing symbolism that would explain the increased number of
rings at Anet. Although this has not yet been done, let us assume that
the quest for symbolic meaning would yield persuasive resuhs. Where
would this symbohsm reside, if not in the envelope of circles, which
was really no more than an expendable piece of formwork for the
transfigured phenomenon of the dome? Would we not be forced to
concede, in the circumstances, that the symbolism was a mere ingre-
dient, lost in form, not carried in it?
What comes out is not always the same as what goes in. Architec-
ture has nevertheless been thought of as an attempt at maximum
preservation in which both meaning and hkeness are transported
from idea through drawing to building with minimum loss. This is
the doctrine of essentiahsm. Such essentialism was held to be paradig-
17. Three projections ofa hippopede: the intersection produced by a cylinder that matic throughout the classical period. It was held to in architectural
passes through, and is tangential to the equator of a sphere. Drawing by the author. texts, but not always in architecture. The notable thing about the
working technique used by de I'Orme, which could only be written
looked sufficiently like the system of intersections to which it had about from within the limits of architectural theory as a way of
given shape (Fig. i8).'' From this we may infer that, for de 1'Orme, in moving truth from here to there, was in the enchanting transfigur-
the end, the desire for perceptible likeness took precedence over the ations it performed. Curiously, the pliability of forms was made
desire to demonstrate the rigorous method through which the visible possible by a homogenization of space. Orthographic projection is
difference had been achieved. The choice to eclipse his own cleverness the language translator's dream. Within its axioms the most complex
by marring the projective equivalence between the two patterns is all figures may be moved at w i h into perfectly congruent formations
the more poignant, given his insufferable tendency to brag elsewhere anywhere else, yet this rigidly defined homogeneity made distortion
in the Tome. measurable. It was this capabihty that de I'Orme exploited.
This is an interesting discovery, because h shows the geometric Orthographic projection played its part as one of numerous tech-
original to be completely expendable, and indeed quhe ugly in com-
parison with its much more wonderful product. ParaUel projection
in this example engendered more potent forms from less, and did
it by an ingenious, regulated distortion of a shape regarded, by
common consent, and by de 1'Orme himself in his writings on archh
tecture, as perfect to start with: the circle.'' Happy results do not of
course occur under guarantee of the drawing technique, also requir-
ing, as they do, an inquisitive mind, a very strong presentiment ofthe
sense within forms, together with a penetrating ability to visualize
spatial relations. This ability was doubtless enhanced by the practice
of projective geometry, but not purchased with it. Stih, it would be as
crude to insist on the architect's unfettered imagination as the true
source of forms, as it would to portray the drawing technique alone as
the fount of formal invention. The point is that the imagination and
the technique worked well together, the one enlarging the other, and
that the forms in question — and there are many more, not only in de
1'Orme's work, but in French architecture through to the end of the
eighteenth century — could not have arisen other than through pro-
jection. A study of de 1'Orme's use of parallel projection shows
drawing expanding beyond the reach of unaided imagination.
This, then, was architectural drawing in a new mode, more abstract
in appearance, more penetrating in effect, capable of a more un-
settling, less predictable interaction with the conventional inventory
of forms of which monumental buildings are normally composed,
destructive also of metric proportionality, the foundation of classical
architecture (see below), and suggestive of a perverse epistemology in 18. Plan ofthe paving of the Royal Chapel, A net, shown as a portion of an
which ideas are not put in things by art, but released from them. expanded version of jigure 16. Drawing by the author.

A A U LES 12
H
 Accordingly, to fabricate would be to make thought possible, not to niques used by artists and architects to counteract the rampant
delimit it by making things represent their own origin (as tiresome a instrumentahty of essentialism,''' which would have art be a form of
restriction in art as in social life). haulage, transporting incorporeal ideas into corporeal expressions.
The pattern of the dome ribs at Anet does have a provenance. A And there is an amusing irony in the prospect of the rigid bunch of
somewhat dubious iconography may even be sketched out. Acker- spectral parallels along which lines were pushed in orthographic
man, in his study of Michelangelo's Campidoglio pavement, found a projection, disturbing the rigidly graded conceptual space through
medieval astronomical chart in a similar pattern of twelve rings, indi- which ideas were pushed into things.
cating lunar revolutions during the course of a year. It is possible, The theme of this article is translation, and I am now talking about
though by no means certain, that an investigation of sources would transportation. There are all those other identically prefixed nouns
reveal links between the solar and lunar charts, other diagrams of this too: transfiguration, transformation, transition, transmigration,
form, the Campidoglio pavement and Anet.'' It may even divulge an transfer, transmission, transmogrification, transmutation, transpos-
informing symbohsm that would explain the increased number of hion, transubstantiation, transcendence, any of which would sit hap-
rings at Anet. Although this has riot yet been done, let us assume that pily over the blind spot between the drawing and its object, because
the quest for symbolic meaning would yield persuasive results. Where we can never be quite certain, before the event, how things will travel
would this symbolism reside, if not in the envelope of circles, which and what will happen to them on the way. We may, though, like de
was really no more than an expendable piece of formwork for the I'Orme, try to take advantage of the situation by extending their
transfigured phenomenon of the dome? Would we not be forced to journey, maintaining sufficient control in transit so that more remote
concede, in the circumstances, that the symbolism was a mere ingre- destinations may be reached. I retain this inane parable, as it gives
dient, lost in form, not carried in it? some idea ofwhat I believe to be the largely unrecognized possibility
What comes out is not always the same as what goes in. Architec- within drawing. One infidelity does stand out, however: these des-
ture has nevertheless been thought of as an attempt at maximum tinations are not like exotic, faraway places waiting to be discovered;
preservation in which both meaning and likeness are transported they are merely potentialities that might be brought into existence
from idea through drawing to building with minimum loss. This is through a given medium.
the doctrine of essentialism. Such essentialism was held to be paradig- But always standing in the way are the pieties of essentialism and
• the intersection produced by a cylinder that matic throughout the classical period. It was held to in architectural persistence (the confusion of longevity with profundity). Whatever
he equator of, a sphere. Drawing by the author.
texts, but not always in architecture. The notable thing about the modernism's much ventilated destructive achievements, it made no
working technique used by de I'Orme, which could only be written mark on these. In the region of drawing they operate ehher through
Tstem of intersections to which it had about from within the limits of architectural theory as a way of insistence on a true and irreducible expressiveness, or insistence on
his we may infer that, for de hOrme, in moving truth from here to there, was in the enchanting transfigur- perspectival realism, or in the demand that only pure geometric forms
tible likeness took precedence over the ations it performed. Curiously, the pliability of forms was made or ratios be employed.
rous method through which the visible possible by a homogenization of space. Orthographic projection is As regards the last, numerous analyses have been published from
The choice to eclipse his own cleverness the language translator's dream. Within its axioms the most complex the seventeenth century to the present day, divulging the secrets of
ivalence between the two patterns is ah figures may be moved at whl into perfectly congruent formations the world's greatest works of architecture in the presence of under-
nsufferable tendency to brag elsewhere anywhere else, yet this rigidly defined homogeneity made distortion lying proportions. Without denying either the presence of or the
measurable. It was this capability that de I'Orme exploited. need for proportionality in architecture, attention might be directed
)very, because it shows the geometric Orthographic projection played its part as one of numerous tech- to certain misconceptions. Not all proportionahty is reducible to
endable, and indeed quite ugly in com- ratio, yet h is only as ratio that h has been admitted into architctural
wonderful product. Parallel projection theory. A ratio is a comparison between two numbers, as 1:^27or VA.
more potent forms from less, and did Since numbers, having no tangible reality themselves, must be wil-
d distortion of a shape regarded, by fully pushed into things, we have to ask how ratio can be made
:'Orme himself in his writings on archi- sensible in architecture; the answer leads back to our point of depart-
h: the circle." Happy resuhs do not of ure, the drawing.
of the drawing technique, also requir- Whhe the simplest means of expressing a ratio outside of number is
mind, a very strong presentiment of the as the division of a line, the second most simple expression is as an
with a penetrating ability to visuahze area, length to breadth. In this surface-making form, ratio resides in
svas doubtless enhanced by the practice architecture. Ratios thus expressed fih a sheet like Lord Nort h fihed a
>t purchased whh h. Sthl, it would be as chair: squarely. And it has to be a sheet of paper with no rucks or
ct s unfettered imagination as the true folds, and it has to be viewed frontally, otherwise the proportionality
portray the drawing tech nique alone as degenerates. The less Euclidean the plane, or the more obhque the
'• The point is that the imagmation and point of view, the more degenerate the form. Nevertheless, as long as
f H r ' "^^^ enlarging the other, and the surface of the building maintains sufficient identity w h h the sheet
^' there are many more, not only in de of paper, proportional ratios may be transferred with little loss. The
^ai chitecture through to the end of the very archhects who used this approximate identity to such advantage,
nave arisen other than through pro- from Alberti to Pahadio and later, were preoccupied w h h estab-
Wy^^T'' °^ parallel projection shows lishing a canon of proportions. They were also keenly aware of the
.1 drawin '"^"^d^agmation. dangers that lurked in the third dimension, ready to degrade the
in'eff" ^ ^ode, more abstract beauty constructed so painstakingly in the flat." But, ahhough this
''"""wi*th v!' ''''P'''^''' °^ a more un- was a perplexing difficulty, it was in accord w h h the entropie account
l^üild- ""^ ^""^entional inventory of value given in the doctrine of essentialism. Things were supposed 'ÖC FÜ EB" ^ " ~T
''^""".nal,"f normally composed, to degrade as they moved from idea to object. It was a difficulty easy
U ' V' of *°"ndation of classical to articulate in theory, whereas the transfiguring capability in the hi' "~é
z^^' cc.' r
'^v J/'P'^'-^erse epistemology m 18. Plan ofthe paving ofthe Royal Chapel, Anet, shown as a portion of an
drawing was a potential advantage that was not.
To judge from the nostalgic and at the same time dogmatic character 19. Three stages ofthe proportional analysis of Helen Wills's face, from Le
'^"t released from them. expanded version offigure 16. Drawingby the author. N o m b r e d'or, byMatila Ghyka, 1931.

A A F I L E S 12
15
 of much twentieth-century hterature on architectural proportion, ah
that has been weh and truly 'lost' is any sense of the intrinsic limit-
ation of the idea, one remarkable demonstration of this regained in-
nocence being the analysis supphed by Matila Ghyka of Helen Wills's
face to prove that her beauty was founded in the golden ratio."" The
analysis is not of the rotund, undulating, folded, punctured surface we
cah a face, but of quite another surface, onto which the face was flat-
tened by the process of photography (Fig. 19). I present this as an
inverted parody of de I'Orme's procedure at Anet. The existing, allur-
ing, complex curvature of Wills's face projected through a camera lens
onto a flat surface upon which is then inscribed an unprepossessing
visor of lines constructed f r om basic elements of plane geometry.
Start from the end and work backwards, and you get the spun fret-
work in the Anet chapel dome. In Ghyka's analysis, basic plane
geometry ended up as a foundation; at Anet h was just the beginning.
De I'Orme's was not the only way; there were others, equally
efficacious. A study of other projects that ruptured the equivalence
between drawing and building — Borromini's S. Carlo ahe Quattro
Fontane or Corbusier's Ronchamp — would show archhects work-
ing quite differently though perhaps, in both instances, more in
accord with our prejudice that archhects of genius (the horse tor-
mented by its bridle, the caged lion, to use Borromini's bestial images
of himself)'^ must wrest themselves free from the restriction of geo-
20. T h e A n c i e n t o f Days, by William Blake.
Frontispiece to E u r o p e : A Prophecy, 1794.
metrical drawing rather than use it. Whhe I have no argument with
this point of view per se, it has left us insensitive to the potency that
has existed — still exists — in the precision of the drawing, which is
also capable of disengaging archhecture from those same stolid con-
formities of shape, propriety and essence, but fro m within the
medium normally used to enforce them.
Two current advertisements: a TV commercial for household paint
which shows the frenetic and messy antics of a barbarous Glaswegian
artist whose studio is all the while being painted spotless whhe by a
meticulous, imperturbable decorator; a newspaper ad for the Youth
Opportunities Programme that shows a lout spraying 'Spurs' onto a
wall, later transfigured into a white-coated apprentice painting a trim
httle nameplate for the same football club. Here is the absurd pubhc
prejudice in favour of neatness: neatness as a sign of civilization.
There is a counter-prejudice, a reaction, a cultured expiation no less
limiting, really, which operates in favour of the unpremeditated and
unregulated as signs of art and feehng. Neither w i l l do. Yet there is
something about the way people work.
It would be possible, I think, to write a history of Western archh
tecture that would have little to do with either style or signification,
concentrating instead on the manner of working. A large part of this
history would be concerned with the gap between drawing and
building. In it the drawing would be considered not so much a work
of art or a truck for pushing ideas from place to place, but as the locale
of subterfuges and evasions that one way or another get round the
enormous weight of convention that has always been archhecture's
greatest security and at the same time its greatest liabihty. This is one
of my ambitions: the history of Blake's architect-geometer has been
written a hundred times (Fig. 20); I would hke to wrhe the history of
Giacinto Brandi's (Fig. 21),'* not, I hasten to add, because she is so
young and pretty, but because of the uncharacteristic expression on
her face and in her posture. It is the kind of expression normally
21. L ' A r c h i t e t t u r a / ? ] , by GiacintoBrandi, seventeenth century. reserved in seventeenth-century painting for prostitutes and cour-
tesans. The picture's subject is uncertain, its title a modern suppos-
ition resting on the fact that she holds dividers, nothing more. One
might ask what such a figure is expected to do with the instrument she
shows us.

16
of much twentieth-century hterature on architectural proportion, all Postscript
that has been well and truly Tost' is any sense of the intrinsic limit- Notes
I wrote this article before visiting Anet and seeing the dome and floor
ation of the idea, one remarkable demonstration of this regained in-
of the chapel. It seems to be as I described it, with the exception of one 1. From the Latin translatio, to remove or carry from one place to another.
nocence being the analysis supphed by Matila Ghyka of Helen Wills's
detail which had escaped my notice in the photographs available to 2. Lawrence Weschler, Seeing is Forgetting the Name ofthe Thing One Sees (Berkeley
face to prove that her beauty was founded in the golden ratio.^'' The
me. After my return, another photograph — one that I had taken - 1982). A study of Robert I r w m .
analysis is not of the rotund, undulating, folded, punctured surface we 3. Michael Baxandall, Giotto and the Orators: Humanist Observers of Pamting m Italy
showed this up. It is easier to discern projective relations between two
call a face, but of quite another surface, onto which the face was flat- and the Discovery of Pictorial Composition 13>0-1450 (Oxford, 1971).
such surfaces in photographs than in the building itseh, where they
tened by the process of photography (Fig. 19). I present this as an 4. T.J.Clark, Tbe Image ofthe People {London, i9yi)- Ue Absolute Bourgeois [London
cannot both be held in view at the same time, and it is only the recok 1972); Ue Pamtmg ofModern Life [London, 1984).
inverted parody of de I'Orme's procedure at Anet. The existing, allur-
lection of apparent simharity that carries the idea of their relation J. Norman Bryson, Word & Image: French Pamtmg of the Ancien fiegwjc (Cambridge
ing, complex curvature of Wihs's face projected through a camera lens
whhin the buhding. (Given the difficulty of direct comparison, de Vision & Pamtmg: Tbe Logic of doe Gaze (London, 1983).
onto a flat surface upon which is then inscribed an unprepossessing 6. Bennington College, Vermont.
I'Orme's modification of projective equivalence to make the two
visor of hnes constructed fro m basic elements of plane geometry. 7. The most stimulating accoum of this slipping-over of the categorical boundanes is
surfaces look more alike is all the more effective and all the more art-
Start from the end and work backwards, and you get the spun fret- Rosalind Krauss's 'Sculpture in the Expanded Field', October 8 (1979); and also in:
ful.) The anomaly in my account of the chapel dome concerns the Krauss, The Originality ofthe Avant Garde and Other Modernist Myths (Boston
work in the Anet chapel dome. In Ghyka's analysis, basic plane
relation of the eighteen ribs looping round the lantern ring. I had 1985).
geometry ended up as a foundation; at Anet it was just the beginning.
thought that they pass across the edge of the lantern ring tangentially 8. Others might include Walter de Maria, Roben Irwin, Gordon Matta-Clark
De I'Orme's was not the only way; there were others, equally — and seen from the floor they give every appearance of doing so — Donald Judd, Robert Smithson, Michael Hiezer, Christo, Robert Morris, Dati
efficacious. A study of other projects that ruptured the equivalence but they do not. In fact the lantern ring cuts a httle way into the edge Flavin, De Wain Valentine, Mario Merz, John Aiken, Sarah Bradpiece, David
between drawing and buhding — Borromini's S. Carlo ahe Quattro Mach, and so on. The question, with these artists, is not whether they use drawings
of the pattern of intersections, ehminating the final circle of half-
(some do), but how they use them and why. Above all, the question is to what
Fontane or Corbusier's Ronchamp — would show architects work- lozenges. Evidently this was another of de I'Orme's modifications of extent the drawing, if used as a means of investigation, imparts significant
ing quite differently though perhaps, in both instances, more in projective equivalence, because the marble inlay on the floor does properties to the thing it represents. Many of the works of the artists listed, which
accord with our prejudice that architects of genius (the horse tor- include this part of the pattern. \t is possible that this particular are geometric and apparently reducible to drawing, are not, smce they possess
mented by its bridle, the caged lion, to use Borromini's bestial images modification had less to do with the forging of apparent likeness properties of substance and luminosity which, though they may be mimicked in
of himself)'-' must wrest themselves free fro m the restriction of geo- drawing, cannot be developed in investigative drawing. To imagine that they can is
between dome and floor than with the technical difficulty of cutting
a truidess illusion now being fostered in architectural schools.
metrical drawing rather than use it. Whhe I have no argument with such acute angles in the more friable stone of the dome.
this point of view per se, it has left us insensitive to the potency that 9. This is true, for Instance, of Kay Larson's otherwise excellem review of Turrell's
Whitney retrospective i n ^ r t f o r a m , January 1981, pp. 30-33.
has existed — stih exists — in the precision of the drawing, which is
I would like to thank Peter Davidson for putting the idea of translation into my head, 10. I am conscious of how similar my description is to Barbara Haskell's description of
also capable of disengaging archhecture from those same stolid con- Laar, another of Turrell's installations, which appeared in Art m America, May
Robin Middleton for comments and cnticism, Mary Wall for her help, and Rodney
formities of shape, propriety and essence, but from whhin the Place for conversations that stirred some of the muck f r o m the bottom ofthe jar. 1981, pp. 90-99.1 knew of Haskell's ardcle before I saw Turrell's work for myself
medium normally used to enforce them. and even read it out in lectures. I then made an effort to write about this type of
installation differently, but found I could not do so to any effect. This is certainly
Two current advertisements: a TV commercial for household paint an indication of my indebtedness to Barbara Haskell. It may also indicate an
which shows the frenetic and messy antics of a barbarous Glaswegian inescapable consistency in Turrell's work. See also: Kay Larson, op. cit., and
artist whose studio is all the while being painted spotless whhe by a Suzaan Boextger, Art Forum, September 1984, pp. 118-19.
meticulous, imperturbable decorator; a newspaper ad for the Youth 11. Pliny the Elder, Natural Histoi-y, vol. xxxv, para. 151. See also: K.Jex-Blake The
Opportunities Programme that shows a lout spraying 'Spurs' onto a Elder^ Plinys Chapters on the History of Art (London, 1896). The story adapted by
the eighteenth- and nineteenth-century painters was recorded by Pliny, not as the
wall, later transfigured into a white-coated apprentice painting a trim
origin of painting (para. 15 et seq), but as the origin of modelling (Diboutades'
little nameplate for the same footbaU club. Here is the absurd public father, a potter f r o m Sikyon, afterwards filling in the outline of the head w i t h clay
prejudice i n favour of neatness: neatness as a sign of civilization. to make a relief).
There is a counter-prejudice, a reaction, a cultured expiation no less 12. Schinkel's painting is uncharactenstic within the genre. Although it is accepted as
hmiting, really, which operates in favour of the unpremeditated and illustrating the Diboutades story, m which Pliny descnbes the shadow falling on a
wall, the complete absence of architecture is, as far as I know, unique, except for an
unregulated as signs of art and feehng. Neither will do. Yet there is
early variant by Joachim von Sandrart, 167J, illustrated by Rosenblum. Schinkel
something about the way people work. seems to have conflated the Pliny anecdote (a woman as the invemor) with
It would be possible, I think, to write a history of Western archh Sandrart's shepherd tracing the shadow of his sheep on the ground. See:
K.F.Schinkel, Architecture, Malerei, Kunstgewerbe (Berlin, 1981), catalogue entry
tecture that would have little to do with either style or signification,
207a, p. 267; and Robert Rosenblum, 'The Origin of Painting , Art Bulletin, Dec
concentrating instead on the manner of working. A large part of this 1957. PP-279-90.
history would be concerned with the gap between drawing and
13. This paradox was pointed out by Lethaby, one of Webb's greatest admirers See-
building. In h the drawing would be considered not so much a work W.R. Lethaby, P M > Webb and his ITor^ (London, 1979), pp. „7-25. 'There are two
of art or a truck for pushing ideas from place to place, but as the locale ideals', he wntes, 'sound, honest, human building, or brilham drawings of
of subterfuges and evasions that one way or another get round the exhibidon designs.'

enormous weight of convention that has always been architecture's 14. A distinction has to be made between those portraits of artists in which the tool
represented the occupation (paintbrush, chisel or dividers) and those in which the
greatest security and at the same time its greatest liabihty. This is one
work Itself was shown. The most notable English exception to the general rule is
of my ambitions: the history of Blake's architect-geometer has been Johann Closterman's portrait of Wren in the Royal Society Collection, which
written a hundred times (Fig. 20); I would like to write the history of includes St Paul's in the background.
Giacinto Brandi's (Fig. 21),'* not, I hasten to add, because she is so 15. G. T. Toolmer, 'Claudius Ptolemy', in Dictionary of Scientific Biography, edited by
young and pretty, but because of the uncharacteristic expression on Gillespie, vol. XI, pp. 186 tt- and Claudii Ptolemaei, Liber de Analemmate (Rome,
1562), which includes many diagrams.
her face and in her posture. It is the kind of expression normally
16. Decio Gioseffl, Giotto Architetto (Milan, 1963), pp. 82-4. Close inspection reveals
reserved in seventeenth-century painting for prostitutes and cour-
some minor inconsistencies in projection. For example, the facets of the corner
tesans. The picture's subject is uncertain, its title a modern suppos- bastions that are decorated with squares of dark marble inlay are shown w i t h the
ition resting on the fact that she holds dividers, nothing more. One vertical strips of inlay registenng the same width in frontal and oblique planes. In a
might ask what such a figure is expected to do with the instrument she few of the lower panels, however, the proportional reduction on the oblique
surfaces was shown correctly.
shows us.
17. It is interesting that the story of the origin of Greek geometry, also mcluded in
Pliny, IS very similar to that of the ongin of drawing. It tells of Thales finding the
height of an Egyptian pyramid by measuring the length of its shadow on the
ground and comparing it to the shadow made at the same time of day by a smaller
vertical of known height. To recognize the formal equivalence between these two

AATILES 12
AA F I L E S 12

17
 things requires, as Meserve points out, imaginary lines to be conceived as joining 28. If a sphere is divided as is a terrestrial globe, with lines of equal longitude and'
the tops of the verticals w i t h the ends of their shadows on the ground, thus latitude, and the diagonals are joined, a pattern of this type emerges (see Fig. 14 in
inaugurating a geometry of abstract lines. The impercipience in the story, claiming this article, from: W.Jamnitzer, Perspectiva Corporum Regularium (Nuremberg,
that the abstract line was discovered i n the measurement of a building, the 1568), series G , plate v). It differs significantly from the arrangement at Anet,
construction of which would have required that knowledge beforehand, is however, in that all the spiralling diagonals radiate f r o m the poles. A t Anet this
discussed by Serres. See: Bruce Meserve, Fundamental Concepts in Geometr)' (New area is completely empty of curves. A simple metric division also gives the pattern
York, 19S3), pp. 222-3; 2nd Michel Serres, 'Mathematics and Philosophy: What of Michelangelo's Campidoglio pavement (see Fig. 11). A n oval constructed with
Thales Saw . . . ' , in//ermes (Baltimore, 1982), pp. 84-96. major and minor arcs of equal length made the division of each of the four
18. One of the finest examples of this type is also i n the Opera del Duomo in Siena, a component curves into six equal parts along the perimeter an easy matter. Radii
drawing of the Siena Baptistery facade made around 1370, probably by Domenico were joined to these f r o m the centre of the oval, and then these radii were
Agostino. The deeply recessed portals and aisle windows depart f r o m themselves divided into six equal parts. Joining these together produced a spider's
orthographic projection. See: John White, Art and Architecture in Italy 1250-1400 web of concentric ovals across radial lines. Alternating diagonals within the
(Harmondsworth, 1966), p. 327 and plate 154. White also discusses the earliest network produced the pattern. Again, all the diagonals converged on the centre.
Italian drawing of this type, the Orvieto Cathedral faipade by Lorenzo Maitani, 29. De I'Orme's stereotomy requires a separate study. His was the first publication of
abouti3io(p. 292). the technique which maintained a distinct presence in French architecture well
19. Edi Baccheschi, L'Opera Completa di Giotto (Milan, 1966), p. 126. Baccheschi into the eighteenth century and was systematically taught well into the nineteenth.
contests the attribution; however. White {supra, p. 172) accepts it as probable while See: J.-M.Pérouse de Montclos, L'Architecture a la Franfaise (Paris, 1982), parts 2
Trachtenberg argues strongly that it is Giotto's scheme and may even be his and 3, especially pp. 80-95.
drawing (M. Trachtenberg, The Campanile of Florence Cathedral: Giotto's Tower 30. The hippopede was one of the few curves, other than the circle and the conic
(New York, 1971), pp. 21-48). sections, that was well estabhshed in ancient Greek geometry, its properties having
20. According to Panofsky, this was the case w i t h not only Egyptian reliefs but also been investigated by the mathematician Eudoxus in the fourth century B C . See:
sculptures in the round. See: Erwin Panofsky, 'The History of the Theory of Carl Boyer, A History of Mathematics (New York, 1968), p. 102. It is an open
Human Proportions as a Reflection of the History of Styles', in Meaning in the question whether this was known to de I'Orme and, if it was, how it contributed to
Visual Arts ( N e w Y o r k , 1955), pp. 60-62. his use of it in the chapel dome.
21. L. B. Alberti, The Ten Books of Architecture, translated by Leoni, edited by Rykwert 31. This interpretation accords with the decreased number of intersections on the
(London, 1955), book i , chapter i . pavement. More importantly, it accords w i t h the morphology of the pavement
22. E . H . G o m b r i c h, Art and Illusion (London, 1972), part ni, especially 'The lozenges. In the envelope of circles I have proposed as the basis of the dome tracery
Ambiguities of the Third Dimension', pp. 204-44. (Fig. I J ) the lozenges are broadest in the middle of the annular ring and flatten
23. 'Those who would take the trouble, w d l understand what I have done with the towards the inner and outer rims. The pavement as built, by increasing the radius
spherical vault that I had to make for the chapel at Anet, with several sorts of of the circles and extending the outer rim of the envelope, plots only six of the
branches inclined in contrary directions and forming by this means compartments original eight rings of lozenges onto the available floor space. The fact that the
that are plumb and perpendicular above the plan and paving of the said chapel'. fourth and f i f t h rings of lozenges (numbered from the oculus outwards) are the
Philibert de I'Orme, Le Premier Tome de LArchitecture (Paris, 1567), chapter xi, same proportion, while the sixth (outermost) ring is noticeably flatter and
p. 112. comparable in proportion to the third ring, supports this conjecture, since this is
24. The most comprehensive recent work on de I'Orme is Anthony Blum's Philibert exactly the property of the full eight-ring envelope of curves. N o t only does this
de TOrme (London, 1958). Blunt noticed and accepted the projective relation in make the floor pattern look more like the expansive dome pattern, but the
question (p. 43). projecting cornice below the dome cuts the lower rim ofthe dome itself f r o m view,
25. De I'Orme, having referred to the projection of the floor into the dome, referred obscuring much of the lowest ring of lozenges and making its observable density
the reader to a simdar proposal, which he proceeded to demonstrate and explain. even more nearly equivalent to that of the floor.
This did involve projection of a plane surface into a spherical surface {supra, 32. De I'Orme, Premier Tome, p. 33. Yet de I'Orme was nowhere near as insistent on
chapter xn, pp. 112-13), but it was a much simpler pattern of concentric squares on a the perfection of the circle as other sixteenth-century writers on architecture,
pavement whose effects in the dome could quite easily be visualized from the plan preferring to concentrate his praise on the figure of the cross.
alone. 33. James S. Ackerman, The Architecture of Michelangelo (Harmondsworth, 1986),
26. The perspective section published by de I'Orme himself, primitive, like most of his pp. 167-8.1 am grateful also to Richard Patterson for information on this.
perspective illustrations, showed the tracery inaccurately and did not indicate the 34. It is now often taken for granted that idealism and essentialism save us f r o m the
floor pattern (although there is one variant of this, the source of which I have not kind of instrumentality that comes with positivism. This they may or may not do.
yet been able to ascertain, which does include the pavement and shows it w i t h the But I would insist that they bring w i t h them other kinds of instrumentality and
same number of intersections as the dome). N o orthographic drawings of the other varieties of subjection just as unsavoury. I would insist also that only some
chapel by de I'Orme survive. The plan published by Androuet du Cerceau in Les kinds of instrumentality are unsavoury.
plus excellents bastiments de France, vol. 2 (Paris, 1607), showed the floor nearly as 35. Thus Alberti, who had done more than anyone to propagate knowledge of
laid (wrong number of arcs, right type of configuration), but the superimposition perspective in his book on paintmg, accused it of distortion in his book on
of the lantern plan obscured the most critical area ofthe pattern. The plan f r o m the architecture (L. B. Alberti, The Ten Books on Architecture (1955), p. 22). In the
survey published by Rudolphe Pfnor in Monographic du Chateau d'Anet (Brussels, ensuing centuries, proportion, its inevitable distortion by the eye, and its practical
1867), to all appearances much more accurate, preserved the topological 'adjustment' to counter the optical deceptions of three-dimensional embodiment
characteristics of the nest of curves, but did not draw them as circular arcs, while in were discussed by numerous authors, for example Caesariano, Serlio, Palladio,
the section both the curves and their intersections in the dome fail to correspond to Scamozzi, Rusconi. Claude Perrault gave a brilliant though highly critical account
the pattern as constructed. of both proportion and adjustment in Ordonnance des cinq espèces de Colonnes
27. Blunt noted the ingenious nature of the coffering pattern, its departure f r o m the (Paris, 168}) {Treatise on the Five Orders, translated by John James (London, 1708)).
usual method of compartmenting domes using lines of latitude and longitude, and 36. Matila Ghyka, Le Nombre d'Or (Paris, 1931), p. 55 and plates 18-20.
the inaccuracy of available drawings, yet-despite these perspicacious observations 37. Joseph Connors, Borromini and the Roman Orator)t {Boston, 1978), p. 3. Connors's
his own account went quickly awry. He wrote: 'through each of the 16 [sic) points lecture on S. Carlo at the A A in 1982 was very informative about Borromini's use of
on the oculus two great circles are drawn, linking it w i t h two points on the equator drawing.
separated f r o m it on ground plan by what appears to be an angle of 180°'. What
Blunt did was to give a description of du Cerceau's plan of the pavement (with its 38. Renato Guttuso, L'opera completa del Caravaggio (Milan, 1967), pp. 108-9.
sixteen pairs of branches, not eighteen, calling the arcs great circles and not taking
into account the superimposition of the lantern plan in the du Cerceau drawing, Acknowledgements
which artefact gave rise to the effects he described i n this passage. In other words, Fig. i : Ethel Redner Scull Collection, Guggenheim Museum, New York. (SDACS
he was describing part of a drawing of the floor, not the whole o f t h e dome (great F I G . 2 ( A B O V E ) : Photograph by M . Lee Fatherree. Fig. 2 ( B E L O W ) : Photograph by Warren

circles could not in any case meet twice on a hemispherical surface unless both Silverman. Fig. 3: Photograph by Michael Trier. Fig. 4: National Galleries of Scotland,
intersections were on the rim). See: Blunt, supra, pp. 39-42. Edinburgh. Fig. 5: Von der Heyt Museum, Wuppertal. Fig. 7: National Portrait
For my own investigations I have used a crude photogrammetry, with Gallery, London. Fig. 8: Opera del Duomo, Siena. Photograph by Grassi. Fig. 9: British
photographs of the dome and floor from the Conway Library, Courtauld Museum. Fig. 10: Bodleian Library, Oxford. Fig. 11: Conway Library, Courtauld
Institute, and in Country Life, 16 May 1908, pp. 702-4. The method, in my hands, is Institute, London. Fig. 13: Photograph by Greff, Paris. Figs. 14,20: British Library,
not foolproof, but it is probably adequate. Certainly it gives far more reliable Department of Prints and Drawings. Fig. 21: Spada Gallery, Rome. Photograph by
results than could be obtained from existing drawings. Scala.

18 A A HLES 12