Architectural Projection ARCHITECTURAL DRAWINGS are projections, which means that organized arrays of imaginary straight lines pass through the drawing to corresponding parts of the thing represented by the drawing. We are all very fa a television screen are projections. Convergit of light reflected from a subject are gathered by a camera lens and focused on a photo-sensitive surface. ‘This is a projection. The resulting image is turned into electrical signals that are transmitted to a cathode-ray tube where they are remustered ina scan- ning electron beam, the pattern of which sketches a duplicate of the original cone of light rays in reverse. ‘When they hit the fluorescent screen they create another image. Thisis also a projection, as are photo- graphs and motion pictures. We are surrounded by these flat versions of embodied events to such an ex- tent that they have long since ceased, in themselves, to be a matter of any amazement, or even of mild curiosity. We are pro ink of them as part of the ‘ever-expanding technology of information transfer. Projection has been incorporated into so many elec tronic and mechanical processes that it no longer needs much space in our imagination. We do not nor- ‘mally have to think spatial relations out this way, and there seems litle point in making anyone do so when ROBIN EVANS it can be done instantaneously with such exactitude and facility in a black box. With the profusion of reproduction techniques, things become flatter. At any rate the vast majority of projections work that way, since two-dimension formation is so much easier to handle than th dimensional things. In practice, projection has be~ come thoroughly directional because of the avaiabil ity of certain instruments and machines for making pictures; but there is nothing in projection itself to suggest directionality. It ean work either way round. Architecture provides an instance of the opposite ten- aking information from flat representations to create embodied objects. “There are, of course, plenty of drawings in ar- chitectural archives that illustrate existing buildings. things as yet un- the difference between the two categories. Whi finished and drawn up ready fo tation drawings (sce cat. no. 30.2). Presentation draw- ings are not supposed to have any effect on the design, ‘Their job is to propogate a completely defined idea, | not to testit or modify it. They should then be classed as records. And yet what they record is not real. To use the word projection in a completely different sense, they are projections ofa plausible outcome for a set of instructions and proposals already defined elsewhere but not yet accomplished. ‘Their status is unclear because they are neither impressions received from a real object, as would be a perspective fro ‘ora photograph, nor are they directly instrumen the making of what they represent. beginning and the end of a process. Temay seem odd to contrast the two ways projec- tion transmits its information only to embark on the description ofan ambiguous case. It may well give the impression that the distinction is uncertain and of litle practical value. My intention, however, is to point out a very common property of architectural drawing in general. Projections—the invisible that relate pictures to things—are always directional. Drawings arrest and freeze these vectors, but even in this fixed state, projected information can be mobilized by the imagination of the observer. When. workman looks ata workshop drawing and envisages ‘what the finished result of his labour will be, he is, byenvisaging it, briefly turning the projection around, suggests that some aspects of the are sufficiently similar to projection to be compared with it, or even confused lar diversions and reversals oceur ata differ~ ent tempo in the making of topographical records, where itis normally assumed that the subject will be unaffected by its portrayal. Draw a building and it will be the same building when you have finished drawing it, neither more nor less. Visual knowledge alights on its subject without taxing it, without expropriating anything from it. Obtaining it can be, and often is, a very gentle, considerate, subtle affair, although there are stories to suggest otherwise, like the one told by ‘Max Enst about his father. Ernst the elder, a scru- pulous realist, was painting a view of his own garden, and finding it unsatisfactory unless a certain tree was omitted, first subtracted the tree from the composi- tion and then afterward removed the tree from the garden. This sounds ludicrous because we are prone Ennst the elder was a bad the garden was at fault. ‘a constant interplay between ive portrayal and the active remodelling of reality? Might this help explain why the accurate rep- resentation of objects, all assimilation and no effect, became so important to western civilization during a period when it was extremely aggressive and rampant, from the fifteenth to the nineteenth century? go ARCHITECTURAL PROJECTION large variety of Buenos Aires. Ie seat works of copy and dup! another to be able to thenon cannot be demolished by drawing, but it can be burgled; its forms stolen and reconstituted by vie- diversions and re-routings. If we subtract ing element, then we would have to judge only which is much easier. The drawings pro- sm Butterfield’ office for St Matthias jon (cat. nos. 39.1~39.2), Nesfield’s shop drawings for furniture (cat. no. 4 Cormiee's drawings for the Palace of Justice in ‘Montreal (cat. nos. 41.1~41-5) then lead ineluctably to their final destinations: house, church, and court- times convenient to do this, but the subtraction should be performedasa temporary measure. Ifthe activating imagination is permanently removed from considera tion, drawing very easily slips into the category of a ‘mere technical facilitator, and this results in two illa- sions: frst, cesno difference to whatisdrawn (unless done incorrectly); second, that drawing can propagate things, but never generate them, These il Tusions will persistas long as we regard good drawing as a simple truth-conveyor. As much can happen in the drawing as out oft. ‘According to ancient wisdom, architects make images from ideas. Theologians were fond of quoting Se Thomas Aquinas on this theme, Anarchitect, wrote ‘Aquinas firsthas an idea of a house and then he builds it. God made the World in similar fashion, Aquins Seer ion Ans He draws the bodyless but full ied ideas the mind and pus them on pas, ure ae tees on rue bee Aquinas's architect is 3 men There may besocherears ba tot be possessed of much inthe way of cra uit the contrary Legining withthe esto Je wold wereheldinthe mind rani, stam. ‘The imagination work wih oe ters and is altered by what is seen. The problem is that f we admit this, then the lag the most uncertain, negotiable position of the main thoroughfare between ideas and things. For this same reason, drawing may be proposed as the principal locus of conjecture in architecture. Most of our knowledge of great architecture comes from pictures. One could therefore imagine a situation in which embodied architecture-not the everyday buildings that we are used to, but buildings in the “great works” category—was hardly more than 4 rumour of an intervening state. We could, if we wished, treat great buildings that way, since they ae ‘anyway so completely surrounded by their own pro- jected images. They are set in an aura of illustration that no doubt alters the way we see them. As eis become more aware of the active role played by pho tography in the propagation and maintenance of ar- chitectural ideas, this intervention becomes clearet? “There is also a growing awareness of the active role ‘of drawing inthe engendering of certain architect forms and in the maintenance of others. Asan instance I would cite Robert Branner’s speculation thatthe attenuated, linear, panel-like character of rayonatt Gothic architecture is ateributable tothe introduction of scaled project drawings on parchment sometime before t240. He thinks Cambrai Cathedral was the first building conceived this way.” We are now wit nessing critical pincer movement that sat once more aware of, more wary of, and more interested in theactive part played by the images on either side of architecture. ‘The modish thing to do would be that makes us insist with some it should maintain the priori always had, or that we think it ought to have. Slightly less modish, but very self-righteous, isthe stand taken against any drawings or pictures because they get in the way of our direct and authentic perception of ar- chitecrure. The first argument is easily tenable, but very disturbing in its implications. The second is ten- able only if one discounts the entire history of western, architecture, which hasalways been dependent on pic- tures for purposes of construction and dissemination. is easy to hold opinions of whatever stamp in the absence ofa full understanding of their consequences. [At present we are only just beginning to investigate the power that drawings and photographs have to alter, stabilize, obscure, reveal, configure, or disfigure ‘what they represent. Whatever the final outcome of these investigations, we can be certain of one thing in the meantime: architecture is reliant on its own pic- tures to a far greater extent than has hitherto been recognized. Inwhat follows I shall try to give a briefsummary of one side of this reliance: the pictures that precede the act of building, ‘The images with which we are most familiar are perspectival. In perspective projection, the array of imaginary lines mentioned at the outset of ths essay all converge on a single point. They behave in exactly the same way as light rays converging on the eye do. “Thus, although these imaginary lines, called projectors, have no real existence, they mimic the pattern of something that does exis, and that is why they can be relied upon to produce pictures that look like or, under restricted conditions, are precisely congruent with, what they represent. They ape the geometry of ‘monocular vision, However, the kind of drawings used in the profes- sional design, production, and even illustration of ar- converge to a point, but remain parallel. Because this is not the way we see things, orthographic drawing seems es easy to place. It does ‘why so many people find such draw read at first sight. The advantage of preserves more of does. Itiseasier to make things from than to see things with. So it is not surprising that orthographic projec tions are more commonly encountered om the way t0 buildings, while perspectives are more commonly en- countered coming from buildings. This gross truth has not prevented a high degree of mixing and slippage between the two, not least because those expert in the ‘one have tended to be expert in the other. Such slip- page cannot be allowed to obscure the fact th architecture, orthographic projection has been the preponderant method for devising, picturing, and transmitting ideas of buildings before they are built So this essay will be principally concerned with or- ‘thographic projection. “The question remains as to how it works. Or- thographic projection is not in the slightest degree ‘mysterious, and yet its employment in architecture raises many imponderable questions, the most press- ing of which have to do with the enigma of how ar- chitectural ideas are given definition prior to being constructed. If we think in terms of art, this anterior definition of the object, whereby all significant deci- sions are normally taken before the thing its begun, is peculiar to architecture. It would be fo itseems to me, to characterize architecture as abstract, is no more abstract than a chair or a nakes a great deal of sense to call the process of its conception abstracted. Architects do not ke drawings of buildings. sly conceived-engineering the different ways projective drawings have been em- ployed, how they are constructed, and, above all, how present. Some has been given, but it been made with a view ‘ways drawings work on the conception of buildings, rather than giving a chronology of techniques or dif- ferentiating the types of projection used. ‘The imagination looms large here, butitis ima- gination construed, I have to admit, in an odd way: an imagination not located solely in the mind of the ar- chitect. Reference has already been made to the active imagination of the observer of the drawing; there is also an active imagination in the drawing itself. This has nothing to do with the mental faculty of imagin- ing. Obviously, drawings do not think. But, because a drawing technique like orthographic projection was itself the product of intense imagination, this massive effort of imaginative intelligence lies dormant in it, animated to lesser or greater effect and to various ends every time the technique is used. some cases, necessarily rare, the imaginative sence of the architect divided between invent- ing the drawing and inventing the thing drawn, Neither can have been taken for granted atthe time, and in such circumstances the relation between pro- jection and the projected is of considerable interest. “The first example is just such a case. eis a plate from the Eiliche underrict 2u bfestigung, by Albrecht Diirer, published in Nuremburg in 1527 (fig. 2), show- ing the plan, ind elevation ofa fortification, printed example ofthese three irawing shown together, amat- ter deemed significant because we have come to re- ARCHITECTURAL PROJECTION 3Fig. 2. Albrecht Durer, Dasign for a Basin at the Ange ofa Tron- ‘Woodeut, page 29.3 «18.9 em. Ca 8216 Cage (etn. 1) ‘gard this set of three as fundamental. They are typical of architectural production, and had been established as such by the middle of the sixteenth century. ever, the following plate in Diirer's book (ig. ) shows something that isat least as significant; itisan enlarge- ment of the elevation of the curved wall ofthe fortress, which, in the woodcut for the smaller drawing on the preceding plate, shows little in the way of detail. ‘There are no projectors indicated on either drawing, but it is perfectly clear that the positions of the in- Fig. 5. Direr, Bleotio of « Bastion, fom Elche andersht 19.1 41.6em. cca 8246 Cage clined and battered arches following the curvature of the wall were determined by projection. Anyone familiar with the established conventions of architec- tural drawing will have no difficulty recognizing that plan until they met the outline simplest procedure would be to n the plan and moment's further refle ‘complex operation is require. surfaces from which the orthogonal. The surface of the fortress w: slice of a cone, curving and inclined at the same time. Drawa simplearch ona sheet of paper with acompas. You can either wrap this around the surface ofthe cone, like a transfer, to get a bowed arch, or you can hold it flat and upright and project the conic surface. In either case the mapped in pl a compass. ‘What we see in Diirer’s fort wall are shapes that important onotice that it is not just the shapes drawn, but the shapesthit ‘would have been bu from the drawings that are de- fined this way. When we envisage wrapping the dra ing of an arch around the conic surface of the forress wall (not a projective operation), we have to do so built. In order to know the shape of he shape of the wall of which it wil the arch we ni bea fundamental part: we cannot find the shape of theLE Speer etecmnmtanategae ik i it wall until we find the shape of the arch. The virtual surfaces constructed through orthographic projection to open this vicious circle: the mea sts can be known before a thi Diirer’s drawing have made any difference to the shape of the just emulate formwork, it could have been made without the draw- ing; if it were cut stone, then it could not. 3, was for several centuries also regarded as one of Europe's great geometers. If judgement on this point has mellowed somewhat,! he must sill be accorded a key place in the development of projective drawing. He was an accomplished practitioner and exponent of perspec tive as well as orthographic projection. In the Under~ ‘weysng der Messung (1525), his bookon the construc- tion of geometric figures, he illustrates a method for doing what he must have done in the fortress draw- ings: plotting information from a circular plan to a conic elevation by orthographic projection. He slices the cone with closely-spaced horizontal cuts, each of which i represented in plan by a circle of eorrespond- ing diameter (Gg. 4). The line or, an oblique cut through the cone, can then be thought of as a series of intersections with the closely-spaced slices. All one has to do is drop these intersection points onto the corresponding circles in the plan below. Diirer then added a third drawing that turned the resulting curve into the same plane as the paper on whi aking the horizontal dimensions from th ions from the oblique line Gr on technique of slicing a solid ‘cuts to facilitate the projection not have escaped the reader that Diirer us set of drawings to represent the cone as he did to represent the fort: plan, section, and elevation. The only difference is thatthe section of the cone is at an ‘oblique angle, not orthogonal ings were published two ye to assume that the set of was used to describe the abstract geometrical figure before it was used to describe the concrete forms of architecture. SRST eas Secemsonmactea Francesca had been exploring the same things in the same way. He left the first exp! orthographic projection in a bil iment of parallel projection, though excel dental. The treatise, De Prospectiva Pingens was about perspective. Diirer must have known of Piero's treatise, either Bologna.” Both artists were investigating a technique of perspective construction much easier to use with ARCHITECTURAL PROJKETION 23CAP KR Fig. 6, Lorenzo Sirgos, Penpecie Projeon of ae omits Plan ‘and Beton, for chapter x1 of La praia di Propeton (58). Engraving, plate 3.0 # 22.0 em. cca wn6ao0 Cage real objects. In fat, Diirer’s famous woodcut showing how to make a perspective image from an existing ‘object—a Iute—with the aid of a weighted line, two frames hinged together, and 2 cursor, indicates the straightforward optical realism from which issued the first accounts of orthographic projection (ig. 5). Piero demonstrated that a similar map of sight lines could ‘be made from drawings alone if the lute was replaced by itsplan and elevation. Similar demonstrations illus~ trating similarly curvaceous instruments can be found in many later works on perspective (fig. 6). It might for real things enabled the invention of othe ‘orthographic projection that brought the imaginary into the scope of perspective without relinquishing anything ofits precision. That is surely a matter of considerable importance in painting, but, since the subject ofthis essay is architectural drawing, attention will be restricted to the architectural implications of “orthographic projection. The first thing to be noticed isthe subordination of orthographic projection to perspective. With Piero this subordination might be attributed to his concen- tration on the latter. Yet the same bias can be dis- cerned not only in architectural writers such as Al- bert, Serlio, and Vignola, but also in the general level of coverage and the characteristic treatment of or- thographic projection asa preparation for perspective, and if not perspective, then asa preparation for some thing else, like making classical buildings, making sundials, making ships, or cutting stone.? While hun- dreds of treatises were published on perspective, there ‘were none dealing exclusively with orthographic pro- jection until the very end of the eighteenth century.” Even now the Engrlopedia Britannica has ninety-six lines on perspective and only nine on orthographic jection. In consequence, while perspective may still be lauded as the great opener of western eyes, ‘orthographic projection is relegated to the status of a technical matter: technical drawing; limited vision. Its usefulness in the pursuit of other tasks is what seems to have impeded its promotion as a form of knowledge. ‘The attention now being devoted to ortho. graphic projection by critics and historians of an- chitecture may, perhaps, change this, although some. times even they bring this same ingrained prejudiee to bear on the material they deal with. “The other ingrained prejudice that in hand withthe above is that orthograp! is either unconnected with imagination, active agency in the formation of images, and itis « very effective agency for the elaboration of imaginary ‘objects. That does not mean that it is good ‘well account for why it has sometimes been bad, us consider Diirer’s woodcut. The imagination radiates through the medium of projection, giving shape to fortifications. As Massimo Scolar there is something a litde sinister in the close assoca- tion between the development of certain kinds of parallel projection and military subject-matter, anas- sociation that was maintained till well into the nineteenth century, when the military applications were overtaken by industrial ones." To my mind these uses do not in themselves signify exhaustion, or ‘even corruption, of the imaginative faculty. Imagina- tion can be unprepossessing, but more usually its ef fects on the world at large are just ambiguous and dependent on circumstances. Diirer himself thought 0, quoting an old adage to make his point: “A sword is a sword, which may be used either for murder or justice." Tei, he says, only misuse that makes things bad: all well-made things, in themselves, are good. ‘The functional fortress may be justified this way no less easly, perhaps rather more easily than some of Diirer’s politically partisan representational projet, like the triumphal arch for Emperor Maximilian of his monument to commemorate the suppression of the Peasants’ Revolt.’ ‘The three drawings, plan, section, and elevation, are also partisan, Although theirs isa different typeof partisanship, it too comes of partiality. We cannot ee from Diirer’s plate what material the forcis made of ‘We infer from the forms drawn that itis masonry; butBut thenitis no lesslikely that, ver the centu: task has been adapted to the drawings. No really tll which. Orthographic projection means perpen projection. [tis c jectors are always perpendicular to the picture plane. ‘This isa relatively abstractidea that, in theory, has no defining or restricting effect on what is drawn. How- ever, in architecture, in practice, where it is tied up with other more obvious orthogonal relations itdoes. In architectural drawings the projectors are not ‘only perpendicular to the sheet of paper but also per- pendicular to the major surfaces of the nit, Buildings are often rectangular, surfaces with the surface of the drawing seems sensi- do; yet this convention of imaginative at way. Whether it does, rolling pin—whether, in other word ings into blocks or sheets~it is a powerful, conserva- tive, forming agency. Diirer’s fortis not a good example ofthis reci- procty between rectangular projection and reetangu- precisely bec is medium than a fish aken froman album of draw- ings asibuted t Androuet Du Cerceau, do, to some extent, show ths easy relationship, in a less Fig. 7. Workshop ofJacques And inthe Antique Mamer (Geeween © technically demanding use of projection. ‘They are also from the sixteenth century, and one can see that Jans, sections, and elevations describe and in describing them, give be assumed to follow. The Du Cerceau instance, is made up almost entirely of front eleva- tions, eighty pages of them (fi. . Fagades like these supply a promin from which the rest of the project may either extend ‘lack ink with Blick sash on vellum, 31.1 x 23.5 em. cea a ys60108017 in consequence or hide behind. Most are far more ‘ment with the page, drums of columns and domes. An alliance had already been struck between the abstractions of orthographic projection and che fun- damental organization of classical architecture. With subtlety bordering on subterfuge the drawing technique conferred properties on its subject; rectan- gularity, planarity, axiality, symmetry, frontality. As painting after the Renaissance was overwhelmingly: perspectival, so architecture after the Renaissance was overwhelmingly orthographie."* ‘The Du Cerceau drawings indicate three ways ARCHITECTURAL PRO}26 ARCHITECTURAL PROJECTION Fig. 8. Works eewcen cs (Offer lithograph, page 24.0 ‘out of thisimmensely stable relation between medium and form. Some of the drawingsin the album are pure orthographic elevations, but many are not. They in- clude il ations of aspects that should pri bee hidden. Sides and undersides are made visibl ting out from the planar fagade su tions and recesses. These have usually been added in such a way that they do not disturb the overall unity ofthe drawing. They do not compromise the planarity of the facade surface, but push and pull it into a thin slab of perspective space, no less frontal than pure elevation. ‘Closely associated with this technique isthe addi- tion of wash shadows inside the ink outlines. The first ten sheets of the album, which show the five orders with no deviation from pure orthographic projection, and with no added indication of shading (ig. 8), may be compared to the fagades in the rest of the album: the former look bodyless, the later, corporeal. This kind of shading had been developed in tandem with perspective during the fifteenth century. Transferred to orthographic drawing, it has the odd effect of dd- hat had been taken away by that ype very strange, [ think, that when any hard to tell the resulting figure from a perspective. Piero della Francesca’s central and parallel prjec- prism and sheet is established that orthographic pro- jection looks so different from perspective. It lols ‘more reticent, more abstract, later, muter. The i- troduction of shading restores what was lost. Once again the effect approaches that of perspectiv. Perhaps the most accomplished drawings ofthis ype from the early period are Antonio da Sangallo the Younger’s sections and elevations of his project for St Peter's made in 1520-1521. Other examplesinclude the late-seventeenth-century elevation/section of Santa Maria della Steccata, in Parma, atributed to ‘Mauro Oddi (cat. no. 6); Marchionni’s 1776 sacristy project for SePeter’s, Rome (cat.no. 7), and Antoine's fountain design from 1752 (cat. no. 9).‘cat normally made by the section would tend to make the middle the most emphatic, because the most fully described, part. The axial organization of much post Renaissance architecture is facilitated by this technical convention. Accordingly, most “correct” classical facades tend towards an aBa rhythm, whereas the Du Cerceau compositions tend to be AbA. This happens across the whole fagade, and also in miniature within lual pavilions and bays, to such an extent that ve three-partstructure is transfigured into «two-part structure joined by a hyphen (fig. to). The album therefore provides a surprising example of centrifugal composition, identified by the architec- Kaufmann as characteristic of lection of fagades all but turn the triparttion, iden tified by two more recent authors (Alexander Tzonis and Liane Lefaivre) as the essential structure ofclass- i co bipartition.'* I do nor say this to show that these scholars are wrong; I include itas an example of the way liberties were taken with the normative schema that was not just classical but orhographicand ‘These three ways out are very different combined presence suggests that while the norms of classical composition may have been supported by the conventions of architectural drawing, architects never conceded everything to the alliance. They would al- ways do something more; they would always extricate themselves. Nor that they were trying to effect means of escape; as one aspect was challenged or denied, another would be adhered to all the more closely: To push one thing you need to pull another. A system like this provides sufficient traction for such manipulation to take place. Next to be considered is an example of a type similar to the last. It reinforces the point that classi- ‘al architecture was not architecture that followed the most exquisite characteristics of which derived, as ‘often as nos, from contrary ideas, thoroughly at vari ance with the principles upon which the whole edifice of classicism was presumed to rest. This may seem an sppropriate place to make such a sweeping claim, the reason for making it isto show, again, how the drawing technique became the agency forthe taking ‘of liberties-liberties of an ingenious and subtle kind that were highlighted in this medium alone, nothing. ever being written about them. Bercrand’s Ombre d'un chapiteau Toscan ( thool work, from the Ecole des Beaux (Gg. 1). Drawings of deals of the classical or- ders were not only produced in quantity within the academies as part of the teaching program but were also the stock-in-trade of publishers. Descriptions of the five orders form large part of architectur ature from the early siteenth century to the nine teenth. The major authors devote either individual works to them or major segments of their major is difficult to find exceptions; even an im- probable candidate like Guarino Guarini, architect of buildings that hardly itinto the classial catex s0."® The orders have the advantage of providing, at ‘one and the same time, the most general and the most particular information abouta classical building, from the shape and size oft tiniest moulding tothe disti- bution of its largest pars. Books on the orders were vastly influential in spreading classical ideas through- ‘out Europe and beyond, changing western architec- ‘ure while giving ita more uniform complexion. Because of this, we might conclude that Ber- trand’s drawing represents the most stolid the most rale-bound, the most circumscribed, the most tic, the least vivacious aspect of the Antique/Re: sance tradition. We might even compare it unfavour- ably with earlier illustrations of the same sort, either ited versions in the works of G.B, Vignola fig. 1), jown architect made between )-ecould be said that Fig. 11. Giacomo Barozi da Vignola, Te Tascam Order, page tt erbnetars (560) Egraving late contrast, the nineteenth-century school drawing was entirely determined by measurement, dessicated by it. The difference is particularly noticeable in the treatment of shadows Incarlier drawings shadows aresketchedin intui- tively, the designer calling on his powers of obser- vation and memory to tease out the form, whereas Bertrand relies entirely on shadow project raphy). The exact lines ofthe east shadows res been chosen. Are we to conclude, ARCHITECTURAL PROJECTION 27fem, ca 197920037500 (at 0.23) this exercise Bertrand was being taught to trust the ‘mechanical procedures of projection rather than to trust his own powers of observation? This does not seem an unreasonable inference. The comparison be- tween Bertrand and Vignola might be regarded as unfair, because Vignola’s illustration was concerned with the column, whereas Bertrand was concerned ‘with shadow-projection, using the capital merely as a convenient working surface to throw thi across. But the fact is that during the I: and nineteenth centuries a larger and larger portion of all illustrations of the orders were like Bertand’s and like Brochier’s slightly later example of the whole ‘Tuscan order from the same school (ig. 12)—drafted with the utmost precision, either tinted with carefully Iaid washes that simulate the smooth modelling of ly dressed masonry in strong glancing light, igraved to similar effect. Through these studies ‘made more fully aware ofthe constitu- exqui Several writershave recently suggested a connec- tion between the development of what scriptive geometry, teaching methods geometry was the brain-child of Gaspard Mong a military engineer, mathematician, and practical sci- entist who rose to prominence during the French Revolution, and who, favoured by Napoleon, was to push through a radical reform of technical Polytechnique in Paris, the model for polytechnigue institutions throughout France, in which architecture ‘was taught side by side with enginees H skills. Only one subject was common toall the courses: ‘mathematics Descriptive geometry is a mathematically rigor- cous formulation of a set of rules, the acceptance of which makes it possible to describe any conjunction or intersection of geometrically consistent forms in space, witha minimum ofinformation and a minum of construction. It also involves parallel projection perpendicular to the picture plane, and could be de- scribed as a more powerful, more al generalized version of architectural dr first taught in conjunction with architecture in the aoles polytechniques, and was later added to the cur- riculum of many other schools.** ‘There are discernible traces of Monge’s descrip- tive geometry in Bertrand’s sciagraphy for a Tuscan capital. Monge required only two projections for even the most complex task. Descriptive geometry.was not from Géometriedesrip- 18 166 em; plate X¥,concerned to show what things were actully like i only to determine relations between ly defined bodies and surfaces, Monge sd that this could be accomplished with the bodily constitution of things drawn disappears leftis a confusing web of dotted and any of which are imaginary, bearing no obvious formal relation to the object re~ Because there are from which they were projected. The fundamen of drawings in descriptive geometry is therefore quite For con- not have to face what is drawn. Monge’s system did away with frontalty as well as substance. [As can be seen in Bertrand’s drawing, the new system could nevertheless be adapted to architectural use. If you imagine the drawing folded up along the horizontal line dividing the elevation of the capital from the half-plan, so the plan: to the elevation, you will see very important.in descriptive geometry because olds the two representations ina fixed relation and canbe used to great advantage. But no use ofitis made in this instance; here we have the format of Monge’ system without its fall exploitation. “The tracing of the shadow lines is done with the sid ofa series of vertical sections through the: lel to the direction of the su ar to Diirer’sslicing of the cone (see above). Having mapped these oblique section lines from the plan into the elevation of the capital, it to find the play of shadows on the double- faces. The sunlight, shining down at a cer- tain angle (represented by the slanted parallels {pink in the original] cast down from the abacus and as- tragal), makes tangents to the section the double-curved surfaces. Above the gency the column isin sunlight; below it in shadow. Join the points of tangency and you get the shadow line. The process is mos easily understood by looking at the astragal moulding at the top ofthe shaft. So even if this is not a consummate example of “Monge’s descriptive geometry, itis an example ofthe ‘more complete determination of architectural draw~ ing by.geometrical means. Ie might be argued that the play of sunlight on stone is nor materially affected by the way we draw it. Unlike the design of the classical orders themselves, itis merely a simulation of what ‘might happen after something is built. This, however, is the reason these drawings are so interesting. The after-effect is more vividly portrayed than the shape ‘of the capital itself. Now, apart from any argument about “scientific” drawis symptom of amod- ‘em malaise as some wi ‘classical architecture, never before or since displayed so clearly ‘There is something alte forbiddin about the combination of two such authorit of knowledge in one drawing. The authority of the classical orders and the authority of geometry leave no space between them for anything else. This then is surely the point at which the argument should be conceded to the critis cited above. Yet while we ‘would expect the integration of cultural norm and ‘mathematical truth to yield a product both unassal- able and moribund, this isnot the result, or soat least i seems to me. would explain its follows: it has to do withthe ‘way mechanical structure is illustrated in one way and simultaneously contradicted in another, The classical ‘orders developed out ofa structural system of columns and lintels. If one desideratum for a treatise on ar- chitecture was a description of the five orders, another ‘was an explanation of the origin of building from the primitive hut. The archetype of our way of building ‘was the Greek temple, and the precursor of the Greek. temple was a rude duelling of timber. Decorative fea- tures in stone buildings that had no obvious utility were traced back to timber constructions, and so legitimized. The historical truth of this interpretation of some if not all elements of classical building is dificult to assess.*8 But its very existence as an expla- nation brings outa bizarre feature of classical architec ture, A structure is shown not only for what i i, but for what it was on top of what it i. Does this not intimate an obsession with stability? Particularly well furnished with recollections of ‘wooden construction are the areas round the top and the bortom of columns. The complex details of torus and plinth, and of astragal, echinus, and abacus, refer to the hoops and pads once necessary to protect these ‘vulnerable points, for it is a fact chat any structure sade from separate columns and lintels will end to fail around these joints. They are the weakest parts. ‘The firs thing written of the Tuscan column in the first book to display the orders asa set (by Serlio) is this: “We find in Antiquities, and also in modern works, many pillars or columnes, which beneath in the joynts at the bases are broken asunder." So the ‘hetorical elaborations of capital and base would seem. to bein complete accord with the real structure ofthe building. They provide a reassurance in sign language that the structure is indeed safe and sound; that the parts in most need of strengthening have been prop- erly held, fixed, and made fast against the possibilty of collapse ‘The sturdy Tusean order not only stands frm ‘but insists on signifying that it does so.%5 There is, however, a third layer of structural interpretation on top of these two, one that has always been visible, though it has not, as far as I know, been discussed at all, Icis this third structural sense that is flaunted im the rote exhibition and studio drawings from the academies and polytechnics. Look once again at the Tuscan capitals of Bertand and Brochier. The shadows, precise as they are, dissolve the structural form. They do so by superimposing a derived pattern, a projection within a projection, which throws onefigure 100 etching with barin, plate v8 35.9 cm contour of the simplest of capitals against its own curved surfaces. Shadows are insubstantial and imper- manent. Their properties are exactly opposite to the properties of the column they glide across. The one thing they share, in this instance, is the frozen sharp- ness of geometric delineation; the indication of a strong sun held in the sky. And strangely enough itis, this one shared characteristic that allows the shadow to take its revenge on the stable column. ‘The shadows on Vignola’s Tuscan column were added in a painterly way to enhance our perception of the rotund but simple shapes that might otherwise By complete co Tuscan column, of circles, are eaten up by shadows-not shaft of the column, buc at its extremities, atthe points of greatest stress. Overlaying and obscuring the clear recognizable geometry and the clear recognizable signs are the distended curves, the sloping lenticular highlights, the sharp, disoriented, flexed triangles: 30 ARCHITECTURAL PROJECTION among the most common), is not of instability: rather lows the observer to imagine the structure as quickened instead of deadened at its crucial poins. These laboured school drawings show us how light can obscure one kind of meaning and supply another— something that was not in the curriculum. e projected drawing is, it should be said, no more aliberating agency thanis classicism, and count- less eases from different periodsex that it may be restrictive and conti may nevertheless be drawn ding examples is tha place to find what we are looking for, it may be mote xctive to look ina less likely spot. The apparently regulated freedom exhibited in the quadrarura de- signs of the Baroque and Rococo periods also berny evidence of an orthogonal order conferred on the de- igns by means of projection, though not as easy 0 cern as in May’s and Hassenpfiug’s work, Drawings by guadratura painters from the seven- teenth centuries vary greatly in 1 of them are vieruoso sketches, flair and vitality even though they were intended oly ry cartoons. Quadratura artists were com- ‘issioned to work on auilding once the architectanl shell was completed. They therefore had the advan tage of surveying the architectural effet in situ, hus enabling the painter to get a clear idea of the often complex envelope of surfaces that he was going to have to treat as a picture plane. By the late seventeenth century perspective had ed throughout Europe, and was sas a matter ofcourse. From the appeared technical treatises that ith the projection of perspective i ages ont al, flat plane of the kind generally favoured by artists, but onto a whole variety of other surfaces: inclined, spherical, cylindrical, and eonie* exists between the knowledge pr- i } _ictory commendation from wpher Wren, John Vanbrugh, and Nicholas Hawksmoor.?2 real space and the imaginary space they An unattributed: den shows a door- ime tuous Serlan frame (fg. 18). The drawing adopts ing. and the Tonic columns around it ue perspective (that is, a fronal ch the vanishing-point far over to one side, in this instance to the right). The question is: Does the sarily arise from our inability m from an effective represen- n che tympanum above the doo ywn as if seen squarely the miniature keystone, at Which begins to lean over to the If, plying, once again, that the observer i looking a itfrom the right-hand side. Above this, the perspective point of view, 17) Am enterprise like ‘a mixture of work de- tural space was, however, Pyrrhic. Viewed eccentri- lex anamorphosis of igures accentuating rts the wall itis probably following reason: if it were i seretched, collapsed, and the observer's consciousness of how the vault the picture, and, more interesting: the fictional space intimated by the less, these effects are achieved mechanic result would be obtained the offensive inconsistency obvious as one walked toward the door rawing represents the comer of also adapted to the a large room as seen from the centre. From this nage to combine leged position both real andillusionistcarcite= these two very different and apparently incompatible tureare inconsistent perspective. The only aberration ys of working? is the adjustment made to soften the effect of distr lar vault in Flaminio Minoz2is drawings, although tion that becomes more and more evident 25 the LL , which was not artist, Enrico Haffner, working on a si j2 ARCHITECTURAL PRsign for the painting of the Capella del Santissimo in San Giovanni in Monte, Bol case, two adjacent walls are folded out into a single plane, while the vault is developed in a broad strip ‘extending from the wall behind the altar (fig. 19). The perspective inthe soffits ofthe two arches above the comice indicates that Minozzi was imagining the altar. The only other indication of a correlation be- ‘ween these two arches is in the section through the ‘moulding at the head of the arch, visible at the left- hhand edge of the strip, asi the unfolded arch beside it had been folded back into place—which would suggest that the drawing, up to that level, is ortho~ ‘graphic. The treatment ofthe transition between the arch and the octagon drum above renders this reading implausible, however. It is impossible to say, from the drawing, exactly what space it refers to; butts posi- ble to say that the drawing must be mule-like and mixed: it cannot be read as orthographic projection, development (the laying out of faceted or curved sur- faces into a fla plane), or perspective, or any consis- tent containment of any one ofthese in any other. ‘The format of these quadratura drawings shows the painters thinking out their work, always contex- terms of unfolded orthogonal surfaces, closely identified with the real architectural shell, but not necessarily identical with it. In other words, they car- ried in their minds a perspective box from which a sequence of fat pictures would be transmitted to the surrounding walls and vaults, Pozzo's procedure of Fig 19, Flaminio Innocenzo Mino, Design or the Desration of Gospel in Son Gianni Mane, Bologna between oad 179) ‘em brow ink with own and rey wash over raphite ond taper id down on ove pape, 15 36) GM CEA DRIGEI005 mapping was a rationalization ofthis procedure. ‘The ‘Minozai drawings show the same format engaged with, and modified by, the shapes and circumstances ‘of specific interiors. Tuave peat thus far with orthographic projection, and to a lesser extent perspective, during the period dominated by classicism. There is no doubt that the essentials of contemporary architectural drawing ‘were mapped out then. What of modern architecture? Should we not expect to find it in mortal combat with these inherited techniques? Perhaps, but it was not that way atall. The question, as faras Tam aware, was never raised. Questions of drawing were raised, but not questions of projection. Whereas in painting vigorous attempts were made by Cubists, Futurists, Suprematists, and Constructivists to destroy the shibboleth of perspective,” and whereas, in architec ture, various other reminders of past practice were under attack—orament, Art, stone, ete.—no such campaign was mounted against orthographic projec tion, which remained the inviolate medium of ar- chitectural thinking. “There have nevertheless been atleast two signifi- cant changes in drawing practice during this century: the increasing prominence of axonometricprojection, with its subsequent incorporation into the conven- tional set of architectural drawings; and the more fre~ quent resort to, and greater investment, the sketch. Neither is universal, yet both warrant attention, The sketch is a peculiar phenomenon. It is impossible to decide, except by dogmatic means, whether itis a projection or not. In 0 far asi is like a scale drawing, itis projective; but its capacity to absorb so many other interpretations, to be whatever ‘one wants to see init, and to multiply ambiguities and inconsistencies, make it work quite differently. So it ‘would notbe right to classify itasan imprecise approx- ‘of a projection. Its relation to its object is far mncertain than with the drawings discussed so ‘more a matter of suggestion than designa- tion. And this is why its increased prominenceback, keeping everything in a state of suspension, of refusing to give in too quickly to the peri, a way of staving off the fixation ofa particular figure or shape. ly applied tothe sketch are those of conception, gestation, and birth. Its norphous, unformed, embryonic character is what etches fora Congress ke smudgesthandraw- ig, show to what extent line and figure may be held abeyance that way (cat 1354) Ie is true that different architects use the sketch different ways. The expressive sketch is familiar enouy ial feeling is recorded in a dynamic calligraphy; and the ensuing architecture tries to follow the original trace as closely as it might, suggesting that all inspiration had been released an 3 frst few seconds. Mendelsohn worked his early career. The drawings by Poelzig for the Grosses Schauspielhaus, Berlin, are also ofthis sort (Cat. nos. 133-1~133:5). Kahn’s are not. His sketches mutated quite suddenly into something els. ‘Out of the blurred charcoal and the eryptic ciphers a ‘complete configuration would erystalize. The model of the Congress Hall shows the same project inthis other, suddenly definite shape (cat. no. 135.4) ‘There is no real evidence thatthe clear configu ration was born out of the indefinite sketch. If any- thing the evidence leads in the opposite direction. When Kahn later described his first idea for the C gress Hall, he described it as a geometric configur tion, much as is found in the completed scheme, Undemeath the blackness of the sketch is the same ‘obfuscated figure. It was already there. Morcover, while the numerous pentimenti in the architectural sketches of Michelangelo and Borromini were used to ‘modulate and modify forms, the form in Kahn's sketch undergoes no obvious variation ‘Kahn's fascination with and expos measurable aspects of architecture are well known, and that is exactly the property that marks the sketch ‘out from other forms of architectural drawing; its in- commensurability. It seems that the sketch was, for m, an illustration of the way order emerges f ofthe un- “ FCTURAL PROJECTION chaos, more than a divining instrument for finding it. Ie was the sign for, rather than the location of, creati ity. It was only incidentally a propedeutic device, a yet gave proof that Art was afoot. I suspect that it was also a constant reminder ofa principle that Kahn felt he had to repeat over and over again (to himself or cothers?): that the commensurable order of arc ture, o overwhelming in his buildings, was only a way to achieve incommensurable effects. He was careful to preserve his sketches, and happy to publish them. ‘The complementarity berween geometric and at- ‘working, And yer the broader development of twentieth-century architectural dra divergence toward similar extre sketch has obtained greater the axonometric projection, and the ax forms of projection, the one most confined $0 too has new kind of space proper to the twentieth cen tury." \etric) projection, so that we look upinto the cei- joined to to sides ofthe put to positive esthetic use. The spatial qualities int ‘mated in Lissitzky’s Prouns and Doesburg’scouner- compositions were unde1 then faced by both these painters who had tured hitecture was not so new, although it mas with unusual force in their work How sic possible to convey the properties so powerfully pres in architectural drawings into the construcions they represent? Such transmission is possible, butia as much as it refers to ambiguous and fluctuating tial registrations, it is not. This kind of flucwation cannot be directly transmitted into three dimension. ‘That is why, in Hejduk’s drawings for the news 2.1 = 12,5), the further elaboration of a similar species of ambiguity keeps the project firmly on the surface of the paper. Votes 1. Marcel Jean, The History of Sureaist Painting lor (London: Weidenfeld and Nicholson, 1960 ind Fiumphrie "Le Corbusier & Phot graphy” (anpublished 3. Robert Branner,“V Origin of Gothic’ ArBeau Are (Paris/New York), 6th series, 1x1 (March 1963), 29-146. ‘iumphal arch as pul engravings in 1515. See Albrecht Direr, Maxim Traumpbal Arch, ed. E. Chmelarz (New York: Dove, ‘The monuments commemorating the suppression of the bridge, Mass: Harvard, does not expresly state this a @ prin throughout the book 18, Aletander Tronis and Liane Lefaivre, Casal Arcit supe: The Paeties of Order (Cambridge, Mass: sr, oa. 19. Guarino Guarini, Arbieaura Civile (Turin, 25, Joseph Rylowert, On Adam's Howse in Paradise (New 26. Jugis (Cambridge: Cambridge University, bor hinecrare for World Reval Diuhosch (Cambridge, Mass: srr, instance. Its ‘Cubist paintersused multiple pers 31. Louis I Kahn, 17Bat Will Be Has Abveys Been, ‘Wiarman (New York: Rizzoli, ARCHITECTURAL PROJECTION 35,Edited by EVE BLAU and EDWARD KAUFMAN With Essays by ROBIN EVANS, EVE BLAU, EDWARD KAUFMAN, WILLIAM ALEXANDER McCLUNG, HELENE LIPSTADT, ROBERT BRUEGMANN DISTRIBUTED BY THE MIT PRESS, CAMBRIDGE, MASSACHUSETTS, AND LONDON, ENGLANDCLE (oe c < ARCHITECTURE ANDITS IMAGE | Four Centuries of Architectural Representation Works from the Collection of the Canadian Centre for Architecture Centre Canadien d’Architecture/Canadian Centre for Architecture Montreal, 1989