“Architects typically produce drawings and other artifacts—words, inscriptions, models, full-scale mock-ups, and so on—that allow buildings to be realized by others, at a distance from their authors. This has not always been the case. Gothic building, for example, was a constructive practice of geometric rules and traditions where master craftsmen operated to some extent as architect and builder simultaneously. In this situation, drawings were not the primary mediator between design and construction; discussions regarding the final form of a building continued throughout the construction process. . . .”
*Fig. 1: Robin Evans, “Projection and its analogues: The Arrested Image,” 1995.
images, clockwise from top left: Fig. 2: David GEORGE, plan drawing; Fig. 3: Laurel CLARK, plan drawing; Figs. 4a–4b: Laurel CLARK, pattern transformations; Figs. 5a–5b: David GEORGE, pattern transformations. All figures the result of design studio projects, Rensselaer Polytechnic Institute (RPI), 2012. Adam DAYEM, instructor.
Architects typically produce drawings and other artifacts—words, inscriptions, models, full-scale mock-ups, and so on—that allow buildings to be realized by others, at a distance from their authors. This has not always been the case. Gothic building, for example, was a constructive practice of geometric rules and traditions where master craftsmen operated to some extent as architect and builder simultaneously. In this situation, drawings were not the primary mediator between design and construction; discussions regarding the final form of a building continued throughout the construction process. Later, during the Renaissance, as the intellectual work of design became separated from the labor of construction, drawing assumed a more primary role in manifesting architectural ideas. This happened first in a relatively unsystematic way,1 and eventually evolved into the contemporary condition, where construction drawings have become legally binding documents intended to convey information to builders with little possibility of misinterpretation. Questioning the role of drawing in architectural education, then, means questioning conventions that have been held by the discipline since the Renaissance.
The widely accepted role of architectural drawings as mediators between ideas and buildings is described by architect Stan Allen in his comparison of autographic and allographic arts.2 Autographic arts, such as painting or sculpture, occur in direct contact with their author. Jackson Pollock’s painting practice, for example, relies on direct interaction between paint, the canvas, and the body of the artist. This is compared to allographic arts, such as music or dance, where the composer or choreographer is generally not the player or performer of the work. Here, a specialized graphic language is used to communicate the author’s intent to the performer. Sheet music and dance notations are examples of this type of graphic language, which is decoded according to accepted conventions and thereby serves as instructions for a particular type of performance. For Allen, “An architectural drawing is an assemblage of spatial and material notations”3 that is decoded according to accepted conventions and also serves as a set of instructions for a performance—in this case, the construction of a building. As such, architectural drawings have traditionally acted as descriptive documents. However, this type of descriptive drawing is disappearing in architectural practice as three-dimensional digital models are increasingly used to communicate with those who construct buildings.
In fact, the practice as a whole may be moving back toward a Gothic model, where construction is more directly controlled by the architect via digital design and fabrication tools,4 or a biological-chemical model, where design emerges from automated growth protocols.5 Shifts in either of these directions, toward the Gothic or the biological-chemical, may tend to diminish the allographic nature of the practice, and diminish the role drawings play as intermediaries between ideas and construction. However, even taking these ongoing changes into account, it remains necessary and desirable to approach architecture as an allographic art. Necessary because the complexity of building systems will, for the foreseeable future, require some sort of communication between human designers and constructors. Desirable because in an allographic art, information exchanged between designers, their design instruments, and their designs creates a complex network through which the products of a creative practice are effectively materialized. Within this network of exchange it seems there will continue to be a role for drawing in architecture.
At the conclusion of his book The Projective Cast, Robin Evans formulates a diagram titled “The Arrested Image.” (Fig. 1) The aim of this diagram, Evans writes, is “to define the different fields of projective transmission that concern architecture.”6 For Evans, projection is the transfer of information from one repository of knowledge to another. The diagram describes four repositories of knowledge: orthographic projection drawings; perspective drawings and images; the designed object;7 and the human designer or observer (which is further subdivided into perception and imagination). Each of these four repositories projects information to, and receives it from, the other three. Design is thus described as a process of continuous informational exchange between the four repositories. Information does not travel smoothly through the network; each type of projection contains some level of contamination or distortion, and the direction in which information travels matters.
“The Arrested Image” appeared in 1995. Since then, digital design and fabrication technologies have altered the status of all four repositories, as well as the nature of information exchanged between them. Outdated though it may be, Evans’s diagram is still an effective framework in which to investigate the evolving roles of the contemporary architectural drawing. Certain projection lines seem to be of greater importance in an academic context, where the designed object is rarely realized, and they will be discussed here: Type 1, connecting orthographic views within a drawing; Type 2, connecting the designer-observer and an orthographic drawing; and Type 6, connecting the designer-observer and perspective images of a project.8
Fig. 6, left: David GEORGE, jointed components; Fig. 7, right: Laurel CLARK, jointed components. All figures the result of design studio projects, Rensselaer Polytechnic Institute (RPI), 2012. Adam DAYEM, instructor.
Projection Type 1 moves information within a drawing, associating various orthographic views and expediting drafting. Digital modeling eliminates the need for these projections by storing all dimensional information in a single three-dimensional construct. When an orthographic view is required for analytical or descriptive purposes, it may be extracted with relative ease from the model. (Figs. 2–3) Information projected for dimensional purposes within a drawing, then, is no longer of much use.
Information projected within a drawing may serve other purposes, though, such as registering change or variation. (Figs. 4–5) As in an orthographic drawing, projection lines in these pattern-transformation drawings describe something other than the edges of geometry; in this case, they connect the beginning and end states of scalar or rotational transformations. By placing graphic emphasis on the projection lines, the drawings register a series of parametrically controlled transformations that vary across a field. Lines that represent change rather than geometry allow difference to be articulated with high levels of specificity without immediately fixing form. This abstraction, combined (in the case of this particular design-studio project) with a forthcoming translative leap into three dimensions, required students to move a strictly parametric drawing into more open-ended material investigations. (Figs. 6–7) In this way, drawings effectively catalyzed design proposals without imposing overly deterministic solutions.
Projection Type 2 is information traveling between the designer-observer and the orthographic projections he or she draws. In the direction from drawing to designer, the information transfer is relatively straightforward. As with any language, reading the drawing is a matter of conventional fluency. Architectural drawings are a mix of intuitively legible spatial dimensions and material notations;9 as long as the conventions of drawing are understood, information passes with relatively little distortion. Information projected in the opposite direction, from the designer-observer to the drawing, is a more complex matter. Evans describes designing as, “[A] performance during which vision maintains a constant interaction between manual movement and resulting inscriptions.”10 Information from the imagination of the designer is registered on the page in a manual performance involving the body and the tools of drafting. The drawing sets up a feedback loop with the designer where information is continually registered, evaluated, and acted upon. In effect, the drawing becomes a device for passing imaginary information across the perceptual threshold (the dotted line in Evans’s diagram) and into the material world.
In this way, “The Arrested Image” describes the generative capacity of drawing—where drawings operate in dialogue with the imagination of the designer and are more instruments than artifacts. Evans considers orthographic projections to be the primary design instrument, and there was certainly cause for assigning them this status when “The Arrested Image” was formulated. In predigital design environments, the speed and precision of an orthographic projection as an instrument for exploring or describing space and form was unmatched. The relative simplicity of constructing and modifying orthographic projections fostered a productive information feedback loop between designer and drawing, which in turn allowed complex works to be studied and refined iteratively. As much as anything else, this instrumental efficiency gave orthographic projections status as the privileged design instrument for approximately 500 years. Of course, digital design environments have altered this situation. There are two main working differences between Type 2 projections in Evans’s diagram as compared to the contemporary design environment. The first concerns the shift from manual to digital production.
Fig. 8, left: Tiffany IP, drawing of cloud formations; Fig. 9, right: Melanie NGAMI, drawing of cloud formations. All figures the result of design studio projects, Rensselaer Polytechnic Institute (RPI), 2012. Adam DAYEM, instructor.
A primary way that digital design environments distinguish themselves from manual ones is through the expanding opportunity they present to incorporate external information into the design process. This type of information includes software-specific drawing operations (Figs. 8–9), parametric inputs (Figs. 4–5), algorithms, and other types of data used to systematically organize form, program, and structure. When digital work associates itself with external resources, it is less under the direct control of the designer. Jesse Reiser and Nanako Umemoto provide a particularly apt description of the role of the architect as “neither a passive observer of determined systems nor a determined manipulator of passive material, but rather the manager of an unfolding process.”11 Media theorist Steven Johnson observes that designers are becoming more tolerant of this oblique control and more comfortable with the partial effect they exert on systems with their own embedded intelligences.12 Johnson speaks specifically of emergent systems, and many architects work with systems that have emergent qualities: algorithms that script low-level rules of interaction to produce higher-level pattern or order. The power of an algorithm, or any other external information resource, “lies in its ability to infer new knowledge and to extend certain limits of the human intellect.”13 Digital drawings that incorporate external resources are relatively new, but design methodologies intended to infer new knowledge are not. Peter Eisenman’s autonomous form-finding exercises in architecture, or John Cage’s compositional techniques in music14 are examples of design methodologies that incorporate externalities to produce new knowledge. Digital tools expand opportunities to incorporate external information into the design process, and drawing is one of the mediums in which these opportunities can be capitalized upon.
The second working difference between Type 2 projections as compared to the contemporary situation has to do with the relation between two– and three-dimensional production. In “The Arrested Image,” Type 2 projections travel between two-dimensional representations and the designer-observer, while Type 6 projections travel between three-dimensional representations (perspective drawings and photographs) and the designer-observer. In Evans’s description of the diagram, exchanges with two-dimensional representations involve active feedback between a design instrument and the designer; exchanges with three-dimensional representations have more to do with the consumption of images, a one-way flow of information from image to designer. For Evans, orthographic drawings are design tools; perspective drawings are not—a reasonable assumption given the difficulty of constructing a perspective view in the predigital world.
Working with three-dimensional digital models produces a different type of projected information that is a mix of Type 2 (orthographic projection to designer-observer), Type 6 (perspective image to designer-observer) and Type 7 projections (designed object to designer-observer). Evans notes that the visual experience of a three-dimensional form like the designed object is relatively straightforward, but unpredictable due to the vagaries and incompleteness of vision. When working with three-dimensional modeling software, perspectival views on the computer screen scale and rotate, and viewing them becomes similar to moving around a physical object. The effect of viewing a design from constantly shifting positions is decidedly different from viewing it from a fixed position, as one would when working a two-dimensional orthographic projection. The perception of three-dimensional forms, in the world and on the computer screen, is never fixed, not only because humans move and eyes move (or a digital model moves), but also because movement is fundamentally present in vision. Brian Massumi argues that in the spirals of decorative motifs one sees not a spiral, but spiraling.15 There is a form and an “abstract dynamic”16—the two cannot be separated. Massumi goes on to describe the perception of volume: a viewer does not infer the hidden faces of an object, but sees them as part of seeing the object. This is the perception of an object rather than the deduction of a surface. Form is full of things we do not actually see: “When we see an object’s shape we are not seeing around to the other side; what we are seeing, in a real way, is our capacity to see the other side.”17
In the digital world, two– and three-dimensional productions are more closely related than in the predigital world. In contemporary parlance, a so-called “drawing” may loosely refer to a construct extracted from a three-dimensional model without regard for a specified projection system. A more precise definition would be that a drawing is worked in two dimensions (regardless of whether it is a two– or three-dimensional projection), specifically to avoid the three-dimensional perceiving described by Massumi. A critical aspect of design is uncertainty—that is, not being able to see what is on the other side. Drawings allow productive uncertainties to enter design in ways that models do not.
“The Arrested Image” shows how architectural design works as an expanded set of dynamics in which drawing has historically played a pivotal role. In order to make assertions regarding how drawing functions in a contemporary design environment, and claim for it potential futures, the diagram has been updated. (Fig. 10) There are four structural differences in this new version. First, the “perspective” is replaced with the “model”—this is assumed to be a three-dimensional digital model. Second, a new repository of knowledge, “external resources,” has been added. This includes information in the categories of software operations, parametric inputs, algorithms, and other data used to systematically inform design. The “external resources” node has been placed across the thresholds of imagination and perception because it becomes integral to how design potentials are imagined and perceived. It has been placed behind the threshold of observation because it becomes a filter through which the designer observes other parts of the design network. Effective incorporation of external resources creates a hybrid human-nonhuman designing and observing apparatus. Projections between the human designer, design instruments, and the design remain—though they are considered to be of secondary importance. Third, the model (previously the perspective) and the design (previously the designed object) are more closely related than before. As digital models become increasingly comprehensive representations of buildings, and digital fabrication tools reduce the significance of Type 5 projection lines, the model is becoming more like the design. Fourth, models and drawings are more closely related than before, but deliberately held apart. Digital tools tend to conflate drawings and models; two-dimensional drawings can easily be extracted from three-dimensional models, and information from two-dimensional drawings can feed directly into three-dimensional models. While acknowledging this relationship, a clear distinction between drawings and models preserves a speculative, abstracted arena where more unpredictable explorations can unfold. Within this updated network of exchange, the following futures for drawing are proposed:
1. Drawings as Abstractions
Apart from a familiarity with ideas to which members of the building trades ordinarily had no access, and a freedom to think about buildings in terms outside the normal conventions of the building trades, the particular skill which such men displayed was the ability to draw, and to represent their ideas graphically, in such a way that they could both be discussed with patrons, and also translated into buildings.18
What most architects fail to realize is that the computer is not a better way of drawing your [sic] projects; it is the instrument that could put an end to drawing altogether in architecture. Drawing is not the best instrument to “think architecture”. It is a pale reproduction of the images that go inside our head. The best way to test a design solution would be to build it, and build it ideally to real-life proportions in a virtual environment.19
As two-dimensional constructs that point to three-dimensional constructs, architectural drawings always contain a level of abstraction. Whether done speculatively as projections into the future, or analytically as descriptions of something that has come before, some slippage between a drawing and its associated third dimension is inevitable. A drawing is decidedly not a model, and translative leaps between two– and three-dimensional work require imagination. Drawings are not the only instruments that provide the freedom to think outside conventions normally held by the discipline or practice of architecture. And they are not the only documents that can be used to graphically represent ideas for discussion with patrons or other participants in the building trades. Models may perform these roles just as well or even better. However, where a model often answers a question, a drawing is more apt to leave productive informational gaps for the designer-observer. In the digital world, drawings are actually more free from becoming trapped into representations of preexisting images, and thus more able to act as instruments for abstract speculation. Overreliance on testing design solutions in “real-life proportions in a virtual environment” would be an unproductive normalizing force in the discipline of architecture. There is a need to preserve an arena for abstract thought.
2. Drawings as Autonomous Documents
“The Arrested Image” demonstrates how drawing gains its instrumental capacities as part of a larger network of design components. However, the autonomy of drawing is also important. As Evans suggests in numerous ways, drawings are not deficient versions of buildings, but worlds unto themselves that create theoretical universes alongside the built environment. Part of their value lies in their potentials to produce a variety of viewpoints in a variety of subjects. Design is not a solitary activity; critical discussions of work require engagement in a larger cultural context. Buildings and models of buildings play a role in these discussions, but the abstraction in drawings is uniquely subjective, and a level of subjectivity is productive in the discourse of any cultural activity concerned with the production of novelty.
3. Drawings as Methodologically Flexible Tools
A methodology is a comprehensive system of techniques governed by a theory for why things should be done a certain way.20 Different approaches to design can operate under different methodologies, which is to say that there is not one right methodology, and a methodology need not become an ideology. The methodologies for producing the work discussed here employ techniques that incorporate information from outside a drawing in an attempt to produce formal and conceptual novelty. While this approach could have operated in other media—manual drawings, three-dimensional models, animations, texts, and so on—which would have led to different results, the work would have still been in the same spirit of experimentation. This is not an argument for or against the usefulness of drawing, but a reminder that the medium of drawing is flexible and capable of operating within a range of different design methodologies.
4. Drawings as Ethical Constructs
The flip side of flexibility is particularity. Drawing is not a neutral medium. No medium is neutral. The nature of any medium influences how ideas arise and are subsequently communicated. Two-dimensional drawings and three-dimensional models pass different information back and forth with the designer-observer. They also communicate ideas differently, both between individuals and into larger communities (architects, builders, critics, and the broader public). All two– and three-dimensional, digital and physical instruments available for use in the contemporary design environment produce value in particular ways: as Alberto Pérez-Gómez has noted, “Value-laden tools underlie the conception and realization of architecture.”21
Digital tools have reconfigured “The Arrested Image,” and it is inevitable that the projected network it describes will continue to evolve into the future. However, as long as the complexity of the projected network remains (as long as architecture functions as an allographic art), questions as to whether the changing roles of architectural drawing are productive or unproductive become ethical ones. Any position established for the role of drawing in architecture must deal with the ethical stance of the designer in relation to cultural norms. Drawings have been a primary design instrument since the Renaissance. This has ascribed particular values into architecture and thus the buildings, cities, and cultures we inhabit. Perez-Gomez has argued that the advent of digital tools in architecture has not improved the quality of built environments22—but improved according to what value system? Without question, digital tools have begun to change our built environments. Any pedagogical approach will bias one type of design environment over another, and in turn this bias will produce projects that ascribe certain values into culture. The responsibility of architectural educators, then, is to evaluate how one’s tools, drawing included, produce value and whether it is appropriate to take ethical responsibility for this value.
Fig. 10: “The Arrested Image Revisited.” Evans’s “The Arrested Image” (1995), redrawn by the author, 2012.
1. Alberto Pérez-Gómez and Louise Pelletier, Architectural Representation and the Perspective Hinge (Cambridge, MA: MIT Press, 1997), 7–8. ↩
2. Stan Allen, Practice: Architecture, Technique and Representation (London: Routledge, 2000), 31–36. ↩
3. Ibid., 32. ↩
4. When digital models control laser cutters, mills, and 3-D printers the need for drawings to act as intermediaries between the architect and built work disappears. For example, R&Sie(n)’s project “Olzweg” proposes integrating the machine that builds with the project itself, and thus collapses the traditional relationship between drawing and building. ↩
5. Consider the work of Zbigniew Oksiuta, who argues that buildings will be grown through biological and chemical processes: “In a world created by human beings, machines, instruments and buildings are hard structures, dead objects. They represent our mechanical view of the universe. Although in the language of civilization, we continue to talk about growth and development, these terms have nothing to do with the natural processes of development. . . . Can life processes, which normally take place in water in the nano-scale, happen in a macroscale? We understand that the shaping of liquid membranes in an architectural scale, or a stretching of a living cell to the size of a house is not possible. It is necessary to develop new ways, until now not existing in nature, which would make it possible for the biological processes to occur in an architectural scale.” Zbigniew Oksiuta, “Biological Architecture. Living Membranes,” paper presented at “Structural Membranes,” the third International Conference on Textile Composites and Inflatable Structures (2009): 1–3. ↩
6. Robin Evans, The Projective Cast: Architecture and Its Three Geometries (Cambridge, MA: MIT Press, 1995), 368. ↩
7. “Designed object” seems to be an inadequate description of what architects do—that is, architects design systems, movements, and performances, things that cannot be adequately described as objects. But for now, Evans’s description is left as is. ↩
8. This essay is being written at the conclusion of a series of design studios that have focused on and celebrated the production of two-dimensional drawings. Drawings in these studios have always been accompanied by three-dimensional work. Sometimes drawings have preceded three-dimensional work and projected into it; at other times they have come after three-dimensional models and functioned as descriptive documents. In either case, a question arises regarding the role of drawing in architectural education today: As digital design and fabrication technologies evolve, will drawing remain relevant? If so, how? ↩
9. Allen, Practice: Architecture, Technique and Representation, 32. ↩
10. Evans, The Projective Cast, 367. ↩
11. Jesse Reiser, Atlas of Novel Tectonics: Reiser + Umemoto (New York: Princeton Architectural Press, 2006), 104. ↩
12. Steven Johnson, Emergence: The Connected Lives of Ants, Brains, Cities, and Software (New York: Scribner, 2001), 177. ↩
13. Kostas Terzidis, Algorithmic Architecture (Oxford: Architectural Press, 2006), 65. ↩
14. “I’ve never thought that, as many people do, that music should be in my head and that I should learn how to write down what I already hear. I really can’t hold a tune and I don’t know solfège at all—so I found ways of writing music to produce sounds that I haven’t heard, and that other people haven’t heard. If I’d studied solfège, if I had a feeling for harmony, which I don’t, I think I would simply write what people have already heard or what I would’ve thought—thought I heard. The result is I have a curious feeling every time I write a new piece . . . I think up until the last minute that maybe it’s just going to be a great mistake or that we won’t discover anything with it.” John Cage, National Public Radio (NPR), 1992. ↩
15. Brian Massumi, “The Thinking-Feeling of What Happens,” in Massumi et al., Interact or Die: There is Drama in the Networks (Rotterdam: NAi Publishers, 2007), 72–74. ↩
16. Ibid., 73. ↩
17. Ibid. ↩
18. Adrian Forty, Words and Buildings: A Vocabulary of Modern Architecture (New York: Thames & Hudson, 2000), 30. ↩
19. “The end of drawing – a new dimension?” GRAPHISOFT, accessed August 2012, http://www.graphisoft.com/community/envisions/2005_02_3dfuture.html. ↩
20. David Ruy, “Lessons from Molecular Gastronomy,” Log 17 (Fall 2009): 35. ↩
21. Pérez-Gómez & Pelletier, Architectural Representation and the Perspective Hinge, 3. ↩
22. Ibid., 377. ↩