The building is the concept / non-linear design

I recently answered a question on Quora about problems with finished student design problems. My answer there is adapted here to interlink two issues – the idea of concept versus building or proposal and the problem of linear design.

I abandoned linear design long ago. This is the idea that you work from the most general to the most specific, from concept to building then to detail, from large scale (urban/site) to small scale (building/room/detail). Although it was initially a reaction to the amount of time students wasted trying to develop ‘concepts’ or doing endless site analysis, in time it proved to go deeper than that. Much of the stalling over concept and site analysis was really procrastination, putting of getting to a building because this is perceived to be the hard bit and students were understandably eager to put that off. However, this approach also freezes over into a belief – that you cannot design (or start design) without a concept or without site analysis.

I’ve implemented a couple of exercises that help the student get out of this trap and rethink what a concept is and where it comes from. In one exercise I ask students to come back with a building proposal section two or three days after the first site visit. I always ask for these to be in scale and to be a building section, not a sketch, concept or massing strategy. These sections are always bad, naïve and based on preconceptions. But they are also based on intuitive responses to both brief and site. They are about something the students sense about the place, its people, history or their own responses to those things. With some of that brought to the surface we can then move on to site, program and precedent research that is relevant to the students’ project intention. The first section also makes immediately clear what they do or don’t understand about the site (views, overlooking, rights to light, ground conditions, drainage, access issues, parking, etc.). The first section, and every subsequent one, along with later plans, models and 3d drawings are part of the simultaneous analysis of site, program and design strategy. It is also simultaneously analysis and design. And in this way project concepts and ideas emerge, they are not imposed and derived from isolated acts of thinking, drawing and data gathering.

Another exercise is to start designing by focusing on a human scale ritual, event or routine and developing an architectural fragment around this. Again, this is done in scale and with material and structural ideas included. This approach foregrounds program development, analysis and ideas about human inhabitation. Like the exercise above it makes clear at an early stage what intentions are viable and what type of research is needed to support the student proposal. Projects are then built up from discreet elements into a larger whole. For example, if a threshold is critical to the idea of a project then one can conceive of a site and building strategy that enables a particular type of threshold and arrival sequence. Again, building ideas and concepts are developed rather than proposed at the outset.

Procrastination is avoided because from the start students are designing via their building rather than abstract visualisations, massing models or sculptural gestures. Solutions to pragmatic problems can be made part of a project strategy rather than something to ‘be solved’ later on. It is true that in professional practices most countries spell out design stages and phases (from programming to schematic design to design development and construction and so on). This has to follow a linear format for administrative reasons. It is not a formula for design.

Two important conceptual issues and a third pragmatic one are addressed by this approach. The first is that we disrupt the fiction of linear design. Good architects work at various scales and level of development simultaneously; they can think about materials and details at the same time as site strategies. The most visible example of this are the drawings of Carlo Scarpa, where plans or sections are littered with various sketches in the margins that correlate different scales and stages of design.

The second issue is that it encourages students to believe that ideas are best explained and demonstrated in a finished product (normally a building, but in school it is a set of final drawings). The building isn’t something separate from its concept and concepts don’t live in concept drawings. Therefore the best way to test the validity of even the most abstract concepts is to draw up your building proposal as early as you can and as often as you can, even when it is a lumpen, ill-formed, and confused mess. When you begin to see your ideas in the stairs, windows, room layouts, movement patterns, wall thicknesses, in cross-sections and in floor plans then you have started to make your idea clear as architecture rather than as symbols of an idea. Layering on details, particularly fiddly ones, or embellishing drawings of an undercooked scheme is like decorating a bland cake. It may look nice but it won’t taste good.

Finally, on the pragmatic front, this approach leads to much fewer issues with completing projects. You will always have a building to hand that you can draw up. And you will spend more time developing the very think that gets examined in the end – the project proposal.

However, you need the support of your tutor and this may not be forthcoming. There is indeed a culture out there that keeps suggesting major design revisions right up to the end of the project. And this is often because they themselves are looking for that magical fully-formed solution that perfectly captures an idea. Nevertheless, most tutors want students to drive their own projects and to be self-motivated and proactive. If it’s up to you, don’t fall for the myths of linearity and for the idea of project follows concept.




The wall between inside and outside: a concept

The French architect Henri Ciriani used to tell the parable of Michelangelo’s square (an imaginary square that he might draw to start off a design). It stated that when Michelangelo drew a square, all sorts of decisions and conditions were already known and decided. michelangeloplan1aThat is, as soon as you know the length of one side of the square, you knew how high you could go because you know what it was being made of – stone or marble, for example. It also suggested mass, the amount of light that came in, how open or closed it was, and so on. This was given by the limitations and knowledge of available building materials and construction technologies. When Mies van der Rohe drew a square, nothing was decided. That is, with modern building technologies the height, density, character, transparency, etc., are all up in the air. Mies-50-01For Mies, even the fact that he might place steel columns at the corners meant very little. The proposed space could still be very high, or low, very open, or closed, massive or light. In this sense, for Ciriani, modern architecture requires that you decide everything, or put another way, that you determining everything. Pre-modern space was of considerably different character than modern space. The expectations of exterior space (generally public) where well-known as were those of interior space. I would add that Michelangelo’s square determines its walls or piers which in turn determine the relationship between interior and exterior – and it did this with a single gesture. The material was often the same on each side. Modern space is less clear about inside and outside. Picture here any canonical modernist project breaking the box with flowing exterior and interior space together here (Wright, Schindler, Rietveld, Mies). The modernist wall then had to qualify an increasingly uncertain and abstract exterior while qualifying an increasingly polyvalent and abstract interior. I have it from a good source that Herzog and de Meuron choose to privilege the inside or the outside. For example, when they want an exquisite flat and abstract exterior surface this means the details on the interior are often horrid (and vice versa). David Adjaye supposedly demands that both interior and exterior work equally well. This makes things much harder for the designers, detailers, builders and clients. It is not a question of a right or wrong approach; this is simply to say that there is a choice, and if there is a choice there is an issue, there must be something at stake.

It then occurs to me that Louis Kahn understood this problem and came up with an interesting solution. Knowing the differing demands of the inside versus the outside and the difficulty of reconciling these Kahn decided to make the wall spatial. More precisely he doubled most dividing surfaces (as did Michelangelo). salk_sectionFloor and ceiling become two planes rather than one. Exterior walls are doubled – one responds to the exterior, the other to the interior. It’s there in the Adler house, the Esherick house, the Bryn Mawr dormitories, Unitarian Church, Hurva Synagogue, Yale Art Gallery (in section), Salk Institute (see above image), Exeter Library, Kimball Museum (you’ll have to use you imagination with this one), Richards Medical Laboratories, and many of his unrealised projects. By the time he gets to Dacca he’s transformed this idea into something of great complexity – somehow entire buildings are treated as thicknesses. kahn daccaIn one swift move (but remember this took him years to develop), Kahn engages with the problem of modern space, the interior/exterior conflict, and with the multitude of choices brought on by modern technologies. That is, the technology of the outer skin does not have to impinge on the resolution of the inner skin. No compromise is necessary. This is, again, in no way meant to say that Kahn’s solution is best – Adjaye and H&M have taken different positions on this. What all of them recognise is a conceptual understanding of the problem of reconciling inside and outside space. This idea was speculative until I thought of Kahn. The clarity and rigor with which he articulates this notion of the wall and pier turned the hunch into a concept.

Michelangelo had it easy.

Data versus Analysis

Many so called ‘analysis’ drawings consist of mapping data – most often during site and context analysis phases of a design project. Both data and analysis are important and critical for understanding how you work with a site, however, it’s important to understand the different between the two. A survey of elements (e.g. building uses, bus stops, site materials, light & noise levels, points of access) tells you what is there and when mapped tells you about their geographical distribution. This information is useful but it is not analysis. To understand analysis, imagine that what you have to show is something that is not immediately obvious or visible. This means trying to answer questions such as ‘why are there x number of y’ or ‘why is x distributed in a particular pattern’. Whereas we can see something like bus stops or commonly used materials you cannot immediately see its causes or effects. There may be economic, bureaucratic, or historical reasons behind what you see. One ‘trick’ to try to get into the realm of analysis is to correlate different spheres of data. You might generate a sun path diagram that demonstrates the distribution of daylight; you might map local area uses. So far this is data. But if you find a link you have begun a form of analysis. An Example: walking to work on a cold winter’s day I noticed that everyone was walking on one side of the street and not the other. They were walking on sunny side of the street, the north side, because it was receiving sunlight whereas the south side was in shadow and much colder. It also turned out that certain kinds of shops seemed to prefer premises on the sunny side rather than shaded side and so all the cafés were on the north side of the street. I noted a possible explanation of how light influenced where people walked and how light, and also footfalls, influenced the location of particular uses. Correlated or linked data leads to analysis and explanation.

Another way to think about this is that data describes and analysis explains. I can describe the colour, dimensions and form of my mobile phone and lists its features (data and description) but I can also explain how its design is influenced by contemporary notions of technology, available materials and current fashions for touch sensitive interfaces (analysis). Getting beyond data and description and into analysis will reveal more about the context or project your are examining and your response and reaction to it will be that much more informed (and less about the surface image of what you are looking at).


Theories of critique and analysis are distinct from theories of making and designing. This seems self-evident but has been often ignored in architectural design. In the 1990s a movement called ‘deconstructivism’ based on deconstruction flourished, helped along by a major show at the Museum of Modern Art in NY. Despite the fact that most of the architects included in the show denied being part of such a movement, architects and students went about using its theories. And critics and theorists churned out many papers about deconstructivist design. Yet the origin of the ideas came from literary theory as a method of pulling apart texts in order to study them and understand their complex (and often contradictory) meanings. It was an analytical theory but was picked up and used as a way of making architecture. Think about this in terms outside of architecture – is the way you critique a text mean that is the way the author put it together? When you analyse a film, photograph, or song, does that then become a pattern to follow when you set out to make one of your own?

This is not to say that what you learn from analysis or critique is not useful in the design studio. It helps you think about issues and to understand complexity. But that is different from telling you how to go about the process of designing.

Deconstructivisim eventually faded but not before a good deal of work was produced that never quite convinced anyone (no one takes it seriously today). However, this confusion continues today with different theories and methods. The computer as a tool is being mistaken for content and the generator of ideas. Think of the computer as a highly sophisticated pencil – was the pencil the origin of ideas? Was the pencil responsible for design decision?

Architectural Drawings

This list is intended to be useful in two different ways: one, if you are asked to draw an idea you can consider the kind of drawing that is necessary to communicate the particular idea you have in mind, and two, as you labour in your design process consider that changing drawing (and design)methods is a useful way of seeing and discovering different aspects of your design. This list is a strictly non-philosophical and non-theoretical overview of drawing methods and types. These are neither definitions nor histories. They are simply a series of points that might get you to think about the diversity of means possible when you decide to draw. There are suprisingly many different kinds of drawing; this list is certainly incomplete.

In no particular order:


A sketch can be a ‘noun’ in the sense that you might draw a quick and shorthand version of an idea (as in a comedy sketch). However, sketch should be largely seen as a verb – it is a way of drawing, as much as it is a specific kind of representation.


Diagrams exercise a distilling process or transform conditions into a code. Diagrams are not about how things look but about underlying structure. They are a way of revealing something hidden.


The doodle is characterized by a sense of distractedness or absence of a clear intention. It is drawing subconsciously, perhaps with intent, but not with one known at the time. Something else occupies the mind while doodling, rather than drawing as the centre of focus.


The cartoon is defined by either a caricatural quality or ironic point of view. The purpose is to exaggerate some aspect, through distortion (formal or content-wise) to make or test a point.

During the renaissance a ‘cartoon’ was a preparatory drawing made for scaling up a sketch up to a painting.

Measured Drawing

This is often confused with a ‘final’ drawing, but is, in fact, different from a ‘final’ drawing. What is important is the word ‘measured’, that is, a testing of dimensions, whether they are lengths, areas or quantities.

Hand Drawing

This is method rather than a type of drawing. Hand drawing does not determine a particular content. However, when used consciously, what is emphasized is the quality of the ‘hand’. It allows for accidents and isn’t concerned with the perfect reproduction of lines or tones. Hand drawings may be measured or sketched.


There are two types and aims: The first is to compare conditions which are not physically proximate. I might overlay the plan of different projects to compare their layouts and scale. The second creates a kind of spatiality in the process of drawing. Drawing one plan over another (on the same sheet or on layered transparent sheets) is a way of testing three-dimensional relationships. Overlays are also used as a compositional technique in presentation drawings, although this can get out of hand, resulting in confusion.

Perspective Drawing

The best way to understand perspective is not as a specific representational type, i.e. a drawing as seen from a human point of view utilizing the laws of perspective with vanishing points, etc. Rather, think of the word perspective and its relation to the following words: position, point of view, aspect. These can be taken geometrically or politically. For example, what is your ‘view’ or ‘position’ on £9,000 fees? When you draw a perspective there should be an intent, view, or position in that sense. You are making a point, and this is more significant than where things vanish to.

Orthographic Projection

These are fictional head-on views in which all dimensions are true and geometrically governed. Plan, section & elevation are the primary projections. They are largely seen as types of drawings required at the end of a project. However, there are reasons for doing these at various stages of a design process. It is important to keep the fictional aspect in mind. These views are never actually experienced; they are, in essence, diagrams or analytical drawings that are taken to represent an underlying ‘reality’.

Axonometric/Isometric Drawing

This is a geometrical projection that combines 3D representation of form while maintaining true dimensions. Nevertheless, axonometric drawings distort your object. This distortion is what is useful about them. Understanding the nature of the distortion and illusion is necessary if you want to use axonometric or isometric drawings in an effective manner. Using them as diagrams is fine, but then you might not be exploiting their potential.

Detail Drawing

While in school this should never be taken as a technical or construction document but as a testing of the design implications of a discrete part of your design.This is the best way of thinking about a detail even when it is a 1/20 wall section.

Hardline Drawings

See ‘Measured Drawing’. In many ways a hardline drawing is defined by its opposition to a ‘hand’ drawing.

Unfinished Drawing

There are good reasons why a drawing might not be finished. A) If the point being tested is met before the drawing is completed. B) If you discover the drawing isn’t addressing what you are trying to investigate. However, as all drawing is a teasing out of ideas and conditions it often makes sense to finish a drawing even if you think you know what the outcome is. Just as often, the drawing will turn out to be different from what you imagined.

There is also a rhetorical way of using unfinished drawings as a way of emphasizing those aspects that are drawn by leaving out parts of the design.

Cutaway Drawing

Cutaways are imaginary views that dissect a building to reveal its inner workings (technical or social). Cutaways have played an important role in the development of architecture, but the drawing type has not received much attention. See, for example, Le Corbusier’s use of the ocean liner or Koolhaas’ use of the Downtown Athletic Club in Delirious New York.

Atmospheric Drawing

This is more a description of a quality than a type of drawing. It is misleadingly used as a type, which too often leads to ‘moody’ imagery. It is up to you to decide when a drawing requires some sort of atmosphere. It is important that you understand what this atmosphere is for and how it is made (e.g. light and shadow, textures of materials, etc.).


Think of rendering as in ‘making’ or ‘fashioning’ – to cause to become. What and how you render requires taking a position and making a statement. Rendering is not a natural outcome of a process (pushing a button). Rendering is often confined to final or presentation drawings, but are useful in the process stage. A rendering should be propositional and exploratory. It’s also worth noting that renderings existed before computers.

Squiggly Line Drawing

Squiggly lines were the product of drawing by hand quickly. At some point it became evident that if you control and to a sense exaggerate the squiggle drawing acquired a kind of texture or vibrant quality. After Michael Graves, squiggly line drawings became a post-modern fashion – the worst thing that can happen to a drawing technique.

Schematic Drawing

There are two definitions here. The first is in a technical or engineering sense and is similar to a diagram. It is, in fact, a specific kind of diagram – it generally outlines procedures or processes. The second definition is in the sense of outline or general ‘scheme’. It is a particular stage in the development of designs in professional practice.

Conceptual Drawing

This is a misused term as nearly all drawings are conceptual. At its worst it is a catchall term for drawings that are not literal in their representation. At its most benign it simply means an abstract drawing. At its best refers to a drawing that is about the concept behind a design. In this sense, such a drawing can be a plan, axon, sketch, diagram, schematic or any other type of drawing.

Hybrid or Mixed-Media Drawing

This is a technique where different media are combined either to speed up the process of drawing or to capitalize on the advantages of different modes of representation. A hybrid drawing might use photography, computer drawing, hand drawing and collage in a single piece.

Final Drawing

There is no such thing apart from the idea that you will have a final deadline on which you will hand in a final selection of finished and unfinished drawings.

Presentation Drawing

The definition of presentation drawings depends on the context but what links them is that they are drawings meant to be presented to someone to make an argument for, or explain, your design. Presenting for a competition jury, to a client, a tutor, a final review or for the RIBA Bronze Medal are quite different things and call for different approaches. Considering the audience and intent of the presentation is the most important thing that can be said about this type. Presentation drawings may draw from any of the previous categories.

Life drawing

This generally means a drawing made by hand in pencil, pen, charcoal or other media, of real objects in space. If it involves naked people it is normally called ‘life drawing’ and if it involves apples and bottles it is called ‘still life’. Often courses in ‘life drawing’ combine both.

Qualitative and Quantitative: Subjective and Objective

The definition of these two terms can be derived from their roots: quali-, as in quality (sensations, experience, ephemeral or perceptual characteristics) and quanti-, as in quantity (things that can be measured, counted, or precisely defined). For example, “That film was grim (qualitative) and was 2 hours long (quantitative).” If you say “…and it was too long” it becomes qualitative. Another way of thinking about this is through the more common notions of ‘objective’ and ‘subjective’. Objective (quantitative) data or information can be definitively defined and defended – “The table is one metre long” can be proven true or false with a tape measure. Subjective (qualitative) data or information cannot be generally proven. The statement “The table is too small” depends on what it is used for and perhaps by whom (a child versus an adult).

Design criteria, research data and site information can be broken down into either of these categories and recording them requires different techniques. For example, qualitative information or data may be recorded through interviews, observation, sketching, photograph, sound recordings, mental maps, and so on. Quantitative data may be gathered by measuring, counting, weighing, documentary photography, measured drawings, and so on. The way you present these in your research or portfolio will therefore differ.

Understanding this difference can help you unravel a design brief. You also avoid making unnecessary assumptions about your site, program or project research. The architect has the difficult responsibility of understanding and satisfying the needs, not just of the client but of the many subsequent users and non-users (passers-by) of architecture. The perceptual point of view of the architect is but one subjective view among many others. The architect’s subjectivity, as such, is no more valuable than anyone else’s. Your task is not to privilege your point of view over any others but to develop skills that allow you to work with viewpoints, interpretations, opinions and other subjective criteria in a measured, appropriate and productive manner. Further, you must weight and balance the relative importance of objective and subjective criteria and information according to the demands of the particular brief. Understanding the difference between these two terms is an important part of this.

Design Thinking

  • Don’t think, then design.
  • Design is thinking.
  • Don’t imagine, then design.
  • Design is imagining, inventing and creating.

Many students have the idea that you must know what you are going to draw before starting to draw. Some think that you must imagine something, and in some cases even know what it looks like, before starting a drawing. This kind of thinking makes drawing a passive activity – you draw in order to put down on paper something you (think) you already know. In the best case the student will allow the idea to change and evolve as they start to draw it, but in most instances they force what they imagine on to the paper.

It is important to understand that architects draw to think – it is an act of imagination and creativity in itself. This does not happen in the mind but with the hand. This is not a rehash of the tired arguments about hand drawing versus computer drawing. The assumption here is that architects understand that both are necessary and that each has its role – they are not opposed merely different.

Consider the following by the American architect and educator Stan Allen:

“[T]he hardest thing to communicate to students is the confidence that you will discover things through the process of working itself. You don’t have to figure it out beforehand…students have this idea that if they think hard enough, work the idea out in advance, somehow the pieces will magically fall together. I have two issues with this way of working. First, it’s a completely false way of thinking about ideas, as if they were abstract entities floating out there in a void […] and second, the implied linearity of this process seems to me false: the idea that you could ever go in a straight line from idea to project. There is always a detour, and it’s precisely in the course of the detour that you discover things.”

You should, in general, know why you are drawing, but the most important lesson to learn at university is to develop the bravery of drawing (or modelling) without knowing or preconceiving the outcome; drawing in order to find something out; drawing and accepting mistakes, ugly outcomes, wrong turns. Only when you are comfortable with this will you be in the position to discover the happy accident, the solution that the mind could not conceive of, the configuration that no one else could have imagined.

Henri Ciriani at work