Lawson B. How Designers Think: The Design Process Demystified

Подождите немного. Документ загружается.

HOW DESIGNERS THINK

208

concept. Should the design process be based on the deliberate

development of one solution or, by contrast, a conscious search for

alternative solutions followed by selection and possibly combin-

ation? Many questions like this to do with the design process

cannot be unequivocally answered, and this one is no exception. It

seems that both ways are used by designers who are considered

successful. Before exploring the idea of generating alternatives and

exploring ways of doing this, let us first examine the case for the

single solution approach.

Many designers dislike the idea of generating alternatives and in

particular the showing of many alternatives to clients. This seems

very much a matter of personal design style and client manage-

ment, but leads to the fear amongst designers that a client may

want to pick ideas from several alternatives that are either impossi-

ble or extremely difficult to combine, or that will result in an inco-

herent and rambling solution lacking in integrity.

The architect/engineer Santiago Calatrava feels that to explore

too many alternatives is a sign of doubt and that since eventually

the designer must develop only one solution and fight to defend

the ideas behind it then it must be believed in to the exclusion of

all else:

You have to let an idea run and proceed with it to be convinced . . . of

course you criticise it and you may leave it and start again with some-

thing new, but it is not a question of options, it is always a linear

process.

(Lawson 1994b)

Perhaps this is similar to what Philippe Starck describes as ‘capturing

the violence of the idea’. Somehow to leave an idea and search for

an alternative may be thought to lose the ‘mental inertia’ which is

needed to develop an idea into a workable proposition. There may

be some parallel here with choosing a name for something, a child

perhaps. You can look through hundreds of alternatives and none

seem particularly to stand out, but when you settle on one and use it

for a while it soon becomes special and feels ‘right’.

However, Santiago Calatrava was certainly not telling us that

he invariably goes straight to this one ‘right’ idea, but that the

process, for him, is based on working on only one solution at once.

The architect Richard MacCormac also believes in both evolution

and revolution during the design process, but is not enthusiastic

about deliberately generating alternatives as a conscious process.

He feels that the designer can sense something in the nature of a

design problem that indicates whether the generation of alterna-

tives is likely to lead to success:

H6077-Ch12 9/7/05 12:34 PM Page 208

DESIGN TACTICS

209

There are certain kinds of design programme that structure the design

very much . . . and you have to have a sense that unless you explore

options you are going to miss some tricks, whereas in other cases, for

example the St. John’s College competition which we won, I rushed

headlong as it were into an idea for the project which enthralled the

client and which was quite different to the other submissions.

(Lawson 1994b)

Unfortunately, Richard MacCormac has not yet been able to express

clearly just how this ‘sensing’ of the problem nature works. Denise

Scott Brown whose practice with Robert Venturi ranges from large-

scale planning right down in scale through architecture to furniture and

even pottery, also seems to have this feeling that the generation of

alternatives works for some problems and not others:

The use of options in planning is to achieve democracy in the process.

You have to accommodate more complexity and confront more political

options in planning than in architecture.

(Lawson 1994b)

There may well be something in what Denise Scott Brown says here,

purely in terms of political expediency, but the idea that there is a

hierarchy of design problems with town planning at the top, architec-

ture in the middle and product design at the bottom has limited

value. In particular the idea that therefore town planning is more

complex than architecture was questioned much earlier in this book

and found wanting. As we shall see very soon, Eva Jiricna working at

the scale of interior design works very much by generating alterna-

tives. It seems, therefore, more likely that while Richard MacCormac

and Denise Scott Brown may feel some problems are more amenable

to the generation of alternatives than others, in reality this may be at

least as much a matter of the personal style and preference of the

designer than an inherent characteristic of the problem.

Generation of alternatives

Let us then explore the use of alternatives and how designers

generate them. In such a process, the designer generates many ideas

each of which have at least some possible advantages, rather than

focusing on one idea too soon. The process then becomes a matter

of eliminating unworkable or unsatisfactory ideas and choosing

between the remainder, possibly combining some features or several.

Two very different advocates of this approach are Michael Wilford,

working at the urban scale and Eva Jiricna working on interiors

(Lawson 1994b). Michael Wilford describes it as ‘a very system-

atic process of investigation of options and selection’ (Fig. 12.3).

H6077-Ch12 9/7/05 12:34 PM Page 209

HOW DESIGNERS THINK

210

Eva Jiricna uses the deliberately calculated stimulus of trying different

combinations of materials which, in an echo of an earlier section in

this chapter, she calls the ‘starting point of the story’:

On the first morning when you start working on the scheme you have

got let’s say ten and they are all equally possible and then you go

through a process of analysing it and develop each of them slightly fur-

ther on, and then you are left with say five. That process goes on and

eventually you are left with one alternative.

It is interesting that these and other designers studied who use the

generation of alternatives, often show them to their clients. This

seems to become part of the briefing process; a way of drawing

more information out of the client about what is really wanted.

However, a more detailed discussion of these issues must wait until

Chapter 15.

For those who wish to practise the generation of alternatives, it

seems necessary to have some basis upon which they are generated.

Figure 12.3a

Michael Wilford describes a

process involving the generation

of many alternatives. These are

just some of the alternative

layouts considered for Temasek

Polytechnic in Singapore. See

also 12.3b

H6077-Ch12 9/7/05 12:34 PM Page 210

DESIGN TACTICS

211

For Eva Jiricna, it is different materials, for Michael Wilford it is much

more about the disposition of major elements on the site. Wilford

warns, however, that this process is not easily performed. He has

taught in schools of architecture and finds that students often have

difficulty producing a range of ideas:

They can’t detach themselves from a particular solution or design to

look at others . . . they are locked into a solution without having a full

spectrum available to judge whether that is an appropriate solution.

Without this the process tends to become ephemeral.

So Wilford is pointing out another benefit to the designer of the

alternative generation approach. He is implying in some way that the

territory is mapped out, that the range of possible solutions is identi-

fied. Of course, designers can never actually know that they have

identified all the major alternative solutions to a problem. Often,

Figure 12.3b

Continued

H6077-Ch12 9/7/05 12:34 PM Page 211

HOW DESIGNERS THINK

212

however, there may be a limited number of basic strategies and an

experienced designer such as Michael Wilford may be pretty confi-

dent that all the major ones have been found. Identifying all these

major alternatives may well be extremely valuable both for discus-

sions with the client and for establishing some firm foundations for

the rest of the design process. The tutor in a design school is usually

only too well aware of this. When setting a class of students the same

problem, it often seems that there are only half a dozen or so valid

and sensible basic solutions with many variants and combinations.

Parallel lines of thought

The development of alternative ideas by experienced designers may

often be rather more sophisticated than the simple generation of a

range of options. When we examine the drawings done during the

design process it is often possible to detect, what we might call ‘par-

allel lines of thought’ (Lawson 1993a). These parallel investigations

represent examinations into different aspects of the design. Thus

Eva Jiricna, who likes to work from materials, also has to plan her

interiors in organisational terms (Fig. 12.4). The design process can-

not simply proceed either from detail to spatial concept or the other

way round; both are developed in parallel:

It is a spatial concept but it goes really parallel to the selection of mater-

ials that do exist and the details and they are all joined together and it

changes.

(Lawson 1994b)

Robert Venturi echoes this with his characteristically ironic aphorism

(quoted more fully in Chapter 3) that ‘sometimes the detail wags

the dog’. What Jiricna and Venturi are both emphasising here is that,

for them at least, design proceeds by investigating both detail and

larger-scale issues in parallel. The central issue here is the designer’s

ability and willingness to allow two or more of these parallel investiga-

tions to take place without necessarily trying to resolve them too early.

However, it is not simply a matter of detail or general. Designers

can be seen to develop and sustain many incomplete and nebulous

ideas about various aspects of their solutions. Sketches done by

Robert Venturi for the famous Sainsbury Wing of the National Gallery

in Trafalgar Square in London show this quite clearly (Fig. 12.5). There

are plans which deal with the problems of circulation, of getting large

numbers of people into the new building and connecting it satisfac-

torily with the axial arrangement of the original Wilkins building.

H6077-Ch12 9/7/05 12:34 PM Page 212

DESIGN TACTICS

213

There are also sketches of the elevations, particularly those seen from

Trafalgar Square where the new and old buildings come together

(Fig. 12.6). The development of this second line of thought about

façade makes a particularly interesting case study for us here:

The main idea for the National Gallery façade, for instance, came on

the second day I was thinking about it in London. I was standing there

in Trafalgar Square and it came like that, and it has lasted, although it

took many months to refine it.

(Lawson 1994)

This comment reminds us of the way an idea can appear suddenly

but then need extensive refinement as we saw in Chapter 9.

However, it is quite clear from Venturi’s description of the whole

design process that much of the refinement is carried out in paral-

lel with that of other ideas without attempting to resolve them too

soon. The sequence of images shows how Robert Venturi and

Denise Scott Brown use a wide variety of techniques for this refine-

ment process. In this case they put the columns from the existing

building on to their computer which enabled them to reproduce

Figure 12.4

Early sketches from Eva Jiricna’s design process showing a line of thought about the junction

between wall and ceiling

H6077-Ch12 9/7/05 12:34 PM Page 213

HOW DESIGNERS THINK

214

Figure 12.5

Robert Venturi developing a line of thought about the National Gallery

extension as plan

H6077-Ch12 9/7/05 12:34 PM Page 214

and transform, them at will, sometimes plotting them for use in

models or collages combined with more conventional sketching.

In a study of design protocols already referred to in Chapters

3 and 6, Rowe’s analysis led him to describe the design emerging

through the use of several parallel primary generators:

In this case study, several distinct lines of reasoning can be identified,

often involving the a priori use of an organising principle or model to

direct the decision-making process.

(Rowe 1987)

DESIGN TACTICS

215

Figure 12.5

Continued

H6077-Ch12 9/7/05 12:34 PM Page 215

In one protocol where the designers were working on a waterfront

site in Chicago, Rowe shows how two primary generators remained

in the designers’ minds for most of the process with one eventually

dominating and partly subsuming the other:

Perhaps the most distinctive feature of the protocol is the attention

paid by the designers to the two large themes of creating a focal

point, or landmark, and extending the grid pattern of Chicago, in a

linear fashion, out into the lake. Throughout, these two themes seem

almost to compete with one another. First one dominates, only to

HOW DESIGNERS THINK

216

Figure 12.6

A parallel line of thought about

the building as elevation

H6077-Ch12 9/7/05 12:34 PM Page 216

recede again as the process unfolds. In the end, design effort was

focused on the proposal of a single landmark building, although even

then its immediate environs were clearly controlled by the idea of the

grid pattern.

If we look further into these parallel lines of thought it seems that

they often reflect quite conventional ways of thinking about the

kind of design under investigation. In Venturi’s case he was think-

ing about a building as plan and elevation. In Eva Jiricna’s case

she thinks about it as a collection of components and as a spatial

organisation. Parallel lines of thought are particularly evident from

the sketchbooks of Santiago Calatrava. It is worth remembering

that we have already seen that he does not use a process of delib-

erately generating alternatives. However, his sketches provide

clear evidence that he is thinking about the design in many ways

simultaneously. Calatrava works by keeping several sketchbooks

open at once. As we shall see in Chapter 14, Calatrava prefers

small rather than large sheets of paper and these sketchbooks

range in size from a small pocket book up to A3. In some he draws

with a pen, in others he works freehand but approximately to scale

in watercolours, and in some he even performs calculations.

Shown here are sketches from two books for his design for

DESIGN TACTICS

217

Figure 12.6

Continued

H6077-Ch12 9/7/05 12:34 PM Page 217