by Nigel Cross
Watching what Designers Do
A more direct form of enquiry into understanding what designers do is actually to watch them at work, observing their activities. Such studies not only report on what was observed to happen, but also try to add another layer of explanation of the nature of designing. Larry Bucciarelli made a series of extensive, very detailed, participant-observer studies of engineering design projects in three different companies. Large projects demand an important aspect of design ability, that of reconciling the variety of interests – technical, financial, social, aesthetic, etc. – that inevitably have to coalesce around a major project. In these cases, designing becomes not just a personal, cognitive process, but a shared, social process. The main conclusion that Bucciarelli stressed is how even engineering design, traditionally seen as a strictly technical process, is in reality a social process of interaction and negotiation between the different participants who each bring to bear their own ‘object world’ – their own specific knowledge and awareness of aspects of the object being designed. His thesis is that ‘the process of designing is a process of achieving consensus among participants with different “interests” in the design, and that those different interests are not reconcilable in object-world terms … The process is necessarily social and requires the participants to negotiate their differences and construct meaning through direct, and preferably face-to-face, exchange.’ The social nature of designing, he suggested, results in acknowledging the inevitability of uncertainty and ambiguity, even within the process of engineering design. ‘Ambiguity is essential to design process, allowing participants the freedom to manoeuvre independently within object worlds and providing room for the recasting of meaning in the negotiations with others.’
Bucciarelli took an ethnographic approach to studying design activity, by participating in the normal, day-to-day activities of the engineers. In another ethnographic study, of graphic designers, Dianne Murray also recorded the social nature of design in practice, and emphasised the openness and shared activities of a design studio: ‘Briefing sessions take place in the studio in clear sight and sound of everyone. Work in progress is left on drawing boards; discarded sketches, photocopies, printouts and transparencies are left lying around on desks or on the light box … Design is not hidden, it is constructed in public so other people can read it, and accepting commentary on it from somebody else is part of a tradition they embody.’ From close-up studies such as these by Bucciarelli and Murray, common features of design thinking and working emerge from quite different fields of design practice.
Peter Rowe made a set of case study observations of architectural design. These studies were of major buildings set in large American cities, designed by leading architects. Rowe noticed in all three studies that the architects’ attention switched regularly between solution concepts and problem exploration – between developing ideas for building form and investigating the implications of those ideas in terms of the design brief and technical feasibility. He characterised the progress of the design activity as ‘episodic’, or as a kind of ‘series of related skirmishes with various aspects of the problem at hand’. This episodic structure was manifest in a number of ways: ‘First, there is the “to and fro” movement between areas of concern … a movement back and forth between exploration of architectural form and evaluation of programme, structure, and other technical issues. Second, there seem to be periods of unfettered speculation, followed by more sober and contemplative episodes during which the designer “takes stock of the situation”. Third, each episode seems to have a particular orientation that preoccupies the designer. We might say that the organising principles involved in each episode take on a life of their own, as the designer becomes absorbed in exploring the possibilities that they promise.’
This sounds like the typically ‘exploratory’ approach of the designer – seeking opportunities that offer ways to progress, pushing ahead along promising avenues, and pausing from time to time to evaluate what has been achieved so far. But Rowe made a criticism of this way of proceeding; it seemed to him to be unnecessarily chancy, and inefficient. ‘These episodes’, he commented, ‘such as the various massing exercises with building volumes, often became very speculative as the designer “pressed on”, as it were, when information from another quarter might have resolved the problem at hand more economically. Such situations often subsequently gave rise to a certain amount of backtracking, as the designer retrenched to what seemed a more advantageous position.’ It is as though the designer adopts a blinkered approach, overly focused on a particular solution concept, and doggedly ‘pressing on’ when a more considered and reflective approach, and consideration of alternative solution concepts, might save time and effort in the long run.
The issue here seems to be to do with the predominance of the ‘primary generator’ in restricting the designer’s thought patterns. There is a ‘dominant influence’, Rowe suggested, ‘exerted by initial design ideas on subsequent problem-solving directions’. Of course, he acknowledged, ‘designers inevitably bring certain organising principles to a problem at the outset’. But there is a danger in clinging to these ideas, a danger of failing to see their inadequacies: ‘Even when severe problems are encountered, a considerable effort is made to make the initial idea work, rather than to stand back and adopt a fresh point of departure.’ This seems to be a weakness in the designer’s attitude and approach – investing too much effort into early, perhaps inadequate, ideas of a solution concept; even perhaps being too attached to a ‘favourite’ idea, rather than being more objective, more concerned to generate and evaluate a range of options.
Why should experienced designers behave in this apparently limited way? A clue lies in an analysis of cases of urban design similar to those studied by Rowe. Peter Levin also found that designers jumped to ideas for solutions (or partial solutions) before they had fully formulated the problem. We know that early solution conjectures offer a way to proceed with ill-defined problems. Levin suggested that, in order to generate these conjectures, some information, or ‘missing ingredient’, has to be provided by the designer himself. ‘The designer knows (consciously or unconsciously) that some ingredient must be added to the information that he already has in order to arrive at a unique solution. This knowledge is in itself not enough in design problems, of course. He has to look for the extra ingredient, and he uses powers of conjecture and original thought to do so.’ Levin suggested that this extra ingredient is often an ‘ordering principle’ and hence we find the formal properties that are so often evident in designers’ work, from towns designed as rectangular grids to teacups designed as regular cylinders. This is the same sort of thing as Rowe saw, such as ‘the various massing exercises with building volumes’, in which the designer seeks an ‘ordering principle’ around which a solution concept can be structured. It could be that designers have to invest some significant cognitive effort in generating these concepts, and so they are reluctant to let go of them.
The most influential study of a designer at work has been that by Donald Schön. The influence of the study is largely due to its being set within Schön’s broader series of studies of professional practice (ranging from psychotherapy to management) that he used to establish his theory of reflective practice, or ‘how professionals think in action’. The study has also been influential because Schön’s analysis of what he observed is acute and sensitive; both designers and design researchers (those with personal design experience) recognise the veracity of the analysis. What is surprising is that such an influential study is based on just one, partial example of design activity – and even that is not a ‘real’ design example, but is taken from observing an experienced designer tutoring a student in a university architectural design studio.
Schön established his theory of reflective practice as a counter to the prevailing theory of technical rationality, or the constrained application of scientific theory and technique to practical problems. He was seeking a new ‘epistemology
of practice’ that would help explain and account for how competent practitioners actually engage with their practice – a ‘kind of knowing’, he argued, that is different from the knowledge found in textbooks. In his analysis of the case studies that provided the foundations for his theory, he began with the assumption that ‘competent practitioners usually know more than they can say. They exhibit a kind of knowing-in-practice, most of which is tacit.’ He identified a cognitive process of reflection-in-action as the intelligence that guides ‘intuitive’ behaviour in practical contexts of thinking-and-acting – something like ‘thinking on your feet’. At the heart of reflection-in-action is the ‘frame experiment’ in which the practitioner frames, or poses a way of seeing the problematic situation at hand.
According to Schön, designing proceeds as ‘a reflective conversation with the situation,’ an interactive process based on posing a problem frame and exploring its implications in ‘moves’ that investigate the arising solution possibilities. A designer, he argued, is faced with a situation of complexity. ‘Because of this complexity, the designer’s moves tend, happily or unhappily, to produce consequences other than those intended. When this happens, the designer may take account of the unintended changes he has made in the situation by forming new appreciations and understandings and by making new moves. He shapes the situation, in accordance with his initial appreciation of it, the situation “talks back”, and he responds to the situation’s back-talk.’
The design example that Schön uses is that of a tutor, ‘Quist’, helping a student, ‘Petra’, with her problem of designing a school on a sloping site. Because Quist has to explain his own thinking to Petra, his words, and the sketches he makes at the same time, give an insight into his cognitive processes, his reflection-in-action. The talking (i.e. the thinking) and the drawing go on in parallel, as other designers have said.
Petra is ‘stuck’ in the early part of her design process: she has drawn a series of connected, L-shaped classroom blocks, but she has a problem fitting them to the site. ‘I’ve tried to butt the shape of the building into the contours of the land there – but the shape doesn’t fit into the slope,’ she explains. Quist suggests that she stops trying to work so closely to the site’s contours, and instead that she should seek to impose her building geometry onto the site. ‘You should begin with a discipline,’ he says, ‘even if it is arbitrary … you can always break it open later.’ Quist starts to sketch in plan and section, exploring the implications of the ‘discipline’ of form that he is now imposing on the site. As he explores, he begins to find some of those ‘new appreciations and understandings’ in the design. With his more aggressive approach to the site, he sees how, in section, ‘We get a total differential potential here from one end of the classroom to the far end of the other. There is fifteen feet max, right – so we could have as much as five-foot intervals, which for a kid is maximum height, right? The section through here could be one of nooks in here and the differentiation between this unit and this would be at two levels.’ The idea of ‘kid-height nooks’ is something that Quist discovers as a potential in his solution concept, not something that was in the design brief; it is an emergent property of his designing. In Schön’s terms, it is an ‘unintended change’ in the situation, which Quist interprets as a positive indicator of the appropriateness of the problem frame that he has set up. A little later, as he continues designing, other positive aspects are spotted and reinforced in the emerging design – and even the qualities of the site come back into play, as here: ‘Then you might carry the gallery level through – and look down into here – which is nice. Let the land generate some sub-ideas here, which could be very nice. Maybe the cafeteria needn’t be such a formal function – maybe it could come into here to get summer sun here and winter here.’
One thing Quist demonstrates is that he has the confidence to ask ‘what if?’ What if we carve the L-shaped blocks more deeply into the site? What if we work with a system of five-feet height intervals? What if we create a gallery-level circulation space? These ‘what if’ conjectures are the ‘moves’ that Schön identified: ‘Each has implications binding on later moves. And each creates new problems to be described and solved. Quist designs by spinning out a web of moves, consequences, implications, appreciations, and further moves … Each move is a local experiment which contributes to the global experiment of reframing the problem … As Quist reflects on the unexpected consequences and implications of his moves, he listens to the situation’s back-talk, forming new appreciations which guide his further moves.’
As designing proceeds, the sketches become a record of the moves and their implications. Many things remain tentative, but some are selected as positive outcomes of the ‘what if?’ conjectures, and are given temporary identities as features to be retained. Schön suggested that these are choice-points within the process. ‘As he reflects-in-action on the situation created by his earlier moves, the designer must consider not only the present choice but the tree of further choices to which it leads, each of which has different meanings in relation to the systems of implications set up by earlier moves. Quist’s virtuosity lies in his ability to string out design webs of great complexity. But even he cannot hold in mind an indefinitely expanding web. At some point, he must move from a “what if?” to a decision which then becomes a design node with binding implications for further moves. Thus there is a continually evolving system of implications within which the designer reflects-inaction.’
What we gain from Schön’s analysis is a clear account of a typical, fast-moving, ‘thinking on your feet’, live example of designing. The initial problematic situation is ‘framed’ by the designer. Quist’s framing adopts the given of Petra’s starting point of a series of linked L-shaped blocks (he is tutoring her in how to develop her solution idea, not starting from scratch with his own idea) and poses the implicit question, ‘How can we make these blocks fit into the sloping site in a coherent way?’ He works through a series of thinking-actions of moving-seeing-moving; that is, of posing a ‘what if?’ move, looking at what results (in his sketches), reflecting on the consequences (good or bad), and making another, related move. One move leads to another, through the medium of the sketches, which not only record the process of moves but also provoke thoughts and initiate new moves.
Something similar must have been happening in Rowe’s studies of architectural design, even though the design projects were on a much larger scale. The designers spin out a complex web of inter-related moves, reflections, decisions, and further moves. They invest a great deal of cognitive effort in spinning and maintaining these webs, and so perhaps it is no wonder that they are reluctant to ‘stand back and adopt a fresh point of departure’. But sometimes it is necessary; a problem frame can prove to be inappropriate, or the designer lacks the ability to maintain a positive sequence of moves within the frame, and so a new departure point, a new problem frame becomes necessary.
An aspect of Schön’s study that helps to make it particularly informative is that it is based on the ‘live’ data of Quist’s talking and drawing. Because he is tutoring, Quist externalises his thinking for the benefit of the student, when normally it would be a silent, internal cognitive process. This therefore provides an example of something like a ‘think aloud’ protocol study, of the type which has come to be used extensively to investigate how designers think. The ‘protocols’ are the sequence of thoughts, reflected in the comments made by the designer. These protocol studies are normally conducted as a laboratory type of study, in which a designer is asked to ‘think aloud’ as he or she works through a short design project. Detailed evidence from these kinds of study will be used frequently in the later chapters of this book.
These various studies of design in action, based on watching what designers do, have tended to confirm what designers say about the nature of designing. There is the need to tolerate and work with uncertainty, to have the confidence to conjecture and to explore, to interact constructively wit
h sketches and models, and to rely upon one’s ‘intuitive’ powers of reflection-in-action.
Thinking about what Designers Do
The criticisms that Peter Rowe made of the way that designers tended to cling for too long to solution conjectures that were proving inadequate have also been reflected in comments by others. This and other early criticisms of the typical ways that designers work led to attempts to provide design methods or guidelines that would encourage designers to work more ‘rationally’. Such guidelines generally outline a systematic procedure of first analysing the problem as fully as possible, then breaking this into sub-problems, finding suitable sub-solutions, evaluating these and then selecting and combining them into an overall solution. It is basically a process of analysis-synthesis-evaluation. However, this kind of procedure has been criticised in the design world because it seems to be based on inappropriate models imported from theories of problem solving and ‘rational behaviour’, and therefore runs counter to designers’ more ‘intuitive’ ways of thinking and reasoning.
Several theoretical arguments have been advanced in support of the view that design reasoning is different from the conventionally acknowledged forms of inductive and deductive reasoning. For example, Lionel March distinguished design’s mode of reasoning from those of logic and science. He pointed out that ‘Logic has interests in abstract forms. Science investigates extant forms. Design initiates novel forms. A scientific hypothesis is not the same thing as a design hypothesis. A logical proposition is not to be mistaken for a design proposal. A speculative design cannot be determined logically, because the mode of reasoning involved is essentially abductive.’