Read an Excerpt
Chapter 6: The Program
Here is the problem. We wish to design clearly -conceived forms which are well adapted to some given context. We have seen that for this to be feasible, the adaptation must take place independently within independent subsystems of variables. In the unselfconscious situation this occurs automatically, because the individual craftsman has too little control over the process to upset the pattern of adaptation implicit in the ensemble. Unfortunately this situation no longer exists; the number of variables has increased, the information confronting us is profuse and confusing, and our attempts to duplicate the natural organization of the unselfconscious process selfconsciously are thwarted, because the very thoughts we have, as we try to help ourselves, distort the problem and make it too unclear to solve.
The dilemma is simple. As time goes on the designer gets more and more control over the process of design. But as he does so, his efforts to deal with the increasing cognitive burden actually make it harder and harder for the real causal structure of the problem to express itself in this process.
What can we do to overcome this difficulty? On the face of it, it is hard to see how any systematic theory can case it much. There are certain kinds of problems, like some of those that occur in economics, checkers, logic, or administration, which can be clarified and solved mechanically.1 They can be solved mechanically, because they are well enough understood for us to turn them into selection problems.2
To solve a problem by selection, two things are necessary.
- 1. It must be possible to generate a wide enough range ofpossible alternative solutions symbolically.
- 2. It must be possible to express all the criteria for solution in terms of the same symbolism.
Whenever these two conditions are met, we may compare symbolically generated alternatives with one another by testing them against the criteria, until we find one which is satisfactory, or the one which is the best. It is at once obvious that wherever this kind of process is possible, we do not need to "design" a solution. Indeed, we might almost claim that a problem only calls for design (in the widest sense of that word) when selection cannot be used to solve it. Whether we accept this or not, the converse anyway is true. Those problems of creating form that are traditionally called "design problems" all demand invention.
Let us see why this is so. First of all, for physical forms, we know no general symbolic way of generating new alternatives, -or rather, those alternatives which we can generate by varying the existing types do not exhibit the radically new organization that solutions to new design problems demand. These can only be created by invention. Second, what is perhaps more important, we do not know how to express the criteria for success in terms of any symbolic description of a form. In other words, given a new design, there is often no mechanical way of telling, purely from the drawings which describe it, whether or not it meets its requirements. Either we must put the real thing in the actual world, and see whether it works or not, or we must use our imagination and experience of the world to predict from the drawings whether it will work or not. But there is no general symbolic connection between the requirements and the form's description which provide criteria; and so there is no way of testing the form symbolically.3 Third, even if these first two objections could be overcome somehow, there is a much more conclusive difficulty. This is the same difficulty, precisely, that we come across in trying to construct scientific hypotheses from a given body of data. The data alone are not enough to define a hypothesis; the construction of hypotheses demands the further introduction of principles like simplicity (Occam's razor), non-arbitrariness, and clear organization.4 The construction of form, too, requires these principles. There is at present no prospect of introducing these principles mechanically, either into science or into design. Again, they require invention.
It is therefore not possible to replace the actions of a trained designer by mechanically computed decisions. Yet at the same time the individual designer's inventive capacity is too limited for him to solve design problems successfully entirely by himself. If theory cannot be expected to invent form, how is it likely to be useful to a designer?
Let us begin by stating rather more explicitly just what part the designer does play in the process of design. I shall contrast three possible kinds of design process, schematically...