In 1964, David Pye published his second book, a slightly longer (91 page) volume called The Nature of Design.
I haven’t tracked an original down to look at it, though I’m curious about the differences between the original and revised edition. There are references to computers in the revised edition that I’m sure probably weren’t in the subsequent expanded edition.
This was followed in 1968 by The Nature and Art of Workmanship, the canonical text that ends up on most woodworker’s reading lists.
One difficulty in looking at mid-century craft theory is the fracture between design and craft. Pye was a pioneer in theories of both, and because he sought to tease out differences he was instrumental establishing these pursuits as separate. Pye’s writing presents, for me at least, the most lucid and structured attempt at a non-economically based theory of work. However, the division is problematic.
Dividing things in this fashion has advantages over the holistic approach of Morris et al., because it allows for fine grain consideration of various aspects of the problems that artisans and designers face, but it also reifies the divide between modern machine culture and earlier forms of craft.
Most productively though, it opens a space for the popular (not just “professional”) interest in emergent trends in design and consumer products. This echoes his earliest writing. Nonetheless, the division presents problems on where to “sort” certain aspects of the artisan/craftsman skill set. Pye’s final words on the problem come in the revised edition of his second book.
The book was issued in a revised and expanded edition in 1978 which adds “aesthetics” to the title and contents, stretching it out to around 160 pages. The growth in his theories is apparent from the frameworks he’s established—aesthetics appears not in the volume on workmanship, but attached to design.
Of course aesthetics is present in varying degrees in all his writings—it is after all “the art” of workmanship— but after reading Bernard Leach, this decision makes more sense.
It may be that workmanship is equivocated with technique, and thus skill; this undercuts the core performance of the designer, who has offered his own aesthetic perspective by his choices, prior to the execution of workmen. Obviously they can work in concert, or at cross purposes, but no workman can really rescue or compensate for ugly designs.
Leaving workmanship aside for the moment, what I’d first like to look at is the development of Pye’s thinking regarding functionalism and aesthetics that began with The Things We See No. 6: Ships.
To summarize his early position, design operates by making changes within constraints; design choices are not logical or rational, but represent aesthetic decisions that have their own mechanism which cannot be elaborated using words or logic. Aesthetic experience is what enables and improves the ability to make these choices. Function presents constraints, but does not negate nor enhance the aesthetic viability of a design. Something that functions can be either beautiful or ugly; that depends on the choices made by the designer.
Sleeping on this, I wonder if it might be more productive to think in terms of affordances rather than constraints, but that’s really a topic that deserves its own essay. It’s best to get through where Pye actually went with this first.
Although the importance of design is realized, the essential nature of the activity seems not to be understood except by designers, and they have not formulated what they know. It is not of the slightest use for us to ask ‘what is good design?’ until we can answer the question ‘what is design?’
The thing which sharply distinguishes useful design from such arts as painting and sculpture is that the practitioner of design has limits set upon his freedom of choice. A painter can choose any imaginable shape. A designer cannot. If the designer is designing a bread knife it must have a cutting edge and a handle; if he is designing a car it must have wheels and a floor. These are the sort of limitations which arise, as anyone can tell, from the ‘function’ of the thing being designed.
Little is ever said which touches on the fundamental principles of useful design, and what is said is often nonsense. Most of the nonsense probably starts at the point where people begin talking about function as if it were something objective: something of which it could be said belonged to a thing.
The dictionary defines function as ‘the activity proper to a thing, the mode of action by which it fulfils its purpose’. What on earth can that mean? Surely if there were activities proper to things, and if things acted, and if they had purposes, Newton might have been relied upon to take note of these facts? ‘Function will not square with physics. And if function is a fantasy, what of functionalism —the doctrine that form follows function? (11-12)
The point of departure, that useful design is born from constraints is unchanged, but the assertion that there is no “activity proper to a thing” is new.
What is the activity proper to a straight cylindrical bar of steel a quarter inch in diameter on cross section and four inches long? What function is this form following, or ought it to follow? What activity exclusively or distinctively belongs to this thing, is in other words proper to it? There it lies on the bench, what are we to say? ‘Well, it isn’t active. You could make it active if you heated it enough. Otherwise it will not do anything unless the bench happens to collapse. Of course you could use it for an enormous number of different purposes, but then for nearly every one of them you could use something different equally well . . .’ the question still has to be answered, ‘what is the function of this thing?’
Now plenty of people do really believe that form can follow function; that if you thoroughly analyze the activity proper to the thing you are designing then your analysis will provide all the information needed, and the design can be derived logically from the function. Plenty of people still believe that ‘purely functional’ designs are possible, and believe that they themselves produce them, what is more! But none of them has yet divulged what an analysis of function looks like and what logical steps lead from there to the design. All you will get from them is talk about the purpose of the thing, which, as we shall see, is a statement of opinion and can never be anything else. (12)
Pye writes this so forcefully and convincingly that it is clear that he hasn’t stopped thinking about these issues since 1950. The clear development here is that function is framed as an opinion; opinions are not logical— they are rhetorical, function is an instance of endoxa, commonplaces that we accept that are not necessarily true or false. It is important for designers to have a clearer sense of what is happening when we evaluate designs. These issues are also interrogated in courts of law with the legal commonplace, “fitness for purpose.” It can be litigated in contract law, but ultimately it is an opinion, not a physical fact.
Someone will reply ‘This is all pedantry. Think out what the thing has got to do, design it in the simplest form which will will do that and there you have a purely functional design; and what is more it will look right.’
This sort of question raises three questions:
How do you determine what the thing you are going to design ‘has got to do,’ and what ‘activity is proper to it’, what ‘it is for,’ what ‘its purpose is?’
Having done so, does the information you have gained govern the design and determine its form, or does it merely guide it, restricting the choice of a form and setting limits within which it can be varied at will?
What does purely functional mean? (12)
The initial answers that Pye proposes are “arbitrary,” for the first question, “it merely guides it,” to the second and for the final question his answer is a bit more flip: purely functional means “‘cheap’, or else ‘streamlined’, or else more rarely ‘light’.” The implications of these answers are treated at greater length in the remainder of the work. The core sentiment, and point for this excursus is to propose something quite remarkable:
Whenever humans design and make a useful thing they invariably expend a good deal of unnecessary and easily avoidable work on it which contributes nothing to its usefulness. Look for instance, at the ceiling. It is flat. It would have been easier to not make it flat. Its being flat does not make you any warmer or the room above you any quieter, nor yet does it make the house any cheaper; far from it. Since there is a snobbism these things flattening a ceiling is called workmanship, or mere craftsmanship; while painting gods on it or putting knobs on it is called art or design. But all these activities: ‘workmanship’, ‘design for appearance’, ‘decoration’, ‘ornament’, ‘applied art’, ’embellishment’, or what you will, are part of the same pattern of behaviour which all men at all times and places have followed: doing useless work on useful things. If we did not behave after this pattern would indeed be poor, nasty, and brutish. (13)
What a powerful description of the human condition: “doing useless work on useful things.” That’s one of the best descriptions I’ve read in a long time.