As I discussed earlier, aesthetics was added to Pye’s Nature of Design at a later date. That’s a thorny problem, and something that has always puzzled me is: just what makes something beautiful?
I grew up in Southern California, and I was surrounded by industrial design. It resonated with me in a way that flowers, trees, etc., never managed to.
To me, warehouses and refineries, parking lots and shopping malls, were all beautiful in strange ways— though few people agreed.
This gave me problems as a photographer, growing up in the shadow of the California nature police like Ansel Adams, Brett Weston, et. al.. I just didn’t get it. I remember being so thrilled to see that Edward Weston photographed refineries near where I grew up, as did Horace Bristol.
Bristol is mostly known these days for traveling to the Central Valley taking photographs on the project that became The Grapes of Wrath with John Steinbeck. It’s possible to appreciate industrial beauty and natural beauty; a few people do. But I digress.
To David Pye’s credit (in my opinion) he first describes the aesthetic element as “doing useless work on useful things.” We’re compelled to, and the reason why isn’t as simple as most aesthetes claim, in my opinion. It’s a convergence of forces, social and natural. In the case of practical design, it’s always secondary. So what are the primary elements of design?
Part of what derailed me from continuing to hash through Pye is his first point in his bulleted list of the six requirements of design:
1. It must correctly embody the essential principle of arrangement.
2. The components of the device must be geometrically related—in extent and position—to each other and to the objects in whatever particular ways suit these particular objects and this particular result (Chapter 4).
3. The components must be strong enough to transmit and resist forces as the intended result requires.
4. Access must be provided (this is a special case of 2 above).
These four together will be referred to as the requirements of use.
5. The cost of the result must be acceptable.
This is the requirement for ease and economy.
6. The appearance of the device must be acceptable.
This is the requirement of appearance. (23)
Obviously, there’s a lot here to unpack (a book’s worth, really). Aesthetics comes in dead last in his principles of design, even though it almost seems like it’s first in one respect: Just what is “the essential principle of arrangement”? As far as I can tell, it’s the rhetorical device that Pye uses to avoid saying “function” when referring to useful objects.
One good example of “the principle of arrangement” would be the knife. Knives come in a variety of shapes and sizes and configurations but there is a sort of essential form. Knives have blades that cut, they have handles, they can fold (or not), etc., leaving us with basic forms pretty much unchanged (beyond embellishments and decorations) for centuries. There need not be a platonic ideal form for a knife that we strive for before we can say that knives have essential forms of arrangement. It’s not an aesthetic thing, it’s utilitarian.
The best way justify approaching it this way has to do with the slipperiness of talking about function. Instead of saying that knives “must fulfill the function of cutting” is that by saying instead, essentially, that a knife must have the potential (in its designed form) for cutting in order to qualify as a well designed knife is that we can’t ever really know what someone will use that knife for.
Perhaps a user will deploy it to pry open paint cans, or turn screws. It may not be well suited for those jobs, but it may be directed towards those ends. Does failing to open the paint can make it a bad knife? Not really, but since we can’t know for sure that a knife will always be used for cutting we can’t simply assume that it will. For example, a good bush knife might also be used to strike a flint and make sparks; the strength of a thicker blade doesn’t help the knife cut, but it does make it possible to use the knife for other things—even as a pry bar in some cases. It’s impossible to anticipate the entire range of functions that an object might be placed in.
The essential principle of arrangement for knives has evolved and moved through a variety of basic forms each one suited not only to a primary task, but also for secondary purposes which almost form new classes. One example that springs to mind is the butter knife. Is it a knife or a spreader? It doesn’t cut well, really, and yet we do cut with it occasionally and still call it a knife. The categories are hardly static.
Pye’s careful choice of words here is admirable. It makes it hard to parse in some cases, but he has chosen very carefully what he labels as “requirements.” I do wonder if, prior to the revision, he included “the requirement of appearance” at all.
A core choice worth noting: notice that he speaks of a “geometric” relationship between objects rather than any sort of mathematical or axiomatic relation? Think of a pair of scissors; we could specify a specific size or shape for that class of objects, but aren’t scissor’s better described by the geometric relationship between the moving parts? I think that’s just brilliant to say it that way.
I was reminded a few days ago somewhere on the interwebs that the root meaning of “tolerance” when it comes to woodworking is that we tolerate errors of a certain amount because due to shrinking or expanding materials (among other factors) it is impossible to clearly specify exactly what size something is supposed to be. We can, however, insist that for a good design the component parts (a drawer for example) be able to function in a precise geometric relationship with the other parts of a piece. That’s good design.
Pye’s discussion, following his list, doesn’t really address these matters in the same way I have here. To place it in context, the list is introduced following a discussion of the difference between invention and design. For Pye, invention is a concept pretty much culminating in a patent-like description of a potential system. Design, on the other hand, is the embodiment of that concept. Thus, it is required that a design follow the pattern and concept of the patent/description but it can deviate substantially and still fit the need and fulfil the concept.
In discussing the “essential principle of arrangement” of the particular design, Pye chooses instead to talk about the differentiation between the way a designer would approach the “shape” of something compared to a layman. Illustrating several wedge forms, he carefully dissects the idea that there is such a thing as a “wedge shape” because it has so many varied forms that are able to function as wedges; wedging is a function of a shape, not a coherent shape in itself. A wedge can come in many different shapes; it’s not a matter for aesthetics, but rather physics.
The only way of closely defining the kind of arrangement of matter that we call a wedge or a hook would be by referring to the way it transmits forces. A hook will pull. A not-hook won’t pull. Shape, individuality, doesn’t come into it.
Shape, for us, is what gives individuality to things. All of us are extremely expert in recognizing the individual character of shape in closely similar things such as human faces and hands of writing. The individuality of shapes is the stuff of art, whether in design, painting, or in any other field. It is our present concern to find out how far the designer has freedom to give devices a chosen character of appearance, of shape. The essential principle which he must embody in the device he is designing sets limits merely to the extent that if the principle requires ‘a hook’ then not-hooks are excluded. But there is precisely and infinite range of possible shapes for a hook. The limitations on freedom of choice, so far, are nugatory. (24-25)
I think I should start using the term “nugatory” in conversation more. The overall tone and care of David Pye rewards constant reading. In essence, it seems as if he is driven to make it clear that much of what we think of as superfluous in design is actually what design is. It’s about choices, within constraints—and those constraints are a lot less severe than we generally think. We choose pleasing shapes because, well, why wouldn’t we? It’s just natural to make a pleasing choice. That’s where the brutalist machine aesthetics of the early twentieth century seems more like an anomaly than the wave of the future.
Though aesthetics are the last item he lists among the “requirements” it is clear that they can be introduced in many of the choices along the way. The shape of something is its primary ornament, and that ornament is not something that is necessarily “applied” along the way; it can be part of the essential formal arrangement of a thing.