“Just measure, cut and bolt together”
Finding out what I don’t know about building things by building things.
A simple potting bench, with shelves on either side. Here’s the design drawing:
It’s for the greenhouse. A small project. The general design is set. But there are many decisions to be made about detailed design, materials, and construction. Those decisions in turn require thinking about tools and skills. What do I need to know, and what do I need to know how to do, to make this bench?
Those are the general questions that building something raises. Knowing, and knowing how. I’m interested in what one needs to know to make things, and how to learn it. This potting bench seems a good case study. As I make the bench, I keep track of the questions that occur, the decisions I make, what I know and what I don’t, and what I learn; and what I can do, and what I can’t. I inventory the tools and skills and the know-how that I have, acquire, and put to use.
Building a bench offers the opportunity to consider the long history of thinking and writing about skills. Skills have a history, and so does thinking about skill. Medieval guilds acknowledged the “art and mystery” acquired through long apprenticeship. The kind of training required for different jobs was a topic of hot debate in nineteenth-century America, and lawyers debated the differences between the “ordinary skill” of a workman and the ingenuity of an inventor. Ruskin and his followers valorized skill as everything lost in the industrial revolution: “Almost the whole system and hope of modern life are founded on the notion that you may substitute mechanism for skill.” Skill is still an art and a mystery, still seems ordinary, and, even more than Ruskin’s time, still seems somehow out of fashion.
There’s a field that might be called the philosophy of skill, attempts to figure out the characteristics of skill and skilled work—how “knowing how” is different from “knowing.” Michael Polyani, in his Personal Knowledge (1958), talks about “rules of art…maxims” that, along with “practical knowledge,” guide the artisan — knowing how, rather than knowing what. My favorite book on skill, George Sturt’s The Wheelwright’s Shop (1923), also uses the term “art,” and contrasts it to “reasoning.” A good wheelwright knew what was right because “He felt it, in his bones.” David Pye, in The Nature and Art of Workmanship (1968), offers us two kinds of workmanship, characterized by risk and certainty. The former requires both skill and judgement, while the latter removes the judgement, replacing with the sort of planning that allows for an unskilled worker, or an assembly line. Ruskin was bemoaning the rise of the workmanship of certainty.
Pye and the others try to answer questions like: What counts as skill? How does one acquire it? How to conceptualize the relationship of skill and knowledge? And how does one learn this? Their answers were based on a life-long career of making. I’m going to try to answer them by building a bench. I am not a skilled craftsman. I’m more like Sturt at the start of his work in the wheelwright’s shop. He was a teacher, fascinated by craft work, influenced by Ruskin, and he took over his father’s shop knowing almost nothing of the work he would need to do. He learned from working with skilled workers, and by doing things himself. My hope is that because I’m figuring things out as I go along, I can ask the questions that are second-nature to the skilled workman.
I’m just going to try to learn by doing, and see if I can write down some of what I know, and what I don’t know. And also, I hope, what I learn.
Design, beyond the design
The basic design is set. It’s pretty simple: a flat top supported by two towers, one with four shelves, the other with three.
What quickly becomes clear is how much of design lies beyond the basic design, beyond the sketch. There are some big issues that need to be settled before I begin: What material? How strong does it need to be? And then there are the small design decisions; design and construction overlap at every step.
Consider just one question: bracing. Structures are strong when all of the forces are straight down; they fail when there are lateral forces that hit materials and corners where materials are weakest. I know enough mechanical engineering to know that triangles are good, offering structural stability. But like so many other general rules, it’s hard to know how to apply it in this particular case. How much force would it take to make the desk collapse sideways? Do I need to worry about that, or can I be sure that the fastenings that hold the structural members together will also offer enough bracing ? And if I want triangle bracing: does it matter if it’s the longest angle, across the back, or would several shorter braces do? And should it be a single large X, or two smaller ones?
A lesson here: knowing general rules is not enough. There’s a long way between the geometry of engineering and the praxis of design, and another long way between that and the nuts and bolts of building. Knowing rules is not enough, knowing how is not enough; I need a sense of materials in the real world.
I chose a new material for this project, slotted angle iron. Why choose this? What does it allow? Stronger, enforces rigidity, easy to make right angles. The online ad reads: “Easy to use… Just measure, cut and bolt together.” It sounds appealing. But it has its drawbacks. Perhaps most important, I have no experience with it — with wood, I can guess strength, I have the tools to cut it, I know how to fasten it together. Steel is new and different. Measuring, cutting, bolting together: each one of those operations requires new knowledge and skill. Working with a new material exposes the skills I lack.
Measurement seems straightforward. In wood, a homogenous material, it is; any piece is the same as another. (There are tricks, like laying out the pieces to get the most use of each 2x4.)
But slotted angle iron, with its holes, is not homogeneous. For shelves, the holes need to line up, and so it’s important that each upright be cut out of its own piece of steel, so that the holes are parallel and the shelves will be straight. Does it matter if I use a slot rather than a hole at the end? That is, does using a slot make it less structurally stable? Or does the pressure of the nut, tightened hard, mean that the bolt won’t slide, even with the sloppiness the slot provides? (Why are there slots, anyway?) And all of the pieces need to be cut so that the space between holes are the right length, not the overall length of the piece of steel. And the extra piece behind the holes can’t stick out too much, or it hits the piece it connects to on the corner.
This is hard-earned knowledge. The first few pieces, I cut to length as I would for wood, and ended up with holes in the wrong place and too much sticking out beyond the hole. I’ve learned: I need to cut to the length of the holes. And I’ve also learned something about iteration. I’m glad I tried to piece things together, and didn’t cut all of the pieces first. Learning by doing.
I need to cut the angle iron. How? Cutting a small amount of steel with a hacksaw is fine, but I will make a hundred or so cuts. There are specialized tools for cutting slotted angle iron, but $400 seems too much for this project. I buy a new saw, a chop saw, to cut the metal. There are some choices here — and, as usual, I don’t know enough to decide. How large a saw do I need? Will I ever need it for a project where I cut, say, 2” steel? And there are two types of saws: abrasive, which cut by grinding through the steel, and cold-cut, tungsten carbide-tipped blades. The latter are more expensive. Worth it?
This is a general problem for the occasional craftsman: what tools, what special purpose tools, what quality of tools, what size, is worth paying for? The right tool makes things easier; but is it OK to put up with less than the best for the day or two this project will take? Money and skill; there’s a separate essay there. When is it worth buying a specialized tool? How to balance time and money and quality of results, versus the time required to choose and learn new tools, and a basement full of used-once tools?
I buy a small saw with abrasive blades (cheap!), and run into another beginner problem. The saw takes blades up to 6 inches, but the store only has 4 ½ inch blades. Buy them, or shop around? Turns out that the smaller blades mean that I have to make two cuts for each piece, not one; take it out of the vice, turn it over, put it back in, line it up again. When does it become worth the trouble to just buy a new blade? The calculation would look something like: an extra minute per cut, 60 cuts (if all goes well) — an hour — versus either an hour’s trip to the hardware store, or an online order that means waiting two days to start up. Again, the problem of a part-time amateur.
And bolt together
Choosing bolts to hold this together should be easy, but even there, there’s knowledge I don’t have. How large in diameter? Exactly fill the hole, or room for slop?
And what kind of nuts? The store has lock nuts, with a nylon rim, that don’t require a lock washer to stay tight. That seems a real convenience. But how many times can they be fastened and unfastened? Does the nylon wear out? If I knew what I was doing, and didn’t need to worry about experimenting, I wouldn’t need to worry about that. But odds are I will need to build and rebuild, and I hope the nylon nuts will hold up.
I know to use two wrenches, one on the bolt head, one on the nut. But which hand should the open-ended wrench be in, which hand the ratchet wrench? A skilled worker would be ambidextrous. I’m not.
How tight should the nuts be? Can they be too tight? Can I strip the threads?
All of these are small bits of knowledge, and I can figure out many of them as I work things through. They’re bits of knowledge that separate a novice from someone with experience, with knowledge of materials. It’s a knowledge that comes from hands-on with materials.
So too is a different kind of knowledge: the muscle memory of tightening those bolts. I know that clockwise tightens, but it’s not always easy to transpose clockwise when I’m working on a nut at an odd angle. And I should be able to remember which way the ratchet selector goes for tighten or loosen, but it seems that that knowledge, like which way is clockwise, is only easy when I’m looking right at the object, straight on. Perhaps there are two ways to look at this: a transposition of coordinate systems, to figure it out each time, but more quickly; or a muscle memory, an immediate knowledge of which way the hand goes. Muscle memory—doing without thinking—is the sign of skill; but do you get there through thinking, or through practice?
Writing about skills offers a challenge: you’re trying to describe non-verbal activity in words. You’re trying to describe challenges that you don’t fully understand, or that seem insignificant as soon as you’ve solved them. And, of course, every challenge identified is admitting what you don’t know.
But thinking about skills is important. Not for practical reasons—there are better ways to learn how to make things than figuring them out this way. It’s important because understanding skill offers insight into the mind, the connection of thought and action. The world is a complex place, and we can’t understand it without understanding how it was made—not just design, but also construction—the measuring, cutting, and bolting together of our built environment.
Even simple things like potting benches capture and present the story of the skills that made them. The larger things—machines, buildings, civil engineering structures—offer their own stories of skills. They’re hidden not only by layers of finish, but also by the assumed and unexamined expertise of their makers.
Making something yourself, at the edges of what you know how to do, reveals the skills hidden in the world around us.
Give it a try.