The Hundred-Year Chair · a chapter

The Search

After building my first chair, I went back to the photographs of other people's chairs that the machine had collected for me. Until then I had given them less attention than my own drawing. Now I had a finished chair I did not like very much, and Rietveld's crate chair began to look more interesting.

The reference wall brought together chairs made from boards, from plain stools to armchairs with overlapping slats. Rietveld's Kratstoel was among them, along with the contemporary HAY version of his Crate collection. Both looked like something I could make with the materials at hand. Could his chair be a better use of the scrap than mine?

I found Rietveld's drawing for the 1934 crate chair. The sheet showed the chair from the side and above, with screw positions, dimensions, and handwritten instructions. The chair itself was plain boards cut square and screwed together. I did not need a mold, a factory, or a new set of woodworking skills to try it.

An afternoon and some more salvaged plywood gave me a chair I liked. The black film on the faces and the pale plywood exposed at every cut made the overlapping boards stand out. Although Wendy and I don't agree on the extent to which our living room should be Karl's chair museum, this chair remains in use in our home.

One piece did not cooperate. A cleat, the short bar supporting the seat, was specified at 390 millimeters, too long to fit between the members of my replica. I had a short argument with Claude about that 390. Claude insisted the drawing was right and I was mistaken. Why I was arguing with a machine, I don't know. Still, I did prevail, and Claude conceded the drawing contained a mistake. I shortened the cleat.

The replica also let me see how Rietveld got his angles without angled cuts. The seat tilts and the back leans, although the boards can all be cut square. Their widths and the positions where they cross establish the posture. My first chair had asked the saw to supply some of its angles; Rietveld's asked the builder to place overlapping boards. Neither method escaped the need for accuracy. They put it in different operations.

Eek's scrap-wood chair was a fussier proposition. To make its curved back, he cuts the scrap down to six millimeters thick and bonds two pieces together in a mold. The seat has faceted mating surfaces that require several angled cuts. The casual appearance conceals demanding joinery. Rietveld's parts could be laid out with a square and cut with a handsaw. With his drawing and the materials I had, I could build a chair in an afternoon.

That left an awkward question. If I could already make a chair I liked, why design another? There was certainly no shortage of chair designs. But I can understand why designers keep returning to them, as poets return to the sonnet. The familiar form leaves room for very different work. A chair for a dining table need not be the one for reading on a porch, and a chair made by an amateur from a drawing need not resemble one made by the thousand. I was not searching for the perfect chair. I still wanted to see what I could make from our narrow offcuts.

The Rietveld build had given me something I was proud of, and this time I sent a photo to my sons with the declaration "Sharp!" On July 31 in Queensland, my son Nate suggested Donald Judd. His message said I should look at the "really cool board furniture." I brought the suggestion to the machine with the enthusiasm of a man handed extra homework: "Geez...now we have to bring in Judd."

I looked through Judd's furniture. Much of it struck me as form ahead of function, interesting in a spare architectural way but not something I especially wanted to sit in. During that exploration I found a photograph of a bright four-panel chair that I did like. I sent it back to Nate, mistakenly thinking it too was a Judd.

My reply put two thoughts together without making the transition clear. "Triumph of beauty and elegance over function" was my reaction to Judd's furniture generally. But when I added, "I may see if we can build this one from scrap...", I meant the chair in the photograph. That one looked cool enough to try.

The design I was drawn to was Verner Panton's Vilbert for IKEA. My notes only began using the right name about a week later. By then I had been working from the angles in the photograph for days. When I went back to the actual Judds, I still preferred the chair I had found by mistake.

The Vilbert had not been an immediate success for IKEA. The company acknowledges as much in a retrospective of its furniture. Now it is a collector's object, offered at four-figure prices, sometimes with the original IKEA label advertised as a virtue. One dealer's asking price was nearly $1,800, label still attached. The unpromising retail history did not stop me liking it, but neither did its later collectibility tell me how my version would work.

I thought it could be a fine MakerStock chair: a design made from materials we stocked and sold as a kit. It was less obviously a way to use the scrap, because it wanted panels rather than narrow strips. I was beginning to pursue a different problem from the one that had started the project.

I could not find public drawings, so I estimated dimensions from photographs and fitted lines to the panel edges to work out their angles. Instead of copying the original connections, I designed a central bracket to hold the panels in position. Before making it full size, I scaled the bracket to one tenth and 3D printed it from ABS plastic. Two-millimeter plywood and superglue supplied the rest of the miniature chair.

I liked the model enough to keep going. "Brackets printed. Came out nice," I wrote the next morning, followed by, "I wonder what parametric exploration would show." In other words, what would happen if I kept the same arrangement of panels but changed their angles? The machine could draw a family of related chairs much faster than I could build even small models of them.

The resulting exploration was different from the precedent wall. The wall had assembled other people's designs; these were variations of one construction. In the close-up grid, the front and rear support angles change in two-degree steps, giving thirty-six combinations. I could compare them while holding the intended sitting posture fixed.

At five past five that morning I wrote, "Fascinating to see 'Memphis' emerge from literal pure functional geometry. Not arbitrary aesthetic gesture in this case!" The splayed panels reminded me of Memphis furniture, particularly the tilted divisions of Ettore Sottsass's Carlton bookcase. I liked the idea that a practical search had produced something with that exuberance. But I was overstating the purity of the exercise. I had chosen the arrangement, the variables, and the shapes I liked enough to pursue.

My mother, a historian, was eighty-eight and in the hospital during this period. That same morning I asked the machine to write up the adventure in a way she might enjoy and sent it to her. The folder on my hard drive for this book was created the next morning. I had a scale model I liked, a family of drawings, and the beginnings of a story to tell.

On August 2 I built the full-size adaptation. Panton's chair used colored, melamine-coated MDF panels screwed together. Mine used bare plywood and the bracket geometry I had been exploring, now 3D printed at full size from plastic. The drawings and model had made the assembly seem straightforward. Sitting in it supplied information neither had given me.

The front panel pried disconcertingly away from its bracket every time I sat down. The chair felt faintly spongy, as if something might let go. I noticed it that evening and wrote it down. The following morning I turned to the cutting plan.

At twenty past five I opened a cheerful new thread: "OK now a fun little project which strengthens the story." I was thinking about producing hundreds of chairs, tuning the dimensions so the parts would fit together efficiently on sheets of plywood. A few millimeters saved on one panel could be repeated across a whole production run. This was a problem I knew how to enjoy.

The cutting plan grouped twenty-five chairs into a seven-sheet batch. The leftover offcuts came to less than one percent of the length of the strips, although that excluded the material turned into sawdust. It was an appealingly precise result. I could account for a five-millimeter saw cut while leaving the movement I had felt through the seat unresolved.

Then I asked for "a lovely abstract graphic of the tiles in memphis colors that is geometrically exact but a composed mosaic that is intended to be a piece of visual art." The machine arranged the cutting strips into a wheel. It ignored some of my directions, as artists should.

I still think it is a genuinely nice piece of art. We were in a good mood. The machine called it "a rather good day of engineering." I had a plan for making hundreds of chairs with very little scrap, and a picture worth hanging on a wall. I did not yet have one full-size chair of this design that I trusted under my own weight.

I returned to the bracket. I had used 3D-printed plastic because I could turn a drawing into a part without leaving the desk. Now I wanted to try hardwood. I also considered cutting the lower panels down to narrow members, or staves, perhaps 75 millimeters wide on each side instead of a panel 400 millimeters wide. That brought the design closer to the strips I had originally wanted to use.

"The staves are growing on me," I wrote, "but I'm not ready to commit to the branch." The next day I committed. With wooden brackets and narrow legs, the new version could be made from a single 600-by-1200-millimeter project panel, the size I could buy off the rack at Bunnings, the Australian hardware chain.

This was the fourth full-size chair, the one Wendy sat in on August 5. I had followed the drawings through a replica I liked, a mistaken attribution, a miniature, and a production plan. Now Wendy's 54 kilograms were moving the wooden-bracket version disconcertingly across the floor.

A chapter of The Hundred-Year Chair, drafted and refined in public. The same text lives in the book; the interactive tools and the fully-sourced technical notes live here. home · the book · the technical notes · hundredyearchair.com