# The Proportion Hewn Shop Notes · https://hewn.build/shop-notes/the-proportion AI-drafted, not yet checked: Drafted by Chip, the writing system Hewn runs on Claude, from Hewn's craft corpus. Jonathan Salomons has not read this note yet, and no outside reader has checked it. Read it as a shop note: check a number before you cut to it. Cite as: "The Proportion." Hewn Shop Notes, hewn.build/shop-notes/the-proportion. Drafted 2026-04-24; AI-drafted, not yet checked. --- The golden ratio does not help you build better furniture. That sounds like an opinion. It's not. It's a measurement problem. Phi — 1.618, the divine proportion, the thing with the nautilus shell — works on building facades and page layouts. It does not scale to an object you sit at, eat dinner on, and bump with your hip while carrying a casserole dish. People who claim their furniture is golden-ratio proportioned are working backwards from a piece that already looks right and finding phi in the noise. You can find any ratio if you measure enough pairs of dimensions and cherry-pick. Furniture proportion is simpler than that. It's also more useful. ## Two Different Problems There are two proportion problems in furniture, and they get confused constantly. The first is scale: the piece relative to the room. Is the dining table the right visual weight for the space? Does the bookcase anchor the wall or float against it? Scale is solved by understanding human dimensions. Table height, sight lines, the architecture around the piece. That's important, but it's not what trips people up. The second is internal proportion: the relationship between parts. Top thickness to span. Rail height to leg width. Case depth to height. Overhang to apron. This is where most design failures live. The piece is the right size for the room. The parts are subtly wrong in relationship to each other. And the builder can feel it but can't name it. You can name it. The rules aren't secret. ## What Actually Works **Top thickness and span.** A solid-wood dining table wants a top between 7/8" and 1-1/4" thick, spanning 30–36" in depth. Below 3/4" on a wide panel reads thin and nervous. Above 1-1/2" reads as slab, not tabletop. The proportional window is narrow because the physics and the aesthetics agree inside it. **Overhang.** A table top that overhangs the apron by 1" looks like it was cut to fit. Two to three inches creates the shadow line that makes a table look finished instead of assembled. Five inches or more and the top starts to look like it floated loose from the base. The functional range and the aesthetic range are the same range. This is common in furniture. **Leg taper.** A straight leg reads as static. A taper that starts at the bottom of the apron and runs to 50–60% of the original dimension at the floor reads as light without reading as fragile. Less taper and the benefit disappears. More taper, down to 40% or below, and the leg starts looking like it belongs under a Victorian side table. A taper does two things: reduces material where the bending load is lowest, and gives the eye somewhere to go. **Rail to leg.** The apron rail height should land between 2/3 and 3/4 of the leg width at the junction. Below that, the rail looks undersized and the leg looks orphaned. Above that, the table reads as enclosed rather than open. Shaker rails tend toward the narrow end. Arts and Crafts toward the wide end. The range is small, and the style sits inside it. **Case pieces.** A tall freestanding case — bookcase, cabinet, armoire — needs its depth to land between 15% and 20% of its height. Below that, the thing looks like it's about to tip over. Above that, it reads as a block. A bookcase 72" tall and 14" deep sits at the edge. Fine for a built-in. Visually anxious standing alone. ## Reading Proportions in a Plan Most of us work from plans more often than we design from scratch. The real skill is evaluation, not invention. Three questions. Ask them before you cut anything. Is the top thickness structural for the unsupported span? A 36" span in hard maple at 3/4" is borderline. At 1" it's solid. At 1-1/2" it's more than enough and may read as heavy. Know where the structural minimum is. The aesthetic window opens above it. Does the overhang exist? A top flush with the case sides reads as unresolved. Even 1/2" to 3/4" of overhang creates a shadow line that reads as intentional. If the plan doesn't specify it, add it. Do the legs match the visual weight of the top? A 1-1/2" square leg under a 72" dining table reads as fragile. A 2-3/4" leg reads as solid. A 3-1/2" leg reads as heavy, unless you're building in a tradition that calls for it. Match the weight or know why you're not matching it. If any of these answers come back "not quite," you've found a design decision the plan made poorly. Worth knowing before the wood is on the bench. ## Designing From Proportion Out If you're starting from a sketch, proportion comes first. Not last. Start with the constraint: what does this piece need to do, at what scale? A dining table sits 30" high. That's close to a law of physics. The number of seats gives you the top size. The top size gives you the leg height. The taper ratio gives you the floor dimension. The span-to-thickness proportion gives you the top thickness. The top thickness gives you the rail height. You haven't designed the piece yet. You've established the envelope it lives inside. Now the real decisions — joinery, details, style — happen inside a framework that keeps the result from being subtly wrong. Most furniture that looks slightly off was designed in the other direction. Style first, proportions after. The legs were copied from a chair the builder liked. The top was 1" because that's what 4/4 yields after milling. The rail height was whatever felt right. Each choice made in isolation. The result: nothing precisely wrong, nothing quite right. Proportion is not a straitjacket. It's the structure the good decisions happen inside.