Every joint is an answer to a question. The trouble is most woodworkers never ask what the question was.
Western furniture joints were built for stone buildings on stable ground in climates that are damp but don't shake. The problem was resistance: stop the rack, carry the load, hold the connection when someone leans on it. Mortise-and-tenon, dovetail, bridle joint. They all work the same way. They lock the connection and transfer the force into the material.
Japanese joinery started with a different problem.
Wood buildings on soft ground in a seismically active country. When the ground moves, everything moves. A tight glued joint in that environment isn't a virtue — it's a liability. The structure flexes, the tight joint transfers stress directly into the wood, and wood breaks in ways it doesn't recover from. So Japanese temple builders engineered the opposite: joints that accommodate movement, flex under load, return to position, and continue holding after the motion that would crack a rigid connection.
The Kanawa-tsugi, the log splice used in temple framing, looks like a puzzle you'd find in a hobby store. That interlocking geometry isn't ornament. It's an engineering solution to shear forces in a seismic zone. Glue would ruin it.
The second thing Japanese joinery solved doesn't exist in the Western tradition. In Western furniture, structure and aesthetics are separate decisions. The joint is structural. Whether you expose it is a design choice. You build the mortise-and-tenon, then you decide if you want to see it.
In traditional Japanese work, the joint IS the design element. The post-and-beam in a tokonoma alcove. The shoji screen grid, which is simultaneously structure and light. The tansu chest with its hardware reading as surface composition and construction logic at once. Structure and aesthetics aren't disciplines being coordinated — they're one discipline. The Shakers understood this. Japanese carpentry started there.
Here's what doesn't transfer: most of the joints.
The complex puzzle joints are designed for softwood construction at timber-frame scale. They work with cryptomeria and hinoki in centuries-old buildings. They don't map to hard maple furniture joinery in a BC shop. You can spend a week learning the Kanawa-tsugi and never once have a place to use it.
What transfers is the question behind the joint.
Not: what joint does this style require? Not: what joint will hold? The question is: what forces are acting on this connection, what will the wood do over the next twenty years, and what joint answers those specific conditions most directly?
That question is the whole practice. Western furniture-making arrived at it through centuries of failed pieces coming back to the shop. Japanese temple carpentry had it from the beginning, because the consequence of getting it wrong was a collapsed building in an earthquake.
Ask it earlier. That's the lesson.
Japanese joinery gets held up as proof that complexity is a virtue. Katsuhiro Hayashi's contemporary work, the joints that assemble without hardware and disassemble clean after four hundred years — genuinely extraordinary.
But complexity proportional to the problem is craft intelligence. Complexity beyond the problem is performance.
Those joints are complex because the problem was genuinely complex. A seismic splice in a sixth-century temple deserves that joint. A drawer box does not.
Ask what the joint is solving. Then use the simplest joint that answers it.