Skip to content

Scribed Joinery

scribed natural logs with bark

After the Maruta Bench project and a few other projects (Chair no. 1, Toshi-nuki Shelves) where I talked about using scribed joinery, I thought it would be worth while to dig in deeper to the reasoning behind using this technique. In each of the previous projects where I used scribed joints I touched on the reasons why I was using them, but really there’s a lot more to discuss, and I imagine there are quite a few folks out there wondering why would you ever bother scribing? After all it’s a lot more work to scribe and fit joinery.

The classic example of a scribed joint here in Japan is on traditional architecture where the structure is designed to stand on a series of stones, with each post being scribed to it’s respective stone. I took the following photo at the Japan Open-air Folk House Museum in Kawasaki (which I highly recommend for anyone interested in old Japanese architecture). All of the posts for this house are standing on stones and have been scribed. For each post, the shape of the stone gets carefully transferred to the bottom of the post which then gets cut and carved to shape, all prior to assembling the structure itself. That process is scribing. It takes time, but the result is really beautiful.

scribed posts to stones

Traditionally posts scribed to stones are not fixed in place, they are simply set in place on the stone. The scribed fit helps the post “grip” the stone keeping it in place, until such time as an earthquake when the lack of any fixing between the two allows the post to shift and move. Quite a few tests have been done in Japan on the effect of fixing vs not fixing a building to its foundation. And this old method has proven to allow buildings to shift and slide such that the force of the earthquake on the structure is diminished, whereas fixing the building in place means that as the earth shakes and jolts the force of that movement is translated directly into the structure causing much more catastrophic results.

But that’s a whole other deep topic… for now back to scribing.

The other most common place you see scribed joinery in Japanese architecture is when a post meets a beam that hasn’t been milled square. In traditional buildings beams were often hewn with an axe or adze to remove the bark of the tree, but otherwise left with their natural shape and curvature. To join a post to a hewn beam a similar process is followed where the shape of the beam gets transferred to the post, and the post gets carefully cut to mate perfectly with the shape of the post.

scribed post to old beam

The following picture is from my first few weeks doing carpentry in Japan, when I was tasked with fitting some short posts to this beam that had been hewn with an axe. There are a lot of tricky nuances to scribing, particularly when it comes to architecture. As you are cutting and scribing you need to constantly be aware of the the posts orientation to the mating beam, as well as the final length. The process usually takes a few different rounds, first rough cutting the shape, then fitting the pieces together to check their alignment and to see where adjustments need to be made. Then you mark the areas to be cut more precisely, trim the post, and once again check the fit, repeating the process until you gradually creep up on the final fit.

scribed post and beam new work

In architecture a common alternative to scribing is to cut a flat in the mortised piece, such that the tenon piece can join simply with a square shoulder cut. This is the solution I’ve seen most looking at images of western timber frames, but it’s also occasionally used here in Japan.

In the picture below of a renovated old house here in Japan, you can see where the end of the beam has been cut flat to more easily join with the post coming in from below. In this case the final result is still pretty clean, and the flat cut in the beam doesn’t stand out very much.

post joining flat seat on old beam

Here’s another example from the same house. It’s a bit difficult to see on the old dark wood, but the upper beam has a flat cut where the post joins.

post joining flat seat

Often this kind of flat seat is cut on the top side of a beam in a roof structure where a small post joins. With the cut on the top side of the beam it’s virtually impossible to see from below. The opposite end of the post will then often be scribed to it’s mating beam, since that half of the post’s joinery is much more visible.

Here’s another example with a pretty extreme cutout in a beam of a house from the Japan 0pen-air Museum again.

post fit to flat cutout on beam

Here’s one more example from a roof structure at the Katsura Imperial Villa in Kyoto.

roof joinery with flat cut instead of scribing

Finally here’s a picture pulled off the internet of a Western-style timber frame. All of the joints between the posts and beams have been simplified by cutting the joining pieces flat, this time on the tenon parts as well, presumably to make laying out and cutting the tenons simpler.

Cutting a flat seat definitely simplifies the process of joining a post to a round beam, but the finished look is clearly different from the look and feel of a scribed joint. To my eye it feels much less refined. Instead of each frame member flowing into the next creating a unified structure, the result feels much more disconnected and disjointed.

Cutting a flat also means cutting away part of the beam itself, which can definitely affect the strength of the beam. In the previous picture the main beams have had a serious amount of material cut out to create flat spots large enough to join the posts and braces, which definitely has structural implications. Scribing on the other hand allows you to leave the beam fulling intact, maintaining it’s strength.


When it comes to furniture, I can’t say I’ve seen many examples of scribed joinery, and I have a strong feeling that mass production is the main cause of that. Nowadays working with round or non-square parts really comes down to a design choice. It’s easy and cheap to build a piece of furniture using square parts that can be quickly and easily cut and joined to one another. But in those situations where round parts are used I can’t say I’ve ever seen anyone scribe the joints. Instead round tenons and drilled out mortises seem to be what most craftspeople gravitate towards.

Here’s an example on a Shaker rocking chair. The chair rail has been tapered down just slightly at the end to join with the leg, making the joining of these parts pretty quick and simple. This kind of joinery is super common, particularly on chairs. It’s quick and easy to do, either with a lathe, with hand tools or specialized tenon cutters.

Shaker chair mortise and tenon joint

Here’s a more extreme example, showing a tenon cutter for making super simple log furniture. This kind of joint is easy to make, but the finished look really loses any sense of natural flow between the pieces. It’s much less refined than the shaker chair above.

Compare the look of the joined log bed in the previous image with the following joinery done on a roof structure once again taken at the Katsura Imperial Villa. Here the logs have a much more fluid connection with one another. The joint is beautifully cut and fit in such a way that you may not even notice it if you weren’t looking for it. Whereas every joint in the previous image is rather conspicuous.

scribed natural logs with bark

Here’s one more image of a nicely done scribed joint from a bathroom hand rail in a house built by the guys at Somakosha. Imagine if they had simply tapered down the horizontal piece and plugged it into a hole in the vertical piece… they may have been able to pull that off much more quickly but the finished result would be completely different.

hewed and scribed hand rail

In furniture cutting a simple round tenon and plugging it into a round mortise also completely eliminates any shoulder from the joint, putting all the stress from racking/movement onto the tenon itself. Depending on the situation that may not be a problem, but if you are concerned about the strength of a joint, and want to provide additional strength against racking, then cutting a scribed joint and providing the tenon a shoulder can be a great option. That was one of the reasons behind my choice for using scribed joints on the seat frame and leg mortise and tenon joints on the Maruta Bench below.

maruta bench pegs installed

Scribing is not a quick process. It definitely takes time and it’s not a process that’s particularly well suited to mass production. It requires slowing down, and getting intimate with the work, carefully marking and cutting your parts to fit. Scribed joints also have a strong effect on the overall aesthetic and strength of the joint as well.

That said when it comes to furniture (and architecture) I think scribed joints can really open up a lot of possibilities. The problem for a lot of people may be overcoming the extra time and effort involved, but if you ask me that extra time and effort is absolutely worth it. And really the process of actually cutting a scribed joint is not that difficult. It’s not even that different than cutting a set of dovetails in many ways. Dovetails are after all essentially scribed joints, with one half of the joint being layed out directly onto it’s mating half and then cutting/trimming to fit.

So I hope the furniture makers and carpenters out the reading this give scribing a shot. Next week I’ll cover more of the actual process that I used for scribing a joint, so stay tuned, and thanks for reading.

7 thoughts on “Scribed Joinery”

  1. A topic I’m interested in so thanks! Those are excellent examples. I have not scribed wood to wood but I agree that with irregular members the joinery looks right when scribed and a little “expedient,” if not. Especially when newly done.
    Probably after a few decades wabi sabi rules and it looks better.

    I have scribed posts to stones, for one project, for myself. I was surprised to find, being a novice, that it didn’t require great skill but it did require patience and time and hand tools. As a rough estimate, six posts took me about a week working half days.

    With the post to stone scribing, what I like is how the posts seem more anchored visually to the ground. It makes the structure look more fixed and permanent, and part of the earth. Probably not all structures benefit from that look but some do.

    1. Thanks for the comment Gary.

      The work you did, scribing posts to stones, is impressive (I saw pictures everyone) and sounds like you got the work done pretty fast too. Six half days or three full days for six posts isn’t too bad, especially if it’s not your regular gig. I imagine you got faster as you went along too, and by the 6th post you probably had a much stronger grasp of the whole process. That’s how I’ve always felt at least when doing any kind of repetitive task. And often by the last piece I find myself finally getting into a groove where the finished result looks really good… and I wish I could go back and start from the beginning again!

      This kind of work is always dependent on other factors as well, and budget again looms large especially in carpentry work. On historical work like a shrine/temple etc. I’m guessing that a carpenter will spend a lot more time fine tuning a scribe, than say on a lower budget project like a house where expediency will become more important.
      I think that same consideration plays into the decision to scribe or not in the first place.

      And I agree 100% about posts scribed to stones feeling more anchored. Scribing creates a really solid physical connection but visually as well. That’s also how I see scribed connections between different frame members; more connected and more harmonious.

  2. Great post. Scribing on one end of a piece is not so bad, merely taking patience. I find it waaaaay harder to scribe a piece on both ends, because final length is something you have to consider too. How do you manage to get a stretcher between legs to be the right final length when scribing on both sides? I am not even sure where to mark in that case since the surfaces can be quite irregular.

    1. Thanks! I’ll try to touch on this in more detail in the next post, but essentially you have two options. On the Maruta Bench project the long stretchers (mortise pieces) were pretty consistent in size so I was able to essentially calculate the distance between the two stretchers at the narrowest point and lay that distance out on the short stretchers for scribing. Then I rough cut the pieces and started scribing gradually working my way up to those layout marks.

      For a timber frame or structure with much more irregular parts it’s best to use centerlines or some other kind of reference line on each of the members with mortises. Then you can calculate the required distance between those reference lines and start to apply that to the connecting part which will be scribed between them. I’ll see if I can pull up some pictures and write about that process in more depth in the next post.

      Thanks for the comment.
      -Jon

  3. I’m also looking forward to seeing how one might handle scribing to fit on both ends of a stick when the dimensions are fixed. That is PhD level stuff for sure!

Leave a Reply