Patrick Daugherty writes,

“I’ve heard it said that there’s no such thing as a glue-starved joint. I suppose it is to imply that, as long as the glue bonds to the fibers, the film thickness between the wood surfaces is not relevant. Do you agree or not?”

Well, it’s difficult to starve a joint. However, if you’ve got a mortise and tenon joint that’s a little too tight, you’re gonna shear almost all of that glue off, during assembly. The tenon is gonna push it down to … the glue you put in the mortise is gonna be shoved to the bottom of the hole. The glue you put on the tenon is gonna be pushed up to the shoulder. Not only do you have a very thin glue joint, or pretty much not enough glue to saturate that surface, but if it’s a real tight joint there, you’re gonna push that joint apart, and you’re not gonna have a good, clamped together surface.

So it’s not just about how much glue is present, or how much glue is applied. It is the way that the two surfaces being glued together approach each other, and whether or not that action of coming together traps glue in there or shears glue off. This starved glue idea, or concern, is exactly why I started working with the taper locking joints, because it does trap glue in there, rather than shearing it off. In traditional joinery, there is a real delicate compromise between avoiding that glue shear, and getting glue bond by having wood-to-wood contact. The tolerance requirement there is very, very tight, and I don’t think can be overestimated or over attended to. So yes, there is such thing as a starved joint, but it’s generally only going to be happening in a mortise and tenon joint.

And I actually have had that. The place that I had that was on a rocker-runner slot that I was fitting tight. And years later, the joint failed, and when I looked in it, it was a…. Even several years after I had built that piece, it was a perfectly fitted joint, wood contact all the way around, no evidence of wood glue whatsoever so it was completely sheared off. So that’s the only time I’ve noticed failure due to starvation, but one time was enough to prove that it actually can happen. So attend to your precision.
Thanks again for your question.

–Transcribed as spoken