Sunday, December 14, 2014

The Science of Sleeves.

There are no shortcuts to making a great fitting sleeve.

When I first started out creating set in sleeves, I began with a very basic, loosely explained method. We'll call this your standard "magical" method. You've come across this method every time the physical act of drawing the correct sleeve head curve is sort of glossed over in the instructions with something along the lines of "until it looks about right".

After drawing out the rest of the sleeve's shape according to the measurements of your arm, and having the armscye measurement in hand, you're supposed to create an S-curve of the armscye length. You're given no real guidelines to work with (except maybe a "don't go higher than this" line), and there's absolutely no information about a sleeve head not actually being symmetrical. So after drawing the totally made up curve, I don't know what. Offer chocolate to the celestial sleeve gods, burn some sandalwood incense, and recite Hamlet's soliloquy backwards, I guess? Hope and pray.

That method is a relatively common starting point for many of us because it doesn't challenge us to understand anything. It's also sometimes touted as a "period method", so we automatically gravitate to it. It produces a sleeve that fits your armscye. Forget the rest.

Until you raise your arm, or try to hug a friend, or just stand there doing nothing, for that matter. The chances that you actually found the right sleeve head curve first try with the worst instructions ever are so slim, they might-as-well not exist. So a sleeve made with this method is not going to be your best attempt at a sleeve.

After getting frustrated with hoping and praying that my sleeves would work, I was really excited to find a method that mostly worked, most of the time. Unfortunately, I made the mistake of believing that that sleeve drafting method, presented in "The Medieval Tailor's Assistant" by Sarah Thursfield, was better because it used math. I figured that a few mostly successful sleeves made from that method (that were much more comfortable than my previous "magical" attempts) meant that Thursfield's drafting system could be relied upon for a great sleeve, every time.

Eventually, however, I began to see that Thursfield presents the formula for drafting a sleeve in such a mathematically precise way that errors in the fit just can't be troubleshot. When half your drafting method requires 5ths and 10ths of your body's measurements, it's pretty dang hard to keep up with exactly which portion of your shoulder joint those relate to.

In essence, applying such mathematical strictures on the process turns the draft into nothing better than an off the rack pattern. Everything is based on proportions that you probably don't really have. Is my arm depth truly 1/5th of my arm length? I'm not a Grecian statue, with perfect proportional relationships between my various body parts. So maybe my arm depth is really 7/32nds the length of my arm. Translate that minor difference across the whole drafting process and the result is an ill-fitting sleeve. And there's no way to correct without a whole other set of adjustments with no relationship to your original draft. [This is not to say that "The Medieval Tailor's Assistant" isn't worth owning- it's a great resource.]

Sleeve drafting is a science, not a shortcut. It can't be made up based on loose or nonexistent limits, nor can it be entirely broken down into formulas and equations. But your crappy fitting sleeve is doing x because of y. Cause and effect. Method and result. SCIENCE!

All good scientific study comes from first understanding the subject on a basic level. When it comes to sleeves, the primary thing to understand before looking further is that just because a measurement is right doesn't mean the shape you create from it is.

The length of a sleeve head and the measurement around the corresponding armscye should be equal, right? In a very technical sense that also means that I could create a flat sleeve head, and it would still fit the armscye as long as the measurements match. That's not to say that a sleeve created and attached in such a manner would be particularly good (or period accurate), just that in a purely mathematical and physical sense, it would work.

So if a flat sleeve head is just as functional as a curved one (for the sake of argument), then that also means that any curve to a sleeve head, as long as the measurements match, is just as technically good as any other curve.

Except that it's not just as good on every other level that matters (aesthetics, comfort, movement, etc.) Even if you can technically make a functioning seam between the armscye and the sleeve head, that doesn't mean that the resulting seam, or the nature of the sleeve beyond that seam, are ideal. If a sleeve head curve of the correct measurement can look however it wants and still fit the armscye, every different version of that curve creates a completely unique sleeve. And, more importantly, only a tiny fraction of those unique versions actually work for your unique body and the fit you are trying to achieve. One might even go as far as saying that, for every single garment armscye, there's only one perfect curve for you, and all the rest are crap. And while that sucks, it also means that if you're still having an issue with your sleeve, something is still not right. Because sleeves are science. You can't just close your eyes, knock your heels together, and stand back as any issues resolve themselves.

The only way to find the right sleeve head shape for your garment, your body, and your fit needs is trial and error. I can send you to at least five sleeve drafting tutorials that will get you technically functional sleeves. You can make a sleeve from any one of them that will fit your armscye. While it may be perfectly passable, though, it will likely not be great. The armpit will probably be too sort, or the top curve too high, or the bicep too tight. Or something.

But since science can be learned, it's not impossible to troubleshoot those issues. You may need to do it one issue at a time. You might even need to revisit an issue after fixing others. You will probably get frustrated and a bit tired from the process. The point, however, is that you don't settle for the shortcut.

I'm putting the blog on winter break while I prepare for and celebrate Christmas with my family. I'll return on January 4th. Season's Greetings to you and yours!


  1. Sleeves are the devil. I recently ran across this link and it makes sense. I'm not sure it will make my next sleeve easier.. but I at least understand the different shapes. Hope it helps.

    1. That is actually the exact blog post that got me started down this path. My post was originally going to be about her information and method. I think she's on to a lot of really good points, but there's still something missing that I can't quite put my finger on just yet. I see a lot of experiments in my future.