Sunday, August 26, 2012

Let's Talk About Sleeves

It's no secret that I hate set-in sleeves. In fact, I would go so far as to say that I *despise* them. They are irritatingly difficult to get right, especially for a frame like mine. My very narrow shoulders, large chest and fleshy upper arms have been a trifecta of evil in the world of sleeve fitting.

I've come to discover recently, however, that my issues with set-in sleeves are probably because I was under the impression, somehow, that creating a set-in sleeve was 2 parts magic and 1 part loosy-goosy laws of nature. There was no way to be precise about it. But after enough frustration, you get to a point when you start to wonder if maybe you've been settling on horrible sleeve construction because you don't know any better.

So I've been taking some time to educate myself.

It started when I used some muslin to create an early 14th century supertunic toile to begin experimenting with a few things. My initial test was pretty basic- I simply wanted to know what a straight-seam, non-fitted gown might look like on me to determine how "early" I would care to take my garb. In the process, it also afforded me an opportunity to try out a different sleeve. Before doing the sleeves on the toile, I leafed through The Medieval Tailor's Assistant and Thursfield's "transitional" sleeve (page 83, figure 9a) caught my eye. Given my hatred of the set-in sleeve, I kind of wondered why I hadn't really noticed it before.

Essentially, the transitional sleeve (as Thursfield presents it) utilizes a curved sleeve head, similar to a set-in sleeve, where the curve is shallow and basically just a formality. (You need a bit of curve in order to match it to a round armscye effectively.) The sleeve is as wide as your bicep, and the shallow S-curve simply caps it. The likelihood that this curved line matches the length of the armscye, however, is pretty low. The length (or rather width of the sleeve head), therefore, is made up with either the addition of a triangular gusset on one side of the sleeve head (a), or a bit of flaring at the top of the sleeve on one side (b). These combinations do not exactly mimic the behavior of the more streamlined set-in sleeve, and they don't give you the same control over the fit, since you're un-scientifically adding width without regard for how that width fits on the arm, but you can still achieve a good fit on the lower bicep, elbow and forearm.

Thursfield doesn't provide a year range for the transitional sleeve (just a "from 1300" notation), but there is evidence of the use of the sleeve gusset on several of the Herjolfsnes gowns, such as Herjolfsnes 38. Since the Greenland finds are typically understood as belonging to a separate and perhaps belated fashion group from those of the mainland, their dating to the mid-14th century probably indicates that transitional gusset sleeves may have still been in use up to circa 1350 or so on the mainland. This also indicates that using a gusset for a psuedo-set-in sleeve would have been considered quite out-dated by the 15th century (especially given the vast amount of fashion upheaval in the late 1300's.)

My personal conclusion is that the transitional sleeve is a nice option for 14th century-specific dresses, but not those to be considered "high fashion". It is a good alternative when the precise look and fit of a set-in sleeve is not critical to the style of the dress.

An alternative method to this that actually does create a set-in sleeve can also be found in The Medieval Tailor's Assistant. Thursfield actually figured out a mathematical method of creating a set-in sleeve, leaving very little to the magical tweaking I had previously relied on. So I decided that since I needed to redo the sleeves on my charcoal gray dress anyway, I would give this method a test.

After removing the old sleeve, I measured the armscye of the gray dress. I used inches and did cm conversions throughout to keep up with Thursfield's metric instructions, but in retrospect, I probably would have been better off using cm all along.

After following Thursfield's instructions (which you can find on pages 34-38, and for the sake of her copyright I'm not really at liberty to lay out for you here), I had a completely new sleeve pattern. Just for the sake of comparison, I laid my old sleeve on the new pattern just to see the difference, and I was astounded by the huge discrepancy between the two.

After stitching the new sleeve toile into place, I put the dress on to see how it fit, and the Thursfield method worked (nearly) perfectly! For comparison, I left the other sleeve on the dress, so I had my husband snap a picture. It may be difficult to tell how poorly the gray sleeve fits in comparison, but if you look at the way the neckline pulls into that side (vs. the toile side), you see that something is clearly wrong. That side of the torso is under a lot of stress, since the sleeve is pulling the fabric unnecessarily into the sleeve seam.

Doesn't this photo kind of remind you of the old Batman villain scenes?
You can also see that there's some extra fabric in the front shoulder area of the toile side of the dress, but fixing that should just entail pulling the excess fabric into the sleeve. For this dress, however, I'm not worried about that. I did have to remove a bit of excess from my sleeve head to fit the armscye properly, but that was an easy adjustment to make (you can see the original adjustment lines in blue along the top of the seam lines below- I basically removed the projected extra I'd put in as a bulge for the bicep.)

The next thing to do was to adjust the seam so that it shifted to the side of the arm for better button placement. To determine that, I marked where my arm side is, and marked the depression on my extended elbow. After removing the sleeve and flattening it back out, I drew a connecting line between the top seam starting point, the elbow mark, and the wrist side line (inner blue markings above).

Then I cut along that line and placed the piece I removed on the other side. The top of the two pieces don't line up without puckering, so it was a bit of work to get the seam lines aligned properly and for the sleeve to still be mostly flat. Then I transferred it to a new toile.

Something went a bit screwy after this step, and I had to adjust the sleeve to be tighter along it's entire length. I also, again, had to remove some from the sleeve head to allow it to fit the armscye. Not a problem to fix, though- I just took it in until it fit again.

The end result is a set-in sleeve that fits (though I still need to make a few very minor adjustments to the precise tightness.)  It did take a bit of time and a couple of large sheets of paper and muslin, but the effort was definitely worth the eye-opener.

So what have I learned here? For starters, I've learned that dismissing Thursfield's mathematical method for as long as I did was probably a stupid move. Though I have serious doubts that a medieval dressmaker would have been as scientifically anal, there's certainly a method here that could have easily be intuitively developed by an individual seamstress over time. I've learned that there's no reason to be stuck with a poorly-fitted set-in sleeve. And I've learned that, while there are some loosy-goosy rules involved, sleeve fitting requires absolutely no magic abilities.

1 comment:

  1. This was eye-opening. I'll be tackling a fitted cotte for myself in the very near future. I have fitted sleeveless kirtles, smocks with straight sleeve/body seams, and one loose cotte with partially-attached sleeves, like the one in the upper lefthand corner of this page from the Maciejowski Bible:
    Guess I should get to work on the toile :)
    - Mathilde