Sewing: Brown Herringbone Smokkr

Tablet weaving at an
event in simple garb.

When getting ready for the 2017 Raglan Fair, I started working on a new Viking smokkr or apron dress. I got as far as drafting the pattern and cutting out the pieces before finding my old smokkr, at which point I decided to spiff that one up instead of rushing to finish the new one. In the weeks that followed, I slowly chipped away at the new smokkr, picking it up and putting it down as my energy levels and free time allowed, and it's now in a wearable state.

I have very mixed feelings about the smokkr as commonly reconstructed in re-enactment circles. It's a garment that has become more or less the uniform for re-enactors portraying women of the Viking period, but all of the information we have on this garment is based on a few tattered fragments, and minuscule fossilized loops of cloth preserved inside of metallic brooches. Those tattered fragments and scraps can tell us a lot about what the smokkr was made from--fibers and dye plants used, thread count, weave, amount of twist in the yarn--but they can't tell us much about the shape, fit, length, or degree of ornamentation of the finished garment. We do have some visual representations of Viking women on runestones and gullgubber, but they are heavily stylized and difficult to interpret. And yet, there are plenty of re-enactors who will happily criticize any interpretation of this garment that differs from their own. I have some very strong feelings about this, which I hope to write up separately one day, but for now I'd like to focus on the garment that I put together as an expedient way of expanding my wardrobe for a week-long SCA event.

Viking-age depictions of women, originally published in a 1981 article by
Flemming Bau entitled, "Seler og slæb i vikingetid, Birka's kvindedragt i nyt lys"
in KUML, Årbok for Jysk arkæologisk selskap. ISSN 0454-6245. Copied by me
from http://urd.priv.no/viking/smokkr.html. 

My reconstruction more or less follows the norm for re-enactment, being shaped more or less like an upside-down funnel with pinafore-like shoulder straps secured by oval brooches. In order to make the size and shape of my pattern pieces conform roughly to the size and shape of the fragment found in Hedeby harbor (taking into account the fact that I'm a bit larger than the average Viking-age woman), I cut my smokkr with five panels: one wide front panel, two narrow side panels, and two narrow back panels, similar to the one used by the Historiska världer, except with a center back seam. (To view the pattern diagram, click the link at the very bottom of the page.)

The whole smokkr is five panels: front (shown folded in half),
left side, right side, left back, right back.

As tempting as it was to squeeze every last inch of fullness out of my fabric, I chose not to use any triangular gores for two reasons: Firstly, we don't have any surviving fragments with evidence of triangular gores.

Cutting out the front and side pieces, with
minimal (but not zero) waste.

In my personal opinion, the reason you see so many re-enactors with triangular gores is because that's what's done on other medieval garments like tunics and tunic dresses. The most economical way to get a big, full skirt out of an expensive piece of fabric (and all fabric is expensive when every step in the manufacturing process needs to be done by hand) is to cut triangular gores in alternating directions like a tangram. I can certainly understand the desire to use every square inch of fabric, but what little evidence we have doesn't seem to bear that out. The Hedeby smokkr fragment appears to have curved seams, which are wasteful of fabric more or less by definition. The near-complete pair of trousers that was excavated from Thorsbjerg is scandalously wasteful of fabric. (If you find a cutting layout for these that doesn't leave you with a giant pile of curvy, bias-cut scraps, please share it with me!) A tunic found in Hedeby shows shaped shoulder seams and sleeve heads. Given that the smokkr can only be worn with the aid of very large pieces of jewelry, it seems reasonable to think that this was a high-status (or at least a special-occasion) garment. High-status garments could afford to be a bit wasteful with their cutting layouts.

Secondly, we're accustomed to seeing big, beautiful flowing skirts on historical dresses, so of course people make smokkrs that look like princess dresses. I don't have enough evidence to make a strong case either way, but when I look at the visual representations of women on runestones, carvings, tapestries, etc., I see female figures that are very columnar, without big ballgown skirts that flare in all directions. The outermost layer on a woman's lower half falls more or less straight down from chest to knees, with any fullness concentrated low and in back where it sometimes extends into a sweeping train.

Of course, as I said earlier, the only surviving pieces of smokkr that we have are small and incomplete, and the visual representations are highly stylized and confusing at best, so any interpretation is going to be based on a certain amount of, "well, I just like the way it looks." I made mine with enough width to go over my hips easily and to walk comfortably, but not so much that it billows.The length of the smokkr is also a big question mark. On Guldgubber and figurines there are often horizontal lines running across the lower portion of the dress, sometimes with closely-spaced vertical lines below. To me, this looks like it could be a representation of a long, full-skirted under layer being partially covered by a less full, slightly shorter over layer (e.g. a smokkr). Of course, those lines don't appear on *all* female figures, the vertical lines seem to start higher in back than in front, and they could just as easily be trimming or embroidery instead of different layered garments, so once again, personal taste and judgment comes into play. On my smokkr, the panels flare slightly less on the front seams and slightly more on the back seams. The length is around the bottom of my calf, well below the knee but above the hem of my under-tunic. The seams are sewn with a running stitch, and then the seam allowances were opened up, folded under themselves on each side, and stitched down with a whip stitch to prevent fraying.

This is how the seams look from the inside.

Since I started with a pattern piece based on the Hedeby fragment, I figured I'd continue in a similar vein. This fragment has what appears to be a narrow dart running vertically down the piece, with a decorative braid covering it. Unlike darts in modern clothing, however, the excess fabric forms a ridge on the outside of the garment instead of being hidden on the inside. This dart is often held up as evidence that garments from Hedeby were closely fitted to the body, but at 2-5mm wide, the dart is so narrow that it would do little to enhance the waistline. It's possible that it was added as a way of adjusting the fit after weight loss, or if the garment was passed on to another person, but again, 2-5mm does not make a huge amount of difference in fit unless the garment is fitted very tightly indeed, which doesn't seem to fit with the very gentle curves of the surviving fragment.

There was, however, a decorative stitch over the dart, extending above and below it. Rather than re-create the dart, I decided to add some decorative stittching over the side front and side back seams, leaving the center back seam plain in case I ever wanted to adjust the fit. I used a vandyke stitch similar to (but not exactly the same as) the one used on one of the textile fragments in the Oseberg burial.

Each of the seams, except the center back, was covered in
a 2-color vandyke stitch.

To make the shoulder straps, I cut strips of herringbone fabric 2" wide and twice as long as I wanted the straps. I folded folded the long edges into the center, and then folded the strips in half, making a strip 1/2" wide with the raw edges enclosed in the middle. I whip-stitched the edges shut, and then folded the straps in half the other way to make a strap half as long. I made another line of whip stitches through both halves of the strap, starting at the raw edges and stopping about an inch short of the fold, and then opened them out flat to make a strap 1" wide with a loop at one end for the pin of the turtle brooch to pass through. I did another line of embroidery stitches (this time, a 2-color herringbone stitch) over the whip stitching to hide it, and then stitched the long straps to the back of the smokkr along the top edge. I made another pair of short straps to go on the front, this time without whipping the two halves together or adding the decorative stitching because they're just barely long enough to peek above the top of the dress.

The pin on the back of the brooch passes through
the loops in the straps.

Eventually, I would like to add a band of blue fabric across the top, similar to what I did with my green apron dress, and maybe some bands of tablet weaving on top of that, but for the time being, it's finished enough to wear to events as a relatively simple, functional piece of garb.

(Note: The more observant among you might have noticed that my fabric doesn't look like wool; it's a jute/linen herringbone twill. Nearly all of the smokkr fragments (or possible smokkr fragments) that have been found so far have been wool, but I only have a few small pieces of wool in my stash, neither of which was appropriate for this particular project. I still haven't downsized my stash enough to buy more fabric, so I used what I had. It doesn't drape like wool, but it's an appropriate color and weave, and it's machine washable. Similarly, the embroidery is done in your basic DMC cotton thread, which is not in any way historically accurate, but it's what I already had in my stash.)