On the Varying Degrees of Accuracy

I have this argument with other folks about the degrees of "accuracy" and "authenticity" in historic clothing reproductions.  It happens quite frequently, so I want to capture my way of thinking of it.
There is a broad continuum of how period or how authentic a repro is, and you could go on and on down the rabbit hole in the argument, to the point where you'd be trying to make a garment that was molecularly identical to the original garment at the time it was constructed.  You can roughly group the aspects of this continuum into four or five degrees. 


First, you can make something that invokes the cultural idea of a costume.  This is where the dreaded bag costumes of places like the Halloween Superstore and Leg Avenue live, and this is how the whole idea of a French Maid went from this:

Courtesy of Project Gutenberg

To this:

Seriously, you don't wanna see the site this came from.

At the second degree, you can make something that's shaped vaguely correctly, but that might actually use modern cuts, textiles, and techniques to get the job done. People who are not engaged in historic pursuits usually chill here, because there's no reason to go further.  Witness The Infamous Butterick Bustle:

From Butterick, but I believe it's out of print.

Third degree is using similar materials, or materials that can be passed off as period (no polyester Renaissance costumes, for example). That gets you a lot further, from this:

"Queen of Scots" dress by Museum Replicas

 To something more like this:

A fabulous gown from Festive Attyre.

 Fourth, you get into using only authentic patterns and techniques from the period in question. This also means sorting out which embellishments belong in which eras - like giant leg o mutton sleeves being period for 1980s, 1890s (EDIT: wow that was an embarrassing typo), 1830s, and (arguably) 1660s. This means not putting pleated lace fabric on a Renaissance gown, and not putting tiered circle skirts in your 1940s outfits.
I also fudge this stage to include patterns redrawn from period sources, though the veracity of these reproductions varies.  However, consider that in some patterns the popular and fashionable shape at the time of the redrawing influences the garment shape, just as much as if not more so than the use of modern construction techniques.  Witness the lovely 1940's line on our dearest Scarlett O'Hara

And the hilariously dropped-waist of this 1920's take on an 18th century costume:

Most people hang out in the early stages of the third degree, though you could also argue that the third and fourth are parallel and can be taken separately - only together lending a real period look.  However, it's in the third and fourth that things can also go a bit crazy.

Hear me out: if they used cotton fabric, how do we know that the cotton varieties are the same now as then?  Is the fiber the same length? How does the difference in hand or mechanical processing affect the output? What about local changes in soil quality or global changes in climate? What types of dyes did they use then? How were these dyes extracted? Does local dye production differ from imported dyes? If the pattern of the textile was printed, was it done by carved block, etched plate, or roller, and how does this differ from the modern printing processes used today?  And then how are the fabrics finished? Is the finish on an 18th century chintz cotton really anything like the glop on modern drapery chintz? How does this change how the fabric drapes, wears, looks, and feels against the skin?

This ends up being a "you can never step in the same river twice" argument.

And *then* you have the modes of production argument - the techniques of hand sewing vary with things like the size and construction of the needle, the twist and material of the thread, and the availability of both. Similarly, if you're sewing something on a home sewing machine from 1950, the techniques you're going to use vary wildly from those of ladies working in factories in 1870, and as a thorough archaeologist it behooves you to at least touch on how those things affect production. What do you do when your machine doesn't have a reverse stitch?  What about when there's no buttonholer?  No zigzag?  And conversely, how does it affect production when you have another lady two rows of machines back from you who spends all day quickly box pleating trim onto the hems of garments you're working on?  At least with these material culture questions you could *possibly* reproduce some factory setting, though you could also go down the path of asking how the worker's wages, food source, and education (or complete lack of all three things) affected the steadiness and quality of the work.  Because how do you think the garment produced here:

Via the Tenement Museum

varies from the ones produced here?

Via Christchurch City Libraries

 Or here?

Via the Smithsonian Museum of American History

 Or while we're at it, here?

Via the Santiago Service Learning blog

 Ironically, this post began as me trying to explain why I also enjoy taking apart vintage sewing machines (more on that in another post, later), and blossomed into the manifesto you now see.  In general, I strive for the best I can, but as my late friend Kayta used to say, I'm not about to raise my own sheep to make sure my wool looks right.  Furthermore, different costumed occasions call for different degrees of accuracy.  Out drinking on Halloween? Most people go with the first degree.  Producing a play or an opera that will be seen, at best, from 30 feet away? Second degree works great.  But if you're doing up close and personal costumed reenactment, in a venue where your goal is to teach as accurately as you can, you should probably invest a bit more time and research into the latter two degrees - to a point.