Hosen in a late 12th century Western European style

My objective for this project is to create a set of hosen based on known evidence for their pattern, execution, and overall appearance. For references, I have used extant garments and period illuminations from as close as my target year as available, 1190CE. I used this information to create a pattern that represents my best guess for a typical pattern of the time and choose a wool which adheres to the appearance which may have been used. Once executed, the hosen were constructed using stitches found in period with linen thread.

Unfortunately, the state of preservation of most textiles of the 12th century and early 13th century is poor, the number available is low, the distribution is less than ideal, and there is little information about the garments that actually survive. Garments I am aware still exist, but quite often the available information for these is very basic. This usually includes one or two low quality pictures and a brief description, rather than a detailed page long analysis. I have included the limited analysis I am able to make with the small amount of information available.

Hosen can be easily divided into two types. The style I will refer to as Type 1(Figure 3) covers the entire leg from foot or ankle to hip, is attached to a supporting garment at the waist, and is depicted in conforming to the body of the wearer. Type 1 is often made of wool and cut on the bias to allow for greater elasticity. The style I will refer to as Type 2 (Figure 2) comes up to just above or below the knee, are usually made of silk, and do not appear to be well fitted to the body. These are also sometimes known as buskins. All of the surviving examples of Type 2 hosen are either ecclesiastic garments or royal regalia. Because of the exceptionally fine and prestigious nature of Type 2 hosen, there are several examples from around the time I am looking at which survive in excellent condition. This project is meant to create examples of Type 1 hosen. Despite the differences in fit, materials, and length between Type 2 and Type 1 hosen, Type 2’s can still inform us of some decisions historical individuals made when faced with the same problems as appear in Type 1 hosen.

The Bocksten Bog man excavated in Sweden is one of few well documented examples of Type 1 hosen. These hose were made from a wool twill and cut on the bias. The hosen each have two selvage edges, the top left and bottom right edge of each leg. This is indicated by bold lines in Figure 3. The date of the find is contested, with the conservation process and state of preservation from the bog potentially making radiocarbon dating of 1290-1430 AD erroneous. The pattern of the hose is angular in design, with straight cuts from thigh to ankle rather than shaped to fit the leg. The top edges of the garment terminate in a right angle at the front of the hip. The leg piece has a triangular portion which extends over the upper foot from the instep portion. The lower foot consists of one piece which is semicircular in form and the back of the foot and heel are formed from the leg piece. The heel is formed from two oblique angles.(Figure 3) This design and that of other garments bears a closer resemblance to examples which are dated toward the early end of the 1290-1430 range, rather than later examples such at the London hose.(Crawfoot p189) The Bocksten man hose also have an assembly of cords which partially wrap around the upper leg, seemingly to add tension to the material and make it fit closer to the body. There are several possible reconstructions of this system and I have decided to not delve into that subject. The suspension system of this example is unclear. There is no indication of eyelets or laces attached to the top of the garment.

The Bremen No.19 hose are Type 1 hosen dated to the 1200s and were excavated from a grave in Germany. The foot assembly is patterned similar to the Bocksten find and was also made of wool. Bremen No.19 also has a triangular piece which extends from the leg piece over the upper foot, has a heel formed by oblique angles, and a semicircular lower foot. The total height of the hose were approximately 178cm from the instep to the top point of the garment. These hose include some piecing at the top of the garment, possibly to add greater height because of a restriction of loom width, like the Bocksten Man examples. Unfortunately I have not been able to determine if any of the edges contain a selvage, which would support that idea. The suspension system of this example is also unclear, with a lack of eyelets, buttons, or attached laces as with the Bocksten Man finds.

Another Type 1 pair of hosen are attributed to Saint Francis d’Assise from the beginning of the 13th century and reside in the convent of Sainte-Claire, Assise.(Figure 5) They are made of linen and cut on the bias. They closely resemble the other Type 1 hosen with the right angle top of the garment and an ankle pattern similar to the Bocksten and Bremen examples, except without the triangular piece extending from the leg piece over the top of the foot. We can see the seam between the foot and leg cutting across the width of the foot before going down. The foot piece is also fairly organic in shape. A godet was added to the top of the garment, but it is unclear what the shape is when only observing the garment from one side. Curiously, this garment has fairly wide woven ties sewn to the top, but not at the apex of the point. (Anderlini p140)

The hose of emperor Heinrich III is from 1053 or earlier.(Figure 6) They are Type 1 hosen, but made of silk. It is unclear from photographs and information I have found if it is a brocade or not, but based on the station of the individual they are attributed to, I imagine they would be. At the junction of the leg and foot pieces, the leg piece continues straight down at the instep instead of cutting across the width of the foot. The entire foot piece is much more angular in design than the Francis hose. Rather than cutting the leg portion on the bias or with flaring sides, a godet was added at the back of the leg, providing the necessary width to fit the thigh. There is an associated band with cords attached which is not described. These cords seem too thin to be cut from strips of cloth, but rather a sort of braid or leather thongs. It seems likely that this is some sort of belt or waistband to suspend the garment, or the waistband of the undergarment itself and the actual undergarment has decomposed. There is no obvious attachment method on the actual hose.

The hose of Robert I de Courtenay is another excellent example of a Type 1 hose, this time from the late thirteenth century.(Figure 7) I have not been able to find dimensions of the garment, but we can find the approximate fit by using body proportions. The distance between the instep and the top of the garment is about two and a half foot lengths, placing the top of the garment at about mid-thigh. I have not found a description of the fiber content, but it is clearly a brocade. The cloth appears to be cut on the bias in the available closeup. Fiber possibilities include silk or fine wool. The pattern seems to follow the Bocksten and Bremen examples with a triangular portion of the leg piece extending over the top of the foot. I can discern no information about the suspension system, but the garments are held onto their form with garters both below the knee and at the ankle. It is unclear if these are contemporary to the rest of the item.

Rodrigo Zimenez was another individual buried with their hose, this time in 1247 in Spain when he was archbishop of Tolède. His Type 1 hose were made of fine wool cut on the bias. The back of the upper hose have considerable piecing, seemingly to bring the top of the garment to be horizontal, instead of the typical angled Type 1 profile. A leather band is associated with the garment, similar to the hose of Heinrich III. This band has cords which are attached to the hosen near the top of the front by means of an eyelet which has been reinforced by a piece of parchment. It is unclear if the current arrangement of cords is contemporary to the garment.(Anderlini p140)

Examples of Type 1 hosen were also found in Greenland.(Figure 9) Norland No. 88 was sewn from a 2/2 twill which was almost black except for some white warp threads. The hosen were cut so that one of the top edges was a selvage. The total height of the garment was 75cm from instep to top, or over three foot lengths. This puts the height of the garment at around the wearer’s hip. Rather than a single semi-circular foot piece, the foot of Norland No.88 is made up of six pieces. Fransen et al in Medieval Garments Reconstructed believed this to be a series of repairs, and I tend to agree. There is no apparent eyelet, button, or cords attached to the garment to facilitate suspension to the supporting garment.

The stocking of Hubert Walters (Figure 2), a late 12th century Type 2 hose, follows the same foot cutting pattern as the Francis example. It is made of a silk brocade with precious metals woven into the cloth. The distance between the instep and the top of the garment is approximately two foot lengths. This equates to a garment which stops just above the knee. There is a garter attached to the top point of the garment, allowing the wearer to secure the hose in place without using any supporting garments. These cords have a zigzag pattern on them, making me believe that they were woven specifically for the purpose.

The hose of St. Desiderius is another twelfth century Type 2 hose.(Figure 10) It is made of a silk brocade and has a series of horizontal red stripes or ribbons applied to the garment. It likewise does not come with adequate dimensions, but by proportions the leg is one and a half foot lengths high. This places the top of the garment just below the knee. The pattern appears to be similar to the Heinrich hosen, with the same angular foot piece. Since this garment seems to not go over the knee, it makes sense that it would not need to flare as much. This garment is also different in that the top is straight across rather than angled forward, and the attached garter is positioned on the side rather than the front.

The stocking of William II who was King of Sicily until his death in 1189 also survives.(Figure 11) It is a Type 2 hose which appears to be made of a brocaded silk. The height is about 1.8 foot lengths, which curiously puts the top of the garment at approximately knee height. A garter of cloth strips seem to be attached to the garment a few inches below the top, below a band circling the leg. These ties are about one and a half foot lengths above the instep, placing them below the knee. If that is the case, it is possible the embellished upper portion of the hose simply sat over the knee while being held up by the garter or was turned down over the garter and lower leg. The foot pattern appears to be consistent with the Walters hose, cutting across the foot width before going down and having a more rounded foot ball area.

A further example of Type 2 hosen is those of Clement II, who was pope until his death in 1047.(Figure 12) These hosen are made of a brocade and do not appear to have foot pieces divided from the leg piece. There may be some imperceptible seam, but the pattern of the brocade seems to match up. The top of the hosen are closed by ties or garters and they are just short of two foot lengths high, likely above-knee in height.

Hosen are visible in numerous illuminations, and are typically shown
fitted to the leg, especially to the calf. In the Psalter of Eleanor of Aquitaine
ca. 1185 on the December page,(Figure 14) an individual is shown picking grapes with his tunic pulled up over his belt and hosen showing. The image depicts the height of the hose to be mid-thigh to waist high at the front, descending from this point at an angle to the back of the leg. This is consistent with the extant examples of Type 1 hosen. In every depiction of hosen where the height of clearly shown, this mid to upper thigh is depicted without exception. Figures 12 and 13 display the range of hosen heights in artwork. Individuals who were likely wearing Type 2 hosen, Royalty and high ranking ecclesiastic figures, are only shown with their feet, toes, or nothing poking out from the bottom hem of their garments.

The type of fabric used is a subject that has sources in both the artistic depictions and extant examples. Of the garments listed above, five (possibly six) of the ten total garments have a brocade pattern woven into the material. Of the remaining examples, St. Francis is famous for adopting a life of poverty and naturally would not have chosen decorated clothing. If the Bocksten and Bremen hosen were ever woven with a pattern it has been degraded by their method of preservation. Patterns may be woven into cloth without even the expense of dying the material, such as by using different colors for the warp and weft for twills, checks, or stripes, so all but the poorest of persons may have been able to enhance the visual interest of their clothing.(Ostergard p72) This practice of hosen being made from cloth with simple weaving patterns is shown in a minority of artwork.

As we have seen in the extant examples earlier, the patterns woven into the hosen of the wealthy and very wealthy could be exceptionally ornate. For example the central individual in Figure 15.5 is King Henry II of England. The hosen he is wearing here has a very simple pattern of broad diagonal stripes. If this was done with real cloth on a loom, it would either be horizontal or vertical stripes that are turned to an angle when the pattern is cut on the bias. As we know from the extant examples, a king would be able to afford much more than this. We can then surmise that the patterns drawn by artists in period artwork do not necessarily equate to the patterns on the clothing they were depicting. The point of this artwork is to communicate the idea of the scene, not to show people how they were in life. There were and are restrictions on the medium, and the closest thing to a faithfully depicting a brocade is in Figure 15.11. All others are simplified designs which fit on the page and add some visual interest. I think that the patterns we see here likely were put on cloth and made into hosen, but in the context of this artwork, some individuals are downgraded to simpler designs that may have been worn by lower classes.

A further detail that I feel is important to the overall look of the hosen is garters. All of the surviving Type 2 hosen have garters attached to them, or used to. The hose of Robert I de Courtenay likewise has associated garters both above the calf and at the ankle. Type 1 hosen in all available artwork show the ideal to be as fitted a possible. A garter at the knee does a fantastic job of establishing this profile and a garter at the ankle would certainly help for an individual who was wearing low shoes. This is a detail that seldom shows up in artwork however. In fact the only times I have found definite depictions of them in an early thirteenth century context they are worn over mail chausses.(Figures 16,17) As civilian wear and armor tended to reflect one another, I feel that this lends some support to the wear of garters over woolen hosen, even though I have not found explicit examples a couple of decades earlier. I have decided that if garters were worn, no attention was really drawn to them, they would blend in or be utilitarian rather than ornamental. If they were seen as an opportunity to show off, I would expect to see them emphasized in artwork.

When it came to selecting the material I would use for this project, I looked at available period artwork for an example of a woven pattern that would be relatively simple to weave but still would imply the wearer had some financial means. I also needed something I could find being commercially woven. The image I arrived at was Figure 1. This is a blue-grey ground with a red windowpane check being worn by an individual who appears to be a foot sergeant. This image is rather early for me, dated to 1136 in Fondi, Italy, but on inspecting the fashion of the period I decided that there was no major change in the garments of lower classes over that span of time. The weaves of Type 1 wool examples when I have information is 2/2 twill. Furthermore, the majority of wool found in areas with high enough concentrations to analyze statistics are 2/2 twill.(Ostargard p62) This is concentration happens to be in Greenland in a very broad time frame. This is not ideal, but in lieu of better information, I must defer to the Greenlandic source.

With this description in hand, I hunted through numerous catalogs of wools and managed to find fabric which matched the depiction I described. It is a medium-light weight 2/2 twill with a blue-grey ground of a color that could be achieved by overdying a grey wool with indigo or woad and a red windowpane check of a hue achievable with madder.(Dean 120)(Figure 20)

The pattern I arrived at is primarily inspired by the Bremen, Bocksten, Norlund, and Robert examples. These are all constructed of wool and I have relatively good information on them. I patterned the hosen on the bias like all of the extant Type 1 examples except for Heinrich III’s silk pair. I used a semicircular foot shape and the triangular leg portion which extends over the foot from the instep. This forms a stronger interface with the foot piece in a high strain area. I patterned the garment to extend up to my hip and did not create an artificial need for piecing. The profile of the garment was intended to have the angular top profile I see in artwork of my late 12th century Italian context rather than the occasionally seen flat top.

The length of the foot piece my foot measurement from the side of the foot where the foot and leg pieces meet to my big toe. The width of the foot piece is simply the circumference of my foot. The width of the leg piece at the ankle can be no smaller than the measurement from the heel around the front of the instep, since this is the widest part of the foot. The length of the leg piece is the measurement from my hip to the instep, plus the length of the triangular foot interfacing. All of these measurements were given an extra 7mm for each seam.

Almost all threads in extant articles from Medieval Europe have disintegrated, only leaving the holes and impressions where the thread passed on some textiles. The information about extant hosen above is the extent of what I have, there seems to be no detailed analysis of stitches and seams used, so I must look elsewhere. For this project I chose seam techniques that result in the thinnest seam possible in the interest of comfort. A number of textiles from the London finds (Crowfoot 153-157) have this evidence, which indicate that a majority of seams were held together by a running stitch or backstitch. In one technique the two edges were then folded outwards and stitched down using running or hem stitches (Figure 19 A). There are a number of hose foot fragments in this source dated to the 14th century that have a stitch which is exclusively used on hose. This is a method where the two sides of the seam are overlapped and edges stitched down, meaning the finished seam is only two fabric thicknesses.(Figure 19 B) The usual overlap of these seams is 4-7mm with stitches that are approximately 3mm long. In order to finish the hems on wool, a single fold seems to have been considered adequate, this being secured with a hem stitch and stab stitch in high wear areas. The stab stitch is similar to a running stitch, except that the needle enters the folded cloth angled backwards, resulting in stitches that are very tight together and provide significant strength and abrasion resistance.(Figure 19 C) I use an unbleached 16/2 linen thread for this project. My reasoning for not matching threads with the material is that the size of stitches is very small, leaving very little visible on the outside, linen does not remain colorfast with natural dyes well, and it would be an expense I do not think is reasonable for my chosen context.

For a fastening technique, I decided to use an eyelet at the apex of each hose reinforced with a small piece of thin rawhide as is described in the Rodrigo hose. Other solutions include cords attached to the top of the garment as on the Francis hose, or the use of a temporary button. This would be done by placing a small object on the back of the hose and tying the suspension cords around the object over the wool. This would not leave a lasting sign, as we see in several of the Type 1 hosen.

My final decision was the use of a garter. After some deliberation and considering of the garters on surviving Type 2 hosen, I decided to use grain cut strips of fulled wool which was wastage of a different project. This is consistent with the woven cloth strips in several Type 2 examples, would not impart any significant cost, and the dark blue pieces blend in with the hosen.

I avoided the selvages of the material as most modern wool has some strands of synthetic fiber at the edges to help automated looms grip the wool. The specific selvage pattern used by automated looms is also not seen in period textiles.(Crowfoot 46) Once cut out, the bottom of the foot and back of the leg should be completed first, creating a stable base for pinning and stitching the foot on.(Figure 21-22) This was done using a running stitch approximately 3-4mm in length with each edge folded outwards 7mm. (Figure 19 A)

The foot was then attached using the stitch found on fragmentary hosen out of London.(Figure 19 B)

When first assembled, the toe had about half an inch of excess material, so I trimmed it back and reassembled the foot. Before and after trimming can be seen below. I adjusted the second hose before assembling it.

With the foot together, the biggest part remaining was the top hem. As I had trimmed the modern selvage away, I had two edges to hem and reinforce with a stab stitch.

The description of Rodrigo’s hosen only states “Les cordons qui les relient au braidel de cuir passent par des œillet renforcés au revers par du parchemin.”(Anderlini p140) Or “The cords linking them to the leather band are going through eyelets that have been reinforced with parchment.” There is no description of the shape or attachment method of the reinforcement. I therefore guessed the use of a square piece of rawhide backing up the wool and attached with the same button hole stitch that creates the eyelet.

This is the product I have arrived at. I am pleased with the execution, materials used, and believe I have given reasonable support for decisions I have made. My interpretation of history may change as I discover more information about extant garments, but I do not expect the main points of this garment to change very much.

Bibliography

Anderlini, Tina, and Beryl-Alexandra Brard. Le Costume médiéval Au XIIIe siècle: (1180-1320). Heimdal, 2014.

Carlson, I. Marc. “Some Clothing of the Middle Ages.” Some Clothing of the Middle Ages, 1999, http://www.personal.utulsa.edu/~marc-carlson/cloth/sites.html.

Crowfoot, Elisabeth, et al. Textiles and Clothing C.1150-C.1450. Boydell Press, 2012.

Dean, Jenny. A Heritage of Colour: Natural Dyes Past and Present. Search Press, 2014.

Fransen, Lilli, et al. Medieval Garments Reconstructed: Norse Clothing Patterns. Aarhus University Press, 2011.

Høibovi, Martin. “European Medieval.” Extant Originals – European Medieval, Part 1, 2006, http://www.kostym.cz/Anglicky/I_01_01.htm.

“Mitre, Buskins, Slippers, Stole and Crozier of Archbishop Hubert Walter.” The British Library, The British Library, 10 Oct. 2014, http://www.bl.uk/collection-items/mitre-buskins-slippers-stole-and-crozier-of-archbishop-hubert-walter.

Østergård Else. Woven into the Earth: Textiles from Norse Greenland. Aarhus University Press, 2009.