Methods of Medieval Maille Weaving: Speed Weaving and Patchwork Construction
If you’ve been around the maille scene long enough to have learned how to weave, you’ve been around long enough to see people claiming that the best method is to make a patch of maille and then to make another patch of maille and then to connect those two patches, &c, &c. until you have a completed garment. You may also have run across people who claim that “speed weaving” is the best method for maille construction. Speed weaving is essentially finishing two rows at once by passing one open ring through four closed rings at once rather than passing one open ring through two closed rings.
Regarding speed weaving, this is certainly the method used throughout much of the Medieval period. However, speed weaving is not always appropriate. Regarding constructing maille in patches, I believe that such constructing fails to recognize the need for tailoring a shirt and the practicalities of doing so. However, I do think that making segments of a shirt, and then adding those segments to other completed segments may have been possible. In other words, one worker would construct the lower portion of the shirt up to the waist, while another would make the top half. Although I am not aware of any direct proof of this method, a shirt from the late fourteenth century offers enticing evidence that this may have occurred.
The method of construction up until the end of the fourteenth century and the beginning of the fifteenth century was to create a shirt with alternating rows of riveted and solid rings. [1] After this period, a number of technological changes in weaponry, armor, and armor production led to maille with only riveted rings. First, swords up to this point were typically meant for hacking and slashing. Their tips were relatively rounded, and not suited for piercing. By the end of the fourteenth century, this had changed and swords became much more powerful at thrusting. Second, armor developed partially in response to this change, but also because armor was becoming much more focused on plate protection. Maille took a secondary role of protecting exposed areas not fully covered by plate armor. Third, the reason for alternating rows of riveted and solid rings was pragmatic. The most labor intensive part of the process is riveting the rings. Speed of production could be increased by alternating. However, having alternating rings meant passing the riveted ring through a top row of whole rings and a bottom row of whole rings. Of necessity, the maille maker was completing two rows at once. As a result of this truth, rings could only be so small before the ability to rivet a ring would be impeded by the two rows of rings. [2] Therefore, a shirt could only become so dense employing this method. However, throughout the Middle Ages, it was easier to create a strip of iron and then stamp it than to draw iron wire and rivet it. Therefore, an all riveted shirt was too impractical. This had changed by the end of the fourteenth century and maille started to become much more dense both because it needed to protect from thrusts and because extra density in smaller patches of maille was not so labor intensive as making an entire ultra-dense hauberk.
Regarding speed weaving, this is certainly the method used throughout much of the Medieval period. However, speed weaving is not always appropriate. Regarding constructing maille in patches, I believe that such constructing fails to recognize the need for tailoring a shirt and the practicalities of doing so. However, I do think that making segments of a shirt, and then adding those segments to other completed segments may have been possible. In other words, one worker would construct the lower portion of the shirt up to the waist, while another would make the top half. Although I am not aware of any direct proof of this method, a shirt from the late fourteenth century offers enticing evidence that this may have occurred.
The method of construction up until the end of the fourteenth century and the beginning of the fifteenth century was to create a shirt with alternating rows of riveted and solid rings. [1] After this period, a number of technological changes in weaponry, armor, and armor production led to maille with only riveted rings. First, swords up to this point were typically meant for hacking and slashing. Their tips were relatively rounded, and not suited for piercing. By the end of the fourteenth century, this had changed and swords became much more powerful at thrusting. Second, armor developed partially in response to this change, but also because armor was becoming much more focused on plate protection. Maille took a secondary role of protecting exposed areas not fully covered by plate armor. Third, the reason for alternating rows of riveted and solid rings was pragmatic. The most labor intensive part of the process is riveting the rings. Speed of production could be increased by alternating. However, having alternating rings meant passing the riveted ring through a top row of whole rings and a bottom row of whole rings. Of necessity, the maille maker was completing two rows at once. As a result of this truth, rings could only be so small before the ability to rivet a ring would be impeded by the two rows of rings. [2] Therefore, a shirt could only become so dense employing this method. However, throughout the Middle Ages, it was easier to create a strip of iron and then stamp it than to draw iron wire and rivet it. Therefore, an all riveted shirt was too impractical. This had changed by the end of the fourteenth century and maille started to become much more dense both because it needed to protect from thrusts and because extra density in smaller patches of maille was not so labor intensive as making an entire ultra-dense hauberk.
The result of this increase in density led to a change in how shirts were manufactured. Instead of passing an open ring through four closed rings, essentially completing two rows at once (what many modern maillers call speed weaving), the shirt needed to be made one row at a time. This was essential to allow riveting the ring in a very dense weave. With alternating rows of riveted and solid rings, speed weaving was a necessity of construction. As rings became denser, speed weaving would not be possible, and only one row was added at a time.
The biggest drawback of adding two rows at a time is that it makes tailoring the shirt more difficult – especially if there are rows of solid links to take into account. Keeping the rows even across the sleeves, completing the armpits, and ensuring that expansions and contractions are evenly spaced becomes that much more difficult because the maille makers needs to account for two rows at a time. This difficulty is made plain from the study of a late fourteenth century shirt. For example, the shirt has irregular hip expansions where the expansions are added and then almost immediately removed. [3] One theory for this mistake is that the shirt was started at the bottom and done in reverse.[4] I have read of modern maille makers employing a similar technique of starting at the hip or torso in order to have a good area to work off of before tackling the more difficult neck, shoulder, and underarm portions. Another thought is that the maker merely added too many and thought better of it, but instead of tearing apart hours of work, just “fixed” the problem a few rows later. I tend to think that this makes more sense, especially since only the keenest eye would ever notice.
Another possibility is that the workshop divided the labor of making a shirt. Think about the practicality of that for a second. If only one worker ever touches a shirt, then the shirt can only be finished in the time it takes for that worker to make it. With two workers, you double the production time of that specific shirt. True, the end efficiency is still the same, but when a product takes hundreds of man hours to finish, it might make more sense to finish one more quickly than to finish two in a longer period.
Also, this theory allows for division of labor according to ability. Making the top portion with the underarm, shoulders, neck, and sleeves, would be decidedly more advanced than making the bottom portion with contractions at the waist and expansions at the hip. All the two workers need do is ensure that they both end at the same size row and attach it later.
This theory is further bolstered by another mistake in the same shirt. The sleeves are of differing length. The left sleeve is 61 rings long and the right is 62 rings long. [5] The sleeves on the shirt have smaller rings at the bottom of the sleeve and are more densely woven. The smaller rings are even at 32 long on each sleeve. [6] At the meeting of the lower portion and upper portion of the right sleeve, the shirt has two rows of riveted rings. This demonstrates that the lower half of the shirt was likely finished and added as a whole piece. [7]
This again supports the idea that there was a division of labor within a shop that handed off certain pieces of armor to be made by different workers. Again, this makes sense from a perspective of trying to get a particular piece done quickly. If a shirt took 500 man hours to assemble, having two or three people work on different parts and then later attach them would drastically cut down production time. For example, assume a ten hour workday six days a week. That amounts to a 60 hour workweek. Assuming a shirt takes 500 hours would mean that worker would take over eight weeks to finish. If you assume that two or three workers collaborated on a shirt, you can cut that time by half or more.
Unfortunately, this method of construction also allows for more mistakes. It is harder to coordinate between multiple people on exactly how to finish all of these pieces. Sometimes mistakes were made. However, the good seems to have outweighed the bad, at least with regard to this particular shirt. It seems as though the method of construction would have allowed for speedier completion and would also likely have allowed less experienced workers to gain some basic maille making skills on easier parts of the shirt.
From the above discussion of medieval methods of weaving maille, the idea that there is any one right way to make maille seems foolish. The best method for making maille depends on the type of maille you are making, the period in which the maille was made (or seeks to emulate), and the nature of the shop or people making the suit. Ultimately, I do not believe that multiple patches were made and later assembled, however, it seems as though segments of the shirt were constructed and later assembled. Finally, speed weaving, or assembling two rows at once, is necessary when using alternating rings. When making a dense weave with thick rings, this method is not possible.
The biggest drawback of adding two rows at a time is that it makes tailoring the shirt more difficult – especially if there are rows of solid links to take into account. Keeping the rows even across the sleeves, completing the armpits, and ensuring that expansions and contractions are evenly spaced becomes that much more difficult because the maille makers needs to account for two rows at a time. This difficulty is made plain from the study of a late fourteenth century shirt. For example, the shirt has irregular hip expansions where the expansions are added and then almost immediately removed. [3] One theory for this mistake is that the shirt was started at the bottom and done in reverse.[4] I have read of modern maille makers employing a similar technique of starting at the hip or torso in order to have a good area to work off of before tackling the more difficult neck, shoulder, and underarm portions. Another thought is that the maker merely added too many and thought better of it, but instead of tearing apart hours of work, just “fixed” the problem a few rows later. I tend to think that this makes more sense, especially since only the keenest eye would ever notice.
Another possibility is that the workshop divided the labor of making a shirt. Think about the practicality of that for a second. If only one worker ever touches a shirt, then the shirt can only be finished in the time it takes for that worker to make it. With two workers, you double the production time of that specific shirt. True, the end efficiency is still the same, but when a product takes hundreds of man hours to finish, it might make more sense to finish one more quickly than to finish two in a longer period.
Also, this theory allows for division of labor according to ability. Making the top portion with the underarm, shoulders, neck, and sleeves, would be decidedly more advanced than making the bottom portion with contractions at the waist and expansions at the hip. All the two workers need do is ensure that they both end at the same size row and attach it later.
This theory is further bolstered by another mistake in the same shirt. The sleeves are of differing length. The left sleeve is 61 rings long and the right is 62 rings long. [5] The sleeves on the shirt have smaller rings at the bottom of the sleeve and are more densely woven. The smaller rings are even at 32 long on each sleeve. [6] At the meeting of the lower portion and upper portion of the right sleeve, the shirt has two rows of riveted rings. This demonstrates that the lower half of the shirt was likely finished and added as a whole piece. [7]
This again supports the idea that there was a division of labor within a shop that handed off certain pieces of armor to be made by different workers. Again, this makes sense from a perspective of trying to get a particular piece done quickly. If a shirt took 500 man hours to assemble, having two or three people work on different parts and then later attach them would drastically cut down production time. For example, assume a ten hour workday six days a week. That amounts to a 60 hour workweek. Assuming a shirt takes 500 hours would mean that worker would take over eight weeks to finish. If you assume that two or three workers collaborated on a shirt, you can cut that time by half or more.
Unfortunately, this method of construction also allows for more mistakes. It is harder to coordinate between multiple people on exactly how to finish all of these pieces. Sometimes mistakes were made. However, the good seems to have outweighed the bad, at least with regard to this particular shirt. It seems as though the method of construction would have allowed for speedier completion and would also likely have allowed less experienced workers to gain some basic maille making skills on easier parts of the shirt.
From the above discussion of medieval methods of weaving maille, the idea that there is any one right way to make maille seems foolish. The best method for making maille depends on the type of maille you are making, the period in which the maille was made (or seeks to emulate), and the nature of the shop or people making the suit. Ultimately, I do not believe that multiple patches were made and later assembled, however, it seems as though segments of the shirt were constructed and later assembled. Finally, speed weaving, or assembling two rows at once, is necessary when using alternating rings. When making a dense weave with thick rings, this method is not possible.