Diamond twill is an ancient weave which has been identified in textiles dating back thousands of years. I think it is a beautiful, symmetrical and versatile pattern that combines well with borders in a point twill, or with variations of beautiful ‘bird’s eye’ designs. (It also works quite well with a stripey warp, as you can see below!)
But the surprising reality is that during the Celtic Iron Age this weave was almost unknown. Lise Bender Jorgensen, in her survey of weave patterns for this period in Europe, describes the diamond twill as “exceedingly rare”. In pre-Roman sites across Northern Europe and Scandinavia, where textiles are found at all, about 10% of them are broken diamond twill – almost the same weave, but with an offset of one thread in each repeat. (Tabby is still the most common weave, followed by diagonal twill, which together account for 2/3 of all pre-Roman textile finds.)
My first thought was … why? Why would this broken diamond twill be so much more popular than a straight symmetrical diamond twill?? On my trusty Louet Spring, I have recently woven both patterns, and the straight diamond twill is certainly a more natural rhythm, and simpler to count out.
During the period of Roman occupation the picture was completely different. In fact, the opposite was true, with fragments of broken diamond twill virtually unknown in the excavation of Roman forts, while evidence of diamond twill is common. Vindolanda, in northern England is great example of this. It was built by the Romans in the first century AD as a frontier fort on the northern edge of their territory. Early levels of occupation have been uncovered which pre-date the construction of Hadrian’s Wall nearby, and it is in these early levels that more than 600 fragments of textile were found. 60% of them were diamond twills.
(Thank you Medieval Textiles Study Group! http://www.medievaltextiles.org/)
My search for an answer was complicated by the terminology of different writers. Some ignore the distinction between the two weaves altogether while others use ‘lozenge twill’ as a synonym. Bender Jorgensen explains that since the true diamond twill is almost non-existent in the Celtic period, she will use the term ‘diamond twill’ to denote broken diamond twill and ‘lozenge twill’ to mean diamond twill …
In the end, an explanation finally emerged from a re-reading of Marta Hoffmann. And of course it leads back to the warp-weighted loom, which was the almost universal weaving technology across Northern Europe and Scandinavia from as early as the 8th BC. In her ground-breaking 1964 book “The Warp-Weighted Loom”, Hoffmann describes the mechanics of weaving on a weighted warp, and explores this very question of diamonds and broken diamonds. She reports that others experimented with weaving these patterns directly on a loom, and discovered that with the string heddles tied as they are onto the hanging warp threads, it is not possible to have a weft float go over more than two warp threads. So that would make the true diamond twill impossible. The Romans introduced the vertical two-bar loom, with the warp anchored at both ends, so this freed them from the limitations of a weighted warp, and meant that it was possible for them to have the longer floats. Hence their symmetrical diamonds.
So that seems to be the answer!