on the fringes

My first attempt at replicating a khipu. I've been working with these artifacts for a couple of years now, and the more I analyze their data the more I am convinced that to truly understand them, I have to replicate some. This tiny example is done with commercial cotton yarn; I haven't yet gotten to the point of spinning my own fine cotton singles, though I may someday. Right now I'm curious about several things:

--what is the most efficient way to make a short cord? By looking at the ends of complete cords, we can tell that they were not cut off of some long piece, but rather each cord was created separately from a bundle of looped singles. The singles were twisted tightly then allowed to ply back on themselves. Not so hard if the cord is arm length or shorter, but what's the easiest was to make a cord of say a couple meters?

--The twist angle on existing artifacts is about 45 degrees. Really. Do you know how tightly you have to twist the singles bundle in order to get it to ply back to 45 deg? More than a lot. A humongous amount. I twisted those experimental cords until you could have opened a bottle of wine with them, but the finished cords still aren't steep enough.

--Some cords change color half-way down the string. Not by painting or dying, but through the cord formation process. How would you do that?

-- How much time does it take to make one of these things? I'm frequently asked that question by people who have just been introduced to khipu. The answer is-- I don't know. I don't think anyone knows. From this tiny experiment my guess is that they had some very efficient working methods, but I also believe that speed was less important than doing it correctly. Cords can get very complex and sloppiness is one thing you don't tend to see in these objects. Btw most khipu have around 30 - 40 dangly cords; some have hundreds.

In general I find that trying to replicate something gives me a new understanding of its construction, even if my replication is unsuccessful. Textiles often suffer from a host of assumptions, among which is the presumption that "it's cloth or string, it can't have taken too much intelligence to do". I am convinced that textiles are vessels of some of the most sophisticated thought to come out of human culture.

Comments

[claudia]

Well, let's see.

It would seem to me that an easy way to make a longer cord is with two people...one on each end twisting?

The changing color baffles me...unless the new color is a different strand twisted in.

How interesting.

[Ian McKinnon]

As an archeology student in Andean studies, I have fun making khipus. So I have a pointer to one of your questions, and if you want more email me at iemckinnon@removethis.ucdavis.edu.
First, khipu strands are 2-ply on the last twisting and not just twisted together. The best way I have found to do this is on a drop spindle where you tie a loose half-hitch in the lead and pass one of the two strands through the resulting loop and then tie that to the lead below the loop with another half-hitch. then the second strand to be plied is passed through the same loop as the first, but not tied off to anything. then tighten the loop. When you drop the spindle and spin it the second string is wound tighter against the first and you can get the nice tight windings you see in khipu leads.