Journey to the East: Ürümchi and the Tarim hats

Please note: This is the story of the journey to examine the Tarim beret. The substantive details of the examination will be published in an appropriate venue, but the journey itself was an interesting adventure. It is quite an epic.

This story begins in 1999 in a small town in Western Washington State and it took me across the Pacific Ocean and across much of Asia, alone, to look at an ~3000 year old hat. It was the kind of journey which apparently was just not done and yet; stories are born.

A map of China with lines showing the route of travel.
Solid lines indicate by train, dashed by plane, and dotted by tour bus.

My quest to learn more about nalbinding throughout the world led to me to a then recently published book: The Mummies of Ürümchi by Elizabeth Wayland Barber. In it she describes the beautifully preserved textiles that had been found on and with the mummies of the Tarim Basin and her trip to Ürümchi in 1995 to study them. Quite early on, pgs. 32 & 33, she mentions the man with the ten hats. One of which has since become known as the Tarim hat, or beret. She describes it as having been “made of dark brown wool in a looped technique that at first glance looks like knitting. But knitting, so far as we know, was not invented for another two thousand years. This hat used a needle and thread method known by the Scandinavian name of nalbinding (“needle binding”).

Fig. 2.6 Little line drawings showing the side view of the shape, the top view showing the ribbed pattern in quadrants, and a little diagram showing a round start of five stitches with a partial second row demonstrating two Z-cross-knit variant wales and a two course deep increase.

She included little line drawings showing the side view of the shape, the top view showing the ribbed pattern in quadrants, and a little diagram showing a round start of five stitches with a partial second row demonstrating two Z-cross-knit variant wales and a two course deep increase. The only other nalbound hat mentioned as such, on pg. 60, is from Tomb 4. “The tomb also contained a dark red onion dome hat (like that in fig. 2.7) done in spiral nalbinding …”

Fig. 2.7 Showing 3 panels of the "tent-shaped" or "onion dome" hat and a helmet shaped hat with two felt horns attached.

This of course peaked my curiosity. What did the hat look like? How was it made exactly? That little diagram and sketch were certainly not enough to know how it worked. Fortunately, that year she was out promoting her new book and had a speaking engagement nearby, for relative terms of near involving over a hundred miles. I, unfortunately, could not attend. However, I knew people that could and they were sent with express instructions to report back any information she might provide. During the Q&A session, they did ask her about the nalbound hats, but they reported back that she said unfortunately, while she knew enough about the technique to identify it, she did not know enough about nalbinding to be able to convey the specifics of its construction. And thus the hats remained a mystery.

1999 was an exciting year for me as I will describe in a later post. Suffice it to say, that summer I found myself heading to Taiwan to continue my Chinese Mandarin studies for a year. Much nalbinding research was accomplished (the Academia Sinica is an amazing library). However, the hats of the Tarim basin were still a mystery and as I was so close, and unlikely to be so again, I decided in the summer of 2000 to travel to the Mainland to see them for myself.

A three quarters top view of the beret style hat with the embedded twig to the front right and the damaged section to the back left.

I flew to Hong Kong and crossed my fingers that I would be able to get a visa. My plane was set to return home a month later and I could not afford to stay in Hong Kong for the whole time. I had no set itinerary, just a few places I wanted to explore and the desire to see the Tarim hats in Ürümchi on the far side of China. It was while I was waiting in Hong Kong that I got my first glance at an image of the Tarim beret. J. P. Mallory and Victor Mair’s book, The Tarim Mummies: Ancient China and the Mystery of the Earliest Peoples from the West, had just been published and there, on page 214, was an image of the beret. I could see the quadrants and ribbing. The stitch was not clear in the image, but it did not appear to be the cross-knit nalbinding variant.

I figured I would take the train as I had loved taking the train as a child and it would let me see the scenery. What I had forgotten was that riding in a moving, vibrating, vehicle has put me to sleep since I was a baby. I got more sleep on that vacation that I have before or since given the on average 35 hour stretches between places I went to explore. Turns out the Chinese are very polite and will pretend you do not exist until they find out you can “talk,” by which they mean speak Chinese. Nine months in Taiwan had increased my speaking ability to where I was quite comfortable carrying on a conversation so I had plenty of companionship along the way. The Chinese also do not travel alone. It’s just not done. So once they figured out I could talk, I tended to get adopted.

The train took me up from Hong Kong to Beijing where I spent several lovely days exploring with a Swede I’d met on the train. From Beijing I traveled to Hohhot, the capital of the Inner Mongolia province in China. I took a tour where they did come through on their promise of an English speaking guide, but as I understood it the first time (and better) in the Chinese, I chose not to subject us to a repeat of the information. I got to ride a pony across the grasslands, climb the dunes of the Gobi desert* and ride a bactrian camel. From there I went to Lanzhou and was convinced to visit Xining by someone that saw me spinning on the train. In Xining I got to see a towering pile of raw cashmere several stories high. I was introduced to the various colors and to a fiber I’d not run into before from yaks; which are quite tasty by the way.

From there I headed to Ürümchi. I arrived shortly after 7 in the morning on the 14th of June and after getting lost trying to find a recommended restaurant that was not where it was supposed to be, I made my way to the 新疆自治区博物馆, Xinjiang Uyghur Autonomous Region Museum, generally referred to as the “Xinjiang Museum” for short. By the time I went to see them in 2000, the mummies were in their then new gallery alluded to in Barber’s book. Though it appears that it was shortly after my visit that the old building was torn down and replaced with the modern building in which the Xinjiang Museum is now housed. The mummies were in separate cases in the middle of a large room. There were large display cases in the walls. The beret was the third item from the left in the long wall case. Negotiations, which progressed much more smoothly in Chinese than English, were made for me to return one week later to examine the Tarim beret. I then took the evening train to Kashgar; planning to return Ürümchi on the morning of the 21st.

The front of the 新疆自治区博物馆, Xinjiang Uyghur Autonomous Region Museum. A rectangular building with flying humanoid figures in slightly darker pink on either side. In the middle a 6 columned overhang with the name of the museum above. All topped with a dome and a golden spire.
新疆自治区博物馆, Xinjiang Uyghur Autonomous Region Museum, as it appeared in June 2000.

My mother, meanwhile, was very curious as to if her experiments matched the hat. I wrote a quick email to the nalbinding list responding to her request on June 16th from an internet cafe in Kashgar. I mentioned that I was hot and dusty, but happy and excited. There wasn’t any air conditioning in the hard bed train carriages and the metal absorbed the heat. We opened the windows for air, but that let in the dust of the Tarim basin. I described that while I could not see well enough through the glass to determine the exact stitch, it clearly was not cross-knit nalbinding. “The light was too dim and it was too far from the glass for me to make out clearly what it was, just that it wasn’t any of my theories exactly.”

It had been a long hard train ride back from Kashgar, I had gotten severely dehydrated and slept 17 hours straight. I arrived to discover they had not yet pulled the beret from the display, so I got to watch them collect it. It turns out the only entrance to the case was a door on the far right wall. Remembering that the beret was third in from the left wall, I watched them carefully step over the items on display to collect the hat. Then, because there was not room to turn around while holding the hat, I watched him carefully step backwards over the displayed items all the way back to the far end of the display case.

They then placed it on a table covered with newsprint for me to examine. I counted the stitches in the beginning row, examined the increase and pattern structure, noted the strategy to decrease towards the underside, and marveled at the beautiful braided edge inherent in the felt roll that supports its shape. While the technical details recorded will be published later, the structure was not cross-knit nalbinding as had been implied by Barber’s diagrams and the quartered and striped drawing. Instead, it is made of patterned blanket stitch with judicious use of spacing between stitches to make the striped pattern. The increasing strategy does mean that there are a few single stitches per row that connect to the previous in a manner to make a single wale of cross-knit nalbinding at each quarter’s edge, but this is a secondary effect, not the primary construction structure.

The photo shows Anne Marie Decker with her arms crossed (one white glove showing) and resting on a table covered with Chinese newsprint on which the Tarim hat is placed, right side up with the embedded stick in the left quarter nearest Anne and the damaged section to the right from the viewer's perspective.
One of the staff was kind enough to take my picture with the hat.

More images of this hat, and the others mentioned below, are included in my presentation: Charting the Nalbinding of the Nile.

I also took notes on three other hats, though they were not removed from their cases for examination. The most notable was referred to as the “witch’s hat” given it’s beret-like base and stuffed point displayed on the skull upon which it was found. The base was covered in a fabric made of loop & twist and the point was covered in a fabric made up of loop & twist and blanket stitch bands. I left Ürümchi that same night taking the train to Xi’an. I did not have internet access until the 26th, at which point I sent a preliminary report to the nalbinding email group list.

After visiting the clay warriors, it was time to make the hurried run back to Taiwan. I took a plane to Shenzhen. Then a ferry across to Hong Kong in order to catch my plane. Three currencies in one day as I was running on my last pennies. I spent three days in Taiwan packing up my stuff and then started the journey home.

——–

* The dunes we climbed rose right up off the river’s edge. Our group decided to avoid the crowds, so we were climbing in fresh undisturbed sand that had been baking in the sun all day. I got about two thirds up the dune and couldn’t go any further because my fingers and toes had swollen so much from the heat retained in the top layer of sand. I had to have someone go before me so that I could climb in the slightly cooler layer that their passage stirred up. There is a reason that people wear wool socks in hot sandy areas and it does have to do with insulation.


For more recent information on the Tarim finds
& updates to the Annotated Bibliography:

Claßen-Büttner, Ulrike. Nadelbinden – Was ist denn das? Geschichte und Technik einer fast vergessenen Handarbeit. Norderstedt: Books on Demand GmbH, 2012. ISBN 978-3-8482-0124-2.

Claßen-Büttner, Ulrike. Nalbinding – What in the World Is That? History and Technique of an Almost Forgotten Handicraft. Norderstedt: Books on Demand, 2015. ISBN 978-3-7347-8775-1.

Hopkins, Heather, and Katrin Kania. Ancient Textiles Modern Science II. Havertown: Oxbow Books, Limited, 2018.

Mair, Victor H. “Ancient Mummies of the Tarim Basin” Expedition Magazine 58.2 (2016). Penn Museum, 2016 Web. http://www.penn.museum/sites/expedition/?p=23790 Accessed April 25, 2019.

Nargi, Lela. Knitting Around the World: A Multistranded History of a Time-Honored Tradition. Minneapolis: Voyageur Press, 2011.

Xin, Xiaoyu. “Research on Prehistoric Hats in Xinjiang (2000 BC-200 BC)” Asian Social Science Vol. 11, No. 7. Canadian Center of Science and Education 2015. ISSN 1911-2017 E-ISSN 1911-2025. http://dx.doi.org/10.5539/ass.v11n7p333 Accessed June 3, 2019.
Note: The article refers to these hats as “knitted” which is not the only evidence of translation issues from Chinese to English.

Dura-Europos Fragments

The most famous of the Dura-Europos fragments is the beautifully stitch patterned cross-knit nalbound fragment, Inv. No. 1933.483, in the Yale University Art Gallery, showing a opposing leaf pattern bordered by repeats of a pomegranate like shape. Its original function is unknown, but the conservation efforts made the three remaining integral lacing loops visible.

Photo credit: Yale University Art Gallery – Public Domain
Source: Knit Textile Fragment | Yale University Art Gallery

The particular fragment pictured above initially caught my eye back in the late 1990’s. I had gotten my first copy of Richard Rutt’s A History of Hand Knitting. The chart he included for a knitted simulation did not match the image of the actual object provided on page 30 with the precision that I desired. I spent many many hours pouring over that image and charting out stitch by stitch the nalbinding pattern the year I was in Taiwan (1999/2000). I also spent a good bit of time consolidating a list of references to track down and discovered that the Academia Sinica library had an amazing Humanities and Ethnography collection. This collection included a copy of R. Pfister and Louisa Bellinger’s 1945 article on the “knitting” in The excavations at Dura-Europos Final Report IV, Part II which included a black and white image that was clearly post cleaning/conservation.

Increase/decrease diagrams from my 2000 cross-knit nalbinding handout.

My class handout I initially created in 2000 included not only diagrams of the possible increases and decreases and my chart for the specific pattern found in Inv. N0. 1933.483, it also included my initial attempts at using the images of the Dura-Europos fragment to illustrate the specific increases and decreases used in extant Roman Era cross-knit nalbinding. It continues to be a favorite piece for this purpose as it includes so many examples thereof in the formation of its stitch patterning.

Reviewing surface structure similarities after my presentation at the 39th International Medieval Congress.

In 2004, I was honored to present “Nalbinding or Not?: Some Structural Differences between Nalbinding and other Textile Techniques” at a DISTAFF session during the 39th International Medieval Congress in Kalamazoo, Michigan. The patterned Dura-Europos piece proved to be an excellent example for demonstrating what cross-knit nalbound increases and decreases looked like in an actual object and how they differed from the corresponding shaping of a knitted object.

Here is a copy of the handout from my presentation.

A sample testing out the pattern I charted from Inv. N0. 1933.483 made while listening to presentations at the 39th International Medieval Congress.
Cross-knit looping being produced by both crossed/twisted knitting and the cross-knit nalbinding variant.

The cross-knit looping structure can be produced by two different techniques, either cross-knit nalbinding or crossed/twisted knitting. They both produce a fabric of the same basic structure. However, they are worked in opposite directions. The clues as to which technique produced the fabric are in the shaping (increases/decreases), pick-ups, and mistakes. The preferred spiral working direction also differs between the two.

More information regarding the stitch patterned fragment, Inv. N0. 1933.483, along with a downloadable full size image is available on the Yale University Art Gallery’s site. The record for the two “ribbed” fragments, Inv. No. 1935.556, that were also found at Dura-Europos is available here. The electronic records were created from historic documentation that does not necessarily reflect their current knowledge about the objects, thus they are still listed as having been knitted.

The Yale University Art Gallery also has a permanent exhibition on the Dura-Europos excavations and as part of that has a very nice online feature outlining the historical background and excavation history with images and maps of the excavation: http://media.artgallery.yale.edu/duraeuropos/

I would like to thank the Yale University Art Gallery for providing such excellent photos in their online collections. I am also very much looking forward to, and very grateful for, the opportunity to view the fragments in person later this month. They continue to play a pivotal role in the study of the nalbinding technique and the structures it produces.