Kumihimo – Saidaiji Temple Braid

June 1, 2009

Saidaiji 1

Saidaiji-gumi, front and back faces

There are confusingly two Saidaiji Temples in Japan. Saidai-ji means Great Western Temple; one was built to contrast with the great temple to its east at nearby Nara, a cultural centre and former imperial capital, south of Kyoto. The other is 100 miles west of Kyoto and is famous today for its matsuri or folk festival, held in the depths of winter each year involving the locals in loinloths in a mad scramble for a stick. I am assuming the braid in question comes from the one at Nara.

Saidaiji is these days a mere western suburb of Nara City and with its four main temples is a mere shadow of its former self. Saidaiji-gumi is a national treasure so is not on show at the temple’s Treasury, though that repository contains other examples of complex braiding. Saidaiji is a natural add-on to any tourist side-trip out of Kyoto to Nara. In Nara, any textile person will want to stand outside the Shosoin Treasury building, visit the Shosoin exhibition in Spring or Autumn, travel deep into the park to see the yoroi armour braids at the Treasury of Kasuga-taishi, see the great bronze Buddha and check out the Nara National Museum. I cannot too highly recommend a trip one train stop west to Saidaiji, perhaps on returning to Kyoto. There are only three temples and a modern Treasury building, so this is no grand garden temple as typified by Kyoto. A ticket for the whole lot costs 2000 yen, which might seem expensive if you’ve been doling out money already for temples, but I recommend sitting in the small dark temple and checking out the brocade sutra-holder in the Treasury building. After a visit, you will have a very clear idea of how the sutra cover (its bamboo or wire struts, its luxurious Chinese fabrics) and its thin, square braid closure actually functions. This will complement the metal tubular sutra containers on show at the Kyoto National Museum.

Saidaiji temple 1            Saidaiji Temple 2

Makiko Tada mentions the braid and the scroll being used in conjunction with an image of the Buddha, a Sakyamuni Tathagata image, in which case it may well originate from the Nara temple since it was one of seven very important temple complexes in the area. Nara and Saidaiji are located on the Yamato Plain, south of Kyoto, very prominent in the early history and development of Buddhism in Japan. Saidaiji-gumi was used not as a braid hidden inside a Buddha statue like Chion-in-gumi, but is believed to have been used as a tie or closure for a pouch containing Buddhist sutras or sculptures. Buddhist sutras, as rolls of paper, seem to have been stored upright in 12-inch tall bronze jars, as displayed in the collection of the Kyoto National Museum, or protected by a mat of bamboo slats within a piece of brocade textile. The original is a mere 3mm in cross-section and the block pattern is repeated some 80 or so times along its entire length.

Saidaiji-gumi is probably the most accessible of the historic temple braids for contemporary braiders since it can be made with 56 bobbins on the taka-dai stand. Braiding is however an extremely slow process – around 10 hours per inch of braid. I found it almost impossible to correct errors once they occurred. The ‘back’ face of the braid faces the braider while working, while the front – as in the illustrations you are trying to ‘follow’ – face downward on the braiding stand. One of the most difficult decisions to be made is the yarn type – do you use the best possible yarn and risk making mistakes or do you use a yarn of lesser quality which might be at odds with the amount of time required to spend on the braid? Perhaps the next most difficult decision is the length of your warps. Knowing that it takes 10 hours to braid an inch, can you afford to have your taka-dai ‘locked-in’ for weeks, even months? To overcome this problem, I know of braiders who have two taka-dai: one for ‘quick’ braids, and one for ‘slow’ braids. I used Praslon commercial 2/20 weaving yarn on my second attempt (as illustrated) because of its subtle shine, but I used many many warps of sewing cotton, in very short lengths, for making a similarly-slow Chuzonji-gumi.

Saidaiji 2

Saidaiji-gumi, side faces.

Tada has published two colourways for this braid, one folloowing the original dark blue/light blue/white of the original (often re-created in blue-green teal) and one where the lighter colour in the bands is duplicated in the centre panel. Her book features a variety of non-tradtional colour schemes. In my experience, a considerable amount of contrast between the colours is required. Yarn in colours which are too similar tonally detract from the overall appearance of the braid.

Saiki has an illustration of a Saidaiji reproduction (braid 4 on page 127) but the Japanese description for braid 1 indicates this is not a Saidaiji though it looks very similar to it. This square briad is in black, white and orange and for all the world resembles the side faces of a Saidaiji. This illustration is used again (slightly enlarged) in the Domyo School Taka-dai book, page 23. From feedback received from Makiko Tada’s UK braiding workshops, I understand that it can be made by two braiders working side by side, using the loop manipulation process. The loop-manipulated braid will be more pliable and flexible than one made on a taka-dai stand.

So what are the relationships between the historic temple braids structurally? On the face of it, it seems terribly unlikely that entirely uniquely-structured braids developed spontaneously around the country. With a little bit of work, I’ve determined that Chuzon-ji-gumi is in fact a ‘multiple’ of Saidaiji-gumi. You simple increase the number of bobbins, change the colourways, but maintain the working method. I’m sure this all explained – in Japanese – in Masako Kinoshita’s Archaic Braids, but I had to discover this by myself by working from Chuzon-ji-gumi backwards to Saidaiji.

Here I’ve played around with potential colour combinations using the standard 56-bobbin Saidaiji-gumi pattern:

saidaijigumi 1

Here I’ve worked backwards from a 56-bobbin Saidaiji-gumi and the pattern is retained using only possible with 40 and 24 bobbins, as follows. Note that the number of bands in both sets of two repeated lozenges increases from 3 bands with 24 bobbins to 5 bands with 40 bobbins to the full 7 bands with 56 bobbins.

saidaijigumi 2


Significantly, if we keep increasing the number of bands (by increasing the number of bobbins) we come upon the “Half” Chuzonji-gumi (Han-Chuson-ji-gumi) and the (full) Chuson-ji-gumi surface patterns. While Saidaiji is near Kyoto, the original location of the Chuzon-ji-gumi historical braid is way up in the north of the island of Honshu.

saidaijigumi 3

My next task, of course, is to braid examples of this mathematical progression.



Masako Kinoshita, Archaic Braids.

Makiko Tada  Comprehensive treatise of braids: takadai book 2. Tokyo, Texte Inc, 1998.

Saiki, Shirou  Chuzon-ji. Tokyo, 2003.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: