Tsuba site, RIP

Last updated 2005 – seems like yesterday… no it doesn’t.

Recently I’ve heard a time or twenty that my old tsuba site is offline. Back in the mid ’90’s there was a thing called a “personal home page” that came for “free” with your dial-up account. For better or worse, that’s where the ancient technology behind Tsuba has been creaking on for all these years. If there was too much traffic the site would be shut down until the meter ticked over again at the start of the month and it would be restored. It turns out that a month ago the meter was turned off for good. I got no warning at all that the legacy sites were going to be shut down. Others got a few days notice to save anything before it was gone forever. I think there were probably about five of us left.

So, it’s gone and it is not coming back. It’s a coincidence that I started this blog at around the same time, but a good one. I have all of the files from the original site and may reprise some text if it still seems relevant. Some of it definitely isn’t, maybe most of it. The photos are pretty much useless other than for nostalgia since the bandwidth upon which the infrastructure was built was such that users complained bitterly about any inline image bigger than a thumbnail.

Update: I have the site running locally on my desktop now so if anyone needs an extract I can get it easily.

What were you doing in the sword and fittings world back in 1996?

Can you name them all?

This was taken by Yoshikawa Eichi out in Yamanashi prefecture. Starting on the left, Michiko and Mamaru Hagihara, Jim Kurasch, Min Shintaku, Yoshikawa Kentaro sensei, Gordon Robson, Mrs. Yoshikawa, Mr. Omino, Jim and Laurine Gilbert and Sam Oyama.


7.17 cm H x 0.49 cm T

After thinking about tsuba finishing from the previous post, I’m adding this tsuba as an example of reaching for the limits of thinness in iron sukashi. The maker is listed in Haynes at H 07513:

NTS: G. Heckmann: TSUBA, 1995, T 39, oval
iron plate carved in the round with Michikaze
with large umbrella by a willow tree and a
frog on the ground, with gold and silver
inlay. Somewhat in the Kyoto style and
with Soten overtones, signed: Nobunori.
SCE. G. Heckmann, 1995

That tsuba from Heckmann is here:

It’s clearly the same guy although the work is a bit less technically extreme. I haven’t found any other examples illustrated but will keep an eye out. Kyo mixed with Soten is about right, but it clearly is his own style. I don’t imagine these were the creation of an amateur who only made a few guards.

So, a closer look at the details of the first guard.

Not only are the lines extremely thin, their width tapers as well and “all” 5 mm of depth is layered in different levels. I don’t see any filemarks (other than the nunome) under more magnification. Was there an easy way to do this? If it was made separately and inlaid or otherwise attached, I’m not seeing any signs. Amazing.

Update: I had a random thought and checked the fine tendrils with a tiny rare earth magnet to seek out possible non-ferrous shenanigans. None were found.

Unfinished business

8.45 cm H x 0.53 cm T

This is an Akasaka or Higo style tsuba with the plum branch and bird motif that was roughed out but never finished. Given the angles at which the hitsuana meet the seppa dai, I’d lean toward Higo, but that’s not really the point. As we’ll see below the blank was worked with a chisel.

Each chisel stroke is easily seen in this part of the rim.

The openwork is chiselled very close to the kebori layout lines defining the shape of the rim and plum branch. Handmade files couldn’t have been inexpensive, so it probably made sense to keep their use to a minimum. Each tooth of a file required a cut with a chisel, so it was probably about as much work to make one as to cut out a tsuba like this.

Given the roughness here it looks like the maker was working quickly and confidently. Opening this up with a jeweler’s type saw would be slow work and hand made blades would probably also have been expensive. I seem to recall reading that they came into use in the Edo period, but don’t remember the source. I’ll have to look for more information. String and abrasive techniques were around well before that, but that would have meant even slower going.

It’s possible that the tip (left) of the upper branch may be the reason the work stopped. The chisel stroke cuts over the centerline of the design. It seems like that could be closed back up without much trouble, but maybe that would be asking for headaches later in the process (or from the boss).

At the right side here there openwork has been refined beyond the rough chiselling. It looks like it may have been scraped down. Comments from metalworkers are welcomed.

It’s interesting that work that’s quite rough is next to spots that are almost finished instead of working everything uniformly from the rough state. The nakago ana, seppa dai and hitsuana were “cast in stone” early in the process.

There are some coarse-ish apparent file marks visible in the most finished section, but they aren’t showing well with either the macro setup or microscope. I wonder if the rim was going to be finished off with a circular cross section or left flat. Interesting choice to have refined the branch tip to that point while the rim is still rough.

I found an interesting video on a likely process for making early files. It’s not about Japanese technique, but from what I’ve seen the methods were probably similar (and the production values here are better).

His whole project on recreating the antikythera mechanism with period methods is well worth checking out. Amazing stuff.

Thanks to Fred G for the subject tsuba.

San Diego Tsuba

At sword shows I’ve seen the occasional puzzled expression at the use of the phrase San Diego tsuba in conversation. It’s a shorthand used by a few for the types of guards seen in an interesting group found at the bottom of the ocean.

The three guards from upper left to center bottom measure 6.5 cm H x 0.2 cm T, 7.0 x 0.25 and 7.3 x 0.25

The Spanish galleon San Diego was sunk just before 3:00 in the afternoon of Thursday December 14, 1600 off the coast of the Phillipines. The wreck was discovered in 1991 and over the course of two field seasons 5,262 objects were recovered.

“Undoubtedly, more than four hundred men crowded the decks – Spanish sailors, natives and even Japanese mercenaries whose presence is attested by their weapons and personal belongings. Twelve hundred pieces of blue-and-white porcelain have been recovered, indicating that the vessel was carrying a rich cargo of china tableware.” (Treasures of the San Diego, c 1996, Association Francaise d’Action Paris, Fondation Elf Paris and Elf Aquitaine International Foundation, New York)

I have an English language copy of Treasures, but if you happen to look for one note that most copies around are in Portuguese. It’s a great book covering the history of the period, details of the ship, its sinking and the European and Asian artifacts recovered. There is a section covering weapons including articles on small arms, artillery and one titled Japanese Warriors by Catherine Delacour. She gives an overview of Japanese history and activity in the region in the 16th C and in the Philippines in particular. Near the end of that century there were around 1,500 Japanese living in Manila including bushi working for Spain, some of whom shipped out on the San Diego and left behind various personal effects when it went down.

No iron artifacts survived submersion, but the group of copper alloy tsuba pictured above were recovered from the wreck. Two of these have fuchi and seppa stuck in place with corrosion and two have just seppa. We find guards like these in circulation today. They’re relatively small with simple but varied decorations. Given that they only have kozuka ana they were probably mounted on wakizashi. It’s interesting that the kiku-gata tsuba has an opening large enough for an o-kozuka. I don’t see any mention of kozuka or kashira being recovered from the wreck. Metal (or any) kashira may not have been used.

So while none of these tsuba have dates, we do know that they weren’t made after 1600, a useful data point. Whether these guards were all brought from home or some made outside of Japan would be interesting to know.

Update: Excavation of an early Edo period archaeological site in Nara yielded casting crucibles and molds including some that would have produced tsuba very similar to the above.

a still captured from the Sankei News video

Note that the lower right object is a clay impression taken from the mold to its left. The video is below:

Thanks to Markus Seksko for his blog post from a few years ago. More detail can be found here:

About those Tensho dated tsuba

I recently came across an iron sukashi tsuba dated Tensho 3 that got me thinking again about the question of whether these early dated guards are legitimate or some sort of tribute (at best). The first of these I saw 20 plus years ago was one Robert Haynes had turned up that appeared in the famous Red Cross catalog. It’s illustrated in his book Study Collection of Japanese Sword Fittings – Gai So Shi.

Bob writes “Since it is the only known example of this signature I entered it in my book as H 03296.0. In Japan such dates are dismissed, but they are wrong. There are a large number of dated tsuba before and just after this one, such as 1532, 1543, 1573, 1582, 1587 and another dated the same year as this one.” Closer views of the signature are here:

February 1573

The light is hitting the strokes a bit differently on the two sides, but the style of the characters look consistent. If adding a spurious early date, why do so with an otherwise unknown name? The workmanship of the rest of the guard is consistent with the period. I’m inclined to agree that it’s more likely genuine than not.

I undertook a search of Haynes’ opus mentioned in his quote above, The Index of Japanese Sword Fittings and Associated Artists for these other dates. I wound up expanding the search through all dates from the 16th century and came up with the following list.

1502 Ranko (Owari) H 07516

1504 Myochin Nobutada H 07188

1523 Miko (Hizen) H 05085

1532 Myochin Yoshifusa H 11476

1532 Muneoku H 06209

1533 Tadamasa (Hizen) H 09102

1533 Terutoshi (Mutsu) H 09651

1538 Myochin Munenori H 06194

1558 Myochin Unkai H 11064

1563 Munenaga H 06172

1570 Shoami Iranken H 01899

1573 Kiyonami H 03296

1573 Masahige (Heianjo) H 04472

1575, 77 Koike Masaie (Echizen) H 04047

1577, 78 Suzuki Shigemitsu (Heianjo) H 08352

1577 Suzuki Shigeyuki (Heianjo) H 08609

1577 Kishiwada Zaisai (Izumi) H 12517

1582 Tojo (Kyoto) H 09720

1587 Mori Soemon H 08877

1591 Mitsumoto H 05282

1599 Shoami Tsuneyoshi (Yamashiro) H 10942

Given that the Myochin appear to have extensively embellished their early genealogy I tend to discount the value of those dates listed above. Some may be valid, but setting them aside still leaves us with 19 tsuba with dates from the 1500’s.

Another Tensho dated example I’ve seen in person was from Alan Harvie’s collection, the illustration here from the 2005 London sale catalog.

Tensho 3 – 1575 8.2 cm H

Unfortunately I don’t have a photo of the other side. This would be the Shigemitsu listed as H 08352 with recorded dates of 1577 and 1578. Since Bob was certainly familiar with this tsuba, there appears to be a slight discrepancy in the record. This is a very finely made and unusual tsuba with the dragon’s eye done in glass. I wish I could get another look at it now. Is it a later worker enhancing his pedigree or an exceptional work? According to the listing there is at least one other dated example by him out there.

Below is the guard I just found.

7.76 cm H x 0.48 cm T
Nobumasa kore o tsukuru / Tensho san nen

So another date of 1575. Checking Haynes, there is a Myochin Nobumasa with a different masa working ca. 1550-1600, H 07118 but no other information. H 07112 with the same characters and also working 1550 -1600 seemed promising, however the example referenced from the Oeder catalog, p. 20 #153 looks quite different. The signature is not easy to see, but certainly not related.

Oeder collection Nobumasa

If my tsuba was signed with the Myochin family name, I’d expect it to likely be spurious. But this is another case of an otherwise unrecorded artist and a date that appears to be in the same hand as the name. I don’t think of mokume tsuba at this early date, but the rest of the workmanship and condition is not out of line with the period.

It still doesn’t quite ring true to me, but what was the goal if it isn’t? Maybe in the shinshinto revival period there was a market for fake early dates as there was for copies of Nobuie, Hoan, Yamakichi, etc. Of course those are all famous names and unmistakable styles. Why bother with a guard like this one.

Well outside the date range I covered above is a never the less interesting ko-kinko guard published in one of the great catalogs from the fittings museum in Sugamo back in 1994: (note that in my research files I routinely add random text notes that do not appear in the original photos)

This is dated much earlier, 1394, and enigmatically signed Botanka rojin. Haynes lists this guard at H 00202 and interprets the signature as perhaps “old Mr. Peony Flower” (botan being peony). The guard certainly looks like it could be from the period and it’s published in that issue of Tosogu Meihin Ten along with other famous ko-kinko and ko-mino works.

In this case the date looks to me to be cut with a thicker chisel. Is it a different hand? There appear to be a couple of uncertain strokes there. Is it all a tribute of some sort? The ring-type nanako punches on the seppa dai are interesting. They aren’t quite the same as what’s on the web of the guard. Was it added as “proof” that the date couldn’t have been added after the name?

I’m afraid I’ve added confusion on top of uncertainty in this post. If anyone has any other examples or clarifying ideas I’d love to hear your comments.

A Hiko-no

7.12 cm H x 0.50 cm T (don’t know how it got to the bottom of the pile so quickly)

I like coming across these early kinko guards that look like they may have been the inspiration for Hirata Hikozo’s work. The color of the alloy combined with the lacquer along with the fukurin and carving have much the same flavor although Hikozo refined them all and built on it.

I had always thought the tagane were also original to these pieces, but a friend pointed out that they may have been added as an upgrade toward Hikozo’s tagane mei. It’s certainly possible, but I also see them on guards around the same age that I think would really take some imagination to try to pass off as Higo. For example:

7.72 cm H x 4.2 mm T

They’re also seen on some iron Saotome tsuba, but the rest of the work is quite different, so probably no direct connection.

Setting Hikozo aside, note that the surface of this one is not covered with nanako or even worked with a ring-shaped punch. Each circle is made up of a series of tiny punch marks. I wonder if this was a country guy’s imitation of nanako or something original and maybe earlier.

Not nanako, but it had to have been a lot of work. Maybe the nanako punch was developed as a time saver.

In this case the fukurin is needed to finish the three layer construction of the plate. Unlike the typical san mai guard, the shakudo here is only a thin foil. There is (of course) a specific term for this that I’ve forgotten. A microscope view inside the kozuka ana is here:

It appears to be quite a tight bond without signs of solder. I don’t know how it was done. Yes, I labeled the photo incorrectly way back when, it really is the inside of the kozuka ana.

A tsuba with two signatures

I recently came across this guard online. It was impossible to make everything out from the photos, but it looked like there were two different hands at work. After some rust removal we have the following.

The front
The back

So, even without reading it, it’s pretty clear that two different people signed.

The easier one first, the back, says Okamoto on the right and Yoshikiyo on the left. Okamoto is a family name found among a number of groups, and very frequently in Choshu. I don’t see a Yoshikiyo with these kanji listed, but Haynes 11517 is a Yoshiharu with the same yoshi, the family name Okamaoto from Choshu, so perhaps a connection.

H 11742 is a Myochin Yoshikiyo with the same characters working in Hizen. Normally that would be a clearly different guy, but given the other side I wonder if there is some connection.

So, the front – the left side is Myochin ki Mune… something. It’s an odd character and partially punched into oblivion. Going through the various Myochin Mune possibilities we find that the last Munesuke signed with that distinctive “xx” shaped suke. He’s recorded in Haynes at 06239 and Bob has handwritten the variant kanji in the listing.

The right side looks like hard work, but it’s convenient to know that it’s something that a number of Myochin smiths added to their work. Haynes reads it shinto go tetsu neru – “forged from five layers of iron.”

A few days ago I was reading Markus Sesko’s blog and found an interesting article on just this phrase:

About shintô-gotetsu inscriptions on Myôchin works

Markus reads it shinto go tetsu ren and suggests that rather than the standard interpretation the meaning may actually be “forged from an old begging bowl” of the type used by mendicant monks. Please read his post for the details and much more information. It’s very interesting.

So, a view of the entire guard with its carving of clouds:

7.52 cm H x 0.57 cm T

The carving resembles Choshu work and given that Munesuke claims credit for forging the plate with either meaning intended, it seems likely that Yoshikiyo worked the chisels. His unlisted status may be because he wasn’t known for making tsuba on his own. Or maybe he is known as the Myochin Yoshikiyo and for some reason at this point was referencing where he came from.

I will have to ask some Choshu tsuba collectors if this apparent collaborative work is a well known thing. I don’t recall seeing it before, but this is outside my usual area and I may have some or all of it wrong.

Update: Not a known thing so far.