The Honjo Masamune

A katana of about 65 centimetres, attributed to Masamune, that stood as the object of the Tokugawa shogunate: after Tokugawa Ietsuna received it in the seventeenth century it was handed symbolically to each succeeding shogun on taking office. It was designated a National Treasure on 29 May 1939. In December 1945 Tokugawa Iemasa, seventeenth head of the house and President of the House of Peers, took it with other family swords to the Mejiro police station in Tokyo under the occupation order to surrender all blades. On 18 January 1946 the police handed them to an American sergeant whose name was written down phonetically in katakana. Nobody has seen the sword since, and nobody has ever identified the sergeant.

The record exists because of what stands behind it. Japanese swords carry the oldest continuous authentication system in world art: the Honami family, appointed by Toyotomi Hideyoshi in the 1590s as the sword-authentication office with an official seal, who issued folded certificates called origami that stated a blade's maker and its value in gold, and whose fee was a percentage of the value they themselves assigned. That is a hereditary authentication board with an interest in the number it writes down, and its documented failures are as instructive as its successes.

The Honjo Masamune is also, legally, no longer a National Treasure. All pre-war designations were reduced to Important Cultural Property in 1950 and had to be re-submitted to regain the higher rank. A sword that has been missing since 1946 cannot be re-submitted. The most important lost object in Japanese art lost its status by not being there.

What is missing

The blade was 65.2 centimetres long with a curvature of 1.7 centimetres, made in shinogi-zukuri with an iori-mune, a wide body and a thin section, an extended point, a narrow shinogi-ji and a high ridge. The forging was fine ko-itame with ji-nie, chikei and tobiyaki. The hardened edge ran in ko-nie forming o-midare and ko-midare with kinsuji and ashi, and the boshi turned back in midarekomi. There were no carvings. The tang was o-suriage, greatly shortened, with a single peg hole.

That description is not a modern reconstruction. It is the text of the sword's own designation as a National Treasure, written when it was last examined by the state, and it is the fullest account of the object that survives. It also records the damage: one large chip in the edge near the base, several smaller chips along the edge, and cuts across the back of the blade. Those marks were combat damage from the sixteenth century, and the polishers deliberately left them in. They are traditionally not shown in an oshigata, which means the only pictures of this sword are pictures with the evidence edited out.

How it was taken, and why it kept the marks

The story is in the Kyoho Meibutsucho, the shogunate's own register of famous objects. During one of the Uesugi campaigns, after a siege, Honjo Shigenaga was sitting on a folding camp stool receiving captured heads when a man named Umanosuke attacked him and cut his helmet in half, wounding him seriously, before Shigenaga killed him. The sword that did it took a couple of chips in the edge from going through the kabuto. Shigenaga survived, kept the sword, and it has carried his name ever since.

How Umanosuke came by a Masamune in the first place is unrecorded. As Guido Schiller puts it in the fullest study of the sword, the original owner was in no condition to answer questions.

What matters for a provenance archive is the chips. Japanese sword polishing removes metal, and a polisher's decision about what to take out is a decision about what a blade is for. Leaving four hundred years of battle damage in the edge of a National Treasure is a statement that the object's history outranks its appearance. The 1939 designation text records the damage precisely because by then the damage was part of the object. And English Wikipedia places the fight at Kawanakajima in 1561, while the Meibutsucho entry names no battle at all. The most famous incident in the sword's life is not securely located.

The register that recorded no value

The Kyoho Meibutsucho was compiled in 1719 by Honami Mitsutada at the command of the shogun Tokugawa Yoshimune. It was the first time objects other than tea utensils, lacquer, painting and ceramics were formally registered as meibutsu, famous things, and it runs to three volumes and an appendix covering swords by the three great smiths, swords by other smiths, fire-damaged blades and further additions.

Its entries are short by design. Schiller describes the form: the register basically states the owner of a sword, its nickname and its length, and then gives a brief history. The Honjo entry keeps to that. It says gyobutsu, which in the Edo period meant the property of the shogun; suriage, shortened; nagasa 2 shaku 1 sun 5 and a half bu; and then mudai, no value given.

Mudai is not a blank. It is a positive entry meaning that no valuation is assigned, made in a document produced by the family whose certificates existed to assign valuations and whose fee was a percentage of them. Whether other entries in the register carry values is not established in the sources consulted, and this record does not claim it. What can be said is narrower and still worth saying: the object that stood for the shogunate was entered in the shogun's own register of famous things with an explicit refusal to price it.

A hereditary authentication board, and how it was paid

The Honami have no counterpart in the European art trade that this archive has found. The family descends from Honami Myohon, an attendant in the cultural circle of Ashikaga Takauji, first shogun of the Muromachi bakufu, in the fourteenth century. They polished swords, maintained them and judged them, generation after generation, through the Ashikaga, the Toyotomi and the Tokugawa.

The authority became formal under the ninth head, Honami Kotoku, who worked for Oda Nobunaga, Toyotomi Hideyoshi and Tokugawa Ieyasu in turn. In 1596 or 1597 Hideyoshi appointed Kotoku token kiwamedokoro, the office of sword authentication, and granted him a square seal reading hon, with authority to appraise swords and to write their value on a folded paper. That grant is the origin of the certificate and, through it, of the modern Japanese word origami.

An origami is a single sheet of high-quality paper, hoshogami or danshi, written on one side and folded in two. It gives the smith, a finding that the blade is genuine, the measurements and characteristics, the date with its zodiac year, the appraiser's name and his written cipher, the seal stamped on the reverse, and the daitsuke, the valuation. Values were written in mai of gold, in oban, or in kan. Five mai of gold. Thirty-five mai. Seven hundred kan. A Soshu Sadamune at fifty oban.

Now the part that a provenance archive cannot leave out. The Honami fee was ten mai of silver for every hundred mai of gold in the valuation. The appraiser set the number, and his income was a percentage of the number he set. Between about 1600 and 1750 the thirteen Honami branch families met once a month to issue certificates in a standardised form, which is a genuine collegiate check, and the certificates of the tenth through thirteenth generations, from Koshitsu to Kochu, are still the ones collectors trust most. But the incentive ran one way throughout, and it ran towards the higher number.

Where the system failed, in its own vocabulary

A tradition that has been authenticating continuously for six hundred years has had time to name its own failure modes, and the names are the evidence.

Tanuma origami is the term for the certificates of the Tanuma era of the 1770s and 1780s, when they were issued in such volume that they lost their meaning. The phrase is pejorative and it is still in use.

Awase origami means deliberately attaching another sword's certificate to a blade in order to raise its value. That is a named, categorised, understood species of fraud inside the system, which tells you it was common enough to need a word.

And then the number that settles the argument. Kotoku himself was known for his integrity and for refusing to judge doubtful blades submitted by powerful people. His successors issued three thousand origami and paper attributions to Masamune. Masamune could not have forged three thousand swords in his life. The board that authenticated the Honjo Masamune also certified thousands of blades as Masamune that were not, and it did so under exactly the commercial and social pressure a hereditary appraiser to the great houses would be under: it would have been extremely difficult to tell a daimyo his sword was a fake.

Modern scholarship has unpicked individual cases. The Kotegiri Masamune was attributed to Yukimitsu while Sano Nobuyoshi owned it and was known as the Sano Yukimitsu; after Maeda Toshitsune acquired it, it became a Masamune. Markus Sesko calls that attribution dubious and argues for Sadamune. The pattern is that a blade's maker improved as its owner's rank improved.

The man who said Masamune never existed

Because so few blades carry his signature, a theory took hold at the end of the nineteenth century that Masamune was an invention. Its author was Imamura Nagayo, whose name is also read Choga, a sword appraiser who served as keeper of swords at the Imperial Household Ministry, and he published it in 1896 as a serial commentary in the Yomiuri Shimbun, which meant the argument ran in the newspapers and the whole sword world had to answer it in public.

His case was specific. He had never seen a genuine signed Masamune. Masamune's reputation as a master appears only after Hideyoshi's era, and earlier texts do not list him among the celebrated makers. His name is absent from the list of one hundred and eighty-two master craftsmen selected in the time of Ashikaga Yoshimitsu. And the Honami family, hereditary polishers and appraisers, had created the Masamune legend under Hideyoshi's political direction. The theory was abandoned as scholarship advanced, and the consensus is that Masamune was a real smith of the late Kamakura and early Nanbokucho periods. Three signed works, the Fudo Masamune, the Kyogoku Masamune and the Daikoku Masamune, are accepted, and two further long-signature blades dated 1314 and 1328 are known from records and are lost.

The detail that ties the argument to the object is this. Imamura is the same man who examined the Honjo Masamune in 1880 and came away unimpressed. The scholar who proposed that Masamune was a fiction had handled the greatest Masamune of all, sixteen years earlier, and thought it ordinary.

The sword of the office

Shigenaga was stationed at Fushimi, ran out of money, and sold the blade to Toyotomi Hidetsugu, Hideyoshi's nephew, for thirteen mai of gold. Hidetsugu was forced to kill himself in 1595. He had already presented the sword to Hideyoshi. It passed to Shimazu Yoshihiro, who gave it to Tokugawa Ieyasu, who gave it to his son Tokugawa Yorinobu, founder of the Kii branch, who presented it to the shogun Tokugawa Ietsuna on his retirement.

From Ietsuna the sword stopped being a possession and became an attribute of an office. Schiller states that it became one of the treasure swords of the Tokugawa and was handed symbolically to each succeeding shogun when he took office. English Wikipedia says instead that it remained in the Kishu Tokugawa family. Both are in print. The succession reading is the better argued and it explains something the other does not: why this particular blade, out of the many fine swords the Tokugawa owned, is the one that everybody in Japan can name.

After 1868 the sword went with the head of the house rather than with the office, because the office no longer existed. Tokugawa Iesato became the sixteenth head. His eldest son Tokugawa Iemasa became the seventeenth, and was President of the House of Peers. Iemasa's only son Iehide died in 1936, so he adopted his daughter's son, Tokugawa Tsunenari, who heads the house today.

December 1945

Under the occupation the manufacture of edged nihonto was banned except by permit, and all swords were ordered surrendered. Enormous numbers were destroyed or dumped at sea. In December 1945 Tokugawa Iemasa took his family's swords, the Honjo Masamune among them, to the police station at Mejiro in Tokyo and handed them in. English Wikipedia says there were thirteen others, fourteen in all. Schiller says only that other swords went with it.

On 18 January 1946 the Mejiro police handed the swords over to an American sergeant. Here the two accounts part company again and neither can be reconciled from anything published. Schiller says the sergeant was reportedly of the 7th Cavalry Regiment. English Wikipedia says he represented the Foreign Liquidations Commission of Army Forces Western Pacific. A record that pretends to know which is lying to its reader.

Swords of artistic and historic merit were supposed to be examined and returned to their owners. Iemasa wrote to the authorities. He was told the sword could not be found and had probably been taken abroad. His letters, and those of other owners of important blades, were answered elaborately and politely. No thorough investigation appears to have followed.

The system that would have saved it arrived five months too late in one direction and two years too early in the other. Art swords had already been excluded from the firearms prohibition decree on 1 June 1946, and the society that exists to preserve them was founded in February 1948.

The name that is not a name

The receipt gave the sergeant's name. It has never led anywhere, and the reason is a transcription chain that nobody at the time thought about.

The Japanese police wrote the name down phonetically, in katakana, as ko-r-de-i-ba-i-mo. Somebody later turned those syllables back into roman letters as Koridie Beimo. Albert Yamanaka rendered the same syllables as Coldy Bimore, and that version is the one that stuck and the one every retelling uses.

Coldy Bimore is not a name anybody has ever had. There is no interpunct in the katakana of the kind normally used between a foreigner's given name and family name, so it may be one name rather than two. The syllables that produced Coldy would also produce Corey, or Cordy, or Goldy. Extensive efforts have failed to identify or locate any serviceman answering to it, and no source consulted documents a systematic search of United States Army personnel rolls.

What the archive can say precisely is this: no American soldier named Coldy Bimore has ever been identified, and the search has been conducted for eighty years against a target that is a double transliteration of a phonetic guess written by a Japanese policeman in January 1946. The sword was not lost because a receipt was missing. It was lost because the receipt was written in a script that could not hold the name.

It is not a National Treasure any more

The Honjo Masamune was designated a National Treasure on 29 May 1939 under the National Treasures Preservation Law. Almost every account of the sword still calls it one. It is not.

When the Law for the Protection of Cultural Properties came into force in 1950, every pre-war National Treasure was reduced to Important Cultural Property, and objects had to be re-submitted to regain the higher rank. Fourteen pre-war National Treasures had gone missing after the war. None of them could be re-submitted, because re-designation requires an object to examine. Their designations are therefore void in the higher category, and they survive only at the lower one.

The Japanese state records this without comment and with complete precision. The Agency for Cultural Affairs maintains a public register of nationally designated cultural properties whose whereabouts are unknown, published under a campaign asking the public to help recover them. Entry 63 reads: katana, unsigned, attributed to Masamune. Important Cultural Property. Designated Showa 14. Length 65.2 centimetres. Owner: an individual. Prefecture: Tokyo. No theft is recorded, because it was not stolen; it was handed in.

The entry does not use the name Honjo Masamune. The state does not make the identification. It is the 1939 designation year and the 65.2 centimetres that make it, and this record says so rather than pretending the government has confirmed something it has not. Two entries away, entry 62 is another nationally designated katana attributed to Masamune, 72.1 centimetres, designated in 1955, stolen in 1982, also still missing. Japan has lost two designated Masamune, by two different mechanisms, and is still asking.

The board that replaced the Honami, and its own bad decade

The Honami monopoly ended with the Meiji settlement. The collapse was not gradual and it was not dignified: the nineteenth head of the main line pawned the official copper seal, and the last head became a shoemaker. What survived was expertise in branch lines, and one of those branches produced Hon'ami Koson, born Kawaguchi Teikichi in 1879, apprenticed as a polisher at twelve, independent from 1907, the man who introduced the gokaden classification of the five traditions and who examined the Honjo Masamune in 1943.

In February 1948, with the occupation still confiscating blades, Homma Junji and Sato Kan'ichi founded the Nihon Bijutsu Token Hozon Kyokai as a government-authorised foundation, explicitly to save Japanese swords from total destruction. Koson became an adviser, trustee and judging member. Homma Junji is also the man who had run the pre-war survey for designating swords as National Treasures, and who wrote that the Honjo's shape and section resembled the Ishida Masamune. The same handful of people ran the old system and built the new one.

The NBTHK now issues four grades: Hozon, worthy of preservation; Tokubetsu Hozon, especially worthy; Juyo Token, important; and Tokubetsu Juyo Token, especially important. Judging is done at the head office. It is, in every functional sense, the Honami office reconstituted as a foundation, with one deliberate change: its certificates state authenticity and do not state a value.

It has had its own crisis, and it was the same crisis. In 1981 it emerged that large numbers of swords certified under the old system, particularly the Tokubetsu Kicho Token grade, were fakes. In September the chairman and all the directors resigned, the certification system was reviewed and the judges concerned were removed. The old grades were abolished in May 1982 and the present four-tier system began that September. Papers issued under the old system are not regarded as carrying force today, and a blade holding one has to be submitted again.

Twenty-five years later the argument came back in a different form. Between 2006 and 2007 members of the Diet raised the fairness of the examination process and the conflicts of interest inside the organisation; the ministry inspected; 344 recorded regulatory violations were documented from 2006; litigation ran through the Tokyo District Court and the Tokyo High Court; and a separate case followed over unregistered blades. The problem an authentication monopoly has is structural and it does not go away by changing the letterhead.

Two Honjo Masamune, and three tracings that do not match

There is a second object called the Honjo Masamune. It is a tanto, unshortened and signed, that belonged to the Honjo family; an oshigata of it is published in Iimura Kasho's Yumei Koto Taikan. It is not a meibutsu-class Masamune and it is not this sword. Schiller identifies it as the probable source of the persistent rumour that the great Honjo Masamune is signed. It is not: it is mumei and o-suriage, and any signature went when the tang was cut.

The visual record is worse. At least three oshigata of the lost blade are known: one by the seventeenth-century smith Tsuguhira, one by a member of the Honami, and one of unknown origin in a private collection in Japan. Schiller describes the other two as much more detailed than the Tsuguhira tracing, and adds that they are probably also not fully superimposable with it; that the Tsuguhira tracing does not appear to be very accurate, oshigata of that period being less exact than modern ones; and that some lines seem to have been redrawn for the 1928 printing because the paper had degraded.

So the position is this. The most famous lost sword in Japan is known from at least three drawings of differing quality, one of which its own publisher calls inaccurate and which was partly retouched, all of which conventionally omit the battle damage that the state's own designation text records as a defining feature. If the blade surfaced tomorrow, the documents that would have to identify it are the 1939 designation description and the chips in the edge, not the pictures.

What this class of object does when it is for sale

No Masamune has ever been traced to a public auction in the sources consulted, and the Honjo Masamune has never carried a price. The nearest published figures are for other blades and they establish the range.

The best documented modern figure in Japan is not an auction at all. In 2018 the city of Setouchi in Okayama prefecture set out to buy the National Treasure tachi Yamatorige, an unsigned Ichimonji blade that had been a chief heirloom of the Uesugi house, from a private owner in the prefecture. The city's own portal states the target: 513,090,000 yen, being 500,000,000 yen for the sword and 13,090,000 yen for facilities, to be raised through the hometown-tax donation system, corporate donation and international crowdfunding. It was reached on 26 January, with more than seventeen thousand supporters in aggregate, and the campaign ran to 31 March 2020. The city's page gives the day and the month without an unambiguous year, and reporting of the achievement differs between January 2019 and January 2020; the record leaves that open rather than picking one.

Half a billion yen, raised in small donations from seventeen thousand people, to keep one thirteenth-century blade in the town where its school of smiths worked. That is the clearest measure available of what a National Treasure sword is worth in Japan, and it is a measure of public feeling rather than of a private market.

On the Western auction side the ground rule matters more than the numbers, and it is this archive's standing rule rather than a claim about any one house: Christie's, Sotheby's, Bonhams and Phillips publish premium-inclusive totals and do not publish hammer prices. Any figure quoted for a Japanese sword at any of them should be read as premium-inclusive unless the house says otherwise, and no hammer figure should be derived from it without the printed premium schedule in force on the day.

What would happen if it turned up

The sword is still on the Japanese government's register, which is a live document with a telephone number on it asking for information. It is not a cold case in the legal sense; it is an open request.

The obstacles to a return are not sentimental. An object surrendered under a military occupation order and passed to a named official is not straightforwardly stolen property, which is why the Agency's entry carries no theft note, and eighty years have run since. Whoever holds it, if anybody does, holds something that cannot be sold openly anywhere in the world and that a Japanese import would immediately re-designate. The rational move for a holder is silence, and silence is what there has been.

What the archive can state is the shape of the problem. The blade is identifiable: the 1939 description is detailed enough to make a match, and the combat damage is distinctive. The name of the man who took it is a phonetic artefact and will not produce him. And the object that stood for the government of Japan for two hundred and fifty years is catalogued today as entry 63 on a list of things the state would like back.