The signature (mei)

What it is and since when

The mei (銘) is the inscription the smith cuts with a chisel (tagane, 鏨) into the tang (nakago, 茎): at the very least his name, and often also where he worked, his title and the date. It is the last thing done to the tang, over file marks (yasurime, 鑢目) that are already finished, which is why the strokes of the signature cut across the file marks and not the other way round.

The duty to sign is very old. The Taihō code (大宝律令, Taihō ritsuryō) of 701 contained a statute on works and manufactures, the Eizen-ryō (営繕令), whose article on the making of weapons ordered the year, the month and the craftsman’s name to be inscribed on them: 鐫題年月及工匠姓名 (nengetsu oyobi kōshō no seimei o sendai seshimu, “let the year, the month and the name of the craftsman be engraved”). Whatever could not take an inscription, such as an arrow, was exempt. The text of the Taihō code is lost; the article is known from the Yōrō code (養老律令, Yōrō ritsuryō), in force from 757, which is held to be almost identical in this statute.

Even so, the oldest smiths’ signatures that survive are much later, from the middle of the Heian period: those of Yasutsuna (安綱), Sanjō Munechika (三条宗近) and Ko-Bizen Tomonari (古備前友成), the earliest smiths with signed work still extant. It is usually said that the custom of signing the tang did not become general until the late Heian period. Many early signatures are the name alone, in two characters, such as 宗近 (Munechika) or 正宗 (Masamune); over time the long signature, which adds the province, the title and the date, becomes common.

Where it goes: omote and ura

The signature goes on the side of the tang that faces outward when the sword is worn, the omote (表); the date, when there is one, usually goes on the other side, the ura (裏). The tachi (太刀) hung from the belt edge down, while the katana (刀) is thrust through it edge up, so their signatures fall on opposite sides of the tang: these are the tachi-mei (太刀銘) and the katana-mei (刀銘).

Some smiths and schools signed the other way round, so the side of the signature is a pointer, but it does not decide on its own what type of blade you are holding. The two sides, their best-known exceptions and why museums display the tachi and the katana with the edge in opposite directions are covered in “Kantei: how to look at a blade”.

The parts of a long signature

The short signature, the niji-mei (二字銘), is just the name in two characters. The long signature, the naga-mei (長銘), almost always follows the same order, and every part except the name is optional:

An example: 肥前国住武蔵大掾藤原忠広 (Hizen no Kuni-jū Musashi Daijō Fujiwara Tadahiro) reads “Fujiwara Tadahiro, Musashi Daijō, resident in the province of Hizen”. It is one of the signatures of the first Tadayoshi of Hizen, one of the great shintō smiths, and it has five of the eight parts: province, 住 (jū), title, lineage and name. Between the two-character signature and the longest one, every combination occurs: province and name, name and 作 (saku), name and date.

Titles: kami, suke and daijō

Many Edo smiths added to their signature an honorary provincial title, the zuryō-mei (受領銘). These are the ranks of the old provincial administration —kami (守), the governor; suke (介), his deputy; and jō (掾), with its senior grade, daijō (大掾)— by then without office or land: the court granted them, for a fee, to craftsmen, merchants and artists who worked for it or for the temples, as a distinction for the house and the trade. The province of the title need not be that of the workshop: a Hizen smith could be Musashi Daijō.

A title could be raised, and that helps with dating. The third Tadayoshi of Hizen received Mutsu Daijō (陸奥大掾) in 1660 and Mutsu no Kami (陸奥守) in 1661: a blade of his signed Mutsu no Kami dates from 1661 or later. Some titles were rare within a lineage: among the Kanesada, only the one called Nosada and the eleventh Aizu Kanesada, in 1863, became Izumi no Kami (和泉守), according to the same article by Markus Sesko cited below. In any case, the title says the smith obtained it, not what a particular blade is like.

The date, and months 2 and 8

The date almost always goes on the ura and reads era, year and month. How to convert it to the Western calendar, and what the first year of an era (gannen, 元年), the “auspicious day” (kichijitsu, 吉日) and the sign of the sixty-year cycle (eto, 干支) are, is explained in “Periods: from jōkotō to shinsakutō”. Numbers may appear in their formal form, the one used in documents: 弐 (ni) for 二 (ni), 参 (san) for 三 (san).

The most curious part is the month. A large share of dates say only 二月日 (nigatsu hi, “a day of the second month”) or 八月日 (hachigatsu hi, “a day of the eighth month”). A count of 57 signed blades by Horikawa Kunihiro, cited by Hara Fumihiko, curator at the Tokugawa Art Museum (2020), finds 18 with the second month and 23 with the eighth: 41 out of 57 in just two months. The explanation that goes with that count comes from the sword scholar Fukunaga Suiken (1975): at the time, smiths wrote “second month” on what they forged in the half of the year from the winter solstice to the summer solstice, and “eighth month” on what they forged in the other half. Workshops also say these were good months for hardening. Whatever the reason, a 二月 (nigatsu) or a 八月 (hachigatsu) is not enough to date a blade to the month. And until 1873 the months are lunar: the second month is not February.

Who cut it: daimei and daisaku

Not every genuine signature was cut by the smith whose name it bears. In a workshop with a master and pupils, two cases are distinguished, both with the master’s permission:

• daimei (代銘), “signature in his place”: the blade is the master’s, and a pupil or a son cuts the master’s name on it;
• daisaku (代作), “work in his place”: the pupil makes the blade, in the master’s style, and it carries the master’s name.

They often go together, and one speaks of daisaku-daimei: the pupil makes the blade and signs it with the master’s name. Markus Sesko relates in his article “The last of the Kanesada” that the eleventh Aizu Kanesada, apprenticed to his father at 14, made many such blades for him in the following years. Since the master authorised it, these are not forgeries: they are workshop products, even if the hand is not that of the name.

What an outsider adds is another matter: the kiritsuke-mei (切付銘), an attribution cut later with a chisel by someone else; the kaisan (改鏨), a signature retouched or altered with the chisel; and the gimei (偽銘), the false signature, put there to deceive.

What is inscribed besides the name

A tang can say much more than who made the blade. These are the most frequent inscriptions, in summary; their details are in “Kantei: how to look at a blade”, and the cutting test in “Ranks, cutting tests and titles”.

The state of the signature

Before reading a signature you need to know what state it is in, and whether it is the smith’s. Each case is explained in more detail in the topic shown:

Inscriptions in gold and in lacquer are an expert’s opinion, not signatures, and are judged as such.

How to read a signature, step by step

A famous example serves as a guide: the Yamanbagiri Kunihiro (山姥切国広), a katana by Horikawa Kunihiro designated an Important Cultural Property. Its omote reads 九州日向住国広作 (Kyūshū Hyūga-jū Kunihiro saku) and its ura 天正十八年庚刁弐月吉日平顕長 (Tenshō jūhachi-nen kanoe-tora nigatsu kichijitsu Taira no Akinaga).

1. The blade first. The signature is judged after the shape, the steel and the hamon, never before: if the blade does not belong to the period and school the signature claims, the signature cannot save it. “Famous smiths, period by period” explains why.
2. The sides. Find which side is the omote: here the signature is on the omote and the date on the ura, as on a katana.
3. The place. 九州 (Kyūshū), the island; 日向 (Hyūga), the province; 住 (jū), “resident in”.
4. The name and the ending. 国広 (Kunihiro) and 作 (saku), “made this”. This signature has neither title nor lineage.
5. The date. 天正十八年 (Tenshō jūhachi-nen) is year 18 of the Tenshō era, which began in 1573: 1573 + 18 − 1 = 1590. 庚刁 (kanoe-tora) is the cycle sign, which confirms the year; 弐月 (nigatsu), the second month, with the formal numeral; 吉日 (kichijitsu), “auspicious day”.
6. What follows the date. 平顕長 (Taira no Akinaga) is the man who commissioned the blade: Nagao Akinaga, lord of Ashikaga castle.
7. How it is cut. With a loupe and raking light, look at the direction and depth of the chisel in each stroke, which each smith repeated in his own way; at the patina inside the strokes, which must match that of the rest of the tang; and at the file marks, which lie underneath.
8. The comparison. Compare it with the oshigata (押形) and photographs of secure signatures by the same smith and from the same period, in a reference book of smiths (meikan, 銘鑑). A smith’s signature changes over the years, and the generations who inherit a name each sign in their own way.
9. The paper. An NBTHK paper transcribes the signature as it is; on an unsigned blade it gives the attribution, sometimes with den (伝, “said to be”); and a blade with a false signature is not papered as signed. How to read all this is in “NBTHK papers”.

False signatures: what they are and what they are not

A gimei (偽銘) is a signature bearing the name of someone who did not make the blade. It is as old as the prestige of names, and the most forged are, as one would expect, the most sought after. What is not a gimei is already in this topic: daimei and daisaku, authorised by the master; kiritsuke-mei, which is an attribution; and the signature cut out or folded over when a blade is shortened, which still belongs to the smith.

On old blades, gimei tend to occur between schools related by the craft, through migration, training or descent, whose blades look alike; a kotō dressed up as the work of another period, by contrast, is given away by its steel and its hamon. On shintō and shinshintō blades, whose steels look very much alike from one province to another, a pupil's work signed with the master's name is common, and the names most targeted are Kotetsu (虎徹), Tsuda Sukehiro (津田助広), Inoue Shinkai (井上真改) and Minamoto Kiyomaro (源清麿). The smiths themselves defended their work by changing the handwriting of the signature, adding hidden chisel marks or making the filing of the tang more elaborate, and today, with the study of tangs and signatures so far advanced, many of those gimei can be recognised from the tang alone.

The most famous forger is Hosoda Heijirō Naomitsu (細田平次郎直光), known as Kajihei (鍛冶平), a pupil of Naokatsu, the son-in-law and successor of Taikei Naotane. According to the study Markus Sesko devoted to him, he was the signature engraver of the Naotane workshop, and on the most widely accepted explanation he turned to forgery after 1876, when the ban on wearing swords left smiths without work. He forged everything: signatures added to unsigned blades, new blades signed with old names of kotō, shintō and shinshintō, gold-inlaid cutting-test inscriptions and Hon'ami attributions, and even tsuba. His obsession was Kotetsu: he compiled a book of 65 oshigata of Kotetsu tangs and, according to Sesko, also made oshigata of his own forgeries. His stroke has habits that allow experts to recognise his hand. He was not the only one: in those same hard years, Gassan Sadakazu (月山貞一), who would become an Imperial Household artist, also made blades with the signatures of old masters.

Two practical consequences. A gimei is not fixed at home: erasing a false signature changes the details under which the blade is registered in Japan, and the NBTHK points out, in the FAQ on its website (touken.or.jp), that the board of education that issued the registration must be told beforehand and the blade registered again. And what fails an examination is not thereby declared false: if the signature or the attribution is doubtful, the NBTHK leaves the blade pending, horyū, and a blade without papers may simply never have been submitted (see “NBTHK papers”).

Unsigned is not worse

A mumei (無銘) blade is not a lesser blade for that reason. Some were never signed, others lost their signature when shortened, and others had it removed; almost all the work attributed to Masamune is mumei. Without a signature, the attribution rests entirely on the blade, which is what kantei is about. The cases, the hypotheses and the signatures deliberately erased are in “Kantei: how to look at a blade”; the full vocabulary of the tang, in “Glossary of the Japanese sword”.