How to Read a Japanese Sword: Mei, Signatures and Appraisal
The most valuable part of an old Japanese sword is not the edge. It is the tang, the hidden root of the blade, because that is where the sword keeps its record: the signature, the file marks, the shape, the patina of centuries. This is a plain-language guide to reading that record, why a signature is a clue rather than proof, and how formal appraisal actually decides what a blade is.
On this page
The tang is the document
Take the handle off a Japanese sword and you reach the nakago, the tang, the unpolished extension of the blade that sits inside the hilt. To the untrained eye it looks like the least interesting part: dark, rough, rusty. In fact it is the single most important part of the sword for anyone trying to understand what the blade is.
The nakago is, quite literally, the sword's document. It carries the signature if the sword is signed. It carries the file marks the smith cut into it. Its shape, its taper, and the form of its end all reflect the period and the maker. And its surface carries a slow, even patina, the dark aged rust that only forms over a very long time, which is itself evidence of age.
Everything else on this page follows from that idea. Reading a Japanese sword means reading the tang, and treating it as an archive to be preserved rather than a surface to be tidied.
Mei basics: signed and unsigned
The signature on a Japanese sword is called the mei (銘). It is not stamped or engraved by machine. It is cut by hand into the tang with a chisel, stroke by stroke, traditionally by the smith or by a trusted hand in the workshop. Because it is hand-cut, the way the strokes are formed is itself part of the evidence, which becomes important later.
Every sword falls into one of a few basic states:
- Zaimei (在銘) means the blade is signed. The tang carries a mei. On its own this tells you a signature is present, not that it is genuine, which is a distinction we return to below.
- Mumei (無銘) means the blade is unsigned. Crucially, mumei does not mean anonymous or worthless. A blade can be mumei because the smith never signed it, or, very commonly for older blades, because the tang was shortened at some point and the original signature was cut away in the process. Many highly regarded blades are mumei.
- Gimei (偽銘) means a false signature, a mei that was not cut by the smith it names. This is far more common than newcomers expect, and it has its own section below.
The single most useful habit to adopt early is this: a signature is where the inquiry starts, not where it ends. The presence of a famous name on a tang is a question to be tested, not an answer to be trusted.
Which side? Katana-mei and tachi-mei
Here is one of the most elegant facts in the whole subject. A signature usually tells you not just who made a sword but how it was meant to be worn, because of which side of the tang it sits on.
The convention is that the smith signs the side of the tang that faces outward, away from the wearer's body, when the sword is worn correctly. The catch is that the two great Japanese long-sword forms were worn in opposite orientations:
- A tachi was slung edge-down, suspended from the belt. Its signature, the tachi-mei, is cut on the side that faces out when worn edge-down.
- A katana, in the later uchigatana style, was worn thrust through the sash edge-up. Its signature, the katana-mei, is on the side that faces out when worn edge-up.
Because the two are worn edge in opposite directions, the signatures end up on opposite sides of the tang. There is a simple practical test used by collectors: hold the sword as a tachi, edge down, and look at the mei. If it reads the right way up in that position, it was signed as a tachi. If it appears upside down, it was signed as a katana. It is one of the quickest ways to read the intended mounting straight off the tang. For how these forms differ in shape and use, see our reference on the types of Japanese swords.
What a full signature contains
A short mei might be nothing more than a smith's name. A longer, fuller signature can read almost like a formal line of address, and each element adds information. In a fuller signature you may find, in varying combinations:
- The smith's name. The core of any mei. Note that smiths often took art names and that names were passed down lines, so a name identifies a working identity, not always a single individual.
- A clan or family prefix. Signatures frequently open with a clan affiliation. Fujiwara (藤原), Minamoto (源), and Taira (平) are all common prefixes, particularly in later signatures.
- A province or place of residence. Where the smith worked, such as the old provinces of Bizen or Mino, often written with the character for province or country, 国 (kuni).
- An honorary title. Some smiths held granted titles that appear in the mei, such as 守 (kami) or the higher 大掾 (daijo). These are honorifics, not job descriptions.
- A date. A dated blade often gives the era name (nengo) and year, sometimes with the season or month. The date is written in the traditional Japanese era system rather than in Western years.
- A making-verb ending. The signature commonly closes with a character meaning made, most often 作 (saku) or 造 (tsukuru), sometimes expanded to a phrase meaning made this.
Read together, a full signature can tell you the maker, his stated lineage, where he worked, any title he held, and when the blade was made. That is a remarkable amount of information cut into a strip of steel, which is exactly why the same information was such a tempting target for forgery.
The gimei reality
This is the section every honest guide has to include and many quiet sales pitches leave out. A large share of signatures on old Japanese swords, especially the big, famous names, are gimei: false signatures, added later, to make a blade appear to be the work of a more celebrated smith than the one who actually forged it.
The pattern is consistent enough to state as a rule of thumb: the bigger and more valuable the name, the higher the chance the signature was forged. Famous smiths were forged precisely because their names carried the most value, so a tang bearing a legendary name should raise more suspicion, not less. The problem is most associated with older blades of the koto, shinto, and shinshinto periods, and is generally far less of an issue with twentieth-century blades, where signatures are usually contemporary and reliable.
None of this is a reason for cynicism, only for method. The existence of gimei is exactly why a formal appraisal system grew up around these swords, and why the tang is read as a whole, signature plus file marks plus shape plus patina, rather than trusting the name alone.
How shinsa appraisal works
Because a signature cannot be trusted on its face, the Japanese sword world relies on formal appraisal, called shinsa. At a shinsa, a blade is submitted to a panel of highly experienced judges who examine it and issue a paper recording their finding. The best known of these systems developed from the 1950s onward under the major sword-preservation body usually abbreviated NBTHK, which we describe in our companion pages on the craft.
At a high level, appraisal does two different jobs depending on the blade:
For a signed blade
The panel judges whether the signature is consistent with genuine work of the named smith. A blade can pass with its signature accepted, or the signature can be rejected as gimei. This is the moment a hopeful name either survives scrutiny or does not.
For an unsigned (mumei) blade
With no name to test, the panel does something more interesting: it attributes the blade. Reading the steel, the shape, the hardening, the file marks, and the tang, the judges assign the blade to a school, a tradition, and a period, and sometimes to a specific smith. That recorded attribution is what gives a mumei blade a documented identity. An entry-level paper broadly confirms the blade is a genuine traditionally made Japanese sword consistent with the stated period and school, without disqualifying flaws; higher-level papers indicate a more specific and more confident judgment, and are not granted automatically. A blade holding a lower paper has to be resubmitted to earn a higher one, and many do not pass the next tier.
Oshigata: the traditional record
Before photography, and still today alongside it, the traditional way to record a tang was the oshigata (押形), a careful rubbing or drawing that reproduces the tang and its signature at life size. In the classic method, thin paper is laid over the tang and worked with ink so that the raised chisel strokes of the mei and the file marks come through as a faithful impression.
The oshigata is more than a keepsake. It is a comparison tool. Because the mei was cut by hand, a genuine smith's signature has a characteristic hand, the way particular strokes are shaped, spaced, and terminated. Laying an oshigata of a questioned signature next to authenticated examples of the same smith often reveals discrepancies that are hard to see on the metal itself, where lighting and rust can hide them. In an oshigata, the strokes are flattened onto paper and easier to compare directly.
For a serious student, building or studying oshigata is how the eye is trained: not by memorizing that a name is good, but by learning what a particular smith's hand actually looks like, so that a false version of it stands out.
Yasurime and tang shape as evidence
Even a completely unsigned tang is full of information, because two features speak without any writing at all: the file marks and the shape.
Yasurime, the file marks
When a smith finished a tang, he dressed its surface with a file, leaving a deliberate pattern of file marks called yasurime (鑢目). These patterns are not random. Different schools and different periods favored characteristic file-mark styles, so the yasurime help place a blade in time and tradition. They are also a feature that forgers routinely get wrong, because faking the correct period file marks convincingly is harder than faking a signature, which makes them a valuable cross-check against the mei.
Nakago shape
The overall form of the tang is evidence in its own right: its length and taper, the curve, and especially the shape of its end, the nakago-jiri. These forms changed across periods and varied between schools, so the silhouette of a bare tang narrows down where and when a blade belongs. When a signature, the file marks, the tang shape, and the patina all point the same way, confidence is high. When they disagree, that disagreement is itself the finding, and usually a warning.
This is the heart of reading a sword: no single feature decides the case. The tang is read as a whole, and the judgment is the sum of agreeing evidence.
Reading the kanji
You do not need fluent Japanese to begin recognizing the building blocks of a mei, and the effort pays off directly in a kitchen too, because many of the same characters appear on knife blades and packaging. The overlap with kitchen-blade script is close enough that our guide to Japanese knife kanji is a natural companion to this section. All characters below are standard Japanese kanji.
| Kanji | Reading | Meaning in a mei |
|---|---|---|
| 備前 | Bizen | An old province, one of the great historic sword-producing regions. |
| 美濃 | Mino | An old province, another major historic center of swordmaking. |
| 国 | kuni / koku | Province or country. Often follows a place name (as in the province of a smith's residence). |
| 守 | kami | An honorary title, roughly guardian, granted and appearing after a place name. |
| 大掾 | daijo | A higher honorary title than kami, again an honorific carried in the signature. |
| 藤原 | Fujiwara | A clan name used as a prefix, very common in later signatures. |
| 作 | saku | Made. A common closing character meaning made by. |
| 造 | tsukuru | Made or forged. An alternative closing character to saku. |
Recognizing even this handful transforms a mei from an intimidating column of marks into something you can partly parse: a name, perhaps a clan, a place, maybe a title, and a made-by ending. It will not make you an appraiser, and it should not, given everything above about gimei. But it turns the tang from a mystery into a text you can begin to read, which is the honest goal of this page.
Frequently asked questions
Should I clean the rusty tang to see the signature better?
No. Never clean, sand, brush, or polish the nakago (tang). Its dark aged patina is one of the main ways an old blade is dated and authenticated, and removing it destroys evidence and value permanently. If a signature is hard to read, the right tools are good lighting and a rubbing (oshigata), not abrasion.
Is an unsigned (mumei) sword worth less than a signed one?
Not necessarily. Mumei simply means unsigned, often because the tang was shortened and the original signature was lost. Many highly regarded blades are unsigned, and a mumei blade with a strong formal attribution can be worth far more than a signed blade whose signature turns out to be false.
How common are fake signatures on Japanese swords?
Common enough that a signature is treated as a claim to be tested, not proof. False signatures (gimei) are especially frequent on famous names, because those names carried the most value. As a rule of thumb, the bigger the name, the higher the chance the signature was added later. The issue is concentrated in older blades and is far less of a problem with twentieth-century swords.
What does it mean when an unsigned blade is attributed to a smith or school?
At a formal appraisal (shinsa), expert judges read the steel, shape, hardening, file marks, and tang, then attribute an unsigned blade to a school, tradition, and period, sometimes to a specific smith. That recorded attribution is an expert judgment, not a signed statement from the maker, but it gives a mumei blade a documented identity and is highly valued.
Can I tell if a sword was made as a katana or a tachi from the signature?
Usually, yes. Smiths signed the side of the tang that faced outward when the sword was worn. Because a tachi was worn edge-down and a katana edge-up, their signatures sit on opposite sides. Hold the blade edge-down: if the signature reads upright it is a tachi-mei, and if it reads upside down it is a katana-mei. Treat this as a strong clue read alongside the other evidence, not as a sole verdict.
Sources
Nihonto reference and glossary material on the nakago, mei, and signature conventions; sword-appraisal and preservation writing on shinsa, attribution, and certificate levels; and reference material on oshigata, yasurime, gimei, and mei kanji. Where sources differ, in particular on the exact wording of appraisal tiers and on how strictly signature orientation should be read, we have described the standard conventions and flagged them as conventions and clues rather than absolute rules.
The same script, on a knife you can use
Okami does not sell, appraise, or broker swords, and we are not dealers or appraisers. But the characters you have just learned to read carry straight into the kitchen, where they appear on Japanese blades and their makers' marks. Our Premium Damascus chef knife brings that lineage into a tool for everyday use: an AUS-10 core, clad in 67 layers, hardened to 60 to 61 HRC, ground thin and hand finished. $199, direct.
Shop the Premium Damascus Read Japanese Knife Kanji