How to Read Japanese Pottery Marks and Potter's Signatures (Mei)
Share
A Japanese pottery mark — mei (銘) — is read in three passes: where it sits, how it was made, and what kind of name it is. The third is the one collectors skip, and it is the one that decides what the mark is actually worth as evidence.
Before any of that, though, an expectation needs adjusting. Most Japanese tea ceramics carry no mark at all, and on the ones that do, the name is more often on the box than on the pot.
Why So Much of It Is Unsigned
In the European tradition, a signature on the work is the normal way authorship is recorded. In Japanese ceramics it is one option among several, and for long stretches of history it was not the usual one.
Tea wares from the older kilns — Karatsu, Bizen, Shigaraki, Iga, much early Hagi — are overwhelmingly unmarked. The pot carries no name because the name was not the point: these were kiln products, identified by clay and firing rather than by hand. Authorship attached itself later, through the box, through the inscription of a tea master, through a documented chain of ownership.
The practical consequence is that an unmarked bowl is entirely normal and not a warning sign, while a marked one has simply handed you one more thing to check. How the box carries authorship instead is set out in How to Read a Tomobako.
Where a Mark Sits
| Position | What is usually there | How much it settles |
|---|---|---|
| Under the foot, or on the foot ring | An impressed seal, or a carved character | The most common place for a potter's own mark |
| On the side of the body, low down | An impressed studio or kiln seal | Often a workshop rather than an individual |
| Inside the lid of the box | Brushed name, seal, and often the poetic name of the piece | Where authorship most often actually lives |
| On an outer box | A tea master's inscription | Standing within tea, not proof of maker |
| Nowhere | — | Normal. Read the clay, the foot and the firing instead |
The Four Kinds of Mark
- Impressed seal (印). A carved stamp pressed into damp clay. Edges are crisp where the clay was soft and the stamp clean, and the same seal recurs across a maker's output — which is what makes comparison possible at all.
- Incised mark (刻銘). Cut into the clay with a tool rather than stamped. The line is angular where a seal's is even, and — unlike a stamp — it differs from piece to piece, because no hand carves the same character twice identically. That variation is why an incised mark is compared for character rather than for exact match.
- Brushed mark (書銘). Painted in iron or cobalt under the glaze, or in enamel over it. Under-glaze marks sit beneath a glassy layer and cannot be altered without disturbing it. Over-glaze marks can be removed.
- Kao (花押). A stylised monogram rather than a written name, usually brushed on a box rather than on the pot. It is the hardest element to imitate convincingly, because it is a gesture rather than a set of characters.
How to Work Through One
- Photograph it straight on, with raking light. An impressed seal is a relief, so light from the side reveals what light from the front flattens. Most unreadable marks are badly lit, not illegible.
- Establish how many characters there are before trying to read any of them. Two, three and four-character marks divide the field very differently, and a miscount sends you looking in the wrong place.
- Identify the script. Block script (kaisho) can be looked up character by character. Semi-cursive (gyōsho) is harder. Full cursive (sōsho) abandons the standard forms entirely — it defeats character-recognition software and most reference books, and is where a specialist is genuinely needed.
- Strip the suffixes before you search. 造 (zō) and 作 (saku) both mean "made by"; 謹製 (kinsei) means "respectfully made". They are not part of the name. Searching for the name plus its suffix is one of the commonest reasons a mark appears to have no match.
- Check for 写 (utsushi). A mark including this character is telling you the piece is an openly made copy after an earlier work, which is a tradition rather than a deception — see antique or modern reproduction.
- Compare against a documented example from the right period. Seals change across a working life. A match against the same name is not a match; a match against the same name at the same stage of that career is.
- Read the pot against the mark. A seal narrows the field, but a seal is also the easiest element to copy. Clay, foot and firing are not. When they disagree with the mark, they are the better evidence.
What a Name Does Not Tell You
The largest trap is not forgery. It is that many Japanese pottery names are inherited.
Under shūmei (襲名), a successor takes the name of their predecessor on assuming the headship of a house. The same characters therefore appear on work made a century apart by different hands. Raku, Eiraku and Ōhi all work this way, and so do many smaller lineages. A name alone identifies the house; only the generation identifies the person, and the generation usually has to be read from the form of the seal rather than from the name.
The Raku line is the clearest worked example of reading a generation from a seal, and it is set out in detail in How to Read a Raku Tea Bowl.
Three smaller cautions. First, not every character on a pot is a name at all. A bowl may carry an auspicious word — 寿 (kotobuki, longevity) is the most common — or a poetic name given to that particular piece. Those are inscriptions, not signatures, and reading one as a maker's mark sends you looking for a potter who does not exist.
Second, a mark may name the kiln or studio rather than the person who made the piece, which for a workshop of any size is a different claim entirely. And third, if a mark reads "Nippon", "Made in Japan" or "Occupied Japan", the object is export ware rather than tea ceramic — those phrases exist because of foreign customs regulations, and they place a piece in a broad period but tell you nothing about a maker.
Pieces are listed with their box inscriptions and marks described among the tea bowls.