Signing for a Lineage: How Museums Describe Inherited Craft Names

In Cleveland there are three small iron objects from Japan on which the name of the maker appears to be the least mysterious part of the record. All three are sword guards. The museum's inscription fields associate all three with Kinai, a resident of Echizen. They entered the collection as gifts of the same donor. The museum dates one to the seventeenth century, another to the middle of the eighteenth and the third to the early nineteenth. Someone looking for the biography of a single artist might conclude that there must be an error. A better place to begin is with the possibility that the three records have preserved exactly what history requires: a long succession of objects, a continuing professional name and more than one person.

The most useful piece of evidence at first has nothing to do with a signature. Look at the accession numbers. The seventeenth-century work is 1919.378, the middle work 1919.311, and the nineteenth-century example 1919.324. The common prefix brings together acquisitions in a museum's history. The last three digits are plainly not a chronology of manufacture. Sort the works by number and the middle object comes first; sort them by the dates assigned to them and a different sequence emerges. Now try sorting by the name in the inscriptions. It will bring all three back together without providing a single new piece of evidence about which generation made which guard.

What Cleveland actually records

The earliest of the three, according to the catalogue, is an iron guard measuring 6.6 centimetres in diameter. In the museum photograph the rim runs around an open design whose broad surfaces make the smaller intervening spaces particularly legible. The record includes a translation of the inscription, gives the work an Edo-period context and credits the gift to D. Z. Norton. These are unusually convenient facts to set out together on a label. They do not all have the same historical source. The name belongs to the inscription field, the period is a catalogue judgment, and the credit line belongs to the history of the object in Cleveland.

Openwork iron guard with rounded shapes inside a narrow rim
Cleveland 1919.378, dated to the 1600s. The museum records the Kinai inscription and a diameter of 6.6 centimetres. Cleveland Museum of Art, gift of D. Z. Norton. CC0.

The second object is a little larger at 7.4 centimetres across. The museum's date narrows the range to the middle of the 1700s. There are similarities between the two designs, including the effect of rounded solid masses against a narrow boundary, but the outlines do not map exactly onto one another. The Cleveland entries do not supply a particular generation number for either. Translating the name on a guard is a way of recording what the institution reads there. Assigning that inscription to the second, third or fourth holder of a name would be an additional act. It would have to be defended on grounds independent of the mere recurrence of the word Kinai.

The published English formulations differ slightly. In some entries the translation includes the idea of making; another records name and residence more succinctly. One should resist the temptation to construct a career out of an English verb that appears in one record and is absent in another. A proper comparison would begin with the original inscriptions and good views of them. It would establish whether a written character is actually absent, whether a cataloguer abbreviated a formula, or whether the difference belongs to translation. There is nothing inconsequential about that distinction. A single editorial omission could otherwise become the supposed identifying feature of an entire generation.

Round iron guard with broad forms and small rounded openings
Cleveland 1919.311, mid-1700s, diameter 7.4 centimetres. The gift credit refers to the collector. Cleveland Museum of Art, gift of D. Z. Norton. CC0.

A professional name on a payroll

Fukui's published prefectural history gives the name Kinai a setting outside collections of guards. Its account of metalworkers links the Takahashi surname to the succession from the second Kinai and cites a domain stipend register from the Tenpō era, in the nineteenth century. The entry identifies a guard maker by the professional name. Here is an issuer of records with a very different purpose from a modern museum. A domain keeping track of allowances needed a way to identify service and payment. A line on a payroll was not intended to illustrate an ornamental style or provide a complete list of signed works. Its very indifference to those modern concerns makes it a valuable independent witness.

This evidence helps explain how a name could remain publicly meaningful over a long interval. A continuing relationship between skilled labour and an institution calls for intelligible occupational identities. One should not infer that the wording of a payroll settles every succession, still less that every guard bearing a particular inscription was supplied to the lord who paid a stipend. The register and the objects bring different things into focus. One records a name in an administrative setting. The others carry words attached to individual pieces of iron. Confronted with both, a historian can ask a much sharper question about the connections than either record would have suggested on its own.

There is a small but important difference between the date range of an era and the date of a specific object. The prefectural history places the cited register within Tenpō, conventionally corresponding to the years from 1830 to 1844. The Cleveland catalogue puts the latest of its three guards in the early 1800s. The two intervals are close enough to make the comparison interesting. They are not enough to put that guard on the desk of a named official in a particular year. A museum visitor has every reason to imagine the larger world in which the object might have circulated. Labelling such a connection as documented would require a matching description or other identifying evidence.

The problem an Oxford cataloguer chose to solve

Albert James Koop's unpublished catalogue of the A. H. Church collection is available through the Ashmolean Museum's Eastern Art Online. Its section on the Kinai school does something a reader might not expect from a discussion of makers: it arranges the guards by subject. Koop explains that the objects are numerous and difficult to classify in other ways. The group includes such subjects as roof tiles, dragons, cranes, shells and plants. A room full of objects has compelled a cataloguer to make a choice. Instead of manufacturing a precise genealogy for each one, the catalogue gives its readers a way to find the objects they want to compare.

Consider what that decision makes possible. Someone studying an image of paper cranes can find it among other birds rather than hoping to guess the name of an individual maker. Someone studying how actual and folded birds differ in metal can find separate entries for them. Such a researcher need not decide which master signed the guards before noticing that the repertory distinguishes a living creature from a human representation of one. At the same time the arrangement does not prohibit research by generation. It merely declines to build the whole catalogue on a distinction that the compiler could not support with equal confidence for every member of a large group.

The Ashmolean section offers several general statements that apply to entries unless a particular exception is identified. This is a familiar device in printed catalogues, where repeating identical material under each number costs space. An online visitor who arrives directly at an individual record may not have encountered the introductory paragraph at all. A careful citation therefore includes both the object identifier and the group introduction when relying on a generalization supplied there. It also distinguishes the compiler's general account of signatures from a new examination of a particular reverse. What a print catalogue can conveniently say once for a shelf of objects ought not silently to become a fresh observation about each digital photograph.

A comparison with Henri L. Joly's catalogue of the Hawkshaw collection complicates the picture in a productive way. Joly noted differences in the characters and in written forms used for Kinai and passed on explanations from earlier sources. In the same tradition of collecting, opinions circulated about the appearance of particular designs and which generations might be associated with them. A historical catalogue is valuable evidence that an opinion was current among its compilers and correspondents. That does not give every statement in it the status of a newly demonstrated identification. When an old attribution is repeated in a modern sale or catalogue, the first question is whether any independent evidence has been added since the earlier publication.

What an early nineteenth-century guard can tell us

The Cleveland guard numbered 1919.324 is seven centimetres in diameter. It is the last of our three in the chronology the museum gives. The photograph shows raised curving forms near the upper rim and open areas around the middle. It has a visibly different balance of ornament and open space from the other two. Even a viewer who cannot read the inscription can compare the ways in which areas of iron and the light around them build a design. The photograph and the museum's reading of the inscription thus furnish two independent starting points for a description. Neither demands that the viewer select a master from a numbered list.

Iron guard with curved raised ornament along its upper edge
Cleveland 1919.324, early 1800s. The inscription record and photographic design record different kinds of evidence. Cleveland Museum of Art, gift of D. Z. Norton. CC0.

A useful thought experiment is to conceal the inscription column and offer the pictures to a class. Ask students to describe the relation between the central opening and the rim. Ask them which edges appear continuous in the photograph and which are interrupted by openwork. Only then reveal the name and the dates. The change in the conversation is usually easy to predict. Students who have been attending closely to individual differences now have a reason to compare the objects as members of a larger family. The best interpretation preserves both insights. The shared name has not made the different outlines less real; nor do the different outlines by themselves identify three distinct people.

The photographs have a practical limitation of their own. Set out on a screen at equal width, the three pieces can look as though they have identical diameters. A diagram drawn to a common scale would show that the largest exceeds the smallest by eight millimetres. That difference is a fact about the catalogue measurements. It does not establish the distribution of sizes across the Kinai tradition. A collection assembled by a donor and represented here by three selected photographs is no substitute for a measured survey. In this instance the modest comparison is worth doing precisely because its limits can be stated in a single line.

The same puzzle in reverse

Another Japanese guard in New York provides a revealing reversal. In the Met's record for 91.1.810, the artist known as Jōi is associated with several names. The museum distinguishes personal names, the signature Nagaharu and art names that include Issandō and Jōi. Here several recorded forms point to the biography of one historical person. In the Kinai group a continuing professional name leads to a succession of people. A database that assumes one name always means one individual will fail in both directions. It may divide one career into several separate people while quietly merging a long line of different craftspeople into one impossibly long life.

Guard with a figure and gold-coloured decoration along the right side
Metropolitan Museum, 91.1.810. The museum links Jōi with multiple names used during one career. The Metropolitan Museum of Art. Open Access, public domain.

The New York guard is also a reminder that recognisable imagery can travel independently of a personal name. Against a broad ground the gold-coloured details and figure occupy a concentrated area on one side. A visitor could inventory that design without having to resolve all of the maker's names; a biographer could make use of the object record without claiming to have identified a school by sight. When the museum offers a possible reading of the subject, the conditional wording matters. There is no obligation to replace it with a definite identification in an English caption written for a wider audience. Precision includes retaining the original degree of doubt.

The Met supplies a second instructive comparison in its record for the guard with paulownia, tendrils and butterfly, accession 91.1.798. Its account describes the name Tō'u on the reverse and notes that the name was used by several generations of the Tsuchiya Yasuchika lineage. The museum gives a broad date rather than a single year or a specific numbered master. A public collection has thus supplied both halves of the problem on its own terms. One entry names several forms associated with one life; another identifies one form that historical practice did not reserve to one life. The examples are far more helpful than a universal rule about how to read Japanese artists' signatures.

The two New York accession numbers also preserve a useful coincidence. Both belong to the Edward C. Moore bequest, which the museum dates to 1891. Placed together under a donor's name they illuminate the interests of a collector and the formation of a museum department. Moved to a list organized under the names recorded on the pieces, they illustrate two distinct relationships between a name and a career. Nothing physical happens to either object in this rearrangement. What changes is the history the visitor is being asked to follow. Keeping the accession and the credit attached allows someone to retrace the route from one comparison to the other.

School, hand and authentication

Two American university museums make a useful small comparison possible through their online records. The Spencer Museum of Art in Kansas uses the Kinai school identification for a Japanese guard. The University of Michigan Museum of Art likewise associates a guard bearing three crests and vines with the school. The latter's discussion raises a possible connection between the crests and the owner. That question concerns a person who may have possessed the work, not the identity of the craftsperson who made it. A school name, an inscription and a heraldic device can occupy the same catalogue entry without competing to name the same individual.

A different institutional practice can be seen at the Sano Art Museum. Its description of a kozuka with an oar design separates the person to whom the work is attributed from the later Gotō master who authenticated it. A name added as an authentication is a historical act in its own right. Its value lies partly in identifying who stood behind a judgment and when that judgment could have been made. It would miss the point to read every written name as an equal claim to have personally made the piece. This is a useful comparison with the Kinai group precisely because the historical relationships are different. No parallel authentication is being proposed for the three Cleveland guards.

Here is one way to write an object label that makes room for these distinctions. Begin with the name of the object and its museum identifier. Put the inscription or the institution's reading of it on a separate line. Describe the dating as precisely as the record warrants, then state whether the museum identifies a person, a lineage or a school. Add donor information under its own heading. If a later scholar has proposed a more exact attribution, supply that scholar's name and reason. The result can be shorter than a paragraph. Its essential virtue is that a reader can see which claim would have to change if the specialist attribution were revised tomorrow.

There is an intellectual pleasure in returning to the smallest piece after this circuit through institutions. The inscription belongs to a particular object one could hold in a palm. The name it carries could connect that object to a province, to a professional tradition that stretched beyond a lifetime, to a stipend recorded by a domain, and eventually to a set of entries compiled a continent away. None of this requires an invented date of birth for the particular person who made the guard. The surviving evidence can reveal how a name worked in public while keeping the individual behind one instance of it just beyond reach. That is an achievement of good cataloguing and an invitation to further historical work.

Questions from the collection

Why do Cleveland’s three Kinai inscriptions not imply a single artist?

The museum dates the three guards to the 1600s, mid-1700s and early 1800s. The Kinai professional name continued across generations in Echizen. The inscriptions do not individually assign a numbered master to these objects.

Which of the three Cleveland guards is the earliest?

Accession 1919.378 is dated to the seventeenth century. The shared 1919 prefix concerns the collection history; ordering the final accession digits does not order the guards by manufacture.

What does the Fukui stipend record add?

The prefectural history cites a Tenpō-era domain register identifying a guard maker as Kinai. It documents the name in an occupational context; it does not identify a particular Cleveland guard as a domain commission.

Why did Koop arrange the Ashmolean Kinai guards by subject?

In his catalogue of the Church collection he describes the problem of classifying a large number of related guards. Grouping them by imagery lets a reader locate designs without relying on a precise generational attribution for every entry.

Can one historical artist also appear under several names?

The Met’s record for 91.1.810 gives several names associated with Jōi, including Nagaharu and Issandō. This is the reverse indexing problem from a succession of artists sharing one professional name.

How is authentication different from a maker’s signature?

An authenticator records a judgment about authorship. The Sano Art Museum distinguishes an attribution to Gotō Yūjō from authentication by a later Gotō master on a kozuka. The two names describe different historical roles.

Museum films

JASA: Japanese Art Society of America — Samurai Splendor: Sword Fittings from Edo Japan (Lecture by Markus Sesko)

The Met — The Japanese Sword and the Japanese Aesthetic

Collection records and further reading

Related Articles