In 1891 two Americans gave the Metropolitan Museum of Art a group of Japanese objects. The surviving catalogue entries repeat their names in the same order, Brayton Ives and W. T. Walters. Beneath that uniform line are things with strikingly different histories. A short sword whose fittings carry old coins appears beside another whose ornaments can be dated across several centuries. A dagger with a fifteenth-century blade shares its donors with a blade made in 1839. One of those blades was centuries old when the gift reached New York. The other was younger than some of the people who could have attended its first display.
To see what the gift brought to the museum requires more than counting the word “sword” in a register. It brought old and comparatively recent work into the same institution. It preserved earlier metalwork in later settings. It made the names of two American collectors easy to recover while leaving many earlier owners hard to identify. Most remarkably, the modern catalogue lets us walk through this group at several speeds. We can move quickly along a sequence of accession numbers, then come to a complete halt at an apparently ordinary entry that contains six different statements about age.
Start with an object you cannot see
The object that prompted this investigation is a dagger, accession 91.2.37. The museum assigns its blade to the sixteenth century and its mounting to the late eighteenth or early nineteenth century. Its materials include steel, wood, lacquer, ray skin and gold. A reader need not guess a decoration, a maker or an owner to understand the significance of the entry. A single catalogue title can preserve three widely separated times, the making of a blade, the making of a mounting, and the transfer of the assembled object to a museum in New York.
This record also exposes a bias built into almost any illustrated account of a collection. Some of the objects in the group have several publicly available photographs and others have none in the public collection data consulted here. The entry for 91.2.37 remains valuable even without a downloadable photograph. Substituting an unlabelled picture of its neighbor would make an elegant page and create a false identification. The image below therefore belongs openly to 91.2.32, a different dagger from the same gift. Its value as an illustration is that the reader can tell exactly what kind of comparison it supports.

The neighboring dagger shares a sixteenth-century blade date, while its mounting is assigned to the nineteenth century. Wood, lacquer and metal recur, yet the lists of materials for the two objects do not coincide. The second record includes a copper-gold alloy and brass. There is no basis here for merging the two into a typical dagger of 1891. In fact the phrase would be a poor description of either work. What happened in 1891 was a change in the place and conditions under which they could be seen, compared, named and eventually reproduced.
Two collectors and two cities
The W. T. Walters of the New York credits was William Thompson Walters. His name deserves to be written out because his son Henry occupies so large a place in histories of collecting. William was active in Baltimore business and built a collection there. In the 1870s he opened his house on occasion to visitors, with admission proceeds benefiting a local charitable association. As the collection grew he added a gallery, opened in 1884. Those dates place the 1891 gift in a world in which the boundary between a collector’s house and a public exhibition could already be crossed by members of the public.
Research published by the Walters Art Museum adds a useful geographical dimension. In London in 1862 the family saw Japanese and Chinese art at the international exhibition. At Philadelphia in 1876 William acquired objects from Japan and China as well as other regions. These fairs encouraged comparisons across categories that are now divided among museum departments. They do not establish where William and Ives obtained the individual works given to New York. A fair that demonstrably informed a collector’s interests is one kind of fact. A receipt identifying a particular purchase at the fair would be another.
William died in 1894, three years after the Metropolitan gift. Henry inherited his collection and continued to expand it. The Baltimore museum opened as a public institution in 1934, after Henry’s death. A modern reader encountering the Walters name on a New York label could easily imagine a gift from that museum. The dates make clear why this would be anachronistic. The institutional history in Baltimore helps us understand one donor’s public life. It cannot replace the Metropolitan credit, which names a joint gift by two individuals and gives a date within the father’s lifetime.
Brayton Ives shares that credit throughout the sample. The order of the names is not evidence for the division of costs, the choice of objects or the part each man played in the transfer. There are many ordinary historical circumstances under which two people might be credited with one gift. To assign one the role of financier and the other that of connoisseur would require a letter, a contemporary list or comparable evidence. Keeping both names together where the evidence joins them is more informative than supplying an attractive division of labor that the labels do not record.
Money on a mounting
Among the most legible objects in the group is a short sword, accession 91.2.52. The maker decorated the fittings with representations of old coins. The Met identifies them as auspicious. The scabbard has a ribbed effect, and the museum connects the form of its end to earlier equestrian traditions. These features allow a remarkably compact object to contain several kinds of historical memory. Coins look back toward an earlier form of currency. The shape of a fitting evokes an older style. The sixteenth-century blade carries another history again within a nineteenth-century setting.

An actual coin, a representation of a coin and a picture of the representation are three different sources. The first might be studied for minting or circulation. The second asks what made an image of old money attractive to someone commissioning personal furnishings. The third allows the design to travel to readers who will never see the metal firsthand. The gift preserved the second of these. The museum photograph supplies the third. A catalogue caption that simply says “coins” leaves the reader uncertain about which of these things is present and can start an entirely unnecessary hunt for evidence of currency embedded in the object.
The old-coin design also demonstrates why the date of a mounting is a historical result in its own right. A reader focused exclusively on the older blade might call everything surrounding it a late addition and hurry on. For someone studying nineteenth-century taste, the priorities reverse. The artist and customer working in that century made a conscious set of choices about which older forms should become visible again. The assembled object preserves those choices at a scale suited to close observation. An institutional collection that kept only blades would have lost much of the evidence needed to ask the question.
Fifty-two years and several centuries
One entry bears an unusually easy date to keep in mind. The blade of accession 91.2.56 is dated 1839. Simple subtraction gives a fifty-two-year interval before the gift. Its mounting is broadly assigned to the nineteenth century. Here a precise year belongs inside a wider chronological category and both are distinct from the acquisition year. A museum visitor in the early 1890s could have regarded this blade as the work of a relatively recent generation. From a distance of much more than a century it is easier to forget how close those dates once seemed.

Set this beside 91.2.40, the dagger with a fifteenth-century blade and nineteenth-century mounting. The Met lists wood and lacquer alongside silver, gold and steel. The person who assembled or received the later form lived at a very different historical distance from its blade than a contemporary of the 1839 work did. The two objects reached New York under the same credit, but a calendar attached only to the gift cannot convey this difference. A collection is a powerful way to compare how the past was present to different people at different times, including a past already centuries old in Japan.
Other dated examples broaden the picture. Accession 91.2.47 carries the date 1626 for its blade. The Kuniyoshi blade catalogued as 91.2.54 is dated 1525. The latter’s public record also distinguishes names associated with the fittings and the scabbard. A sequence ordered by the date of the steel therefore diverges sharply from a sequence ordered by the museum numbers. That divergence matters even for a short exhibition label. It permits a curator to organize one case around historical chronology and another around a moment of American collecting without pretending the two cases tell precisely the same story.
Six dates in one entry
Read the catalogue date of 91.2.48 aloud and it starts to sound like a compact exhibition. The blade is assigned to the middle or later fourteenth century. Grip ornaments are assigned to the late sixteenth. A collar and an accessory are placed in the seventeenth; the guard spans the late seventeenth and early eighteenth centuries. The assembled mounting is placed in the late nineteenth. None of those pieces would by itself give a fair date for the experience of seeing the whole. The entry takes the time to identify what each date belongs to.
What should a museum do with an object like that in a chronological gallery? A case headed Fourteenth Century could invite a visitor to treat the ornaments as material evidence for a much earlier period than their own. A case headed Nineteenth Century could conceal the significance of the oldest part. One elegant solution would be to display the assembling of periods as the subject of the case. A drawing keyed to the dates could make centuries visible in space. This is a proposal for interpreting the surviving record. It is not a reconstruction of a display known to have been installed in 1891.
For a researcher there is a further consequence. A database search for the late sixteenth century might legitimately find this work because a pair of small ornaments belongs there. Another search for nineteenth-century mountings might find exactly the same accession. Counting search results without recording which component was matched could count a single assemblage twice or classify it under the wrong century. The museum’s prose date field contains enough detail to avoid either problem, provided its reader does not flatten it into a single number before asking what the data should measure.

Who entered the later history of the gallery?
The museum’s retrospective exhibition Discovering Japanese Art described a collection that began with a relatively small and varied group of decorative objects and was transformed by successive collectors and curators. Bashford Dean occupies an important place in that account of arms and armor. His work in Japan, his exhibitions and his publications helped shape what later visitors could see and read. The Ives and Walters credits give that broad history a valuable earlier layer. A visitor may encounter the surviving gift in an installation shaped by scholarship that developed only after the objects reached the institution.
That sequence is familiar in museums and easy to reverse in a short account. A curator whose research first made a subject widely legible can become so prominent that the earlier arrival of objects seems merely a prelude to the curator’s career. Records like 91.2.37 permit another emphasis. Two donors made the transfer; contemporary Japanese makers had already contributed layers of interpretation through their mountings; subsequent curators made further distinctions among the surviving parts. Today a photographer and a cataloguer shape the terms on which a remote reader encounters the same work. Each belongs to a different part of its public history.
One more name is conspicuous by its absence. These credits do not identify every Japanese owner who valued and preserved the objects before 1891. The silence cannot be repaired simply by identifying a likely American place of purchase or naming the city in which a documented maker worked. Origin, production, ownership and acquisition each require their own evidence. A research note that leaves a space open for a former owner is more likely to recognize a future discovery than a smooth narrative that assigns every missing event a plausible participant.
Return finally to the dagger without a readily available photograph. Its donor line is brief, its date line precise about the limits of precision, and its unillustrated entry comparatively easy to overlook. Yet it leads to a gallery in Baltimore, an international exhibition in Philadelphia, a group of old coins in miniature and a short sword whose separate elements span much of Japanese history. That is what a joint museum gift can bring to a city. It gives later readers the chance to find relationships across objects without forcing them to share a single date, a single previous owner or a single reason for surviving.
A small exhibition that could be made on paper
There is an illuminating way to arrange these objects without moving any of them. On a sheet of paper, put each accession in the order in which it appears in the museum catalogue, then write the earliest date and the latest component date beside it. Some lines on this imaginary chart would be relatively short. The blade dated 1839 and its nineteenth-century mounting would occupy much the same neighborhood. The line for the short sword with the early blade and several generations of ornaments would stretch across a large part of the page. The uniform donor credit would run down the sheet like a margin. What looked like a single group at the start would acquire a distinctive historical profile for each entry.
The same chart could support an entirely different exhibition about certainty. A dated blade provides one sort of anchor. A broader attribution such as sixteenth century gives another. Accession 91.2.31 goes one step further: the museum treats the fourteenth-century date of its blade as a possibility. A picture caption with space for only a date might quietly lose that qualification. On paper it would be easy to draw the corresponding part of the line in pencil or use a pair of brackets. Historical uncertainty has a shape when a designer allows it to have one. The result gives readers more information than a row of equally definite century labels.
A third arrangement would begin with the parts least likely to receive a full-page photograph. One display could gather the collar, ornaments and accessory catalogued within 91.2.48. Another could bring the old-coin fittings into a comparison with representations of money elsewhere in the museum, making explicit that the coins in this work are images. A third could keep the name of the donor constant while changing the place and period of every illustrated component. All three would offer legitimate ways of exhibiting the collection, and each would bring a different set of makers and viewers back into focus.
This is where the modest unillustrated entries do some of their best historical work. They prevent the collection from being defined exclusively by the objects that reproduce well on a small screen. They also give a reader a way to test an apparent pattern in illustrated pieces. If all the images on a page show a dark scabbard or a dated blade, one can look at neighboring entries before attributing that choice to the two donors. Sometimes the selection tells us about nineteenth-century collecting. Sometimes it tells us about which photographs were ready to publish. The surviving register provides a means of asking which explanation fits.
Questions about the New York gift
Who are the two donors named in the 1891 gift?
The Met credits Brayton Ives and W. T. Walters. The latter was William Thompson Walters, the Baltimore collector and father of Henry Walters. Henry later inherited his father’s collection. His subsequent role in Baltimore should not be substituted for William’s name in the New York gift record.
What did the gift include besides daggers?
The same credit appears on short swords and longer swords with their mountings. One short sword carries old-coin designs and another has dates for six different components or groups of components. The dated 1839 blade is a further example. These records document variety without constituting a complete inventory.
Which number identifies the sixteenth-century dagger featured here?
The Met accession is 91.2.37 and the museum’s digital record identifier is 21874. The blade is assigned to the sixteenth century and the mounting to the late eighteenth or early nineteenth century. Use the accession when comparing catalogues and retain the digital identifier when linking directly to the online description.
Can every date in one accession be reduced to a single century?
Doing so would discard valuable evidence. For 91.2.48 the dates attached to the blade, grip ornaments, collar, accessory, guard and later mounting span centuries. A chronological account can follow each part in turn while retaining the single accession under which the composition entered the museum.
Did the gift travel to New York from the Walters Art Museum?
The New York gift dates to 1891. William was then alive; the Baltimore institution opened as a public museum in 1934. Its history provides a context for the donor, but the joint gift should be cited from the New York credit and not treated as a twentieth-century transfer between museums.
Is a missing online image evidence of a lost object?
A catalogue entry can describe a work without supplying a downloadable photograph. A neighboring accession with an image may be used for a clearly identified comparison. Establishing the loss, disposal or physical whereabouts of an object would require evidence beyond the presence or absence of an online photograph.
Museum films
The Met — The Japanese Sword and the Japanese Aesthetic
The British Museum — How the samurai sword evolved over 1000 years | Curator's Corner S10 Ep11
Collection records and further reading
- Dagger 91.2.37
- Dagger 91.2.32
- Dagger 91.2.40
- Coins on the mounts of 91.2.52
- Dated 1839, 91.2.56
- Six component dates, 91.2.48
- Dated 1626, 91.2.47
- Kuniyoshi accession, 91.2.54
- The Walters Art Museum, history of the founders
- Journal of the Walters Art Museum, volume 59
- Discovering Japanese Art: American Collectors and The Met
- Dagger with a qualified date, 91.2.31
- The Gotō Family Across Generations: A Tokugawa Daishō as an Archive
