A pair that lives at two addresses
In Kyoto National Museum's exhibition on Kitano Tenjin, the two blades occupied consecutive entries. Number 131 was a tachi from Kitano Tenmangu Shrine, known as Onikirimaru or Higekiri. Number 132 was a tachi from Daikaku-ji Temple, known as Usumidori or Hizamaru. Both were identified as Important Cultural Properties. On the exhibition page the photographs sit close enough for a reader to compare the whole outlines. At the end of a loan, however, the two objects have different homes. Their ability to make sense as a pair across that separation is one of the most interesting things about them.
The word twin in the familiar story is a claim about a relationship in cultural memory. The exhibition labels do not describe two pieces made to a common drawing, signed by the same person or acquired by the same institution. Instead they introduce two surviving tachi around which traditions of the Minamoto house have gathered. Visitors meet the relationship long before they learn the difficult details of either object. The names travel together in theater, images, commemorations and contemporary exhibitions. To understand the force of the pairing it helps to spend some time with the two records separately, then ask what happens when a curator reunites them.
In the Kyoto exhibition, the tachi appeared within a history of devotion to Tenjin. The setting matters. It put works familiar to audiences through the names of warriors into an account of the cultural life of shrines. The museum discusses offerings of swords and tells of patrons whose association with Tenjin affected what they gave. An object might enter sacred custody because its donor's family relationships, religious commitments and public identity converged on a particular institution. Reading the blades only as illustrations from a chronicle of warriors would leave out the people who kept, named and displayed them through long intervals of peace.

How to read four names for two tachi
Higekiri and Onikirimaru identify the Kitano object in the exhibition. Hizamaru and Usumidori identify the Daikaku-ji object. These pairs of names are not interchangeable with the names of smiths. The inscription field of a museum record answers one question, the familiar name another. Even a visitor who reads no Japanese can keep the distinction straight by noting where the name appears in the label. A phrase introducing what an object is commonly called carries a different weight from the line where a curator transcribes an inscription on its tang. A single exhibit can have both kinds of names and a history attached to each.
Usumidori supplies a striking example. The word evokes a pale green color. Daikaku-ji relates a transmission account in which the tachi received that name in connection with Minamoto no Yoshitsune. The account is evidence for how the temple presents the history of the name. It does not require the visitor to look for a green tint in the steel or expect every photograph of the object to reproduce a particular hue. As with a personal nickname, the sound of the name can preserve a story long after the conditions under which it first made sense have become difficult to reconstruct.
In writing about these objects it is useful to retain the name used in the source for as long as possible. A modern museum caption calling the temple work Usumidori can be connected to Hizamaru by a carefully placed parenthesis. A historical narrative that mentions a Minamoto heirloom cannot be made into an accession register by silently replacing the narrative name with a museum's present title. Doing so would decide in advance precisely the historical problem under investigation. The Kyoto museum's phrasing leaves room for an association sustained by tradition. That is a more informative description than an unqualified claim to have identified two objects from a story.
A damaged character at Daikaku-ji
The temple dates its tachi to the thirteenth century, in the Kamakura period. A crucial part of the inscription is damaged. The second character can be read as tada, but the entire personal name does not survive clearly. Kyoto's English exhibition reproduces the gap before the surviving ending. On the Japanese page a square does the same work. Far from being a typographical nuisance, that small sign is among the most informative details on the page. It tells readers exactly where the capacity of the object to answer a question about its maker comes to an end.
The temple has considered a possible identification with Chikatada of early Bizen. Its published explanation compares the compact grain of the steel and subtle details of the hardened margin with features associated with that group. It also describes the difficulty with the proposal. There is no securely signed example by Chikatada available to settle the comparison of inscription and workmanship. The surviving syllable narrows a question without supplying a unique answer. This is a particularly generous piece of museum interpretation. A visitor can follow why an attribution has been suggested and why the institution preserves the incomplete transcription rather than filling the missing character in its label.
Look also at the features the temple can describe without recovering the name. It calls attention to a small point and a curve most evident toward the base, as well as fine grain and a restrained pattern with small variations. These observations provide the reader with a picture of the object apart from any decision about personal authorship. The attribution could be revised tomorrow and the observations would remain available for discussion. An account that treats a missing first character as the end of the inquiry understates how much a medieval work can reveal about its date and visual relationships even after some of the writing has been lost.
The distinction extends to the explanation given for the damage. The temple associates it with an iron collar. That is an interpretation of a physical history, separate from the transcription of what remains legible today. It should be credited to the institution rather than retold as an incident witnessed at a specific date. Records of injuries to historic objects rarely offer the tidy narrative found in a tale about a famous owner. Sometimes what survives is a contact between materials, a visible loss and a thoughtful reconstruction of how the two may be related. Certainty about the loss can coexist with caution about its cause.
The box that belonged to another blade
Among the most revealing episodes in the history of Hizamaru occurred in recent museum work. Daikaku-ji reports that a survey preceding the Kyoto sword exhibition of 2018 identified a mismatch. A box regarded as belonging to its tachi turned out to belong to a different blade. This is a small sentence with large consequences for the writing of object histories. A box is precisely the kind of companion item that can carry a name or preserve an arrangement after the original circumstances of gift and collection have been forgotten. A false association can acquire authority each time the same pieces are described together.
The result does not make boxes unreliable as a class of evidence. It shows why a box and its contents should be identified independently before a researcher builds a story from their association. Objects used for storage can be moved, borrowed, reallocated or relabeled. Their surfaces can preserve information about a collector's habits that the object inside does not carry. In this instance the temple has made the correction public. The visitor gains a more accurate picture of what belongs to the tachi and a rare view into the collaborative work undertaken before an exhibition opens to its first audience.
Imagine encountering a photograph of a blade laid above a box on a table. The photograph would establish that the two occupied the same place at the time of the exposure. A later label might add that they had usually been kept together. Neither fact alone would prove that the box had been made for this particular tachi or that a document once enclosed in it had accompanied the blade through every stage of its history. The temple's account gives a concrete reason for asking such questions. They protect the value of associated materials by specifying the kind of history each can responsibly support.
What the Kitano exhibition actually says
The case of the Kitano blade offers a different difficulty. Visitors will encounter the name Yasutsuna in accounts of Onikirimaru. A cultural report on the earlier Tokyo exhibition uses that reading. The later Kyoto National Museum presentation explicitly labels its shrine loan with the inscription Kunitsuna, both in the Japanese caption and in the English one. The important first step is to identify the disagreement accurately. The name of the loaning shrine, the familiar names Higekiri and Onikirimaru, and the comparison with the Daikaku-ji object make clear which object each exhibition is discussing. There is no need to invent a second shrine tachi to make the two labels agree.
Equally, the change in wording by itself supplies no explanation of when the inscription was cut, whether it was altered or whether published scholarship arrived at a consensus on the smith. Those propositions would require closer specialist documentation. In a caption accompanying a photograph from the Kyoto exhibition the reproducible choice is to record Kunitsuna and identify the catalogue consulted. In a history of public interpretation, retaining the earlier Yasutsuna reading with its source is equally important. A reader can then see what changed in public descriptions without being asked to accept a theory of physical alteration unsupported by the material cited.
There is an additional trap for English-language readers. Similar names of smiths recur in histories of famous Japanese swords, and a celebrated attribution can make it tempting to transfer dates or provinces from a better-known example. Sharing a reading is insufficient to establish that two inscriptions identify one individual. It is especially precarious when the reading of one inscription is itself reported differently. The fixed points in this account are the shrine's custody, the names associated with the tachi and the wording of the two exhibition sources. Those three facts do more historical work than a neat but undocumented genealogy.
The Minamoto household and the arts of retelling
The account associated with Daikaku-ji introduces Minamoto no Mitsunaka as the holder of two tachi. It also gives Yoshitsune a place in the story of the later name Usumidori. The two men should not be compressed into a single legendary generation. A transmission narrative can connect distant figures because its concern is precisely the persistence of a household and its treasured things. A museum history has to ask how that narrative moved between people, records and objects. Without the story, a visitor could still study an early blade from its shape. Without the surviving object, people could still have told a story about a treasured pair.

A print depicting uncanny events at the mansion of Minamoto no Yorimitsu, also known as Raikō, illustrates how the wider cycle of Minamoto legends grew in visual culture. The artist can place a frightening assembly before the viewer at a glance, giving physical presence to creatures that inhabit the realm of storytelling. The print belongs to the British Museum, which records it as an individual work. It is reproduced here to show the pictorial afterlife of such stories. It is neither a medieval eyewitness account nor a portrait from which to identify the tang of the shrine or temple blade.

A different print, in Los Angeles, locates Yoshitsune in the story of Dannoura and Kenreimonin. Its identification of participants depends on a visual and literary tradition intelligible to its audience. Compare the questions raised by the two prints. Who could recognize a figure without a written name? How could a change of setting signal that the image concerned a particular episode? Which details would a viewer expect an artist to repeat because earlier artists had made them memorable? Those questions belong to the history of storytelling and spectatorship. They can be pursued without assuming that an artist enjoyed privileged access to the contents of either Kyoto treasury.
How a shrine scroll belongs beside a sword
The illustrated legends of Kitano Tenjin are essential to understanding the institutional setting of the shrine blade. A surviving handscroll fragment dated to 1277 in Tokyo National Museum carries the narrative in images on paper. Another example of the illustrated tradition in the Metropolitan Museum appears below. Such works brought the origins and efficacy of a sacred place before audiences through scenes composed to be read in sequence. An exhibition devoted to Tenjin can therefore bring together documents, paintings, offerings and metalwork while asking a coherent question about how a cult acquired a material history.

The relationship is more interesting than the observation that the scrolls and swords are Japanese antiquities. A scroll can instruct viewers how to understand extraordinary events within the history of a sanctuary. A donated treasure can give worshippers a material connection with a celebrated family or a remembered patron. Both can draw people to a place, but their records have different shapes. A scroll can contain an episode without ever naming a particular sword. A shrine can preserve a named tachi without its label serving as a key to a specific painted scene. Keeping those freedoms of interpretation open makes the exhibition richer.
Daikaku-ji offers a related lesson from its own commemorations. The temple records the creation of commemorative seals on occasions associated with the two tachi and calls attention to their exhibition together in Tokyo. In that account it is the pairing itself that merits commemoration. Seeing the objects together gives a visitor an experience unavailable from two isolated collection records, however detailed. Later visitors can recall that shared encounter through an image or a seal, adding another layer to the history of the pair. Such evidence is particularly valuable for tracing the modern life of an old relationship, an area too easily dismissed as mere publicity.
A reader returning from the exhibition might keep four short entries in a notebook. One names the object and its custodian; another records the inscriptions exactly as presented at that particular exhibition. A third identifies the source for the household tradition, and a fourth identifies a later image or commemoration in which the tradition appears in a new form. Each entry can be expanded when new evidence emerges. If the inscription is reexamined, the notebook can change without displacing the shrine or temple. If a new illustrated version of the story is discovered, it can enlarge the history of reception without moving either object on the map.
One detail the visitor can record with unusual confidence is the museum numbering. These two works were neighbors by curatorial decision. Quoting the adjacent numbers identifies a specific moment when people in Kyoto could walk from one to the other, with their shared reputation and their differences equally available for consideration.
The result is a pairing grounded in several kinds of evidence rather than dependent on a single daring leap from legend to steel. One can take the religious institutions seriously as custodians, recognize the literary achievement of the Minamoto narratives, and still acknowledge that one character is lost and another inscription has been presented under different readings. For a museum visitor the reward is the chance to see two famous names acquire separate material identities. The two tachi return to different institutions after a shared exhibition. Their next meeting will have a history of its own, as did every earlier occasion on which someone chose to place them side by side.
Frequently asked questions
Where are Higekiri and Hizamaru kept?
Kitano Tenmangu Shrine in Kyoto preserves the tachi associated with Higekiri. Daikaku-ji Temple preserves the tachi associated with Hizamaru.
What is another name for Hizamaru?
Daikaku-ji identifies it as Usumidori. Its transmission account associates that name with Minamoto no Yoshitsune.
Are both Kyoto blades National Treasures?
Kyoto National Museum lists each of these two tachi as an Important Cultural Property.
What can still be read on the Daikaku-ji blade?
The final character reads tada. The damaged first character leaves the complete name uncertain.
What did the survey before the 2018 Kyoto exhibition uncover?
Daikaku-ji reports that a box previously associated with its tachi belonged to another blade.
Have the two objects been exhibited together?
Yes. Kyoto National Museum presented them as works 131 and 132 in its Kitano Tenjin exhibition. The temple also records their joint display in Tokyo.
Films and talks
kaomaru42 — 「鬼切丸・髭切」「薄緑・膝丸」 (スライドショー) 京都国立博物館 特別展「北野天神」にて 2026年6月12日
日テレNEWS — 【両陛下と愛子さま】大覚寺の国宝など紹介の展覧会鑑賞
