A dragon looks out of a groove
One of the best reasons to abandon the English nickname “blood groove” is a dragon. A tachi in the Tokyo National Museum carries an inscription dating it to the fifth month of 1322 and naming Kagemitsu of Osafune in Bizen. On one face a dragon entwined around a sword appears in relief within a channel. On the other a Sanskrit character rises from its recessed ground. Generations of viewers have found the first image so distinctive that the sword has a name evoking a dragon peering out. A visitor can read the two faces as sculpture, writing, and devotion on the same piece of metal.
The e-Museum catalogue identifies the figure as Kurikara, associated with the Buddhist deity Fudō. That identification changes what we look for in the photograph. A line that seemed to belong to the dragon alone may belong to the sword around which it coils. The frame formed by the recess gives the creature somewhere to look out from. Its head and the background seem to occupy different depths even though they share the narrow confines of the blade. On the reverse a character performs a quite different visual task in a comparable space. Few objects reveal so economically how a cut in a surface can also create a place for an image.
English vocabulary often bundles these observations together. A fuller is a groove. A relief is a figure distinguished from its ground by variations in height. An inscription can be legible without being pictorial, and a sacred character can be pictorial without ceasing to be writing. Japanese sword records use hi for grooves and horimono for carved decoration. These categories can overlap on an individual object. The presence of a dragon in a groove does not turn every groove on a Japanese blade into a religious design. Equally, the dragon is poorly served by an explanation concerned only with the amount of metal present.

One channel, a companion line, and a pair
A bō-hi is the long, comparatively broad groove commonly encountered in English discussions of Japanese blades. A slender subsidiary groove alongside a larger one is called soe-bi. Futasuji-hi, also transcribed futatsuji-hi, describes two narrow grooves. The English word “double” loses an important distinction. Two similar lines have a different visual relationship from a broad band accompanied by a much finer one. Imagine cataloguing a drawing in ink. A pair of strokes of similar width reads like a repeated motif. One broad stroke edged by a fine one gives a different emphasis to the space beside it. Engraved channels can establish both kinds of rhythm.
| Catalogue term | What the viewer sees | What to check in the record |
|---|---|---|
| Bō-hi | One relatively broad channel | Check where the channel begins and ends in the photograph. |
| Soe-bi | A narrow channel alongside a larger groove | Read both grooves on the same face. |
| Futasuji-hi | Two narrow lines running in parallel | A pair of comparable channels differs from a main groove with a companion. |
| Bonji | A Siddham-script character | Use the catalogue identification for the syllable or associated deity. |
| Kurikara | A dragon entwined around a sword | Check the orientation and the surface illustrated in the collection record. |
Read the illustration and the face description together. A single photograph need not show all the features recorded for a blade.
Sources · e-Museum, Kagemitsu and the two faces of the 1322 tachi · Cultural Heritage Online, Akihiro dagger with lotus and paired grooves
Those words support quite modest and useful observations in a gallery. Where does the decoration start relative to the portion of the blade exposed above its mounting? Is the field empty, or is a figure visible within it? Does the other face repeat the composition or answer it with something different? On a narrow blade a line that begins as a background can become the dominant visual element once the figure ends. A complete view gives that rhythm a scale that an enlarged detail cannot preserve. A detail in turn may reveal the base of a lotus or a turn in a character that the complete view compresses to a dot.
The Cultural Heritage Online record for a signed dagger by Akihiro offers a concrete comparison. Its description places a sword and lotus pedestal on one side, a sacred character and a pair of grooves on the other. Here the two sides share a devotional vocabulary while differing in composition. It would be easy to see a single photographed face, recognize a double groove, and miss the relation entirely. By recording the objects and signs on both sides, the catalogue preserves a feature of the design that would vanish from a list of measurements. The attribution of this vocabulary to a tradition is also grounded in a particular dated cultural context, not in a generic belief about all engraving.
The limits of the I-beam comparison
The most elementary statement about a channel concerns mass. Compare two otherwise identical pieces of metal, one with some material removed. The latter has less mass by exactly the amount removed. It need not follow that two historic objects, one grooved and one plain, will rank in that order on a scale. They can differ in length, width, thickness, and fittings, and both may have changed during their histories. Even a careful museum weight is useful only when it is clear whether the measurement includes the blade alone or an assembly. Those conditions are often missing from captions attached to photographs circulating online.
The familiar comparison with an I-beam expresses one insight about cross sections. How material is distributed matters to bending, and the same mass distributed differently need not behave identically. A museum blade is a tapering, curved object of changing section. A small channel with a dragon in it is hardly an industrial beam of uniform section. Removing material from a given solid piece does not magically make that piece stronger. No individual stiffness, toughness, or original balance can be read from the mere presence of a bright groove in a photograph. The analogy is helpful to the extent that it prompts a better question about the physical evidence.
Balance requires particular care with language. The centre of mass belongs to the object as a whole in its specified condition. A mounting, the loss of part of a tang, and a change of fittings all affect what object is under discussion. The visual balance of a composition is something else. A small dragon near one end can command a disproportionate share of a viewer's attention. A curator can meaningfully describe that visual effect without performing a measurement. Keeping these senses apart allows a physical explanation and an account of artistic intention to coexist without borrowing authority from one another.
The nickname about blood adds a purpose that neither the museum description of this tachi nor its physical form establishes. A recessed line is no proof of a story about its use. Repetition in captions and casual translations can give such a story a spurious historical depth. There is no comparable mystery about the fact that this groove houses a figure. It is visible, it recurs in a public collection description, and it explains the name by which viewers have long distinguished the object. For this particular work, the attested account is more interesting than the nickname precisely because it is more specific.
Kurikara and a sword that signifies wisdom
In Japanese Esoteric Buddhist art, Fudō Myōō is a Wisdom King with a fierce appearance and a protective role. Representations commonly give him a sword and a lasso. The sword signifies the wisdom that cuts through ignorance. Its religious meaning travels across media. A sculptor can put it in the hand of a standing figure. A painter can reduce it to a red outline. A metalworker can make an image of a sword on the surface of another sword. In the Kagemitsu carving, a dragon winds around that interior sword, giving a recognizable form to a presence that might elsewhere receive an entire painted body.
A ritual ken in the Metropolitan Museum of Art offers a valuable counterpoint. Its association with a vajra hilt gives the object an explicitly Buddhist context. A viewer can compare the symbolism of the ritual sword with the image carved into the blade in Tokyo. The relation is one of religious vocabulary. The two are different kinds of object with different uses and histories. A small depiction can borrow the authority and associations of a ritual implement without making the surrounding tachi into precisely that implement. This is a familiar operation in religious art, where a few attributes can evoke a figure too large or elaborate for the available space.
Bonji is the Japanese term commonly encountered for Indic script in this setting. A seed syllable, or bīja, can stand for a deity in condensed written form. Chikai's image of Fudō with attendants in the Metropolitan Museum makes the meeting between a written sign and a pictorial body unusually vivid. Rows of characters take part in the construction of the image itself. When a collection record identifies a single character on a blade, the amount of surface it occupies is therefore a poor measure of its significance. One character can invoke an extensive world of worship, learning, and visual association.
The photograph must also be read with a little humility. Reflections interrupt fine strokes, and a reversed image changes their orientation. Even a reader familiar with modern Japanese script may have little experience of the forms used for sacred syllables. A curator's identification in an object record is better evidence than a passing resemblance in a cropped photograph. Where the Kagemitsu catalogue identifies a Sanskrit character but leaves its reading unstated in English, a reader can preserve exactly that level of certainty. Printing a confident transcription for every partly visible stroke would lend the picture a precision it does not possess.

A maker, a carver, and a pair reunited on paper
The Metropolitan Museum catalogues two blades by Naotane that make a rewarding comparison with Kagemitsu. One is a short sword, the other a long sword. The museum identifies a Sōshū stylistic reference for the shorter blade and a Bizen reference for the longer one. It attributes the engravings on both to Yoshitane, an artist associated with the decoration of Naotane's blades. The records thus distinguish at least two sorts of authorship within each object. They also distinguish the making of a pair from the mechanical repetition of one historic style in two sizes.
The short sword illustrated above has an accession number of its own, and the long sword below has another. Following both records prevents a common slip. What the museum says about the engraving may apply to both. What it says about their historical stylistic references differs. A catalogue entry for one member of a pair may therefore carry essential evidence for reading the other. When two digital images appear side by side at the same height on a screen, the interface can flatten the real difference of scale. The object titles and measurements restore the distinction between companion works and two views of a single blade.

An attribution is part of the history of knowledge about an object. “Attributed to Yoshitane” preserves a scholarly judgement about the engraving. It does not tell us that his full name is inscribed beside each motif, that he personally chose the programme, or that a particular patron ordered it for a known ceremony. Each of those stronger claims would need additional evidence. The distinction also works in the other direction. A named smith is not automatically the hand responsible for every image on a blade. Recognizing a second artist changes the history of the work from the solitary biography of a smith to a relationship between specialists.
It is worth imagining how the pair would be described in two different galleries. A gallery about the history of Bizen and Sōshū styles might emphasize the different references adopted by Naotane. A gallery about the history of engraving might put the attribution to Yoshitane first. Neither has to suppress the other attribution. The same surface carries evidence for both histories. A short online description, however, can silently lose one of them when a database exports only its first maker field. Following the sustained prose in the original record often recovers a collaboration that the abbreviated credit line has made almost invisible.
Albums, relief, and the problem of an unseen flaw
An album of designs for metal carving in the Met collection changes the scale of this question again. On an album page the draughtsman can develop an outline independently of any particular steel surface. The finished carving must fit a narrow field, be seen under shifting reflection, and stand in a relation to the broader shape of an object. Comparing paper and metal helps explain why two motifs can belong to the same visual family without being identical in every contour. The album reproduced here is evidence of a means of recording and circulating designs. It is not presented as the working drawing for the Kagemitsu dragon or for the Naotane pair.

Relief is sometimes mistaken for an additive process because the finished figure seems to rise toward the eye. In descriptive terms it means that the figure stands above its immediate ground. In the recessed field of a groove that figure can still lie below the adjacent undisturbed face. This nesting of relative heights is central to the impression of a dragon peering out. It also explains why a photograph taken under flat light can make a complex surface look like an outline drawing. A second exposure, or the catalogue's description of the reverse, can restore an understanding of the depth that a single view has flattened.
A persistent collectors' story proposes that decoration was added to conceal an imperfection. A later addition and a concealed defect are two separate historical propositions. Even if a carving could be dated later than the blade by inscription or by a securely identified early drawing, the date would not by itself establish why it was commissioned. A change in devotional preference, patronage, or the wish to associate an object with a named carver could also be compatible with a later decoration. Where a condition report or an early description documents a particular intervention, it belongs to that object's biography. The records cited here do not diagnose a hidden flaw under Kagemitsu's dragon.
An old drawing or rubbing can help reconstruct such a biography, but it has limits of its own. An artist may record an inscription very carefully and abbreviate the rest of a blade. A rubbing gives a different sort of evidence from a free sketch. A modern reproduction can trim away precisely the portion needed for a comparison. Before treating a missing ornament in a drawing as evidence that it had not yet been made, ask whether the maker of the drawing intended to record all the ornament. An apparent absence may tell us about editorial purpose instead of the state of the metal.
Reading the object from both sides
A productive final comparison is between Kagemitsu and Akihiro. On the former the two faces support a dragon entwined around a sword and a character. On the latter, the official record pairs a sword on a lotus pedestal with a character and double grooves. Listing their motifs reveals a shared devotional language. Listing their arrangements reveals individual decisions. Neither inventory can be condensed into a claim that all medieval blade engraving carried one meaning. The evidence is richer. Within related conventions, one artist could organize a face around an animal figure while another gave comparable prominence to the support beneath an emblem.
A gallery visitor can leave with three distinct ways of seeing a single recess. The physical observation concerns missing material and the limits of conclusions about an object's mass. The sculptural observation concerns the relation of a figure to its ground. The historical observation concerns who named an image, attributed its making, or preserved its devotional meaning. The most memorable witness to all three is still the dragon in the blade from Bizen. Its head looks out of the channel, the sword inside its coils supplies an attribute of Fudō, and the answering character waits on the face that a single photograph cannot show.
Frequently asked questions
What is a fuller on a sword?
A fuller is a recessed channel in the blade. Japanese collection records distinguish several forms of hi and sometimes describe relief sculpture in a channel.
What is a katana bō-hi?
Bō-hi is a long, broad groove. A slimmer companion groove is called soe-bi, while futasuji-hi refers to paired narrow grooves.
What does horimono mean?
In a Japanese sword record horimono means carved decoration on the blade. The subject and the identity of the carver require separate evidence.
Who is Fudō Myōō?
Fudō is a protective Wisdom King in Esoteric Buddhism. His sword signifies wisdom, and a dragon entwined around a sword can evoke him in another form.
What is a seed syllable?
It is a sacred syllable that can stand for a deity in Buddhist ritual and art. A single character can therefore function as a devotional image.
What is relief carving?
In relief, a figure rises from its surrounding ground. The dragon on Koryū Kagemitsu rises within a recessed field on one face of the blade.
Films and talks
JASA: Japanese Art Society of America — Samurai Splendor: Sword Fittings from Edo Japan (Lecture by Markus Sesko)
JASA: Japanese Art Society of America — Dr. Sinéad Vilbar: Kami in Cleveland: Creating the exhibition Discovery of the Divine Japanese Art
Collection records and further reading
- e-Museum, Kagemitsu and the two faces of the 1322 tachi
- Cultural Heritage Online, Akihiro dagger with lotus and paired grooves
- Metropolitan Museum, Fudō outlined in seed syllables
- Metropolitan Museum, ritual ken with vajra hilt
- Metropolitan Museum, Naotane and Yoshitane
- What Is a Fuller on a Sword? The “Blood Groove” Myth Explained
