One dagger, three descriptions
Atsushi Tōshirō is a useful blade with which to enter a Japanese sword gallery. In the Japanese cultural properties catalogue it is a tantō, it is hira-zukuri, and it has an iori-mune. These are three answers to three different questions. The first names the object as a short blade. The second describes the organization of its broad faces. The third describes its back. A change of lighting can bring out one of these properties and all but hide another. The catalogue keeps them together because none can do the work of the other two. The small Yoshimitsu blade offers a better introduction to Japanese blade shapes than a row of silhouettes arranged by length.
Tsukurikomi, often rendered as blade construction or form, refers here to the arrangement of surfaces in the finished object. It helps to imagine two drawings that a museum illustrator might make. The first records the outline of an entire blade seen squarely from the side. The second looks across a short stretch of it and records the changing direction of its faces. Objects with similar outlines can give quite different second drawings. A third drawing might be needed close to the point, where a ridge can change direction or a second edge can appear. Those changes explain why one diagram headed “the shape of a katana” can leave so much of a collection unexplained.
The four names in the title each preserve a different habit of looking. Shinogi-zukuri directs attention to a ridge. Shōbu-zukuri invokes an iris leaf. Hira-zukuri describes the broad face in the absence of the characteristic longitudinal ridge. Kissaki-moroha draws attention to the termination of the blade. A fifth term, kiriha-zukuri, belongs alongside them because it asks where a ridge lies in relation to the edge. Knowing the words gives a visitor a way to return to a photograph with a question that the photograph may actually answer.

Following the ridge in shinogi-zukuri
Look first at a diagram with the Japanese words still attached. The Kyoto National Museum marks the shinogi as the line dividing the main face toward the edge from the shinogi-ji toward the back. In its diagram the ridge runs much of the length of the object. Viewed obliquely, the two areas catch the light differently. From another angle both can take on the same tone. There is no need for a dark stripe or a permanently bright line to exist in the material. The boundary one sees is an effect of the way two surfaces meet and reflect their surroundings.
| Catalogue term | Feature to locate | What a photograph can establish |
|---|---|---|
| Shinogi-zukuri | A ridge divides the side planes. A yokote defines the point area. | Look for the longitudinal and transverse lines separately. |
| Shōbu-zukuri | The ridge runs toward an iris-leaf-shaped point. | A missing transverse line in one poorly lit image is inconclusive. |
| Hira-zukuri | The side lacks a distinct shinogi. | A profile shows the outline. A second angle reveals the surfaces. |
| Kiriha-zukuri | The bevel begins at a ridge relatively close to the edge. | Look for the museum’s description of the bevel and its position. |
| Kissaki-moroha-zukuri | Part of the back near the point is also edged. | A view of the back helps establish how far this feature extends. |
A shape term describes the visible blade. Accession records may separately identify the maker, date, mounting, and later alterations.
Sources · Kyoto National Museum, illustrated anatomy of the blade · Nara National Museum, kiriha-zukuri in the Shōsōin exhibition glossary
At the point the illustration becomes especially informative. The transverse division called the yokote marks the beginning of the distinguishable tip area. A shorter ridge outlines part of that area. Kyoto labels the meeting of the long ridge, the shorter one, and the yokote as the mitsu-gashira. A reader can follow the three branches outward from one junction and discover that they delimit areas of the object. By contrast, a line of writing can cross a surface without changing its direction at all. The distinction matters in a photograph where an inscription, a reflection, and a physical ridge can all be printed in the same shade of grey.
A shinogi-zukuri description therefore supplies more than the observation that a blade has a pointed end. It places the point in a particular relationship with the body. The change can be legible in a modest photograph whose color tells us almost nothing about the steel. A brightly lit detail of surface grain might offer the opposite bargain, rich in texture and poor in information about the overall arrangement. Neither view is defective. They were made for different questions. When a catalogue supplies several images, the full-length view is the natural place to start tracing the junction before moving to a detail.
The iris leaf and the missing transverse division
Shōbu is an iris. The botanical image takes the reader to the end of a leaf, where the form runs toward its extremity in a continuous sweep. In the museum vocabulary of shōbu-zukuri a ridge is present, but the conventional transverse separation of the tip area is absent. This combination gives the term its precision. A smoothly tapering outline by itself does very little to identify the form. Many objects with a distinct yokote have an equally graceful outer contour when seen in profile. It is the relationship between that outline and the surfaces inside it that makes the iris comparison memorable.
There is a practical difficulty for anyone studying a digital collection. Absence is much harder to record than presence. A very small photograph can show that an outline tapers, while containing too little information to show a shallow change of plane. Reflected light can merge two adjacent areas in one exposure. A hilt, a museum support, or a decorative title over a reproduced image can hide still more. If a collection identifies an example as shōbu-zukuri, that identification carries the authority of direct examination. If an anonymous photograph simply appears to lack a yokote, the most accurate description may be that the junction cannot be seen in that view.
The Samurai Museum Berlin includes shōbu-zukuri in its account of tantō forms. This placement corrects a persistent expectation about Japanese terms. A blade need not change its class from dagger to sword when a catalogue gives it a ridge. Nor does a blade with a pointed leaf shape automatically belong to one period or one maker. A collection can place a recognizable convention in the work of a particular atelier using an inscription or a documented attribution. The convention itself is available to other makers. The word is most valuable when it allows comparison across those boundaries rather than closing the question of date prematurely.

Hira-zukuri and a particularly informative back
Return to Atsushi Tōshirō. The official record identifies a hira-zukuri blade with very little curvature and a slight inward inclination near the end. These observations describe independent features. A reader could recognize the inward inclination in a silhouette but would not yet know whether the faces were ridged. The word hira directs attention to the uninterrupted expanse of the face, which lacks the characteristic shinogi of a ridged form. It does not imply that every part of the object is flat like a sheet of paper. The tip still narrows, the blade has thickness, and the profile of the back can be described in its own right.
The iori-mune of the same record is an especially good reminder of that independence. The architectural comparison is with a pitched roof when seen end on. The meeting of the two slopes along the top belongs to the back. It does not supply a shinogi across the broad face or convert the catalogue's hira-zukuri into a contradiction. English readers sometimes meet the translations “ridge” and “ridged” in several contexts on the same page. Asking which part of the object the English word refers to will usually settle the apparent inconsistency before any appeal to an exotic exception is needed.
The attribution to Yoshimitsu makes the example historically specific. A visitor can compare an actual work associated with the Awataguchi tradition in Kamakura-period Kyoto against a general diagram assembled for a modern audience. The diagram has the advantage of perfect visibility. The photographed object has the advantage of preserving particulars that a designer of diagrams might omit, including the changes in outline close to the point. Neither resource can replace the other. The knife in the collection also carries a history under a personal name. That history accounts for part of its prominence without turning its precise outline into a rule for every dagger by the same maker.
Kiriha-zukuri in the vocabulary of the Shōsōin
The Nara National Museum glossary for the exhibition of Shōsōin treasures describes kiriha-zukuri with an unusually exact observation. The longitudinal break lies markedly toward the edge, while the area between that break and the back maintains an even thickness in the terms of the definition. A side view gives a long boundary relatively close to one margin. A cross-sectional drawing explains why that placement is of interest. The broad area and the comparatively narrow area beside it meet in a way quite unlike the division in the standard modern diagram of a shinogi-zukuri blade. Both descriptions can contain a ridge and yet describe different forms.
Nara's glossary was prepared for people looking at a particular exhibition. That fact matters. The Shōsōin preserves the material culture of an early court familiar with objects and designs circulating across East Asia. The vocabulary needed to describe such a collection includes forms that can seem peripheral in an exhibition devoted to the later history of the curved sword. A visitor who begins in Nara and a visitor who begins with an Edo-period collection may consequently come away with different ideas about which diagram looks familiar. Their impressions say something about the samples they have seen as well as about Japanese history.
An early context is not a date stamp. Recollections of prestigious old forms and later works made in conscious reference to them belong to the history of collecting. An artist or a patron can take an interest in a particular silhouette long after its first period of prominence. To move from a classification such as kiriha-zukuri to a calendar date, read the rest of the entry. A dated inscription, an early inventory with a persuasive identification, an excavated context, or a specialist's attribution can each supply information of a different order. The line itself continues to describe the visible object whichever of those kinds of evidence survives.

Kogarasu Maru and the double-edged point
Kissaki-moroha-zukuri poses a different problem for a diagram. A section drawn at the middle of the body may miss exactly what the name describes. Toward the point a portion of the back becomes a second edge. Further down the object the organization changes. The famous Kogarasu Maru, associated with the imperial collection, has given its name to the form in such expressions as Kogarasu-zukuri. It is possible to convey this distribution in a pair of drawings taken at different places. A single symmetric triangle marked “tip” or a single transverse section chosen for convenience conceals the transition that prompted the name.
The example has attracted origin stories, including an attribution to Amakuni. A named work can offer a vivid way of remembering a form even when its traditional attribution does not yield a secure biography. Imagine a gallery label in two sentences. The first says where the second edge occurs. The second introduces a tradition about an early smith. The first sentence can be checked against a sufficiently complete set of views or an examination of the blade. The second asks for a history of transmission and attribution. A careful reader is entitled to accept the physical description while retaining the qualification attached to the name of the maker.
It is tempting to place Kogarasu Maru halfway along a straight line of development from straight to curved blades and to let its outline do all the historical work. A diagram of that sort is beautifully economical and historically demanding. It would require dated objects on both sides, evidence that one workshop inherited the relevant convention from another, and some reason for treating parallel experiments as a single succession. A museum can responsibly exhibit early forms side by side without claiming to possess every link in that chain. The rarity of the celebrated example is part of its interest and a reason for caution when it is used to stand for a whole age.
How photographs, fittings, and captions affect a comparison
The mounted dagger reproduced here brings a second history into the discussion. A mounting provides a visible environment around a blade and may belong to a different date. The broad curvature one sees in a lacquer scabbard is the curvature of an exterior object with its own surfaces and fittings. When the museum displays the blade beside its mounting, the photograph places those histories in conversation. The class of the blade remains a matter for the exposed metal and the catalogue record. A lacquer ridge reflected in a showcase can be arresting and irrelevant to a question about the location of the shinogi.
A mounted tachi offers an even clearer lesson. Suspension fittings, the shape of the furniture, and the arrangement of the scabbard can help explain why the record calls the ensemble a slung sword. If the metal is shown separately, a reader may then ask about its planes and its point. If only the closed mounting is available in the image, a class name in the caption cannot supply a missing view of the blade. A good collection record is generous about such differences. It may date the blade and its mounting separately, identify different makers, and provide more than one accession component under a single object title.

There is another line that frequently enters the wrong conversation. The bōshi concerns the continuation of the hardened pattern into the point. A white curve associated with it can occur close to a physical ridge in a museum image. The question asked by the word bōshi concerns the pattern of the hardened region. The question asked by yokote concerns the transition into the tip area. A photograph optimized for one can make the other ambiguous. Kyoto's arrangement of the anatomical labels and the patterns of the hardened edge on the same sheet is an invitation to study their relationships while preserving their separate names.
Consider how a caption could describe one image to a reader who cannot see it. “A curved sword with a bright line” conveys little that distinguishes the object. “A long ridge separates a narrow area beside the back from a broader face, and a transverse boundary marks the point” describes a visible organization. If the image does not resolve that last boundary, the caption can simply stop after the first observation. This economy has a historical advantage. A later reader comparing the caption with an earlier drawing will know exactly which features the observer recorded and which remained outside the description.
Looking across periods without flattening their differences
The early court collections, the celebrated Kamakura-period works, and the composite ensembles of later collectors give us three quite different encounters with blade form. One preserves evidence within a treasury. Another brings particular smiths and named works into focus. The third may bring together metal, lacquer, textile, and fittings across several generations. It would be surprising if a single schematic sequence explained the survival of all three. Similar forms can recur under different conditions of patronage, and dramatically different forms can coexist. A label records an instance of those possibilities. A visitor gains more by comparing securely identified instances than by counting exceptions to a simplified family tree.
The best test is to return to the little Yoshimitsu blade after looking at the others. The roof-shaped back, the quiet extent of the broad face, and the turn of its outline near the end can now be expressed in separate words. Compare that description with Nara's explanation of the ridge close to the edge, then with the iris analogy, and finally with the extra edge at the end of Kogarasu Maru. The terms do more than sort objects into boxes. Each preserves a particular observation that a viewer can carry from one collection to another. Together they restore depth to objects so often reproduced as narrow strips of silver on a page.
Frequently asked questions
What is a shinogi line?
The shinogi is the longitudinal ridge dividing the broad face toward the edge from the area toward the back. The yokote is the transverse division at the point.
What distinguishes shōbu-zukuri from shinogi-zukuri?
Shōbu-zukuri has a ridge that continues toward an iris-leaf-shaped tip without the distinct transverse yokote characteristic of shinogi-zukuri.
What does hira-zukuri mean on a tantō label?
The broad face lacks the usual longitudinal shinogi. The back can still have its own ridges, as it does on Atsushi Tōshirō.
What is kissaki-moroha-zukuri?
A second edge occupies the back near the point while the lower body remains single-edged. Kogarasu Maru is the celebrated example.
Where is the ridge in kiriha-zukuri?
The Shōsōin exhibition glossary places the ridge markedly toward the edge, leaving a broad area of even thickness toward the back.
What is a tachi?
A tachi is a Japanese long sword associated with a slung mounting. Its designation and the arrangement of the planes of its blade describe different features.
Films and talks
The Met — Art of the Samurai—An Introduction
The Met — The Japanese Sword and the Japanese Aesthetic
Collection records and further reading
- Kyoto National Museum, illustrated anatomy of the blade
- Nara National Museum, kiriha-zukuri in the Shōsōin exhibition glossary
- Samurai Museum Berlin, tantō forms
- National Institutes for Cultural Heritage, Atsushi Tōshirō
- Following the Bōshi in a Museum Photograph: Kissaki, Yokote and the Limits of Light
