The Book of Five Rings (Gorin no Sho): Surviving Manuscripts, Higo Relics, and Musashi's Ironwork

A book whose author never saw a modern edition

A roll of paper just under eighteen centimetres high would scarcely interrupt the view across a reading desk. Unrolled to its full length, one such manuscript of the Fire book of Miyamoto Musashi extends more than twelve metres. Both measurements belong to the same object. The reader who knows The Book of Five Rings from a paperback may recognize the title immediately and still have to learn how this book once occupied a room. Its history is partly the history of the distance between forefinger and thumb, between one column of handwriting and the next, and between the display case where a scroll lies open and the much greater length that remains concealed.

The five scrolls in question belong to Eisei Bunko, the collection associated with the Hosokawa family. Kyushu National Museum published their dimensions when the set appeared in an exhibition of Hosokawa treasures. In Kumamoto a different institution, the Shimada Museum of Art, identifies a manuscript of the Wind book particularly close in date to Musashi's lifetime. A third collection, at Kumamoto Prefectural Library, records damaged papers whose very identification as parts of the work carries a question mark. A good history of Gorin no Sho needs all three. They give us a text recognized under one title, bodies of paper with strikingly different histories, and the names of people who made transmission possible.

Historic portrait of Miyamoto Musashi, traditional attribution
Historic portrait of Miyamoto Musashi, traditional attribution. Artist unrecorded (via Wikimedia Commons).

A guest in Hosokawa country

The five years Musashi spent in Kumamoto toward the end of his life give these objects an unusually precise setting. Hosokawa Tadatoshi invited him there in 1640. Higo was the province, Kumamoto the castle town, and Hosokawa the name of the ruling house. Keeping the three distinct helps explain why papers and pictures linked to the same man now turn up in a local art museum, a prefectural library and a family collection in Tokyo. A province cannot inherit a manuscript. Families, students and institutions can, and they do so by quite different routes.

Musashi's fame as a duellist can make this last residence sound like a quiet appendix. The evidence for the writer and artist is especially rich here. The Shimada museum associates the Kumamoto years with another text in thirty-five articles as well as with Gorin no Sho. Treating them as separate works restores some time to the story. A teacher could formulate his thinking more than once, for different occasions and at different lengths. A reader disappointed that an early statement does not match the arrangement of a later one may be asking it to perform a task for which it was never written.

The Hosokawa name also changes meaning according to the source. In a biography it identifies the connection that brought Musashi to Higo. In a loan exhibition it can identify a collecting history that lasted for generations. A seventeenth-century relationship does not by itself date the sheet in a nineteenth-century mounting. A useful parallel is a family photograph album containing a much later print from an early negative. The album can offer outstanding evidence for what the family preserved without making each impression contemporary with the sitting.

What survives of the Wind book

Japan's official multilingual guide to the Shimada collection identifies its Wind manuscript as a copy made in 1651 by Terao Magonojō, the student who received Musashi's text. Musashi had died in 1645. Six years is a short interval for the transmission of an early modern work, yet long enough to distinguish the act of composing from the act of making this witness. The guide reports that no manuscript of the work in Musashi's own hand survives. That statement concerns the physical writing of this particular work. It does not erase the museum's separate holdings of brushwork associated with him.

Four witnesses to Musashi’s time in Higo
Witness What survives What the date refers to
Wind scroll A manuscript copy at the Shimada Museum 1651 dates the surviving copy.
Book of Five Rings A text transmitted through copies A printed edition can be much later than the text it transmits.
Brush painting A work catalogued by subject and attribution A catalogue or exhibition date belongs to the record unless it also dates the painting.
Sea-cucumber guard An openwork design associated with Musashi The object and later examples of its design have separate histories.

For manuscript citations, record the scroll, copyist, date of copying and present institution whenever the catalogue gives them.

Sources · Kumamoto exhibition notes · Shimada Museum of Art

On the Shimada collection page the short Japanese labels repay a slow reading. The name of the writer, the identification of the piece as a copy and the era date sit close together. An English caption that carries over the name but drops the word for copy changes what the visitor thinks is in the case. If there is room for just one line, give it the work, the surviving section and the status of the sheet. With two lines, the name of the copyist and the date bring the document into sharper historical focus.

Wind has a further ambiguity for an English reader. It is one of the elements giving the five divisions their familiar names, but the character also belongs to a vocabulary of styles, ways and inherited manners. An English title can preserve the elemental symmetry beautifully while losing that secondary association. The collection's identification of a Wind scroll has the advantage of being perfectly specific as a catalogue description. It tells the visitor which physical portion survives at Shimada. It does not require that every occurrence of the word in every passage be read in exactly the same way.

The length of the five books

The Eisei Bunko set makes a surprising comparison possible. Its Earth and Fire rolls each extend roughly twelve metres, Water and Wind each run a little over nine, and the surviving Void roll is under four. The heights remain within a few millimetres of one another. Laid out in a table, the measurements might suggest five standardized volumes containing very unequal quantities of writing. A person looking into the display case gets almost the opposite impression. Five exposed sections of similar height can appear much more alike than the complete objects actually are.

These dimensions belong to the set recorded in the exhibition checklist. They have no direct claim on the dimensions of the sheets in front of Musashi. Script size, generous margins, blank paper at either end, the way separate sheets were joined, and later mounting can all affect length. One can reasonably ask why Void looks so short beside Fire without converting the ratio of their lengths into a count of Musashi's words. The checklist gives a broad Edo-period dating for this set. A precise length to the nearest millimetre can coexist quite comfortably with uncertainty of more than a century about a date.

The museum noted that different scrolls would be shown over the course of the exhibition. Rotation has consequences for the history a visitor can write. Two people who saw this exhibition on different days might both accurately remember seeing Gorin no Sho and have seen different handwriting, titles or passages. A photograph labelled only with the exhibition name cannot establish which roll was open. It helps to copy the object label while the relationship between label and paper is still visible. Once the picture leaves the gallery, a missing book name can be surprisingly hard to recover.

Yagyu-school iron guard photographed at Tokyo National Museum
Yagyu-school iron guard photographed at Tokyo National Museum. Ihimutefu (via Wikimedia Commons). Image licence. Earlier catalogue comparison: Shrike on a Withered Branch, ink painting by Miyamoto Musashi.

Two doubtful entries that tell us a great deal

Far from the headline cases, the catalogue of the Tominaga martial documents at Kumamoto Prefectural Library offers a different introduction to scholarship. Entries fifty-one and fifty-two have no surviving title that the cataloguer could confidently supply. Both are described as writing in ink on rolled paper; both are damaged. The catalogue tentatively associates them with the Five Rings. One begins with text already in progress, and a reference to the Wind book occurs in the opening words transcribed for the other. These scraps of information are more instructive than a silently completed title would have been.

The library records a height of just over twenty-four centimetres. A reader comparing this with the Eisei Bunko objects is comparing two distinct groups of physical evidence. There is no need to invent a rival literary school to explain the difference in the height of a roll. Nor is there a reason to force all surviving forms into the five-part presentation familiar from a modern edition. The particular damage recorded in the inventory, including insect damage and losses, limits what the librarian could establish from the beginning and end of the papers. Conservation details here have an immediate bearing on literary interpretation.

The catalogue transcribes names and dates from the end matter. Seventeenth- and nineteenth-century dates appear together in the record of the first paper. That is precisely the sort of detail that disappears when a digital interface allows only one date per object. A copying or transmission statement can preserve the date of an earlier handover even when the paper carrying that statement belongs to a much later transaction. For a family keeping a record, the earlier date was part of the document's value. For a historian, retaining both dates exposes a chain that a single number would conceal.

Reading a colophon with a pencil and a separate sheet

One way to read such a catalogue is to rule a sheet of paper into three columns. In the first write what the historian can see or what the librarian physically measured. In the second write what the manuscript says about its own history. In the third write what later editors or family tradition have inferred. The doubtful Tominaga papers fill all three columns unevenly. Their height fits easily in the first. A name in the end matter belongs in the second. A proposed title for the damaged text sits in the third with the question mark intact.

This modest exercise also answers a common question about copies. A copy can be the earliest available evidence for a passage and at the same time a late witness to a claim about ownership. Copyists preserve spelling, modify it, fill a gap from memory or leave a blank they cannot resolve. Where two copies disagree, the visually older one does not automatically preserve the earlier reading. Two late copies might descend from a source older than the surviving early witness. Establishing that relationship requires comparing repeated agreements and distinctive errors across a substantial stretch of writing.

For readers in translation there is a corresponding experiment. Set two translations of a short passage side by side and look first at the editorial apparatus. Does either translator identify the Japanese text used? Is the break into paragraphs the same? Does one put a heading where the other leaves the line uninterrupted? A difference can arise before the translator makes a single disputed word choice. The physical break between rolls, the division supplied by a modern editor and the blank line on a digital screen represent three different moments in the work's history.

An iron fitting and a brush painting

Musashi is visible in the Shimada collection through objects that involve no sustained sequence of words at all. Among the best known are a brush painting of a shrike perched on a withered branch and an iron guard with openwork conventionally read as paired sea cucumbers. The museum credits the painting to Musashi and describes the guard with the qualification den, an attribution received through tradition. The difference is easy to miss when the objects are introduced together under a famous man's name. It changes how one can use each as evidence for his artistic practice.

Consider the distribution of blank space. On paper the bird occupies a small part of the available field. The spectator's eye has to travel along the branch and through untouched areas of the sheet. In an openwork guard, the wall or cloth behind the object becomes visible through actual holes. What looks like the unmarked part of a composition is physically present paper in one case and the absence of metal in the other. The comparison is worth making on its own visual terms. A shared taste for economy remains a comparison; it cannot identify the person who drew a brush across paper or commissioned an iron plate.

Sea cucumber is an unexpectedly precise name for a motif that can look almost abstract in outline. The name gives an observer permission to recognize a living form without demanding naturalistic details such as a particular skin texture. A display against a pale backing invites the eye to read the openings as shapes. Against a dark backing, the iron outline may predominate instead. Two catalogue photographs could give a reader quite different intuitions about the design while representing the very same piece. Neither lighting arrangement resolves the traditional attribution.

Historic site of Reigando, Kumamoto
Historic site of Reigando, Kumamoto. Grattepierre (via Wikimedia Commons). License.

A cave, a portrait and the expectation of a likeness

Reigandō outside Kumamoto occupies a prominent place in local memories of the writing. A visitor can still experience the transition from landscape to enclosed stone. What survives at the site makes the geographical tradition tangible. It supplies very different information from the colophon of a copy. Weathering of rock cannot establish at which stage the five books took their present form. A view of the site is best captioned with the place and the date of the photograph. It can then stand alongside the manuscripts without impersonating an image of the author at work.

Portraits create a related problem through the opposite visual means. The visitor sees a face and instinctively expects a person who sat still before the artist. Kumamoto City's account of the Shimada portrait preserves the tradition that it passed through the Terao family and was understood as a self-portrait. The museum's listing leaves the painter unidentified. Both details matter. The family history explains why the portrait had authority and why later images looked to it. The museum's restraint keeps open a question that inheritance alone cannot settle.

Other items in the collection carry the image of Musashi well beyond the lifetime of the students who knew him. Paintings of him in youth, later prints of an encounter on Ganryūjima, and much later representations of earlier incidents make a small history of audiences. Dating these works to their own periods gives them more interest. A print can reveal what a late Edo public expected a celebrated encounter to look like, including what parts of the action required explanation in an inscription. No seventeenth-century witness could have foretold all of those later expectations.

Painted eagles and bamboo at Kanchi-in, Toji, Kyoto
Painted eagles and bamboo at Kanchi-in, Toji, Kyoto. Miyamoto Musashi (via Wikimedia Commons).

What one can take home from Higo

Even the expression Higo relics deserves attention. It can mean things preserved in the region, things attributed to a man who spent his last years there, or things that a modern institution assembled to tell that story. The Shimada is a private art museum, while Eisei Bunko preserves collections associated with a great family, and the prefectural library inventories a named donation. Their different origins help explain both the remarkable range of evidence and the omissions. Someone searching only the category of fine art might miss the most instructive doubtful fragments in a library list.

The next time a bookseller or an exhibition poster shows the five elemental titles together, imagine three objects beside the familiar cover. There is the Wind copy, close to the people who transmitted the writing. There is the long Fire roll whose exposed passage could change during a museum exhibition. There is the damaged sheet whose missing beginning prompted a librarian to print a question mark. Each preserves a part of the material history of reading that a beautifully typeset edition leaves out. Together they explain how Musashi's book could keep a recognizable identity while changing in its handwriting, dimensions, ownership, audience and language over nearly four centuries.

There is a practical consequence for describing the book in English. Give a manuscript two places in a catalogue when the evidence allows it: one for the setting in which the work was composed and another for the institution that cares for this particular witness. Keep a third place available for the collection from which it came. Higo, a family name and a Tokyo collection can then all appear on one label without contradiction. The same discipline helps with a printed reproduction of a painting. Record the artist attribution as it stood in the lender's entry and identify the photographer or publisher independently. A century from now a reader may be as interested in the history of that reproduction as we are in the history of a paper copy made after Musashi's death.

Frequently asked questions

Does the original manuscript of The Book of Five Rings survive?

The official guide to the Shimada collection reports that no manuscript in Musashi’s own handwriting survives.

Which part of Gorin no Sho is in the Shimada Museum of Art?

The museum lists a manuscript of the Wind book. Its government collection guide identifies it as a copy made by Terao Magonojo in 1651.

What are the five scrolls called?

Earth, Water, Fire, Wind and Void are the conventional English names for the divisions. Eisei Bunko preserves an Edo-period set of five scrolls.

How long is a historic Five Rings scroll?

In the Eisei Bunko set the Fire roll exceeds twelve metres and the Void roll is under four. Other copies have their own dimensions.

Did Musashi make the Shimada sea-cucumber guard?

The museum records a traditional attribution to Musashi. The Japanese qualification den is important to understanding the label.

Why is Kumamoto central to the history of the book?

Musashi settled there at the invitation of Hosokawa Tadatoshi in 1640. The local museum and library preserve different witnesses to the transmission of his work.

Films and talks

熊本県博物館ネットワークセンター — 宮本武蔵と熊本の深い関係!?【島田美術館】

【公式】熊本市西区チャンネル「まちセンTV」・「なんさま にしくで ごゆるりと」 — 明治の町屋と小さな美術館〜島田美術館〜

Collection records and further reading

Related Articles