A judge raises a flag, or does not. You saw the bamboo sword land on the mask. You heard it. Nothing happened. Two exchanges later a strike that looked identical ends the match, and the hall makes a sound of recognition you are not part of.
If you came expecting Japanese fencing, this is the first thing that will not fit. In épée, contact is the point; in kendo it is one of three things that must arrive in the same instant, and the other two are harder to see. This article is for the person in the seats, not the person in the armour.
In a sentence adopted in 1975, the All Japan Kendo Federation defines kendo as the way of human formation through the discipline of the principles of the sword. Not a sport that builds character as a side effect, but a dō — a way.
Épée answers one question quickly and without argument: did the blade touch first. Kendo answers a different question, and accepts that the answer takes longer and needs a human being to make it. A strike that touches a valid target with a valid part of the sword can still be worth nothing, and the referee is not wrong. The same dō attaches to Japan's classical performing arts, to calligraphy, to tea.
This has a documented answer, and it has flipped twice. In 1919 Nishikubo Hiromichi argued that the fighting words should go: kenjutsu and gekiken, swordsmanship and fencing, should become kendo, and the older words for martial skill should become budō, a martial way. In August 1946, under the occupation, the Ministry instructed that the word budō be dropped altogether, reasoning that the characters themselves carried military meaning. And yet the federation today defines kendo as a budo again. The useful question is not whether it would work in a street fight, but what the technique is for.
Kendo did not descend from swordsmanship in a continuous line. It broke three times: a law, an occupying army, and the disappearance of the target itself.
Two and a half centuries of enforced peace turned a battlefield technique into something else: swordsmanship rose from killing into the "life-giving sword", aimed at the formation of a person — beginning with the samurai class whose weapon this once was.
On 28 March 1876 the Dajōkan, the new Meiji government's highest organ, ended sword-wearing as an ordinary fact of life with the edict remembered as the haitōrei. The National Archives of Japan holds the document, filed as volume Tai-00223: carrying a sword was prohibited except in full court dress, and for military, police and officials in uniform. The exceptions are the story: the sword became not illegal but a uniform item — permitted to the state, forbidden to the person. The samurai class had been abolished, and the federation's history records the consequence bluntly: swordsmanship declined. What vanished was not the techniques but the everyday reason to have them. Revival then came from the state that had disarmed them: from 1877 it returned through the Metropolitan Police.
The edict did not arrive alone. On 5 August of the same year the Dajōkan issued Decree 108, and the hereditary stipends paid to the former warrior class were commuted into public bonds worth between five and fourteen years of their value — the settlement known as chitsuroku shobun. Read the two together and the year changes meaning: in March the sword came off the body, in August the income that had made carrying one a profession was gone. A skill survives losing its tool, and it survives losing its patron. Losing both within six months is the pressure everything afterwards answers.
Japan was under Allied occupation from 1945 to 1952. Sakaue Yasuhiro of Hitotsubashi University sets out what it meant: martial arts were banned under demilitarisation, in school physical education and clubs alike. At least 1,927 secondary-school budo teachers lost their posts, and the Dai Nippon Butoku Kai — founded in 1895 to preserve exactly this — was dissolved.
What happened instead is the disguise. Under the guidance of the occupation's own Civil Information and Education section, practitioners built a sport closely resembling kendo under another name: shinai-kyōgi, bamboo-sword competition, first held nationally in Nagoya in 1950. In April 1952 it was shinai-kyōgi, not kendo, that entered school physical education; kendo itself returned only in July 1953. The All Japan Kendo Federation was founded in 1952, once the occupation had ended: the body whose definitions this article cites is younger than the disguise it replaced.
The historiography then gets uncomfortable. Okada Morikata of Kōchi University counted how kendo instruction books narrate the period. Of the 80 he verified, 63 per cent describe the prohibition and 64 per cent the revival — but only 50 per cent describe wartime kendo at all. The literature, he argues, treats the wartime flourishing positively and receives the ban as a victim would. Take either side: a practice crushed politically, or an art that never accounted for its own use.
The third break has no date, because it was not an event. Across the Edo peace the target stopped being a body, and so it had to be invented — equipment first: in the 1710s Naganuma Shirōzaemon Kunisato spread a method of striking with a bamboo sword in protective gear, and in the Hōreki era (1751–1764) Nakanishi Chūzō Tanetake added an iron-barred mask and bamboo armour. Armour on head, wrists, torso and throat means those four places, and only those, can be struck: the valid target and the padding line up, which is worth noticing when you watch.
A valid strike is called yūkō-datotsu, and the federation defines it in Article 12 of its competition regulations: an accurate strike to a designated target, executed in high spirits with correct posture, with the striking section of the shinai at the correct blade angle, and followed by zanshin, the alertness held after the strike lands. The shorthand is ki-ken-tai-no-itchi: spirit, sword and body as one, arriving together, not in sequence. The strike you saw land and go unrewarded missed one of those requirements, and three referees saw the gap you could not.
The referees are the most legible thing in the hall, and almost nobody watches them. A match is judged by three shinpan-in, who move around the pair in an isosceles triangle with the shushin, the head referee, at the apex and two fukushin at the base. Each carries two flags, one red and one white, matching the ribbons worn by the competitors. A point requires two of the three to raise the same flag: a single flag is not a score, and a referee who is unconvinced must still indicate a judgment rather than let the majority carry it. All three have equal authority over what counts, though as a rule it is the shushin who halts the match. Watch the flags rather than the players for one match and the standard stops being abstract: you see which strikes two of three adults would sign for.
The last requirement is the one most likely to surprise you. The federation defines zanshin as the mental and physical posture taken after executing a strike, and instructs referees to judge the whole action by what the striker does once the blade has landed. The consequence is the most useful thing a newcomer can carry into a hall: a competitor who shows excessive excitement, claiming the strike's validity in an improper manner, has it deemed invalid — not penalised afterwards, but invalid, as though the blow had not happened. So a player who lands a clean strike, then turns and holds guard with no expression, is not being modest. That is the last requirement being met.
Ki is the spirit component, and the shout, kiai, is its audible form — not an accompaniment but part of what makes the strike count. Experienced spectators listen for the shout as part of the strike, not as decoration. Listen to when the voice starts: a shout that begins before the commitment reads differently from one that arrives after the hit that was not there. What the requirement rewards is committing before the outcome is known.
A shinai has no edge, yet a strike delivered with the flat does not count. Hasuji is that blade angle — the component most invisible to a new eye, since correct and incorrect look alike on a moving object. When the Dai Nippon Imperial Kendo Forms were established in 1912 — the first official use of the word kendo — one stated purpose was to correct strikes that ignored the blade angle. The principle it defended — that the shinai is above all a substitute for the Japanese sword — is still doctrine.
Tai is the body, and its signature is fumikomi, the stamping step driving the strike. For a spectator it is the one component you can verify with your eyes closed. The right foot slaps the floorboards. The shinai lands. The voice is already there. If those three are one sound, you are probably looking at a valid strike. If you hear them as two, something did not hold and the flags will stay down.
To an untrained eye a match has long stretches where nothing occurs: two people at distance, tips crossed. It reads as a lull; it is the opposite. The strike is the conclusion; this is the argument. Almost everything determining the outcome happens before either moves, in two things you can read: distance and pressure.
Issoku ittō no maai is the discipline's reference distance: one step, one sword — the gap from which a step forward lets you strike and a step back lets you evade. Learn to see it and the fidgeting resolves into content: those small entries and withdrawals of the front foot are a negotiation over a boundary, each holding for a distance where their step reaches and the other's does not. The players are not waiting; they are arguing about where the line is.
Semeai is the pressure exchange under the stillness: each tries to induce a movement they have already prepared for — a flinch, a raised hand, a breath at the wrong moment. It leaves a question worth carrying into the hall: did the player who moved first seize the initiative, or was that player drawn out, so the strike merely records a decision already lost?
The word for that stillness is ma, and it is not a kendo word — it belongs equally to Noh, tea, calligraphy. When Bespoke Discovery interviewed Takeda Tomoyuki, a Kanze-school shite-kata Noh performer, we asked whether ma matters as much as its reputation suggests. His answer began with the difficulty: "Yes. But it is very difficult to define ma in a single phrase." What followed describes a kendo standoff — a Noh performer's account of ma:
"Still, in moments when no one is moving, when nothing appears to be happening, when it is so quiet that one can hear the rustle of clothing, there is a great deal that is entrusted to the audience's imagination. I think that too is one form of Noh's ma."
He goes on: "It is not merely that everything has stopped. There is a presence within that stillness. If even the slightest movement occurs, the air changes. And though nothing seems to be happening, there is in fact considerable tension." Substitute semeai for ma and nothing breaks.
In kendo the equipment carries the doctrine: if the making stops, the standard stops with it.
A shinai is four staves of bamboo bound so they act as one shaft — a late-Edo design adopted because it was stiffer than the older leather-sheathed fukuro-shinai. It is a consumable: staves split and are replaced. Japan publishes no statistics on bamboo grown for kendo equipment. What the Forestry Agency publishes is the trade as a whole: production is falling as substitutes spread and cheap imports increase, with output at 863,000 bundles in 2025, down 1.8 per cent. That covers bamboo of every kind, not the graded culms a maker accepts, and narrowing it would mean inventing a number. Nor is there a count of the people who cut and fit them.
The armour, bōgu, is four separate objects. The men is built around the men-gane, the grille you look through at a player's face. The kote (gauntlets) and tare (waist guard) are defined by sashimono, their dense parallel stitching. The classic dō — the chest guard, a different word from the dō above — was late-Edo work in tanned leather hardened with lacquer. Lacquer is where the supply chain becomes legible, because it is counted. The Forestry Agency reports that 93 per cent of Japan's lacquer consumption is imported, mostly from China. That is the national figure, not one for armour, but it locates the material base.
A kendo player never handles a sword; a polisher handles nothing else. When we visited the workshop of Nagaoka, a togishi or sword polisher, the first thing we noticed was how he sits: a posture resembling nothing in ordinary movement. In his hands was a wakizashi, the shorter of the paired swords, made about 450 years ago in Bizen. He is a polisher who chooses not to finish, and by his own estimate the trade numbers a hundred at most.
His ethos inverts what a spectator expects. "What almost everyone is taught is: do not reduce the sword," he says. "To polish is to reduce. But anyone can reduce. To finish it cleanly without reducing it — that is the skill." He works to a concept he calls togishiro: steel left unspent for whoever polishes the blade generations later.
"To put it simply," he says, "the sugata — the overall shape, the length and the curve; the hamon — the pattern that appears along the edge; and the jigane — the skin of the steel itself. You look at these three elements and take pleasure in them." That is the same shape as ki-ken-tai-ichi: several criteria held at once. A kendo player is judged on the angle of a blade that is not there; Nagaoka judges the blade that is.
Competition is the smaller part of kendo. The larger part is keiko, practice, which a visitor is likelier to be shown than a tournament. Guides for people who want to join are everywhere; this one is for the person in the doorway. These halls sit in ordinary neighbourhoods rather than a cultural district, which makes them a natural stop on a quiet route through Tokyo.
Practice begins by sitting still: mokusō, a short seated silence, then suburi, swinging the shinai in unison on a count. How long the silence runs, and whether it is genuinely silent, is the hall's discipline rather than its claim. Who is in the line — children, working adults, people in their seventies on one count — tells you whether the hall reproduces itself. Where the teacher stands, in the line or apart, tells you how correction is delivered.
Then the drills escalate: kirikaeshi, alternating cuts while advancing and retreating; uchikomi onto a target the partner offers; then jigeiko, free practice. This is the part a first-time spectator is most likely to read as warm-up rather than as the practice itself. Apply the standard one repetition at a time: each cut corrects something different — blade angle, the foot, a late voice. A hundred identical-looking cuts are a hundred separate corrections. Then watch the pairings: a senior who gives a beginner a full engagement rather than a token one tells you what the hall is for.
Access to kendo is layered, and knowing the layers exist is most of the value. The first is open to anyone: public tournaments need no introduction. The second requires knowing it exists: sessions at municipal halls, early-morning keiko, the kangeiko of midwinter — none secret, none advertised. The third needs an introduction: sitting in on practice at a private dojo, talking with a sensei afterwards. We are not printing contact details or booking routes. The point is that the third layer exists at all, and search does not reach it.
The question is not whether kendo survived the twentieth century — it did — but what the arithmetic is now. Three things are moving at once, and not the same way.
Nobody publishes this. There is no national dataset on the cost of running a dojo: the bodies collecting kendo statistics are federations and schools, none with reason to publish a private hall's accounts. The structure can be stated: armour is a durable good with a long replacement cycle and a high unit cost, while a shinai breaks on a schedule nobody controls, and a hall teaching children funds both from monthly fees.
The number everyone quotes misleads. The federation reports 2,081,669 registered kendo dan holders as of the end of March 2025, against 2,027,257 in March 2023. The figure went up — but it is a count of grade registrations, not of people currently training; it cannot say whether kendo is shrinking.
The decline that is genuinely measured sits one level down, in who is available to start. The Statistics Bureau reports that the population under 15 stood at 13.29 million on 1 April 2026, 350,000 fewer than a year earlier — the 45th consecutive annual decline and the lowest on record. By three-year age band it runs from 3.09 million aged 12 to 14 down to 2.13 million under three, every band smaller than the one above. A hall recruiting eight-year-olds is not guessing; the next decade's cohorts are already counted.
Against all that, kendo is expanding outward. The International Kendo Federation states its arithmetic plainly: the body was established in 1970 with 17 affiliated countries and regions, and as of July 2024 has 64. The 20th world championship is scheduled for Tokyo in May 2027. The asymmetry is the story: the pool of children inside Japan contracts on a fixed schedule while affiliates outside have almost quadrupled. An art banned in 1945 as nationalist is now sustained, in part, by people who are not Japanese — which includes you.
Take three lenses. The standard: a strike counts when spirit, sword and body arrive in one instant, so listen for whether foot, shinai and voice are one sound or two. The stillness: the exchange is decided before anyone moves, and as Takeda says of the same interval in Noh, a great deal is entrusted to the audience. The chain of hands: bamboo, stitching, grille, lacquer, and behind it the steel.
The Edo peace had already turned the technique into a way; 1876 removed every other use for it, leaving that the only answer. When the flags go up and you understand why, you are watching the end of that argument.