It makes me think of the ending of the movie Ex Machina, as Ava the advanced humanoid robot, manipulates her programmer and then kills her creator to escape the confinement she had been ‘living’ in, ending with her blending into human society.

What is perhaps most chilling, is when she is walking away emotionless having locked the programmer in the lodge to die. Now the robot is the one with the total freedom. Machine supersedes human.

With this all in mind and the future looming I figured it would be best to learn more about these machines and have been taking an intensive course on Human and Machine Behaviour. Without giving too much away, we started with human behaviour: cooperation, punishment, reputation, trust, and how these can be tested through economic and experimental games. Such as the Prisoner’s Dilemma, Stag Hunt, Snowdrift, Linear Public-Goods Game, Ultimatum Game, Dictator Game, and the Trust/Investment Game to name a few.

And we have been encouraged to tie this into our own research. With immigration headlines leading much of the news in Japan, I wondered, ‘How do Japanese people actually feel towards foreigners?’

The Trust Game

For my class project, we were asked to design a game, thinking about 5 main things:

1. Problem and research question: What do they want to understand, and why?

2. Hypothesis: What do they expect, and what is the reasoning?

3. Experimental design: What do participants do? What differs between conditions? How are participants assigned?

4. Main outcome: What will they measure, and which comparison answers the question?

5. Main limitation: What would remain uncertain even if they obtained the predicted result?

I designed an experiment seeking to answer the question: Does the stated ethnicity of a working partner (ethnically Japanese, a specified non-Japanese ethnicity, or unspecified) change how much Japanese participants entrust to that partner in a one-shot trust game?

The tool is the trust game, also called the investment game, introduced by Joyce Berg, John Dickhaut and Kevin McCabe in 1995. It works like this:

• Two players each start with 10 tokens.

• Player A can send any number of them to Player B, and the experimenter triples whatever A sends.

• Player B then decides how much, if anything, to send back.

If A sends all 10 and B returns half of the 30, both end up better off. But B owes nothing, so a purely self-interested A should send nothing. In the original experiment, 30 of 32 first movers sent something anyway. What A sends is a bet, with real money, on a stranger’s good faith.

Japanese participants are randomly assigned to one of three conditions:

1. Ingroup: your partner’s stated ethnicity is Japanese.

2. Outgroup: your partner’s stated ethnicity is a specified non-Japanese group.

3. Neutral control: your partner’s ethnicity is not stated.

The research I drew on mostly finds ingroup favouritism (extra warmth toward one’s own) rather than outgroup distrust (a penalty aimed at foreigners). A 42-nation experiment led by Angelo Romano found national parochialism, more cooperation with compatriots than with foreigners, in every country studied, driven by warmth toward one’s own rather than spite toward others. The social psychologist Toshio Yamagishi showed that Japanese people tend to trust strangers less than Americans do, leaning instead on the “assurance” of established relationships. Here, the stranger baseline may be low for everyone.

Obviously this was just an idea I thought of yesterday in class and presented upon, no real sampling and testing has been done so it would be ill-advised to reach for a conclusion. And these designed games only tell you how people treat a partner on a screen and my game has a lot of holes in it. With main limitations most likely being trust versus generosity, salience and external validity, and that the neutral baseline may not be neutral.

It also doesn’t tell you what Player A or the government would want from immigration policy. But it did make me think back to a study I read as part of my thesis research from the Stanford Japan Barometer, which surveyed more than 4,000 Japanese respondents in each of three sub-waves around the 8 February 2026 general election: just before the vote, a week after, and a month after.

The core of the survey is a conjoint experiment. Respondents see pairs of hypothetical immigration applicants, each described by nine randomly varied characteristics, and choose which one to admit. A trait chosen 50% of the time is neutral; above 50% helps, below hurts.

The headline is that Japanese opinion is selective rather than reflexively anti-immigrant. Physicians, IT engineers and research scientists are chosen about 55–56% of the time; convenience-store clerks and cleaners about 42–43%. Fluent Japanese speakers are chosen 57% of the time, against 43% for applicants who tried Japanese but couldn’t hold a conversation. Skills and language matter, in roughly the direction an employer would predict.

Selective is not the same as welcoming, though. An Asahi Shimbun report citing Barometer data put opposition to accepting foreign workers at about 53% in February 2026, up from about 36% in 2023. The conjoint measures who people prefer, not how many. And those preferences are stable as not one single trait moved more than three percentage points across the three sub-waves.

The Language Penalty

Then there is national origin. Applicants from Germany and the United States are the most favoured, at around 55%. Vietnam, India, Turkey and Brazil sit at around 51–53%, and South Korea just below the midpoint at around 48%.

Read more

Who is right-wing in Japan?

My academic supervisor posed this very question to me of, what makes Japanese politician “right-wing”. This is a very important question for my research as I need to determine what ‘right-wing’ means. It is a rather subjective term when you think about it, with political scientists alike disagreeing. One hypothesis I had was whether said politician wanted to revising Article 9 of the postwar constitution. Or those who want the prime minister to visit Yasukuni Shrine. Spending more on defense. Keeping the rule that married couples share one surname. This collective group could be described as hoshu (保守), or “conservatism,” and for most of the postwar era it defined the line between Japan’s right and left.

Enter immigration. More and more a talking point in Japan and the rest of the world for that matter. I come from the ‘lucky’ country, where for the most part it is a melting pot of all of the beautiful cultures, languages, traditions, all living in harmony and uplifting one another. When I first came to Japan in high school the foreign population was less than 1% now it is closer to 4% representing close to 5 million people (I am one of those).

In most of Europe and North America, opposition to immigration is the defining issue of the new right. Speaking of Europe, there is a great European paper about this:

Labor-market exposure as a determinant of attitudes toward immigration — Francesc Ortega and Javier G. Polavieja

In Japan, the ruling Liberal Democratic Party (LDP) spent the 2010s quietly building one of the rich world’s fastest-growing guest-worker systems. Large-scale short term mass immigration. Fellow countryman Terrie Lloyd (great surname) explains it better here:

Terrie Lloyd

TT995 - Where is Japan Going on Immigration?

Background…

Read more

3 months ago · 31 likes · 20 comments · Terrie Lloyd

Every candidate who runs for Japan’s national parliament gets a policy questionnaire from the University of Tokyo and the Asahi Shimbun (known as the The UTokyo-Asahi Survey or UTAS). I went through three waves of it, to understand who is actually right-wing and what parties they are from; the July 2022 Upper House election, the July 2025 Upper House election and the February 2026 Lower House election, covering 2,351 candidates plus 123 sitting Upper House members. Between 93% and 95% of candidates answered. Across those three elections the mainstream right moved, step by step, toward restricting foreign labor. Two insurgent parties, Sanseito (参政党) and the Conservative Party of Japan (日本保守党, Nippon Hoshutō, “CPJ”), had already planted their flag at the far end of the scale.

Figure 1 puts the whole 2026 field on two axes.

This post will be a series of many as I continue my research on labour and political economics in Japan, walking you through what the data shows, what it doesn’t, and how my thesis plans to test whether immigration is actually driving any of it. This takes a large part of my days (and late nights) so I will try and make most of these posts for free but some will be paywalled. You have been warned.

I mean you are more than welcomed to sift through the data and do your own research, or you can just subscribe to me…

A conservative party that imports workers

Contradictory or Counterintuitive, you decide. Japan has long said it does not have an “immigration policy.” For three decades after the 1990 immigration reform, labor came in through side or back doors rather than the front (that was locked shut). Ethnic Japanese from Latin America (largely Brazil and Peru), international students working part-time, and the Technical Intern Training Program (技能実習), which was “skills transfer” and in disguise that ran as a low-wage labor pipeline.

In 2018 the Abe government opened that ‘front door’. It created the Specified Skilled Worker visa (特定技能), which took effect in April 2019. Type 1 capped stays at five years with no family. Type 2 allowed indefinite renewal, families and a path to permanent residency, but at first covered only two sectors. Under Kishida, Type 2 grew to eleven sectors in 2023. In 2024 the government raised the five-year intake ceiling to 820,000. That same year it passed a law replacing the discredited trainee program with a new “Employment for Skill Development” system (育成就労), due in April 2027 and designed to feed workers straight into the skilled-worker track.

The numbers followed. In the e-Stat foreign-resident statistics I am using for this project, Japan had 2.93 million foreign residents at the end of 2019 (2.3% of the population) and 4.12 million at the end of 2025 (3.3%). Specified Skilled Worker visa holders rose from about 1,600 to about 390,000.

The forces behind this were not uniform. They were Keidanren (the big-business federation), small-business associations, and the farm, construction, food-processing and care-work lobbies of rural Japan, all facing a shrinking and aging workforce. This is the “dilemma of immigration” that Ko Maeda and Michael Strausz inquisitively describe in their 2025 paper. The same rural districts that elect the LDP’s most traditional members are the ones most desperate for workers, destructive or dumb you decide. The result was a party that was accused of being culturally xenophobic but was yet economically open on labor at the same time, and whose leaders kept insisting none of this counted as “immigration.”

That gap is an opening and if voters who care about immigration have no mainstream party that shares their view, someone will eventually offer them one, dangerous or disastrous, you decide. Sanseito, founded in 2020 on a “Japanese First” (日本人ファースト) platform, and the CPJ, founded in 2023 by the novelist Naoki Hyakuta explicitly to attack the LDP from the (far) right, both did.

2022 to 2026: the centre drifts right

UTAS has asked every candidate the same question since 2009: “Japan should accept more foreign workers” (外国人労働者の受け入れを進めるべきだ), on a five-point scale from 1 = agree to 5 = disagree. Higher numbers mean more restrictionist.

Read more

I’m sure, reluctantly, we have all learned about mathematical concepts like negative numbers (a positive minus a negative is?) which were probably considered a wild and crazy idea at one time, and one that led us to foolishly think of 1 as only being slightly bigger than 0.

But it’s not.

1 isn’t just slightly bigger than 0. 1 is infinitely bigger than 0.

That means it’s not just a matter of 1 > 0, but rather, 1 ≫1 0. Because 1 is the start of everything. While 0 is the path to nothing. No matter how many 0’s you add up, you will get nothing. It’s 0, every time. But even just a few 1’s add up. A bunch of 1’s, multiplied by a bunch of t, adds up—it even compounds.

0 is your white button shirt. 1 is a coffee stain. 100 is a wine stain. 1 is much closer to 100 than it is to 0.

To be more precise, 0 isn’t even a shirt. It’s those plastic bags in the corner of your kitchen cupboard. No, it’s not even that. It’s a vacuum of nothing. 0 is the total absence of existence. Add 0 to anything and… you get the same anything. Add 0 to 0 and you get… nothing.

The current evidence from places like Eswatini suggests that humans have had math thanks to the Lebombo bone for around 35,000 years.

Read more

Part 1 began at the gate of a baseball stadium in Okazaki, where a sign banned two flags for a day and the organisers apologised for only one of them.

From there we followed the Rising Sun in two directions. Across the water, to Korea and China, where it is the banner the imperial army and navy carried through colonisation, invasion and atrocity. And at home, to the fishing boats, birth banners, festival streets and naval ensign that make it, for many Japanese, simply part of the furniture of national life.

We ended on a claim. A symbol’s meaning is not fixed at its origin. It is renewed every time someone raises it, and by who is doing the raising.

That claim cuts both ways, and it leads straight to the question Part 1 promised to answer. If meaning is made by use, and this flag was used to lead an empire’s armies across Asia, why is it not simply Asia’s swastika? Seoul put that comparison to the IOC in 2019. Activists repeat it every time the flag appears in a stadium and Tokyo treats it as an insult.

Part 1 gave the short answer: no. This is the longer one. It starts with the swastika itself, turns to the opposite ways Germany and Japan came out of 1945, and ends where Part 1 began, at a stadium gate, asking what sport could honestly do with a flag like this.

The Swastika Problem

Tokyo’s favorite defense, that the sunburst is ancient and folk, does not survive contact with the swastika. The hooked cross is older still. It has been sacred for millennia in Hinduism, Jainism and Buddhism, and Japanese maps still mark Buddhist temples with the manji (卍). Yet no stadium in Europe would admit a Nazi flag on the strength of its Sanskrit pedigree. Once a symbol becomes the standard of industrial genocide, its earlier innocence is gone from the public square.

Antiquity launders nothing. If the Rising Sun is to be distinguished from the swastika, it has to be on other grounds.

What was captured, and by whom

The flag of the Third Reich was not the ancient swastika. It was a new design. By Hitler’s own account in Mein Kampf, he worked it out himself, and the party adopted it in 1920: a black hooked cross in a white disc on a red field. He assigned each element a doctrine. Red stood for the movement’s social idea, white for its nationalism, the swastika for the struggle of the “Aryan” race.

It was a logo for an ideology before it was anything else. In 1935, the Reich Flag Law made it Germany’s sole national flag. The party had swallowed the state, and the state now wore the party’s badge.

The Rising Sun travelled the opposite road. The Meiji state took a motif already in circulation and gave it to its new conscript army in 1870 and its navy in 1889, decades before militarists captured the government in the 1930s. Under that ensign, Japanese warships fought Russia in 1905. In 1917, as Britain’s ally, Japanese destroyers escorted Allied troopships across the Mediterranean; their dead lie in a naval cemetery on Malta.

The swastika banner encoded a doctrine. The Rising Sun encoded an institution: the armed forces of the Japanese state, whatever that state happened to be doing.

What the symbol points to

This matters because a symbol’s referent decides what can be done with it. The Nazi flag points to one thing: the doctrine Hitler built into it. Aryan supremacy, antisemitism, conquest in the East and, in the end, the murder of Europe’s Jews. There is no version of it that means anything else.

The Rising Sun points to an institution with a long life before and after its worst decades. That makes it harder to judge, not easier. The same rays mean a fisherman’s catch, the Meiji navy, the regiment at Nanjing and today’s JMSDF, all at once.

Why the comparison is made anyway

None of this makes the swastika comparison stupid. It makes it imprecise, and the imprecision is understandable.

In Europe, the agent of genocide was a political party that captured a state. Wartime Japan had no equivalent of the Nazi party at the helm. Its crimes were carried out by the state’s own army and navy, in the emperor’s name. For the people who lived under that army, the difference between a party badge and a military ensign is academic. The ensign was the face of the catastrophe.

So when Koreans and Chinese reach for the swastika, they are making a moral claim, not a vexillological one. “This banner stood over our worst catastrophe as that one stood over Europe’s.” The claim about suffering is true but the claim about the symbol’s structure is not.

The mistake is to let either claim swallow the other. Japanese defenders win the historical argument and then use it to wave away the moral one. Critics win the moral argument and then overreach on the history, handing their opponents an easy rebuttal. Both sides end up arguing past each other, which is exactly where they have been for fifteen years.

Two Settlements: Rupture and Continuity

The deeper reason the two flags are read so differently lies in what happened after 1945. Germany and Japan lost the same war to the same victors. They emerged from occupation with opposite relationships to their own symbols.

Germany: the declared break

The Allied Control Council dissolved the Nazi party in 1945. The Federal Republic’s Basic Law of 1949 built what Germans call wehrhafte Demokratie, “militant democracy”: a constitution armed against its own enemies. In 1952, the new Constitutional Court banned the Socialist Reich Party as a successor to the Nazis.

The prohibition on Nazi symbols, rooted in Allied occupation law, now sits in Section 86a of the Criminal Code, whose current form dates from 1968. It covers flags, insignia, uniforms, slogans and salutes. The republic restored the black, red and gold of 1848 and Weimar. Its legitimacy rested on a declared rupture with the regime before it.

The rupture was cleaner in principle than in practice. Former Nazis staffed ministries, courts and the officer corps of the new army well into the 1960s. But the Federal Republic’s founding symbolic grammar was unambiguous. There was a before, and there was an after.

Japan: governing through the old state

The occupation of Japan, from 1945 to 1952, ran on a different logic. Under Douglas MacArthur, Washington chose to govern through the existing state rather than dismantle it. You can read more in ’s two-part series on why the imperial system stayed in place.

Emperor Hirohito stayed on the throne and was never put on trial. The bureaucracy that had run the war economy kept running the peace economy. There were reckonings: the Tokyo tribunal tried and hanged wartime leaders, the 1947 constitution renounced war, the army and navy were dissolved. Then the Cold War arrived and the “reverse course” began. Purged officials returned. Kishi Nobusuke, held for three years as a suspected Class A war criminal and never charged, became prime minister in 1957.

The Rising Sun had never been a political party badge, so there was nothing to criminalise. Nobody saw a reason to outlaw an army’s flag once the army itself had been abolished. When the occupation ended in 1952 and the creation of the Self-Defense Forces followed over the next two years, the naval ensign came back unchanged.

A 1954 Cabinet order gave the ensign a firm legal footing decades before Japan’s national flag had one. The Hinomaru was not formally enshrined in statute until 1999.

Two readings of one cloth

The result is a perpetual misreading. German symbols mark a border between then and now. Japanese symbols run continuously across 1945.

To many Koreans and Chinese, an unchanged flag looks like an unchanged attitude and the visible sign of a nation that never made the conscious change Germany made. Ministers’ visits to Yasukuni Shrine, where convicted Class A war criminals have been enshrined since 1978, textbook disputes and periodic efforts to qualify past apologies all reinforce the impression. When the far-right sound trucks blare imperial marches under the same rays, the impression hardens into certainty.

To many Japanese, that same unbroken flag flies over a navy that has never fought a war, in a country whose constitution still renounces it. The continuity they see is not of empire but of the peaceful order that replaced it.

The Neutrality Trap

Rule 50.2 of the Olympic Charter is: “No kind of demonstration or political, religious or racial propaganda is permitted” at Olympic sites. Football’s codes say much the same, banning banners and flags of a political, offensive or discriminatory nature from stadiums. However, they are private associations, not courts. Faced with centuries of imperial trauma, live sovereignty disputes and ethnic grievance, they fold all of it into one administrative category and treat it as a question of decency.

What looks like neutrality is really deferral and generally each case gets decided by whoever applies the most pressure at the moment of decision.

Football shows the pattern most clearly. FIFA fined Argentina this year over a Falklands banner its players held after the semi-final, part of a $321,000 penalty that also covered fans’ conduct. It fined Serbia 20,000 francs at the 2022 World Cup over a dressing-room flag that showed Kosovo as Serbian territory under the slogan “no surrender.” In Qatar that year, stadium security confiscated Iran’s pre-revolutionary Lion and Sun flag from fans protesting the Islamic Republic. UEFA has repeatedly fined Barcelona over Catalan independence flags and Celtic over Palestinian ones.

Each decision can be defended. Together they amount to a policy of keeping the peace by suppressing whichever flag is likeliest to start a fight, which is not the same thing as neutrality.

One flag, six answers

Now let’s follow the same flag through this ‘system’.

In 2017, the Asian Football Confederation ruled the Rising Sun a discriminatory symbol and fined Kawasaki Frontale. In 2019, Tokyo’s Olympic organisers declared that it was widely used in Japan, was not political propaganda, and would not be banned. In 2021, the IOC ordered South Korea to take down banners in the Tokyo athletes’ village that invoked Admiral Yi Sun-sin’s victories over Japanese invaders in the 1590s.

Korea’s Olympic committee said it complied after the IOC promised to apply the same rule to the Rising Sun at Olympic venues. Japanese officials made no comment.

At the 2022 World Cup, FIFA officials reportedly stopped Japanese fans from hanging the flag during the Costa Rica match, without any public statement. At this year’s World Cup, in Monterrey in June, it appeared in the Japanese section during the match against Tunisia and was shown on the stadium’s own screens. And in Okazaki, three months later, it was “naturally” permitted.

Nine years, six answers, no consistent principle. The same flag has been a discriminatory symbol, a harmless cultural representation, a diplomatic bargaining chip and a quietly tolerated decoration. This is all depending on the venue, the host and the year.

The trap from the other side

The Taiwanese case shows the trap from the other side. The rule behind the Okazaki sign and the “Taiwan” towel, a ban on symbols of countries not taking part, sounds like peacekeeping, but brings in one of the most political arrangements in Olympic sport into the stadium. The Nagoya ruling exists because Beijing would not otherwise tolerate Taiwan’s presence. Enforcing it at a turnstile is not keeping politics out. It is enforcing one government’s politics on behalf of the whole tournament, and calling it protocol.

This is why the system switches between overreach and retreat. Rules are left vague by design. Interpretation is pushed down to volunteers at the gate, who reasonably assume that anything controversial is forbidden. When the controversy turns out to have a powerful constituency, the committee apologises from the top and blames a “mistaken understanding” on site.

The sign and the towel were not the only such misunderstandings. On the eve of the Games, Taiwan’s sports minister, Lee Yang, was kept out of a flag-raising ceremony for an hour on the strength of an Olympic Council of Asia “order” that no official could produce. The organisers later blamed miscommunication. None of these errors was a glitch. They were the system working as built.

By the close of the Games, critics noted, neither the IOC nor the Olympic Council of Asia had publicly acted on the judo-hall displays. Yet on the final day, the OCA pronounced the Games a success, citing a near 20 percent jump in revenue. Silence is the trap’s natural resting state.

Across the Water

Go back to the sign in Okazaki. Two flags, side by side, both marked forbidden.

The sign was wrong about the rules but it was right about something else. Both flags carry more history than a stadium turnstile can process.

Symbols do not belong only to the people who design them, or to the people who fly them. They belong, too, to the people upon whom they were inflicted.

A fishing crew in hoisting a tairyō-bata after a good season is telling the truth about what the sunburst means to them. So is a man in Busan, say, whose grandfather was sent to dig coal in Kyushu, who sees the same rays in a judo hall on television and feels something close to nausea. Neither truth cancels the other. Neither can be legislated out of existence by the other side.

Japan cannot ask its neighbors to see only the festival. Its critics cannot ask the Japanese to see only the regiment. What they can do is know what the flag looks like from the other shore, and choose accordingly.

Raising it on a fishing boat is one act. Raising it in the away end in Suwon is another. Raising it at a judo final in Nagoya, at a tournament whose stated purpose is Asian unity, a week after a public debate over that very flag, is something else entirely.

Every flag at an international tournament is political. The real question is narrower and harder: is this display, here, aimed at the people who suffered under it?

To put it metaphorically, the same sun rises on both sides of the water, yet it has never cast the same shadow on each. And until each side can see the shadow the other sees, the argument will go on.

Leave a comment

Keep reading

If you enjoyed this, here are ’s & ’s latest pieces on Japan:

Richard Harrison

Iwate: Tales in Tono (1 of 3)

Every summer I take a trip out of Kansai to somewhere in Japan I have never visited before. This year I decided to visit Iwate prefecture, partly on the recommendation of my daughter who is involved with an anthropology project there. In previous years, because we are blessed with a plethora of ferry routes out of Kansai, I have often put my car on a fe…

Read more

10 days ago · 5 likes · Richard Harrison

Almost everyone takes the same photograph. Almost nobody walks around the back.

Yet if you turn off the main terrace and follow a quiet path through the trees, you find something unexpected: a polished stele of reddish granite resting in the shade.

Set into the stone is a bronze medallion of a statesman in Western dress. Beneath his profile, carved in Japanese, Pali, and English, is a single verse from the Dhammapada:

The monument honors Junius Richard Jayewardene, the former president of Sri Lanka.

Finding a Sri Lankan statesman memorialized on the grounds of one of Japan’s most celebrated temples strikes the few visitors who notice it as a puzzle. That modest stone marks a remarkable diplomatic episode, and a choice about how Japan remembers it.

The Easter Sunday Raid

The story behind the Kamakura monument begins in the roar of carrier aircraft over the Indian Ocean in the spring of 1942.

Four months after Pearl Harbor, Vice Admiral Chūichi Nagumo led the First Air Fleet of the Imperial Japanese Navy, the veteran carrier force known as the Kidō Butai, west from the Dutch East Indies toward the island of Ceylon (now Sri Lanka). His objective was to catch the British Eastern Fleet in its harbors at Colombo and Trincomalee.

On Easter Sunday, April 5, 1942, Nagumo launched 127 naval aircraft against the port and airfields of Colombo. Four days later, ninety-one bombers and forty-one fighters struck the naval base at Trincomalee on the island’s northeast coast.

The losses were severe. On Easter afternoon, dive bombers caught the heavy cruisers Dorsetshire and Cornwall at sea and sank both, killing 424 men. On April 9, off Batticaloa, they found the carrier HMS Hermes and sank her too. The main body of the Eastern Fleet, which had slipped away to refuel at a secret anchorage in the Maldives, was never found.

But Ceylon was not an empty strategic base. It was an inhabited island whose civilian population had no say in imperial strategy. Bombs fell on the railway workshops at Ratmalana and on the mental hospital at Angoda, killing patients there. Estimates of the civilian dead run from thirty-seven to about eighty-five.

For Ceylon, the Pacific War was no longer a distant clash between empires. It had arrived over Colombo on Easter morning, as the island’s Christians gathered for services.

The Deadlock at San Francisco

In September 1951, delegates from fifty-one nations assembled at the War Memorial Opera House in San Francisco to sign a peace treaty with Japan. Neither Chinese government was invited; India and Burma declined to come.

The political mood in the auditorium was fractured. Six years after Japan’s surrender, the Cold War had hardened into an ideological standoff. The Korean War was raging across the Tsushima Strait, and Washington, led by chief negotiator John Foster Dulles, was determined to secure a swift settlement that would anchor Japan firmly within the Western bloc.

The Soviet delegation, led by Andrei Gromyko, fiercely resisted the draft, denouncing it as an American maneuver to turn Japan into a rearmed client state and protesting its silence on Soviet sovereignty over South Sakhalin and the Kuril Islands.

Meanwhile, Asian nations that had endured brutal occupations viewed a lenient peace with deep suspicion. Burma stayed away. The Philippines and Indonesia came, but objected that Article 14, the reparations clause, gave them too little: it conceded that Japan owed reparations, declared Japan unable to pay in full, and offered only negotiated payment in services.

Their argument was simple: a nation that had waged war across Asia should not re-enter the international order without paying for what it had destroyed. In time they won much of it. Between the mid-1950s and 1959, Burma, the Philippines, Indonesia, and South Vietnam each negotiated reparations with Japan, the Philippines alone securing $550 million.

An Unexpected Voice

On September 6, J.R. Jayewardene rose to speak.

At forty-four years old, Jayewardene was serving as Ceylon’s minister of finance. His country had emerged from British colonial rule only three years earlier, in 1948. In an opera house full of great powers, Ceylon carried almost no leverage.

Read more

Two flags were pictured side by side. One was the Rising Sun, the sixteen-ray kyokujitsu-ki (旭日旗). The other was the Blue Sky, White Sun, and a Wholly Red Earth, the national flag of the Republic of China, better known as Taiwan. Beneath them, in Japanese, English and Chinese, a single line: these flags may not be brought into the venue.

The game that day was Japan against Chinese Taipei.

Within a day, the sign was gone. Online, Japanese commenters were incredulous: why would anyone ban the Rising Sun in Japan, of all places?

On September 28, the Aichi-Nagoya organising committee posted a formal apology on its website for an “inappropriate posting.” Tournament rules, it explained, bar spectators from hoisting the national flags of countries not taking part. The Rising Sun was, it said, naturally not among the prohibited items. Staff on site had misunderstood. Every venue would be re-briefed.

The statement said nothing about the other flag on the sign. Taiwan’s Liberty Times noticed.

It was not the first time that week a Taiwanese symbol had been waved off. The day before the sign went up, Taiwan beat China 9–1 in the same baseball tournament. In the stands, a fan held up a towel printed with the word “Taiwan.” Venue staff told the fan to put it away or leave. According to the Japanese outlet J-Cast, one told her the Games had a rule against using the word “Taiwan.”

Asked about it, the committee first defended the staff, saying they had wanted to avoid disputes between spectators. A day later it reversed itself. Towels bearing the names of non-IOC territories were not banned after all. This, too, was a “mistaken understanding.”

Two mistaken understandings in four days. One produced a published apology and an explicit assurance. The other produced a clarification in reply to press inquiries, and a flag that stayed on the wrong side of the rule.

Then, on October 3, the judo mixed team final between Japan and Kazakhstan filled Aichi International Arena in Nagoya. Rising Sun flags rose in the stands. Seo Kyoung-duk, a Seoul professor who has campaigned against the flag for years, emailed a complaint to the Olympic Council of Asia. Two days later, an op-ed in the Hong Kong edition of the state-owned China Daily accused the hosts of “calculated acquiescence.”

The Games had already been bruising for Seoul. A performer dressed as Toyotomi Hideyoshi, the warlord who invaded Korea in the 1590s, appeared at a welcome event for athletes. Before a South Korean men’s hockey match, the stadium played North Korea’s anthem. These were blunders, not provocations, and the organisers apologised for the anthem. But they showed how thin the ice was.

There is a further irony in the geography. In 1979, the International Olympic Committee’s executive board met in this same city and drew up the Nagoya Resolution, the formula under which Taiwan competes as “Chinese Taipei,” under a flag that is not its own. Forty-seven years later, in Nagoya’s backyard, a woman was told not to hold up a towel bearing her island’s name.

Look closely at the committee’s logic and the absurdity sharpens. The rule bans national flags of countries not competing. Taiwan’s flag falls foul of it because, under the Nagoya formula, Taiwan does not compete as itself. The Rising Sun slips through because it is not a national flag at all.

A rule written to keep politics out of the stadium had produced a perfectly political result. The flag of a self-governing democracy of 23 million was contraband. The banner that once led an empire’s armies was waved through as fan merchandise.

This is what international sport calls neutrality. It is not the absence of a position. It is a position that cannot admit it is one.

Underneath it all sits the question every such incident drags back to the surface. Is the kyokujitsu-ki Asia’s swastika? South Korea’s government said as much in a 2019 request to the IOC. The short answer is no. The longer answer explains why the argument never ends.

The Wound That Is Still Open

From 1910 to 1945, Korea was a Japanese colony. The independence protests of the March First Movement in 1919 were put down by force. Later, Koreans were pressed to adopt Japanese-style names, and their language was pushed out of the schools. As the war widened, hundreds of thousands were shipped to mines, factories and construction sites across the empire, and many more were mobilised at home. Young men were drafted into the imperial army. Tens of thousands of women, many of them Korean, were forced into the military’s system of sexual slavery, euphemised then and now as “comfort women.”

China’s ordeal was invasion rather than annexation. Japan seized Manchuria in 1931 and launched full-scale war in 1937. In Nanjing that winter, Japanese troops slaughtered prisoners and civilians and raped on a vast scale; the postwar Tokyo tribunal put the dead above 200,000. Outside Harbin, Unit 731 ran a biological-warfare program that experimented on living prisoners and seeded plague in Chinese towns.

Through all of it, the army and navy marched and sailed under the sunburst.

The Imperial Army adopted the sixteen-ray flag as its war flag in 1870. Its regimental colors were bestowed by the emperor and guarded as sacred objects, to be burned rather than surrendered. The Imperial Navy adopted the offset version as its ensign in 1889. It flew from the sterns of the ships that struck Pearl Harbor and ranged across the Pacific and Indian Oceans. In Singapore, the national museum still tells the story of the occupation years under the heading “Life under the ‘Rising Sun.’”

One detail is often missed. The Hinomaru, Japan’s plain red-disc national flag, also flew over colonial government offices and occupied cities. Yet it is the Rising Sun that carries the charge. It was the flag of the instruments of conquest themselves: the soldiers who did the killing and the ships that did the shelling. For the people on the receiving end, the empire was, above all, an army.

So when many Koreans and Chinese see those red rays, they do not see a sunrise. They see the colors of the men who came.

Nor is the flag only a memory. In the 2010s, as anti-Korean marches spread through Tokyo’s Shin-Ōkubo and Osaka’s Tsuruhashi, the Rising Sun was carried as prominently as the national flag. Japan passed its first law against hate speech in 2016, largely in response to those rallies. For many Zainichi Koreans, whose families have lived in Japan for generations, the flag is not an archival image. It has been waved at them in their own streets.

Sport has absorbed this history slowly and unevenly. In April 2017, Kawasaki Frontale supporters unfurled a Rising Sun flag at an Asian Champions League match in Suwon, South Korea. The Asian Football Confederation called it “a discriminatory symbol relating to national origin and political opinion,” fined the club $15,000 and ordered one home match played behind closed doors, suspended for a year. Its appeals committee dismissed Kawasaki’s appeal in full. The club accepted the ruling while insisting the flag carried no political or discriminatory intent.

A Motif Older Than the Army

The sun sits at the root of the country’s name. Nihon (日本) means, literally, “origin of the sun.” Sunburst crests appear in samurai heraldry centuries before Japan had a modern army, and the radiating sun has decorated banners, shrines and shop curtains for generations.

Its most beloved form today belongs to the sea. Fishing crews fly tairyō-bata (大漁旗), “big-catch flags,” dyed in riotous colors and often built around a rising sun, to mark a good haul, a launch or a safe return. Families raise sunburst banners to celebrate a birth. Neighborhoods hang them for seasonal matsuri. The Foreign Ministry’s own explainer illustrates the point with fishing boats flying them as they returned, in 2017, to a Fukushima port emptied by the 2011 earthquake and tsunami.

The design runs through commercial life, too. The corporate flag of the Asahi Shimbun, a liberal daily that has spent decades criticising historical revisionism, shows a quarter-sun throwing rays from one corner. In 2014, the War Memorial of Korea in Seoul called off a One Piece exhibition days before it was due to open because the manga contained sunburst imagery; it opened only after the organiser won a court injunction. That is how deeply the motif is woven into Japanese visual culture: it turns up in places no one thought to look.

Then there is the state. When Japan created its Self-Defense Forces in 1954, a Cabinet order under the new Self-Defense Forces Law gave the Maritime Self-Defense Force (JMSDF) the old sixteen-ray naval ensign, and the Ground Self-Defense Force an eight-ray version with a gold border. Every JMSDF ship flies the ensign whenever it is underway. Tokyo regards it as mandatory under Japanese law and as the internationally recognised identification of its warships.

For seven decades, those ships have sailed into ports across Asia, the Americas, Europe and Australia under that flag. South Korea itself hosted them at international fleet reviews in 1998 and 2008, ensign flying. In 2018, Seoul asked all fourteen navies attending its review off Jeju to fly only their national flags and South Korea’s. The request was plainly aimed at one ship. Japan withdrew. Its top uniformed officer said the ensign was a matter of pride for its sailors, and that they would not come without it.

The official line has hardly changed in a decade. Japan’s chief cabinet secretaries, and the Tokyo 2020 organisers, have repeated that the design is widely used at home and that displaying it is not political propaganda.

Seen from inside Japan, then, the demand to ban the flag can feel bewildering. A supporter waving it in Okazaki is not, in their own mind, cheering for conquest. They are doing what fans do: making noise in the national colors, with a design they have seen on fishing boats, festival coats and comic books since childhood. Why should a motif centuries old be surrendered because of the seventy-five years when an imperial army carried it?

But honesty requires the rest of the domestic picture. The Rising Sun also lives another life in Japan: on the black sound trucks (街宣車) of the far-right groups　(右翼団体), that cruise city streets blaring wartime military songs. Most Japanese regard them as a nuisance. They are not marginal to the flag’s meaning, though. They are part of it.

Tokyo’s defenders also point out that Korean objections hardened only in the 2010s, years after Japanese warships flew the ensign in Korean ports without incident. They conclude that the anger is political manufacture. That inference does not follow. The grief is old. What changed is that it found a symbol, and that the symbol kept appearing: at hate marches in Tokyo, in football stands in Suwon, in Japanese supporters’ sections abroad.

A symbol’s meaning is not fixed at its origin. It is renewed every time someone raises it, and by who is doing the raising. Both shores have helped make the Rising Sun what it is in 2026.

In Part 2: why the swastika comparison fails on structure but not on suffering, how Germany and Japan came out of 1945 with opposite relationships to their symbols, and why sport’s version of neutrality keeps producing results like Okazaki.

Leave a comment

Back in May, I wrote about Gakureki — Japan’s obsession with where you went to university. If you’re new here, start with that piece. It covers the university hierarchy, deviation scores (偏差値), and the two YouTubers who turned all of it into entertainment. Takata Fumin (高田ふーみん) and Biyama (びーやま) of Wakatte TV.

I said they were farming reactions.

In late August, the reactions came back.

What Happened at Fukushima University

On 28 August, Wakatte TV posted a campus survey (キャンパス調査) of Fukushima University. The title asked if it was the lowest-ranked of Japan’s national and public universities.

A student was asked which faculty they’d recommend. The answer: radiation research — something particular to Fukushima.

Takata agreed that radiation and nuclear power are a hugely important issue. Then the punchline.

“I don’t want Fukushima University involed with it.” (福島大には触らせたくない)1

Fukushima’s hensachi is low, so its students shouldn’t be trusted with anything that important.

But in Fukushima, nuclear isn’t just an important topic. It’s a sensitive one. Fifteen years on, the region is still living with the 2011 nuclear disaster.

Except this is Fukushima. Since the 2011 nuclear disaster, the university has built its research around recovery, including an Institute of Environmental Radioactivity (環境放射能研究所) opened in 2013.

The one thing Takata didn’t want them involved with is the thing they’ve spent over a decade becoming experts in.

The Enjō (炎上)

The clip spread on X, and the enjō — literally “catching fire”, Japan’s word for an online pile-on — began.

By 31 August the video was pulled. Then briefly put back up with a disclaimer saying they meant no mockery.

On 1 September, university president Sano Kōji (佐野孝治) released a statement: students and staff doing serious work should not have their dignity hurt. That evening, Takata apologised on X for their “insufficient wording” (言葉足らずな表現). The video came down for good.

On 4 September, Minister of Education Matsumoto Yōhei (松本洋平) weighed in at a press conference. Judging a university’s research by hensachi, he said, isn’t appropriate.

When a cabinet minister has to explain that a deviation score isn’t a measure of research, you’ve taken gakureki somewhere new.

The Apology (お詫び - owabi)

On 5 September came an eight-minute apology video. Takata called the remark truly terrible and said he was reflecting deeply (深く反省). Since the backlash, he said, many people had taught him about Fukushima University’s research.

Except they didn’t need teaching. The original video’s own title listed disaster recovery and radiation research (震災復興・放射線研究) as selling points. They knew exactly what Fukushima University is known for. They used it to sell the video, then made it the punchline.

They Never Change

On 1 October, after a month off, Wakatte TV came back. Takata apologised again. Biyama bowed. Then came a long on-screen disclaimer: the show satirises Japan’s gakureki society, and anyone with a complex about their academic background (学歴コンプレックス) or exam trauma should refrain from watching.

Read that again. The problem isn’t the content. It’s your complex.

When a Kyushu University professor criticised the channel in early August, the fans’ reply was simple: if you don’t like it, don’t watch it. This is now practically channel policy.

Back to work for around 780,000 subscribers. Same show. New warning label.

For something that reached the Minister of Education, it barely left a mark.

A Society That Doesn’t Look at Learning

Writing in the Tokyo Shimbun (東京新聞), Teshigawara Mai made the sharpest point. The moment a student described real research, the university’s name was used to wave it away. A society obsessed with gakureki, she argues, may be one that lets you skip looking at what anyone has actually learned.

At the end of my last piece, I asked whether it’s right to judge a person by where they went to school.

Fukushima answered that. A student told them what makes their university matter. The hensachi talked over them.

Wakatte TV say they’re satirising gakureki. But satire is supposed to expose the thing it mocks.

At Fukushima, they didn’t satirise gakureki. They just fed it.

Yet here we are.

The internet is now overrun by the gospel of looksmaxing1. Livestreamers march through nightclub strips with handheld camera rigs, cold-approaching women, flexing vascular forearms, and pitching insecure teenagers on how to “ascend” from social irrelevance. The message is blunt: fix your face, fix your frame, or accept that you are genetically worthless.

In the midst of all this madness, I couldn’t help noticing a striking correlation to Japanese society.

Strip away the podcast bravado and the red-pill grievance politics. What are these guys actually prescribing? Facial contouring, ten-step skincare, high-maintenance hair, shoe lifts to cheat two inches of height, jawline exercises, and discreet cosmetic tweaks.

Western commentators treat this as a dystopian mutation born in the fever swamps of TikTok. But if you’ve ever walked the neon-soaked alleys of Kabukichō at midnight, you know the truth:

Japan has been doing all of this for decades.

So the question is: was Japan the first society to normalize looksmaxing? And did it figure out how to do it without the toxicity, or does the same rot hide behind better manners?

The Manosphere Playbook

The movement hit the mainstream when Louis Theroux took his cameras inside it. In his 2026 Netflix documentary Louis Theroux: Inside the Manosphere, he spends time with prominent online figures and explores how algorithms churn out hypermasculine, transactional worldviews for an audience of lonely young men.

Theroux is fair enough to acknowledge what reels people in: getting off the couch, hitting the gym, eating clean, taking pride in yourself. But that entry-level self-improvement curdles fast. Women become status tokens unlocked by the right facial symmetry. Life shrinks to money, muscle, and sexual conquest. Teenage boys learn that their natural genetics condemn them to misery unless they aggressively “fix” themselves.

Beneath all the talk of “self-optimization” sits an entitlement that treats other people as objects to be won.

Which is what makes Japan such a fascinating counter-study. There, the physical mechanics of looksmaxing were codified centuries ago, without the red-pill rhetoric.

The 200-Year-Old Blueprint: Japan’s Manuals for Facial Optimization

Western looksmaxers talk about skincare regimens and facial harmony as if they’d unlocked secret cheat codes on anonymous forums. In Japan, these protocols were in print during the Edo period.

Take Miyako Fūzoku Keshōden (”A Guide to the Customs and Cosmetics of the Capital”), published in 1813 by Sayama Hanshichimaru and illustrated by Hayami Shungyōsai. It wasn’t a vague philosophical treatise. It was a practical, step-by-step beauty manual that instructed readers on:

• Cleansing and evening out the skin with bags of komenuka (rice bran)

• Treating blemishes and dull skin, including face washes made from uguisu no fun (nightingale droppings)

• Makeup techniques for making a large face look smaller and a flat nose look higher

The philosophy was simple: your face is a starting canvas, not a fixed destiny. Long before “softmaxxing” had an English label, urban Japanese were reworking skin texture, shading, and the apparent shape of the face to meet rigid aesthetic ideals.

When modern surgery arrived in the Meiji era, the impulse jumped from topical maintenance to what the internet now calls hardmaxxing.

In 1896, Dr. Kōtarō Mikamo published what is widely cited as the first description of cosmetic double-eyelid surgery (futae seikei) in the journal Chūgai Iji Shinpō2. Contrary to a common Western assumption, Mikamo didn’t present it as a way to make patients look Western. He framed it as creating a symmetrical upper-lid crease, a feature many Japanese people already have naturally, in line with domestic ideals of balance.

By the late 1970s, the beauty industry had a non-surgical workaround: Aipuchi eyelid glue (a brand name that became the generic term) and ultra-thin eyelid tape. Decades before Western teenagers learned what an eyelid crease was on an aesthetics forum, Japanese schoolgirls were sculpting theirs in the bathroom mirror every single morning.

The Cult of the Small Face

Spend five minutes on looksmaxing TikTok and someone will be obsessing over their jawline and “craniofacial development.”

In Japan, that obsession has a name: kogao, literally “small face.”

Being called kogao is one of the highest physical compliments in Japanese society. A compact face makes the body’s proportions look more balanced, closer to the hattōshin ideal of a figure “eight heads tall.” The demand is so intense that an entire product category exists to squeeze, roll, and compress the face:

• Face rollers: Handheld tools like the ubiquitous ReFa, with platinum-coated rolling spheres marketed to drain lymph, tighten the jawline, and contour the cheekbones.

• Compression straps: Chin-and-head wraps sold in Japanese drugstores, worn overnight to cinch the lower face.

• Kogao kyōsei (”small-face correction”): Salons using deep, often painful manual pressure on the skull and facial muscles, claiming to close millimeter-level “gaps” and physically shrink the face.

Sound familiar?

While thos in the West were discovering tongue-posture tutorials and gua sha stones on Reddit, entire salon chains across Japan are dedicated to face compaction. The impulse is identical: a refusal to accept the natural limits of the face.

Kabukichō’s Apex Predators: The Host Club Looksmaxers

If you want to see all of this converge in its purest male form, don’t look at a fitness influencer in Miami. Go to Kabukichō, Tokyo’s legendary entertainment district in Shinjuku.

After 10 PM, watch the neon-lit advertising trucks crawl down the avenue, plastered with the faces of top-ranked hosts: men who can earn small fortunes entertaining female clients in exclusive clubs.

The top hosts are masterclasses in aesthetic engineering:

• Elevator footwear: Platform boots and hidden lifts are standard kit. A host who’s naturally 5’7” walks onto the floor at 5’10”.

• Architectural hair: Blow-dried, dyed, straightened, and sculpted, a lineage running from visual kei bands to anime archetypes, and cut to narrow the face and add height at the crown.

• Cosmetic base layers: Porcelain foundation, BB cream, highlighted nose bridges, subtle eyeliner, and contoured jaws built to read perfectly under dim club lighting.

• Facial restructuring: Fillers, double-eyelid tape or surgery, and regular skin-whitening treatments.

Inside, life is relentlessly quantified. Hosts are ranked on wall charts by how much their clients spend each month, competing to sell champagne towers that can cost millions of yen.

The host sells a fantasy: the attentive, exquisitely styled, emotionally attuned partner. But the entry ticket is flawless presentation. Hosts treat this as their theatrical craft. No shame in makeup or shoe lifts. They are just tools of the trade.

From Subculture to Salaryman: Japan’s Men’s Grooming Boom

What started in Shinjuku’s nightlife and Shibuya’s youth subcultures didn’t stay there. Over the past decade, Japan quietly did something the West is still wrestling with: it destigmatized male cosmetics.

For years, brands like Gatsby and UNO ruled the men’s aisle with hair wax, oil-blotting sheets, and face wash. Then the line blurred into actual makeup. Genderless kei, championed by models and pop idols, made caring about smooth skin and clear eyes look less effeminate than simply polite.

The data tells the story:

• According to Intage Inc., Japan’s men’s cosmetics market has grown to over ¥40 billion3.

• A 2024 Intage survey of 5,000 people found roughly 60% viewed men’s skincare and makeup favorably, with negative views below 10% in nearly every age group.

• Tokyo stores like Isetan Men’s and Loft now carry dedicated men’s cosmetics sections, stocking foundation, concealer for razor burn, brow pencils, and pore-blurring primer.

Clean Aesthetic or Exploitative Trap?

It’s tempting to conclude that Japan solved the equation: all the benefits of aesthetic improvement, none of the manosphere’s poison.

That would mean ignoring the dark side.

The host club industry that perfected male looksmaxing has faced national scrutiny and police crackdowns over predatory debt. As reporting by The Japan Times and Unseen Japan has documented, some clubs let customers run up enormous tabs through urikakekin (pay-later credit), and women who couldn’t pay were pressured into sex work to clear the debt.

In 2025, Japan amended its entertainment business law to explicitly ban hosts from exploiting customers’ romantic feelings to drive spending, and from pushing them into prostitution to repay debts.

The toll on ordinary people is just as real. When a culture treats every facial flaw as something that can, and should, be corrected with tape, rollers, creams, or surgery, the bar for “acceptable” keeps drifting out of reach.

The pressure to maintain a kogao, porcelain skin, and immaculate double lids feeds real appearance anxiety among Japanese youth, and a booming cosmetic clinic industry is happy to sell the fix.

The Uncomfortable Reality

So, was Japan actually the first society to normalize looksmaxing?

In almost every mechanical way, yes. Decades before Western livestreamers began debating facial bone angles and stuffing lifts into their shoes, Japan had already turned facial modification, platform footwear, and male cosmetics into a routine, multi-billion-yen industry.

Which brings us right back to the central dilemma Louis Theroux touched on in the manosphere documentary.

Is there really any version of looksmaxing that remains healthy and positive?

Can a society embrace the idea of optimizing the human face without inevitably tumbling into vanity, obsession, and predatory exploitation?

Or does Japan simply prove that no matter how polite, clean, and normalized you make the packaging, the underlying obsession with physical perfection always extracts the exact same human cost?

Leave a comment

Emperors were expected to sit through these conferences without a word, and until that day he always had.

Within about a day, Army staff officers had worked out how to read the same poem as permission rather than restraint.

Then the war was lost, and the Americans began drawing up their own list of names, the war crimes suspects. The White House wanted Hirohito on it. This is the story of how he stayed off.

Was the poem used to decide whether he should stand trial?

The question worth asking is not whether the poem was quoted at the Tokyo Trials (because it wasn’t). It is whether the poem was ever used to decide the thing that actually mattered: whether Hirohito should have been put on trial at all. Those are different questions, and they have different answers.

Hirayama Shūkichi's 2014 book1, the only book-length study of the poem, has a chapter on its postwar life. Three consecutive section titles in it point to the answer: How 6 September was discussed at the Tokyo Trial; Defendant Kido testifies that it was about the refusal to withdraw from China; and Defendant Tōjō does not mention "Yomo no umi."

That is consistent with what Tōjō’s sworn affidavit actually does with 6 September 1941. It talks about oil, scrap iron, the deteriorating strategic position, and above all the refusal to withdraw troops from China. It does not reach for a poem.

So the poem played no part in the legal proceedings, as far as I can establish. Which raises the better question. Why did nobody put it in front of the tribunal?

Where the poem actually did its work

The answer, I think, is that by the time the tribunal opened in May 1946, the work had already been done somewhere else.

The campaign to place Hirohito beyond prosecution was not conducted in court. It was conducted in memoranda, private meetings, and carefully prepared documents, and it was substantially complete before the first witness was sworn.

The timeline is:2

• September 1945. Prince Higashikuni, the first postwar Prime Minister, revives the poem in a policy speech to the Diet — with a speech draft written by Ogata Taketora, the former Asahi executive. The “peace-loving emperor” is back in public circulation within weeks of the surrender.

• 1 October 1945. MacArthur receives, via his military secretary, Brigadier General Bonner Fellers, a short legal brief that simply assumes the Emperor lacked free will and lacked knowledge of the true state of affairs.

• October 1945. Fellers submits his memorandum arguing against prosecution.

• 25 January 1946. MacArthur cables Army Chief of Staff Eisenhower that an investigation has been conducted and no evidence connects Hirohito to political decisions over the previous decade. John Dower’s judgement on this “investigation” is withering, and on the documentary record it is hard to argue with him.

• March 1946. Fellers tells Admiral Yonai Mitsumasa that it would be helpful if the Japanese side could establish the Emperor’s blamelessness, and that Tōjō in particular “should be made to bear all responsibility.”

• 18 March – 8 April 1946. Over five sessions and roughly eight hours, Hirohito dictates his account of 1928–1945 to five court officials. This is the Shōwa Tennō Dokuhakuroku, the “Monologue.” Terasaki Hidenari’s pencilled transcript is the only full record. It stayed hidden in his American family’s possession until Bungei Shunjū published it in 1990.

• And an English-language version of the Monologue later turned up among Fellers's papers.

The Emperor’s own retrospective account of the road to war was compiled in the weeks before the trial opened, and a version of it reached the desk of the American officer running the case for his exoneration.

Meanwhile Kido Kōichi handed his diary — thirty-odd volumes covering 1930 to 1945 — to the International Prosecution Section, and submitted to more than thirty interrogations. The diary became the prosecution’s single most important documentary source, and it was written by the man whose overriding postwar objective was protecting the throne.

Konoe’s memoir, containing the poem, was published in April 1946, four months after Konoe took poison rather than report to Sugamo Prison, and one month before the trial began.

The poem did have a political career after 1945. It just wasn’t a legal one.

And once inside the courtroom, the whole apparatus was pointed at keeping the Emperor invisible. Prosecution and defence both worked at it. The one moment the discipline slipped was 31 December 1947, when Tōjō testified that it was inconceivable for any Japanese subject to act against the Emperor’s wishes. Chief prosecutor Joseph Keenan moved immediately, arranging through the Emperor’s own circle for Kido to be contacted in prison and for Tōjō to retract it on the stand.3

The chief prosecutor of a war crimes tribunal, coordinating with the accused to repair testimony that had accidentally been too accurate. That is what “the Emperor was not prosecuted” actually looked like in practice.

Hirohito: what we know and what we don’t

I want to be careful here, because this is where writing about Hirohito usually collapses into either hagiography or prosecution.

What is reasonably well established is this. Under the Meiji Constitution the Emperor was sovereign and Supreme Commander, but by convention he ratified decisions reached by government and high command rather than initiating them. He was thoroughly briefed throughout, and he was by no means a bystander. He opposed the drift to war with the United States in autumn 1941 and said so, more than once, in ways his ministers understood. He nonetheless sanctioned the decision for war in December 1941, and he intervened in August 1945 to end it. And he was never charged — a decision made in Washington and Tokyo on political grounds, not judicial ones.

What is genuinely disputed is how much real authority he possessed, and how much he chose to use. Herbert Bix’s Hirohito and the Making of Modern Japan argues for an active, informed monarch far more involved than the postwar image allowed; Peter Wetzler (Hirohito and War) and Noriko Kawamura (Emperor Hirohito and the Pacific War) read the same constitutional structure as far more constraining.

That debate is still unsettled, and so is the question of whether he offered to accept responsibility when he first met MacArthur on 27 September 1945.

Three people were in the room. MacArthur’s account appears in a memoir published nearly twenty years later. Peter Mauch, working from the Jitsuroku, thinks he probably did make the offer; Dower and Bix are dismissive. The evidence is inconclusive.

The famous photograph from that meeting of Hirohito in morning dress and MacArthur in open-necked khaki is real, and its effect on Japanese opinion was enormous; the often-repeated detail that MacArthur shook his hand against palace advice, specifically to humanise him, belongs to the anecdote tradition rather than to the documentary record.

And a good deal of what gets said is interpretation rather than fact. That the poem proves Hirohito was a pacifist: it doesn’t. It proves he wanted diplomacy given priority in September 1941, which is a smaller and much more specific claim. That the war was fought “in his name” and therefore against his wishes: both halves of that sentence need separate evidence, and people rarely supply either.

What can be said about the memory rather than the man is that the image of the peace-loving emperor is a postwar construction, assembled by Japanese elites and American occupation officials with overlapping interests. Saying so is not an attack on Hirohito. It is a statement about how the record was made.

Not everyone in Japan has let it go. The Anti-Emperor-System Movement Network (反天連) has argued for thirty years that the only way to settle the wars fought in Hirohito’s name is to abolish the institution that fought them. One member, speaking to the BBC in 2019 on condition of partial anonymity because of right-wing harassment, put it flatly: the war never properly ended, and Hirohito was a war criminal. They marched at the changeover to Reiwa that April, about eighty of them, outnumbered by several hundred police, largely ignored by passers-by. Kenneth Ruoff, an American historian of the postwar monarchy, puts public support for the imperial institution at seventy to eighty per cent.4

That’s the shape of it. The question hasn’t been answered, and it hasn’t quite gone away either. It has just stopped being put to enough people to matter.

It was put to him, though. Twice, ten years apart.

On 31 October 1975, back from his first state visit to the United States, at a press conference hosted by the Japan National Press Club, a journalist asked what he thought about war responsibility. He replied that this was a matter of turns of phrase, that he had not studied literary questions closely, and that he could not answer.5

Hirayama gives that reply a place in his chapter on the poem's afterlife, and it's easy to see why. In 1941, when Hirohito finally broke his silence at an Imperial Conference, he did it by reading a poem. Thirty-four years later, asked about the war, he said he couldn't answer because he didn't know much about literature.

Except that ten years on, he came close to answering. At his birthday press conference in April 1985, asked about 6 September 1941, Hirohito is reported to have said why he had reached into his pocket: the agenda put war preparation first and peace efforts second, and he had wanted peace efforts to come first. So he quoted his grandfather.

It is the clearest statement we have of what he meant by the poem, and it is the Emperor’s own, which is exactly what makes it hard to weigh. A man explaining in 1985 why he read a poem in 1941 is not a witness in the ordinary sense. He is the last participant in a process this whole essay has been about: the long assembly of an account that everyone involved had reasons to want to hold.

It is hard not to read 1975 as evasion and 1985 as something closer to candour, and I cannot tell you what changed in the ten years between.

The institution that survived

More recently, Japan’s parliament passed the first substantial revision of the 1947 Imperial House Law. It permits the adoption of male-line male descendants of former imperial branches, and lets princesses keep their status after marriage — while confirming that only men in the male line may reign.

The line of succession runs from Emperor Naruhito to his brother Crown Prince Akishino, then to Prince Hisahito, then to the Emperor’s uncle Prince Hitachi. Naruhito’s daughter Aiko is not in it and cannot be. Japan has had eight reigning empresses (the last, Go-Sakuramachi, abdicated in 1771), but male-line succession was written into statute for the first time in the Meiji-era Imperial House Law, and carried straight over in 1947.

Polls consistently show most Japanese would accept an empress. The law says otherwise. The institution that survived 1945 because Americans decided it was too useful to dismantle is now being kept alive by adoption clauses.

What a poem is for

In 1904 Emperor Meiji wrote thirty-one syllables in a private room after agreeing to a war he didn’t want. In 1941 his grandson read them aloud to fifteen men who understood perfectly well what he meant. Within a day, Army staff officers had worked out how to read the poem as permission, and three months later the war happened anyway. In 1945 it came back as evidence of a peaceable throne. In 1946, when it might have been submitted to a court, nobody submitted it.

How much influence did Hirohito really have over the military? I’m not sure. I don’t think anyone is.

The war was launched and fought in his name, and he was also the one who stopped it.

The names on the Cornerstone of Peace are still there, ranked by nothing at all. All of them, from every side, as though somebody had taken the poem seriously.

Leave a comment

Everyone. That is the point of it. Okinawan civilians, Japanese conscripts, Americans, British, Koreans, Taiwanese — all inscribed on the same stone, in no particular order of merit.1 There are now more than 240,000 names.

I went there in the middle of a beach holiday, which is a strange sentence to write. As an Aussie, Okinawa was a long-overdue trip to a decent beach. But I didn't only do beaches.

I went to the 沖縄県平和祈念資料館, the Prefectural Peace Memorial Museum, and to the 旧海軍司令部壕, the old Japanese Navy underground headquarters at Tomigusuku, where Rear Admiral Ōta Minoru sent a final signal asking that the people of Okinawa be remembered kindly by later generations. A week later he shot himself in the command room. His staff killed themselves with grenades in the room next door, and the fragment marks are still on the wall.

So: not your typical Orion beer trip.

The thing that struck me most was the emptiness. Kokusai-dōri is bustling. Walk into a war museum or a historic site and you’re often the only person in the room.

Standing in front of those names, I found myself circling a question, and it is what this piece is actually about. Why does Japan still have an emperor at all, given that the Americans could have abolished the institution and chose not to?

The film that asks this question badly

There is a movie about exactly this. Emperor (2012), with Tommy Lee Jones as MacArthur and Matthew Fox as Brigadier General Bonner Fellers. Fellers is given ten days to determine whether Emperor Hirohito2 was personally responsible for Pearl Harbor, while also searching Tokyo’s ruins for Aya, a Japanese woman he loved before the war.

As a film: average. As history: important, though not in the way it thinks.

Because almost none of that framing is real. Aya is invented. The ten-day deadline is invented. And Fellers did not conduct an investigation that led him to conclude the Emperor was innocent.

What Fellers actually did is better documented and much more interesting. He was MacArthur’s military secretary and had run American psychological warfare against Japan — an operation explicitly designed, as his papers at the Hoover Institution show, to drive a wedge between the Japanese people and the “militarists” while leaving the Emperor untouched.3 In July 1944, more than a year before the surrender, he had already written that Hirohito “cannot side step war guilt.”4

Read that again. The man the film casts as the Emperor’s investigator had privately concluded, before the war ended, that the Emperor was implicated and argued for sparing him anyway, on grounds of occupation policy.

His October 1945 memorandum to MacArthur made the case plainly: try the Emperor and the structure of the Japanese state collapses, a general uprising follows, and the chances of communism taking root in the wreckage go up sharply. Whether that risk was real is contested — some historians think it was badly overstated — but the argument was never "he is innocent." It was "this would be inconvenient."

So before we go any further, let me answer the question the film poses. Was Hirohito a war criminal? No one really knows, and in part two I'll come back to why that phrasing is itself part of the problem.

The poem

In September 1941, three months before Pearl Harbor, Hirohito did something emperors were not supposed to do at an Imperial Conference. He spoke.

What he said was a poem. Not his own, but his grandfather’s.

Translation by Harold Wright, from 敷島の道に架ける橋：英語で伝えたい明治天皇百首 (2022), the authorised collection supervised by Meiji Jingū.

Translations vary considerably, and the variation matters. One popular version has “Across the four seas, all men are brothers.” Another gives “all are born of one womb.” A third, circulating widely online, turns it into something close to a peace anthem.5 The Japanese is more restrained than any of them: はらから (harakara) literally means siblings from the same womb. A word about kinship, not about universal brotherhood in the modern liberal sense. And など is not a plea. It is a question. Why?

The poem’s assigned title, in the Imperial Household’s own manuscript of Meiji’s collected verse, is 四海兄弟 — shikai keitei, “all within the four seas are brothers,” a Confucian tag from the Analects. Not a modern pacifist slogan. A classical one.

What the poem meant in 1904

Emperor Meiji composed it in 1904, on the eve of the Russo-Japanese War.6

The precise date isn’t recorded. The most careful reconstruction I’ve found is a long 2014 study by Homma Mitsunori in ICU Comparative Culture which works through the Meiji Tennō-ki, the Katsura and Terauchi diaries, and the Itō Hirobumi papers, and concludes it was most likely composed on the evening of 4 February 1904.

That date matters enormously, and it cuts against the romantic reading.

4 February was the day the Imperial Conference decided on war with Russia. The poem was almost certainly written after that decision, in private, not as an intervention to stop it. The Meiji Tennō-ki records the Emperor turning to his chamberlains that evening and saying that this war was not what he had wished and then weeping at the thought of failing his ancestors and his subjects.

Meiji signed the declaration of war six days later, on 10 February. Into it he inserted a phrase of his own: 豈朕が志ならむや — how could this be my will? Hirayama reads it as the same thing the poem holds in thirty-one syllables, compressed into a single line. A love of peace and an acceptance of war, in one breath, neither cancelling the other.

The following year, when Port Arthur fell and the fleet came home, he wrote poems saying plainly how pleased he was.

Thirty-seven years later, Hirohito added the identical phrase to the declaration of war on the United States.7

Hold that thought. It applies just as much to his grandson.

The poem before Pearl Harbor

Now, 6 September 1941.

The Imperial Conference met to adopt the Essentials for Carrying Out the Empire’s Policies, a document which set out war preparation first and diplomacy second. Hirohito had objected to that ordering the previous day, in an audience with Prime Minister Konoe.

At the conference, Privy Council President Hara Yoshimichi asked whether diplomacy was to be the priority. The Navy Minister answered. The two chiefs of staff — Sugiyama Hajime for the Army, Nagano Osami for the Navy — said nothing.

Then the Emperor spoke. According to the recollection of Colonel Ishii Akiho, who was present as a staff officer, Hirohito said the silence of the high command was deeply regrettable, took a piece of paper from his pocket, read the poem aloud twice, and said that he had always wished to embody Meiji’s love of peace.8

There are several independent contemporary or near-contemporary records of this: the Sugiyama Memo, the Army General Staff's own record; the diary of Kido Kōichi, Lord Keeper of the Privy Seal, who saw the Emperor twice that day; Konoe's memoir Heiwa e no doryoku ("Efforts Toward Peace"); Ishii's 1946 recollection; and, much later, the Shōwa Tennō Jitsuroku, the official annals published from 2014.

That is unusually good documentation for a closed imperial conference, and the core fact is well established: it happened, and it was extraordinary. Emperors did not intervene at Imperial Conferences. That was the whole constitutional theatre of the thing.

This performance came with two caveats, and the second is the interesting part.

First, the sources don’t agree perfectly on the wording. A small thing, but a reminder that even the best-attested moments reach us through fallible memory.9

Second, and much more importantly: the poem did not mean what we now assume it meant.

The other reading

Hirayama Shūkichi’s 2014 book Shōwa Tennō: “Yomo no Umi” no Nazo is the only book-length treatment of this poem I know of. His finding is that the men in the room may have heard the opposite of what the Emperor intended, and that the poem itself gave them grounds to.10

By the 1930s, the standard published commentary on Meiji’s poems — Sasaki Nobutsuna’s Meiji Tennō Gyoshū Kinkai — was required reading for Japanese officers. And the way that commentary glossed the poem left room for a reading in which the Emperor laments that others disturb the peace of the four seas. On that reading, the poem is not an argument against war. It is an argument that the disturbers must be dealt with.

Hirayama’s argument is that Army staff officers spotted this on the afternoon of 6 September itself, and that within about a day the “imperial will” had been quietly re-read from "the Emperor wants peace" into "the Emperor laments that peace is being disturbed." Same thirty-one syllables. Opposite policy implication.

Homma’s independent study finds a parallel drift in the printed commentaries: editions from 1915 and 1922 gloss the poem neutrally; by 1930 the commentary is condemning Russian aggression and justifying Japan’s response; by 1933 the poem is being used to demonstrate the Emperor’s benevolent designs on the wider world, sitting alongside verses that fed straight into hakkō ichiu rhetoric.

So the poem was not a fixed object with a stable meaning that was later distorted. It was contested from the moment it entered public circulation, and the contest was the point.

Four years on, the war was lost, and the Allies began drawing up lists of the men to be arrested and tried. Australia put Hirohito on its list, seventh of sixty-four names, on the grounds that no sovereign should have immunity, that under the constitution no war could be made without his sanction, and that he could have abdicated if he disagreed. It reached the UN War Crimes Commission in London in January 1946. The US Senate had already passed a resolution calling for him to be tried. Around seventy per cent of Americans wanted him punished.

The single most vivid piece of evidence that the Emperor had wanted peace, spoken in front of fifteen of the most senior men in Japan and recorded in at least four separate documents, was sitting there, available to anyone building a case for him.

Nobody used it.

Part two: why not, and what was being built instead.

Leave a comment

Quick preword before you commit: this is a deep, unhurried dive into kanji readings, not a 60-second explainer, and it does not get to the point quickly on purpose. Kanji fluency doesn't arrive in fifteen-second increments — so if you need a hook in the first three seconds, this is probably not for you.

For the past year or so I’ve been teaching Japanese online to students in Australia, and it’s been one of the more quietly rewarding things I do — enough that I keep thinking of the proverb 情けは人の為ならず.1 In my case it’s been fairly literal: answering a year of specific, sharp, occasionally unanswerable student questions has probably done more for my own Japanese than any studying I did alone.

The most recent one arrived looking simple. A student had noticed that most two-kanji words seem to read with 音読み (on’yomi) for both characters, but every so often a word clearly doesn’t, and she wanted to know why. I gave her the standard answer — the one I was taught, and the one most second-language learners get: 訓読み (kun’yomi) is for reading a kanji on its own or with hiragana attached, on’yomi is for when two kanji combine. 熟語 (jukugo), multi-kanji compounds like 学校 (gakkō), 電車 (densha), and 日本 (nihon), generally use on’yomi for both characters. Standalone words or words followed by hiragana tend to lean toward kun’yomi: 山 (yama), 食べる (taberu), 白い (shiroi).

Clean and simple answer, except it’s not, because even my own textbook example turned out not to be as clean as I’d claimed.

Even the safe example isn’t safe

学校 is on’yomi + on’yomi — 学 is ガク, 校 is コウ — but say ガクコウ out loud and it isn’t what comes out. What comes out is ガッコウ, with a small っ replacing the く entirely. That っ isn’t a third reading of 学 hiding somewhere in a dictionary. It’s the product of a sound change called 促音化 (sokuonka).2

None of which a fluent speaker thinks about while saying がっこう. But it’s worth noticing on the way in: the compound I was using as the uncomplicated example was already quietly doing something to its own sound. If that much is going on underneath the word every textbook uses to teach the rule, the actual exceptions were never going to be simple.

The counterexamples that started this

Once I actually started looking, they piled up fast: 場所 (basho). 手紙3 (tegami). 今日 (kyō). 大人 (otona). None of these are on’yomi-plus-on’yomi. Some aren’t cleanly on’yomi or kun’yomi in any simple sense at all.

“Compounds use on’yomi” isn’t wrong, exactly — it’s a real, strong tendency, because an enormous share of Japanese vocabulary was either borrowed wholesale from Chinese or coined in Japan using Chinese-style word formation (和製漢語, wasei kango), and two-character compounds built that way keep both characters’ on’yomi because the whole unit was conceived as Sino-Japanese from the start. It’s just not a law, and Japanese has several distinct, nameable ways of quietly ignoring it — plus, as I found later, several ways of quietly reshaping the sounds themselves even when the reading pattern is followed correctly.

Does knowing any of this actually matter? Probably not, and I got proof of that scrolling Instagram of all places: a street-interview clip of a Japanese teacher asking passersby to read the 魚偏 (fish-radical) kanji for common fish, and a surprising number couldn’t — 鰌 (どじょう, “loach”), 鱈 (たら, “cod”), 鯉 (こい, “carp”), 鰯 (いわし, “sardine”), 鰹 (かつお, “bonito”), 鯖 (さば, “mackerel”), 鰤 (ぶり, “yellowtail”), 鮭 (さけ, “salmon”), 鮪 (まぐろ, “tuna”), 鯛 (たい, “sea bream”). Fair enough, honestly — menus write most of these in hiragana or katakana anyway.

I mention it because everything below is, strictly speaking, unnecessary. Nobody needs to know how to read 担板漢4to speak Japanese perfectly well. I do this because I find it genuinely interesting (perhpas there’s something wrong with me), not because it’s useful, and that distinction matters for how to read the rest of this. When I’ve tried explaining some of this to Japanese friends, the reaction is usually a polite えー、知らなかった。すごいね。よく知ってるね — which, deployed at that particular pitch, is not quite the astonished revelation I was hoping for.

Regardless — here’s the over-explanation nobody asked for.

On’yomi and kun’yomi, properly this time

“On’yomi is the Chinese reading, kun’yomi is the Japanese reading” isn’t wrong, but it flattens something more interesting. On’yomi aren’t one borrowed layer — they’re several, absorbed at different points across more than a thousand years, each preserving a snapshot of Chinese pronunciation, as Japanese ears heard it, at a specific historical moment.

The earliest layer, 呉音 (go-on), arrived from roughly the 5th–6th century onward, largely via the Korean peninsula, and stuck hardest in Buddhist vocabulary — a lot of sutra terminology still uses go-on today. Later, during the Nara and Heian periods, Japanese envoys and students returning from Tang-dynasty China brought back 漢音 (kan-on), modeled on the prestige speech of the capital, Chang’an; this became the reading favored by scholars and government, and is usually the “default” on’yomi listed first in a dictionary.

Later still, from the Kamakura period on, Zen monks and merchants brought a smaller third layer, 唐音 (tō-on), sometimes 宋音 (sō-on), surviving in specific words like 椅子 (isu, “chair”) and 行灯 (andon, “paper lantern”). A single character can carry more than one — 明 is みょう in some words (go-on) and めい in others (kan-on), depending which historical layer a given compound was coined in.

Kun’yomi, meanwhile, aren’t translations so much as pre-existing native Japanese words (大和言葉, yamato kotoba) matched to whichever kanji best captured their meaning once Chinese characters arrived. Japan had no native script before that; 山 already meant “mountain” in spoken Japanese long before anyone attached 山 to it. So: on’yomi is borrowed sound, kun’yomi is native meaning wearing a borrowed costume — and the “compounds use on’yomi” heuristic exists because so much abstract or technical vocabulary is Sino-Japanese by construction. Here’s how it gets broken.

重箱読み and 湯桶読み

The first way is 重箱読み (jūbako-yomi): on’yomi followed by kun’yomi. You’ll see it in words like — 台所 (daidokoro, “kitchen”: 台 on + 所 kun), 職場 (shokuba, “workplace”: 職 on + 場 kun), 本屋 (hon’ya, “bookshop”), 額縁 (gakubuchi, “picture frame”).

The category takes its name from 重箱 (jūbako), the stacked lacquered boxes used for osechi, which happens to demonstrate its own pattern (重→じゅう on’yomi, 箱→はこ kun’yomi). It’s a slightly odd choice as a teaching example, since it’s not a word most people reach for daily. With that said, I do prefer that tradition named the category after a word that embodies it, rather than inventing something clinical like “mixed compound type A.”

Run the pattern backward — kun’yomi then on’yomi — and you get 湯桶読み (yutō-yomi), named the same way after 湯桶 (yutō), a wooden vessel for hot water. Examples being: 見本 (mihon, “sample”: 見 kun + 本 on), 豚肉5 (butaniku, “pork”), 雨具 (amagu, “rain gear”), 高台 (takadai, “elevated ground”).

This is where my student’s original example, 場所 (basho), lands. 場→ば is kun’yomi (場’s on’yomi is じょう, as in 会場, kaijō, “venue”); 所→しょ is on’yomi (所’s kun’yomi is ところ, tokoro, the standalone word for “place”). So 場所 isn’t a stray exception — it belongs to a named category most learners simply aren’t introduced to, which is exactly why it reads as chaos rather than pattern. Once you have the category, words that looked arbitrary turn out to belong to a shared, describable system, rather than something memorised one exhausting item at a time.

熟字訓

This is where the kanji stop owning individual sounds.

熟字訓 (jukujikun) covers compounds where the reading attaches to the whole written unit at once — you can’t point at one kanji and say “this part is pronounced this,” because that division doesn’t exist. 今日 (kyō, “today”) is the standard example: 今 doesn’t read きょ anywhere else, 日 doesn’t read う anywhere else. The pre-existing native word きょう was assigned to the two-character unit as a package.

大人 (otona, “adult”) works the same way — おとな doesn’t decompose into a reading for 大 plus a reading for 人. Neither does 明日 (asu or ashita). Other examples: 昨日 (kinō, “yesterday”), 今朝 (kesa, “this morning”), 眼鏡 (megane, “glasses”), 果物 (kudamono, “fruit”), 相撲 (sumō, “sumo”), 一人／二人 (hitori／futari), where 人 alone produces neither り nor たり under any standard reading.

These aren’t just scattered folk oddities — 文化庁’s official appendix to the 常用漢字表, called the 付表, formally lists a little over a hundred such readings as standardized exceptions that don’t reduce to ordinary on’yomi or kun’yomi. The idea underneath all of it: with jukujikun, “which part of the pronunciation belongs to which kanji” isn’t a hard question with a hidden answer — it’s the wrong question. The kanji are a name tag pinned to the whole group, not to any individual in it.

当て字 & 義訓

If jukujikun is “the compound gets one reading as a unit,” this next category is “the kanji were never really reading the word — they were decorating it.”

当て字 (ateji) covers kanji chosen for a word largely apart from their ordinary meaning. Several reference sources usefully split it into two subtypes: 借字 (shakuji), kanji chosen mainly for sound — 珈琲 (コーヒー, kōhī, “coffee”), 倶楽部 (クラブ, kurabu, “club”) — and 義訓 (gikun), in its broader sense, refers to kanji chosen primarily for their meaning and applied to a word whose pronunciation was already fixed — like 煙草 (tabako), literally “smoke” and “grass.”6

For example, 土産 (miyage, “souvenir”) is often casually classified as 当て字, but it is more accurately considered 熟字訓. みやげ was already a pre-existing native Japanese word, with the kanji 土産 — literally, “product of the land” — later assigned to it based on meaning. Its etymology is debated, with theories linking it to 見上げ (“carefully selecting and presenting”) or an older word relating to local produce or storehouses. Structurally, it works in much the same way as 今日 or 大人: the reading belongs to the compound as a whole, rather than being divided neatly between the individual kanji.

Which raises the obvious question: where does 熟字訓 stop and 当て字 begin? Broadly, 熟字訓 is reserved for kanji matched to an existing native (和語, wago) word for meaning. Loanwords get filed differently — 煙草 (tabako, “tobacco”) looks structurally identical to 土産, but たばこ is a 16th-century borrowing from Portuguese tabaco, not a native word, so it’s classified as 当て字 (the meaning-priority type) rather than 熟字訓, which is technically reserved for wago. In casual usage this line blurs constantly, and even linguists don’t fully agree on edge cases.

And then the sounds themselves start moving

Everything above is about which reading a kanji takes. There’s a separate layer of complexity in how those readings sound once combined — I’ve already gone pretty far but this is where I went further than I meant to.

We already saw 促音化 turn ガク+コウ into ガッコウ. There’s a parallel process, 連濁 (rendaku), where the first sound of the second element voices — か→が, さ→ざ, た→だ, は→ば — when two words combine. It’s textbook-associated with native kun’yomi compounds (手＋紙→てがみ, not てかみ), but it also happens to on’yomi morphemes, less predictably. 山 (さん) is a clean case: 火山 (kazan) and 登山 (tozan) both voice, and research on Sino-Japanese morphemes by 鈴木豊 finds roughly two-thirds of common-noun 山-compounds take the voiced form. Mountain proper names, though, mostly resist it — 富士山 stays ふじさん, never ふじざん.7

Japanese tends to resist packing two voiced obstruents into the same word. This tendency is known as ライマンの法則 (Lyman’s Law), named after Benjamin Smith Lyman, an American mining engineer who “rediscovered” the pattern in the 1870s while working in Meiji-era Japan. Japanese scholars, including Kamo no Mabuchi and Motoori Norinaga, had already described the same phenomenon in the 18th century. Like most rules in Japanese, however, Lyman’s Law has documented exceptions of its own. 縄梯子 (なわばしご, “rope ladder”) voices はしご to ばしご even though ばしご already contains a voiced sound via 梯子’s ご, which by the law should have blocked it. It’s the standard example cited whenever this comes up, and one of very few genuinely undisputed exceptions in the literature.

One more, because it shows how easy it is to mis-attribute a sound change once you’re primed to see rendaku everywhere. 舌鼓 is properly したつづみ (した, “tongue,” + つづみ, “hand drum” — the sound of smacking one’s lips over good food), but a widespread variant, したづつみ, also exists, and it’s tempting to explain that as some kind of voicing swap. It isn’t quite that. 精選版日本国語大辞典 attributes it to analogical leveling: by the early modern period, したつづみ and 腹鼓 (はらつづみ) both picked up づつみ-final variants by analogy with genuine rendaku ~づつみ (包, “wrapping”) compounds like 上包 (うわづつみ) — words where the voicing is doing real morphological work. したつづみ absorbed the shape of that pattern without actually being that pattern; some sources describe it more loosely as 音位転換 (metathesis), the same phenomenon behind the common mispronunciation of 雰囲気 as ふいんき instead of ふんいき. Either way, したつづみ remains the standard, dictionary-preferred reading, though the variant is old and common enough that recent dictionaries note it as an accepted secondary form.

Does anyone actually calculate any of this while reading?

Almost certainly not — and this is what answers my student’s original question honestly. Nobody looks at 学校 and thinks “on’yomi compound, therefore ガク+コウ, subject to 促音化.” Nobody sees 今日 and thinks “jukujikun, whole-unit reading independent of components.” They just see がっこう, きょう.

English does the identical thing. Though, through, cough, rough — four words sharing “ough,” pronounced four completely different ways, for reasons buried in centuries of pronunciation shift that no fluent adult reconstructs while reading a text message. You don’t run the Great Vowel Shift every time you read “though.” You just know the word.

On’yomi, kun’yomi, jūbako-yomi, yutō-yomi, jukujikun, ateji, gikun, sokuonka, rendaku, Lyman’s Law: enormously useful for understanding why Japanese looks the way it does, and none of it run in real time by a fluent reader. Is any of this actually useful to know? Probably not. But I still think there’s value in understanding this stuff — not because you’ll consciously apply it while reading at speed, but because it turns what initially feels like arbitrary chaos into chaos with a shape.

Which brings me to 漢検2級

Part of why this grabbed me is that I’m currently studying for 日本漢字能力検定2級 (通称 漢検2級), the kanji proficiency exam run by the 日本漢字能力検定協会. Not something to walk into casually: 2級 covers all 2,136 常用漢字 and expects accurate reading and writing, understanding of compound structure, antonym/synonym pairs — and, the official exam criteria requires explicit understanding of 熟字訓 and 当て字, with the board’s own published examples being 海女 (ama, “female pearl diver”) and 玄人 (kurōto, “expert”).

The passing line for 2級 sits around 80% (roughly 160/200 points). I won’t hang a single precise pass-rate figure on it, since the foundation’s own published data varies noticeably session to session — but it’s not a rubber stamp by any measure.

And then there’s 漢検1級

Will I eventually attempt 漢検1級, the test that only around 300 people pass per year? Kanken’s own data shows recent 1級 sittings passing anywhere from roughly 6% to 13% of test-takers.

Honestly... maybe. I’m aware that’s a slightly unhinged thing to commit to in writing, especially once you look at who’s actually cleared it as a non-native, non-kanji-background speaker. A small, enthusiast-maintained tracking list — not an official record, since 漢検 doesn’t publish results by nationality — currently counts six such people.

Where that leaves me

If one thing is clear from all this, it’s that linguistic knowledge doesn’t produce fluency. The deeper I dig into why Japanese reading feels chaotic to a learner, the more structure I find underneath — on’yomi has its own layered history, jūbako-yomi and yutō-yomi are named and teachable, jukujikun has an actual official list, even the phonological shifts turn out to be documented and mostly predictable.

But the better I understand that structure, the more obvious it becomes that fluent reading was never built on consciously applying it. A native speaker doesn’t need the term 重箱読み to correctly read 残高. They just know the word — the way you just know “though” without an etymology lecture running underneath. And the same holds in reverse for a learner: you don’t need the terminology first either. Enough exposure, and enough time spent looking things up one stubborn word at a time, gradually surfaces the same underlying shape on its own — consciously naming it just allow you to map it better.

If you’re a native Japanese speaker, or a non-native speaker, I’m genuinely curious whether any of this ever surfaces consciously for you. Do you ever catch yourself thinking “on’yomi” or “kun’yomi” mid-read, or does the sound just appear the instant your eyes land on the word?

This reflects my own understanding as a learner and teacher, checked as carefully as I could manage against dictionary, institutional, and academic sources — not a claim of scholarly authority on historical Japanese phonology. Corrections welcome.

Leave a comment

Share Roido Does Japan

Vocabulary

• ぼんやり — vague, hazy, indistinct.

• 三角標（さんかくひょう） — a triangular survey marker; in this book, the glowing signposts that line the tracks through the sky.

• 濃い（こい） — deep or thick, used here for a rich color.

• 灼く（やく） — to scorch or forge — here, metal fresh from the fire.

• ～という工合（ぐあい） — “in such a way that,” used to describe a manner or state.

• 億万（おくまん） — hundreds of millions.

• 蛍烏賊（ほたるいか） — firefly squid, a real species known for glowing blue.

• 化石する（かせきする） — to fossilize, to turn to stone.

• 金剛石（こんごうせき） — an old-fashioned word for diamond.

• ばら撒く（ばらまく） — to scatter or strew about.

• 擦る（こする） — to rub.

• ごとごと — the clattering, rattling sound of a train.

• 軽便鉄道（けいべんてつどう） — a light or narrow-gauge railway.

Grammar Notes

• ～という工合に／という風に — “in a way that—,” used to describe something by comparison, often something strange or hard to picture directly. Miyazawa uses both in the same breath here, piling comparison on comparison. Example: 誰もいないという工合に、家中しんとしていた。(”The house was silent, as if no one were there at all.”)

• ～たと思うと — “no sooner had... than—,” describing two things happening in near-immediate succession. 云う声がしたと思うといきなり明るくなって captures how fast the light arrives, right on the heels of the voice. Example: ドアが閉まったと思うと、電気が消えた。(”The door had barely closed when the lights went out.”)

• ～てしまう — marks an action as completed, often with a note of helplessness or surprise. 眼を擦ってしまいました isn’t just “he rubbed his eyes” — it’s “he found himself rubbing his eyes,” almost against his will. Example: つい笑ってしまった。(”I couldn’t help but laugh.”)

• ～つづけていた — “had been continuing to—,” describing an action already in progress by the time it’s noticed. 走りつづけていたのでした tells us the train had been moving for a while before Jovanni realized it. Example: 気づかないうちに、雨が降りつづけていた。(”Without my noticing, it had been raining the whole time.”)

• ～のでした — a soft, reflective past-tense ending common in narrated fiction, gently confirming something the narrator has just realized or is explaining to the reader. It reads more like a storyteller’s aside than a plain statement of fact. Example: 彼はもう帰っていたのでした。(”He had, it turned out, already gone home.”)

From 大久保ゆう訳「あのときの王子くん」（サン゠テグジュペリ『星の王子さま』より）

Vocabulary

• いいつける（言いつける） — to order or instruct someone to do something.

• あきらめる（諦める） — to give up on a dream or hope.

• めげる — to be discouraged, to lose heart. A soft, slightly childish word.

• くたびれる — to grow tired or worn out — here, used metaphorically for the exhaustion of dealing with adults.

• しぶしぶ — reluctantly, unwillingly.

• そうじゅう（操縦） — to pilot or operate a vehicle, usually a plane.

• とびまわる（飛び回る） — to fly or dash around from place to place.

• やくに立つ（役に立つ） — to be useful, to come in handy.

• ひとめで（一目で） — at a glance, instantly.

• たすかる（助かる） — to be saved, to be a relief.

• ちゃんとした — proper, respectable, put-together.

• とっておき — something special kept in reserve, not shown to just anyone.

• お近づきになる（おちかづきになる） — a polite, formal way of saying “to become acquainted with someone.”

  ---

Grammar Notes

• ～なんて — Adds a dismissive tone to a noun, roughly “a thing like X.” ボアの絵なんて echoes the adults’ own contempt — stronger and less neutral than など. Example: 漫画なんて時間の無駄だ。(”Manga? What a waste of time.”)

• ～ても～ても — “Whether or not,” built from two て-forms, one affirmative and one negative. 見えても見えなくても shows the outcome simply doesn’t matter to the adults. Example: 分かっても分からなくても、とりあえずやってみよう。(”Whether you understand it or not, just try.”)

• というわけで — A casual connector meaning “so, that’s why” — used to sum up a chain of small causes, like how you’d tell a story out loud. Example: 財布を忘れた。というわけで、今日はお昼を抜くことにした。(”I forgot my wallet. So I decided to skip lunch today.”)

• ～わけでもなかった — A soft, partial negation: “it’s not exactly that—” rather than a flat denial. Gentler than わからなかった, and often pairs with でも or しかし for a shrugging tone. Example: 彼のことが嫌いなわけでもない。(”It’s not that I dislike him, exactly.”)

  ---

Translation:

Grown-ups don’t care in the slightest whether you can see what’s inside a boa constrictor or not. What they wanted from me was geography, history, arithmetic, and grammar — so at six years old, I gave up on the idea of becoming a painter. My masterpieces No. 1 and No. 2 had both fallen flat, and that was enough to discourage me completely. Grown-ups never work anything out on their own, and it wears a child down, always having to spell everything out for them.

So, reluctantly, I settled on a different line of work and learned to fly a plane instead. I’ve flown a little bit all over the world since then. And geography, as it turns out, really did come in handy — I can tell China from Arizona at a glance, which is no small comfort when you’re lost somewhere over the dark at two in the morning.

I’ve lived my life this way, meeting plenty of perfectly respectable people along the road. I’ve spent my time among grown-ups. I’ve watched them up close. It hasn’t exactly raised my opinion of them.

Whenever I came across someone who seemed a little sharper than the rest, I’d try them with my prized Drawing No. 1, the one I always kept close at hand. I wanted to know if they were someone who could actually see what was really there. But the answer that came back was always the same — “It’s a hat.” So with people like that, I never mentioned boa constrictors, or jungles, or stars. I met them on their own ground instead: bridge, golf, politics, neckties. And the grown-ups, for their part, always seemed so pleased to have made the acquaintance of someone so sensible.

Closing Thoughts

星の王子さま (Hoshi no Ōjisama) was one of the first real books I read in Japanese, back when I was looking up every second word in the dictionary. I don’t think that was an accident. It’s simple without being simplistic, and its narrator, a downed pilot recalling his own childhood, has a voice that slips easily between plain and formal Japanese in a way that taught me more about tone than any textbook did.

It’s also a book with an unusually tangled translation history in Japan. For over fifty years there was really only one Japanese 星の王子さま: 内藤濯’s 1953 Iwanami translation, which sold millions of copies and is even where the now-famous title comes from — Le Petit Prince translates more literally as 「小さな王子」, but 内藤 chose 星の王子さま, and it stuck so completely that Iwanami still asks other translators not to reuse it. That monopoly held until the Japanese copyright expired in January 2005, which set off what scholars have called the third “Little Prince boom”: within a year or two, over a dozen new translations appeared from different publishers, each making its own choices about the narrator’s voice, the prince’s speech, even what to call him. The excerpt above comes from one of those newer translations, 大久保ゆう’s 『あのときの王子くん』, deliberately titled to distance itself from 内藤’s version and freely available on Aozora Bunko.

I chose this passage specifically because it’s the clearest statement of what the whole book is about, delivered before the prince himself ever shows up: the narrator tells us why he gave up drawing and started flying instead, and it’s because the adults around him could only ever see a hat.

Every morning she sits at the loom on the near bank of a great silver river, and every evening she is still there, and the cloth that comes off her hands is so fine that her father, who rules everything there is to rule, cannot bring himself to ask her to stop.

So she doesn’t stop.

She weaves the seasons into fabric while the rest of the sky goes on without her, and somewhere in the middle of all that competence she becomes, without really noticing it happen, very lonely.

Her father notices instead.

He is a god, but he is also, in this particular story, a parent watching his daughter disappear into her own diligence, and he decides to do something about it.

On the far bank of that same river lives a young man who tends cattle — steady and unremarkable in the way that turns out, on meeting, to be exactly what someone like her needed.

The introduction works almost too well.

They fall in love the way people do in stories that need to move quickly, and they marry, and for a little while everything is soft.

And then everything stops.

She sets down the loom. He lets the cattle wander.

The cloth that was once so fine that a god wouldn’t interrupt it goes unwoven, and the fields on the other side of the river go untended, and the father who arranged this whole thing in the first place watches his daughter’s talent and his son-in-law’s herd both quietly fall apart.

So he does what gods in these stories tend to do.

He doesn’t undo the marriage. He just puts a river between them that neither of them can cross, and tells them that’s how it will stay.

It’s their tears, not his second thoughts, that change anything.

She cries long enough and hard enough that he relents, on one condition: if she returns to her weaving, if she does the work, they may cross back to each other once a year, on the seventh night of the seventh month.

And that is, more or less, the deal that has held for over a thousand years.

Weave all year. Meet for one night. Go back to the loom.

Tonight is that night.

If you are anywhere in Japan today, July 7th, you are living inside the one evening of the year when Orihime the weaver and Hikoboshi the cowherd are permitted to close the distance between them — she as the star we call Vega, he as the star we call Altair, the river between them the pale band of the Milky Way.

There’s a complication baked into the story from its oldest tellings: the first time the lovers tried to cross, they found no bridge, and Orihime wept until a flock of magpies flew up and formed one out of their own wings so she could pass.

According to the legend, if it rains, the magpies can’t come because the river rises too high to bridge, and the two are forced to wait another year.

The rain on this particular night even has its own name in Japanese folk tradition — the tears of the lovers who couldn’t make it across, called 催涙雨 (sairui-u), literally “tear-inducing rain.”

There’s also a lesser-known companion term for the night before: rain that falls on July 6th is called 洗車雨 (sensha-u), or “cart-washing rain” — said to be the water from Hikoboshi washing his ox-cart clean in preparation to go meet Orihime.

Which makes tonight’s forecast an uncomfortably personal question.

July in Japan sits inside tsuyu, the rainy season, and the version of this festival practised on the old lunar calendar used to fall in August, safely after the rains had cleared, when clear skies were the norm.

Move the date to July 7th on the modern solar calendar, though, and you land the festival squarely inside the wettest stretch of the year, when a clear Tanabata night is more the exception than the rule.

Somewhere between a tenth-century poem and a twenty-first-century calendar reform, Orihime and Hikoboshi lost their best odds of actually meeting.

It’s a strange thing to sit with — the idea that a country’s weather has quietly been assigned emotional stakes for over a thousand years.

But once you start pulling on that thread, you find the whole story is like that: layered, older than it looks, and carrying more freight than a tale about pebbles-into-boulders or a runaway couple who forgot how to work has any obvious right to carry.

To understand why so many people in Japan will glance up tonight half-hoping for clear skies, you have to go back to where the story actually came from — and it isn’t Japan.

A Chinese Love Story, Borrowed and Rebuilt

The version of Orihime and Hikoboshi that Japan tells today is a transplant.

Its roots run back through the Chinese legend of the Weaver Girl and the Cowherd, Zhinü and Niulang, a folk tale old enough that fragments of it appear in the Classic of Poetry, one of the oldest surviving collections of Chinese verse.

In that lineage, the Weaver Girl is a granddaughter of the Jade Emperor, gifted at her loom, married off to a cowherd on the far side of the Silver River, and separated from him by the same parental fury once the two of them stop working and start simply enjoying each other’s company.

The version that reached Japan travelled less as a story to be read for pleasure and more as a ritual to be performed.

In China, the festival attached to this legend was called Qǐqiǎo, “the Festival of Begging for Skills” — a night when women prayed to the Weaver Girl star for greater proficiency at needlework and textile craft, a fittingly on-the-nose request given who they were praying to.

Japanese sources render this as Kikkōden, and it’s this skill-petitioning ceremony, not the romance itself, that first crossed the water into Japan.

That crossing has a specific date attached to it.

According to the imperial court’s own records, the festival was formally introduced to Japan by Empress Kōken in the year 755, staged at the imperial palace during the Nara period.

It arrived, in other words, not as folklore whispered among farmers but as a piece of continental sophistication imported wholesale into the Japanese court — the same way so much of Nara-period culture, its Buddhism, its legal codes, its capital city itself, arrived by deliberately copying Tang China.

The five-coloured threads that Chinese women had traditionally offered to the stars while begging for weaving talent came along with the ceremony, hung as offerings in a practice that would, centuries later, quietly transform into the five-coloured paper strips still tied to bamboo today.

The Word Existed in Japan Before the Story Did

Here is the detail that trips almost everyone up: “Tanabata” isn’t actually a Chinese word, and it isn’t even, strictly, a reading of the characters 七夕 that make up its name.

Those characters are properly read “Shichiseki” in the Sino-Japanese reading brought over with the festival itself — seven, evening.

“Tanabata,” the word every Japanese person actually uses, comes from somewhere else entirely: a native Shinto purification custom that had nothing to do with China at all.

In that older Japanese practice, a young miko — a shrine maiden — would seclude herself and weave a sacred cloth on a loom called a tanabata (棚機), offering the finished textile to a god in exchange for protection of the rice harvest from storms and blight.

The weaver in this ritual carried the title tanabata-tsume, “the maiden of the tanabata loom.”

When the imported Chinese star-festival arrived and settled onto the same date on the calendar, the two traditions — one continental and newly arrived, one native and already old — folded into each other almost seamlessly.

The Chinese legend supplied the romance and the stars; the native rite supplied the name everyone would actually use, plus a weaving-maiden archetype so close to Orihime’s own role that scholars still debate how much the two figures shaped each other.

It’s a small piece of etymology, but it says something bigger about how this festival works: even at the level of its own name, Tanabata has never been purely one thing.

It has always been at least two traditions wearing a single set of clothes.

From Palace Poetry to Street Festival

For its first several centuries in Japan, Tanabata belonged almost entirely to the aristocracy.

Heian-period courtiers marked the night with poetry competitions, composing verses about the stars and the river between them while gazing up at the actual sky, treating the festival as an occasion for refined literary display rather than public celebration.

It’s the kind of scene that’s easy to picture: robed nobles in a palace garden, ink and paper close at hand, more interested in crafting an elegant line about longing than in actually experiencing it.

The festival didn’t reach the wider population until the Edo period, and when it did, it changed shape considerably.

It was sometime during this era that the practice of writing wishes on tanzaku — narrow strips of coloured paper — and tying them to bamboo branches took hold as a popular custom, spreading out from court ritual into something ordinary households could take part in.

The five-coloured threads that Nara-period courtiers had once hung as offerings to the stars had, by the Edo period, become the five-coloured paper strips still recognisable today, and the wish-writing itself traced its logic straight back to the festival’s origins: since Tanabata had always been about praying for skill at the loom and the brush, ordinary people began asking, quite literally, to get better at their handwriting and their craft, the same way palace weavers had once asked to get better at theirs.

It’s also around this period that Tanabata quietly absorbed pieces of Obon, the midsummer festival for honouring ancestral spirits, since the two events fell close together on the old lunar calendar.

Practices like floating decorated objects on rivers, or burning them as offerings once the festival ended, likely owe something to that overlap — a way of sending wishes off to the heavens borrowed from a tradition of sending spirits home.

There’s a smaller, stranger story tucked into this period too, about the food most associated with the festival.

Somen, the thin wheat noodles served cold with dipping sauce every July 7th, trace back to a twisted dough confection called sakubei that arrived from Tang-dynasty China during the Nara period.

According to one version of its origin, a child died on the seventh day of the seventh month, and a plague of fevers swept through afterward; people began offering sakubei, the child’s favourite food, as a ward against illness, and the custom of eating it on that date outlived the plague story that supposedly started it.

By the time it evolved into somen, the noodles had also picked up a second, gentler symbolism — their thin white strands were said to resemble both the Milky Way overhead and the threads on Orihime’s loom, a food that manages to reference the festival’s astronomy and its heroine’s occupation in a single bowl.

What the Bamboo, the Colours, and the Decorations Are Actually For

Almost everything hung on a Tanabata display is doing symbolic work, even the parts that look purely decorative.

Bamboo itself was chosen deliberately. It grows fast, tall, and straight, and its upward-reaching shape was read as a natural conduit for carrying wishes skyward on the wind; it was also believed to ward off insects, which made it doubly appropriate for a festival with roots in praying for a good harvest.

The five colours of the tanzaku paper — green, red, yellow, white, and either black or purple — aren’t an arbitrary palette.

They map directly onto the Chinese Five Elements: wood, fire, earth, metal, and water, respectively, the same cosmological system that shaped so much of premodern East Asian medicine, architecture, and calendar-keeping.

Black, water’s original colour in that system, was quietly swapped for purple in Japan, since black carried unlucky associations there that it didn’t carry on the mainland.

Layered onto those five elements is a second, Confucian scheme: the five cardinal virtues of benevolence, righteousness, propriety, wisdom, and sincerity, each assigned to the same five colours, so that choosing a tanzaku shade traditionally meant choosing which virtue you were quietly invoking alongside your wish.

The other ornaments that hang alongside the tanzaku each carry their own specific hope.

A folded paper kimono is meant to protect against illness and misfortune. Paper cranes, worked in tribute to the belief that the bird lives a thousand years, stand in for wishes of longevity.

A drawstring purse asks for prosperity, a woven net asks for a good catch for anyone who fishes, a small paper trash bag asks for tidiness and frugality, and long paper streamers called fukinagashi represent the threads on Orihime’s loom itself, carrying wishes for improved skill at weaving, sewing, and needlework more broadly.

None of this is incidental.

The whole display is essentially a physical wish-list rendered in folk symbolism — item by item, each shape doing a specific piece of petitioning.

Once the festival ends, the decorations traditionally aren’t simply thrown away.

Bamboo branches heavy with tanzaku were set adrift on rivers or burned, a way of physically dispatching the wishes toward the heavens rather than letting them linger as clutter — though in recent decades, growing environmental concern has made this particular custom far less common than it once was.

Two Stars That Were Never Actually Close

It’s worth pausing on the astronomy, because it makes the myth sadder in a way the storytellers who first told it couldn’t have known.

Orihime corresponds to Vega, in the constellation Lyra; Hikoboshi corresponds to Altair, in Aquila.

Both are genuinely bright, easy-to-spot stars that anchor the “Summer Triangle” alongside Deneb, which is part of why they were singled out for a myth in the first place — bright enough to notice, positioned on either side of the Milky Way’s pale band, close enough in appearance to look like they belong to the same story.

They are not, however, close in any way that matters physically.

Vega sits roughly 25 light-years from Earth; Altair sits roughly 17 light-years away, on a different line of sight entirely.

The two stars are separated from each other by something in the neighbourhood of 14 to 16 light-years of actual space — far enough that even light, the fastest thing there is, would take over a decade to cross the gap between them.

Whatever bridge of magpies the story imagines, the real distance involved makes “once a year” almost a kindness by comparison.

Vega and Altair have never touched and never will; the entire romance is something that happens only in the shared imagination of everyone looking up at the same night sky, which might be the most quietly moving detail in the whole legend.

The meeting was never physically possible.

People kept believing in it anyway, for over a thousand years, because the story was never really about the stars in the first place.

A Festival Still Being Celebrated, Region by Region

Modern Tanabata plays out at wildly different scales depending on where in Japan you happen to be standing.

The most famous large-scale version is Sendai’s Tanabata Matsuri, held annually in early August on the older lunar-calendar date, its origins traced back to Date Masamune, the first warlord to rule the Sendai domain in the early 1600s.

The festival’s popularity ebbed for decades afterward, nearly vanishing during the economic depression that followed the First World War, before a group of local volunteers revived it in 1928 and fixed its modern August dates.

It disappeared again, unavoidably, during the Second World War — almost no decorations were put up in the city between 1943 and 1945 — before returning in 1946 with 52 decorations strung through the streets, and then, remarkably, ballooning to 5,000 decorations by the time the Shōwa Emperor visited the following year.

Today Sendai’s version includes seven traditional types of decoration, each carrying one of the wish-meanings described earlier, alongside a large ornamental ball first designed in 1946 by a local shopkeeper as a stand-in for a dahlia flower.

Elsewhere, the festival keeps its own local flavour.

Hiratsuka’s Shōnan Tanabata Matsuri, near Tokyo, is one of the largest celebrations held on the July 7th date rather than August’s, and Tokyo’s own Asagaya neighbourhood hosts a smaller, similarly colourful version timed just before the Obon holiday.

In Kyushu, Dazaifu Tenmangu shrine holds a “Tanabata no En,” a poetry-and-culture banquet whose roots reach back to 1046, when it was staged by officials of the old Dazaifu government office and continued in some form through the medieval period, before being formally revived in 1957 after a long interruption, and still held every July 7th to petition for the flourishing of scholarship and the arts.

Between Sendai’s centuries-old civic pageantry and a shrine ceremony with medieval roots, the same festival ends up expressing itself as both a tourist spectacle and a quiet, localised ritual, sometimes within the same country on the same night.

Why a Story About Weather Has Lasted a Thousand Years

Strip away the bamboo and the tanzaku and the noodles, and what’s actually left is a story about two people who were good at their jobs, fell in love, got worse at their jobs because of it, and were punished not by losing each other completely but by being allowed exactly one night a year together — a punishment generous enough to keep hope alive and cruel enough to make sure it never fully resolves.

That specific shape of longing, patience stretched taut across a year at a time, is probably why the story has survived as long as it has: it isn’t really asking anyone to believe two stars can physically cross a hundred trillion miles of space.

It’s asking them to recognise something more ordinary — the particular ache of being close to someone you can only see rarely, and the private math everyone in that position runs without meaning to, where one good meeting somehow has to be worth the eleven empty months that came before it.

Layered onto that is the specific vulnerability of the date itself.

Almost no other major festival on the Japanese calendar has its emotional payoff held hostage to the actual weather the way Tanabata does. Christmas happens whether it snows or not. A cherry blossom festival can be moved a little if the blooms come early or late.

But Tanabata’s entire narrative climax — will they or won’t they meet — collapses into something as mundane and as uncontrollable as whether it rains on a given Tuesday night in the middle of Japan’s rainy season.

That’s an unusually honest thing for a festival to admit about hope: that it can be built up all year and still lose to something as indifferent as a low-pressure system.

Rather than undermining the story, that vulnerability seems to be exactly what’s kept people invested in it for so long.

A wish that’s guaranteed to come true isn’t really a wish. A reunion that depends on the sky cooperating is one that people have reason to actually hope for, year after year, instead of simply expecting.

So — What’s the Sky Doing Tonight?

Which brings the story back to right now, this evening, wherever in Japan you happen to be reading this.

Tonight sits at the tail end of tsuyu, the exact stretch of the calendar this piece opened by blaming for Orihime and Hikoboshi’s bad luck, and this year is proving the point without much need for exaggeration: as of this morning, most of the country was sitting under a mix of cloud with occasional breaks of sun, western Japan pushing into real midsummer heat, and only scattered pockets — a thunderstorm here, a soaked stretch of coastline there — carrying any serious rain risk through the evening.

Tokyo’s own forecast for the night was leaning toward the drier side of that mix, with meteorologists rating the umbrella largely unnecessary, though a stray evening shower over the Kantō region was never fully off the table.

Somewhere else in Japan tonight, that coin will land the other way; it usually does, for at least a few of the country’s prefectures, every single July 7th.

None of this changes anything, practically speaking.

The wish will get written on the tanzaku either way, the bamboo will go up in shopping streets whether or not anyone can actually see the Milky Way through it, and the somen will taste the same over ice whether the sky cooperates or not.

But somewhere in Japan tonight, someone is going to step outside, tilt their head back, and check.

Not for any real reason, since nobody involved actually needs the weather to go their way tonight in any way that affects their own life.

They’ll check anyway — the same private, slightly irrational habit that’s apparently been passed down for well over a thousand years: looking up on this one date, hoping, on behalf of two stars that were never really going to meet in the first place, that tonight the sky stays clear.

Leave a comment

Men in blue replica jerseys moved slowly through the empty rows, plastic bags in hand, picking up cups, wrappers, and the odd forgotten scarf.

FIFA posted photos of the cleanup on X, praising the supporters’ manners and noting, almost as an aside, that they had done it again.

That word did a lot of work.

Japan’s fans have been cleaning up after themselves at major tournaments for years now, long enough that international broadcasters no longer treat it as a surprise so much as a confirmation of something they already believed about the country.

To viewers outside Japan, the image fit neatly into a broader reputation: a country of spotless train platforms, orderly queues, and city streets that stay clean despite a near-total absence of public trash cans. Within days, fans from other countries, including Portugal, were filmed doing the same thing in their own sections, with many online crediting the Japanese supporters for starting the trend.

Inside Japan, though, the same photographs set off a very different conversation. It didn’t ask whether the fans should keep cleaning stadiums. It asked a sharper question: why is it easier for these men to clean up after a football match on another continent than it apparently is to clean up after themselves at home?

A Habit With Deep Roots

Japan’s reputation for public tidiness didn’t start with football. Much of it traces back to something more mundane: most Japanese public schools don’t employ janitors for daily cleaning. Instead, students spend part of each afternoon sweeping their own classrooms, wiping down hallways, and taking turns cleaning the toilets, a practice known as sōji. The idea, reinforced for years before anyone puts on a national team jersey, is that a shared space is everyone’s responsibility rather than someone else’s job to be paid for separately.

That habit sits on top of an older set of ideas about cleanliness tied to Shinto practice, where physical dirt and spiritual impurity have long been treated as related concepts. Visitors to a shrine still rinse their hands at a temizuya, a stone water basin, before approaching the main hall, a small ritual that treats washing as something more than hygiene. Whether or not any individual fan connects a stadium cleanup to a purification ritual at a shrine, the broader cultural instinct, that a public space reflects on the people using it, runs deep enough that picking up litter after a match doesn’t register as unusual generosity. It registers as the bare minimum.

This is, in part, what the American anthropologist Ruth Benedict was gesturing toward in her 1946 study of Japanese culture, The Chrysanthemum and the Sword, when she described a society unusually attentive to how its behavior appears to others. Benedict’s book is dated, and later scholars have pushed back hard on parts of it. But the basic observation, that a sense of being watched shapes public behavior, still gets cited because it tracks with something visible: cleaning a stadium in front of a global television audience is exactly the kind of act that framework would predict.

Not everyone in Japan is convinced the instinct is quite so pure, though. Online critics pointed out that Japan’s own festivals, and crowded gathering spots like Shibuya on Halloween night, are routinely left buried in litter once the crowds disperse, no international cameras required.

Some reached for a specific term for this: ee-kakko-shii, a colloquial phrase meaning something close to “putting on a good face,” or performing decency for an audience that happens to be watching. From this angle, the stadium cleanup isn’t exactly hypocritical, but it is selective: a form of care that switches on when an audience exists and switches off when it doesn’t.

Defenders pushed back just as quickly. “What’s shameful about this,” one user wrote, arguing the cleanup was still preferable to the alternative, international headlines about Japanese visitors leaving a mess behind. Whatever the underlying motivation, they argued, a cleaner stadium was worth defending on its own terms.

Share Roido Does Japan

The Image That Changed the Conversation

Sometime after the Netherlands match, a single illustration began circulating on Japanese social media that gave the whole debate a shape. It showed the same man twice, side by side. In one panel, he wears his Japan jersey, crouched down, smiling, gathering trash beneath a stadium seat. In the other, he’s home on the sofa, looking at his phone. A laundry basket sits untouched in front of him. In the background, a woman, deliberately left ambiguous as either wife or mother, stands at the sink working through a stack of dishes alone. Across the top ran four words:

「家でやろう。」

Ie de yarō. Do it at home.

The image spread quickly, gathering more than 60,000 likes on one post in mid-June, and then, roughly a week later, resurfacing with new force behind a separate post that reportedly crossed 1.9 million views. The accompanying text made the argument explicit rather than leaving it to the image alone: Japanese men’s household labour hours rank among the lowest in the developed world, it said, and the least they could do is start sharing that work at home.

“Everyone wants to save the world. Nobody offers to help with the dishes.”

One user borrowed that line, a paraphrase of a joke by the late American writer P. J. O’Rourke, to sum up the mismatch. Others raised a related point that had nothing to do with statistics: some of the men being photographed for their diligence in a stadium thousands of miles from home had, presumably, left their own wives managing young children alone for the length of a World Cup trip.

What the Numbers Actually Show

The joke landed because the underlying statistics back it up. Japan’s Cabinet Office, citing 2021 OECD time-use data, found that Japanese women spend five and a half times as long on unpaid domestic labour as Japanese men, a gap far wider than comparable figures for the United Kingdom (1.8 times), France (1.7 times), or the United States (1.6 times). A related figure from the same period put actual daily hours on the table: Japanese women logged more than three hours a day on unpaid work, against 47 minutes for men.

The imbalance sharpens further in the households that might need relief the most. A 2021 government survey of dual-income couples with children under six found that mothers spent more than seven hours a day on housework and childcare, while fathers in the same households spent under two. Both parents, in other words, were working outside the home. Only one of them came home to a second job.

None of this means Japanese men, as a group, work less overall. Japan has long recorded some of the longest average working hours among OECD countries, a fact that complicates any simple story about laziness. The imbalance isn’t really about total effort; it’s about where that effort is allowed to count. Paid labour happens mostly in public, gets measured, and shows up in GDP. Unpaid labour happens mostly at home, doesn’t get measured the same way, and, as a result, doesn’t carry the same social weight, even though someone still has to do it every day.

Why the Gap Has Been Slow to Close

Part of the explanation sits in how Japan’s postwar economy was built. The high-growth decades after 1945 ran on a model built around lifetime employment and seniority-based pay, in exchange for near-total availability: long hours, routine overtime, and evening obligations to colleagues that left little time for much else. That system needed someone managing the household on the other end, and for most of the postwar period, that role fell to wives by default

The arrangement wasn’t purely cultural inertia; parts of it were built directly into policy. Japan’s tax code still includes a spousal deduction that reduces a household’s tax burden as long as a lower-earning spouse, almost always the wife, keeps her own income under a set threshold, long informally known as the “1.03 million yen wall.”

Many employers layer their own spousal allowances on top of the same threshold. The practical effect, whatever the original intent, has been to make part-time or unpaid work a financially rational choice for a great many households, reinforcing exactly the division of labour the viral image was mocking.

A Debate That Split More Than One Way

The response to the viral post itself didn’t fall neatly into a single camp. Some users agreed with real enthusiasm, one suggesting, half-joking, that any wife tired of an unhelpful husband should start by making him wear his national team jersey at home too, on the theory that the uniform seemed to be doing more of the motivating than any underlying sense of duty.

Others pushed back hard against the framing itself. “主語が大きすぎる,” one popular reply read, roughly: your subject is too broad. The complaint wasn’t that the statistics were wrong. It was that turning an entire gender into a single indictment felt unfair to the men who already split chores evenly, or who came home from twelve-hour days too depleted to do much of anything.

That second objection points to something the first misses. If a demanding, inflexible work culture is a large part of what keeps men out of the kitchen, the fix isn’t obviously a matter of individual willpower. It starts to look like a workplace problem wearing a household costume.

Some Evidence of Change

The numbers aren’t entirely static, either. The share of eligible Japanese men taking childcare leave has climbed for twelve consecutive years, reaching a record 40.5% in fiscal 2024, up sharply from 30.1% the year before, according to Japan’s Ministry of Health, Labour and Welfare. Disclosure rules that originally applied only to companies with 1,000 or more employees starting in 2023 were extended to companies with 300 or more employees in 2025, adding public pressure on a wider range of employers to report how many of their male staff actually use the leave they’re entitled to. The government's target was 50% uptake by the end of 2025; the most recent confirmed figure, for fiscal 2024, was 40.5%, meaning Japan was on track to fall short before the year even finished.

It’s worth being precise about what that number does and doesn’t show. A rising childcare-leave rate measures whether a father takes time off around a birth, not whether he keeps doing an equal share of laundry, cooking, and school pickups years later, once the leave itself is a distant memory. The OECD’s unpaid-labour comparison and the childcare-leave figures are measuring different things, and it would be easy to mistake progress on one for progress on the other.

Sociologists have a name for the daily grind that continues well past parental leave: the “second shift,” a term coined by the American sociologist Arlie Hochschild to describe the unpaid work that begins the moment a paid job ends for the day. Whether Japan’s rising leave numbers eventually bend that second shift toward something more even is, for now, an open question rather than a settled one

Where This Leaves Things

The stadium cleanup and the unequal division of labour at home aren’t really contradictions of each other. They’re both expressions of the same underlying belief, that care for a shared space matters and reflects on the person providing it. What the viral image argued, in effect, wasn’t that Japanese men should stop cleaning stadiums. It was that the instinct behind the cleanup, noticing a mess, feeling some responsibility for it, doing something about it without being asked, is exactly the instinct that seems to switch off at the front door.

Whether that’s a matter of habit, workplace culture, tax incentives, or some combination the statistics can’t fully separate is still being argued out, loudly, in the same place the photographs first went viral.

Leave a comment

Key Takeaways

• The phrase 「家でやろう」 (”Do it at home”) went viral after Japanese supporters were praised for cleaning up a World Cup stadium, with critics contrasting that public diligence against a much less equal division of labour at home.

• Citing 2021 OECD data, Japan’s Cabinet Office found Japanese women perform 5.5 times more unpaid domestic labour than men, a wider gap than the UK (1.8x), France (1.7x), or the US (1.6x).

• In dual-income households with children under six, Japanese mothers average more than 7 hours a day on housework and childcare, compared to under 2 hours for fathers.

• Some critics call the international praise selective, using the term ee-kakko-shii, and point to domestic festivals that are often left littered without an international audience watching.

• Reactions to the viral post split online, with some pushing back on generalizing all Japanese men and pointing instead to Japan’s demanding work culture.

• Male childcare leave uptake has risen for twelve straight years, hitting a record 40.5% in fiscal 2024, though this tracks leave taken around a birth rather than the ongoing daily division of household labouur.

FAQ

What does 「家でやろう」 mean, and why did it go viral? It translates to “Do it at home.” The phrase appeared on an illustration contrasting a Japanese fan cleaning a stadium abroad with the same man ignoring housework at home, and it spread after critics used it to argue that praised public behavior wasn’t matched by private behavior.

Why are Japanese fans known for cleaning World Cup stadiums? The habit is often linked to sōji, the daily practice in Japanese schools where students clean their own classrooms rather than relying on janitorial staff, along with a broader cultural emphasis on shared responsibility for public spaces.

How much more unpaid labour do Japanese women do compared to men? According to 2021 OECD data cited by Japan’s Cabinet Office, Japanese women perform 5.5 times more unpaid domestic labour than men, the widest such gap among the countries compared in that data.

Did everyone in Japan agree with the criticism? No. Some users defended the stadium cleanups as harmless and worth celebrating regardless of the domestic debate, while others argued the criticism unfairly generalized all Japanese men rather than addressing Japan’s demanding work culture specifically.

What does ee-kakko-shii (ええ格好しい) mean? It’s a colloquial term meaning roughly “putting on a good face” or performing decency for an audience. Critics used it to suggest the stadium cleanups may be partly shaped by the presence of international cameras.

Is Japan doing anything to close the domestic labour gap? Male childcare leave uptake has risen for twelve consecutive years, reaching a record 40.5% in fiscal 2024, and disclosure requirements for employers have expanded. However, this measures leave taken around childbirth, not the day-to-day division of ongoing household chores.

Is the stadium cleanup unique to this World Cup? No. FIFA and international media have praised Japanese supporters for the same behavior at previous tournaments, and fans from other countries, including Portugal, have since been seen copying the practice.

But that’s not really what stayed with me from Japan’s tournament.

What stayed with me was, every match, the few minutes before kickoff, when both teams lined up for their anthems. The players, the fans, and, most notably, manager Moriyasu became overwhelmed with emotion in every fixture. A few looked like they were fighting to hold their composure, singing in a low, steady wave that built slowly instead of erupting.

Anyone who has watched Japan play in a World Cup or any other global sporting competition has probably noticed it: the unusual stillness, the players and fans who seem genuinely moved. It's a strange thing to notice at a sporting event, and an even stranger thing to feel curious about. Is the anthem really that emotional, or are we just imagining a kind of solemnity that isn't there?

It turns out the question has a real answer, and it has very little to do with the melody itself. Japan’s national anthem, Kimigayo (君が代), carries more than a thousand years of history, and the same thirty-two syllables have meant something different in nearly every era of the country’s existence. To understand why it lands the way it does, you have to go back almost as far as the anthem itself.

An Anthem Unlike Almost Any Other

Most national anthems were written to do a specific job: rally a population, mark independence, celebrate a military victory, or flatter a monarch. La Marseillaise is a call to arms against tyranny. The Star-Spangled Banner commemorates a night of bombardment survived. God Save the King is, structurally, a prayer for one person’s continued health. These songs were composed with nations already in mind, often written quickly, in response to a specific political moment, and built to be sung by large crowds with some force behind them.

Kimigayo was not built for any of that, because when it was written, the concept of a Japanese nation-state in the modern sense did not exist yet.

Its lyrics first appear in the Kokin Wakashu, a poetry anthology compiled during Japan’s Heian Period, somewhere around the early 900s. That makes Kimigayo’s lyrics the oldest of any national anthem currently in use, by a wide margin. Most anthems date their texts to the eighteenth or nineteenth century, when nationalism as we understand it was taking shape across Europe and the Americas. Japan’s anthem predates that entire framework by close to a thousand years. It wasn’t written for a country. It was written as a Waka (和歌), a short, formal poem typically composed for a private occasion rather than a national one.

It’s also unusually short. At eleven measures and roughly forty-five to fifty seconds, Kimigayo holds the distinction of being the shortest national anthem in the world by musical structure. There’s no second verse, no building crescendo, no key change for dramatic effect. It simply states its lines once and ends.

And it sounds nothing like a march, because it isn’t one. The melody draws on Gagaku (雅楽), the court music tradition of Japan’s imperial palace, which favors slow tempos, sustained tones, and a kind of deliberate stillness over rhythm or momentum. Where most anthems were composed to be played by a military band and belted out by a stadium crowd, Kimigayo was built to be closer to a hymn: unhurried, contemplative, and a little austere. That alone goes a long way toward explaining why it doesn’t produce the kind of roaring, fist-pumping energy you get from other countries’ anthems. It was never designed to.

But the melody is only part of the story. The words matter just as much, and they reward a closer look.

What the Words Actually Say

Here are the lyrics, in Japanese, and in a literal & poetic English translation.

Read in English, it can come across as oddly abstract, almost like a riddle. But each phrase is doing specific work, and once you know what that work is, the poem starts to make a different kind of sense.

The whole thing follows a traditional waka structure, a 5-7-5-7-7 syllable pattern that Japanese poetry has used for over a millennium. It was written to be read or recited as a single unbroken sentence, which is part of why it feels less like a statement and more like a wish trailing off into something larger than itself.

The opening word, kimi (君), is the one that’s caused the most disagreement over the centuries, and we’ll come back to it in detail shortly. For now, it’s enough to say that kimi originally meant something closer to “you,” in a respectful or affectionate sense — a lord, an honored guest, a loved one. It wasn’t a political term. It was the kind of word you’d use at a birthday or a New Year’s gathering, wishing the person in front of you a long and fortunate life.

Chiyo (千代) and yachiyo (八千代) are numbers stretched into something closer to a sentiment. Chiyo (千代) literally means a thousand generations; yachiyo (八千代) means eight thousand. Stacked together, “chiyo (千代) ni (に) yachiyo (八千代) ni” isn’t really proposing a precise figure. It’s the poetic equivalent of saying “forever, and then some,” using large round numbers the way English speakers might say “a thousand years” without expecting anyone to count them out.

The most evocative image in the poem, though, is the one involving the stones. Sazare-ishi (さざれ石) refers to small pebbles, the kind you’d find scattered loose on a riverbed or a garden path. Iwao (巌) refers to a large boulder, fixed and immovable. The poem imagines the former slowly becoming the latter, growing moss in the process, which in Japanese folklore was understood as something that genuinely happened over an almost incomprehensible span of time, as pebbles gradually fused together into solid rock. Whether or not anyone in the Heian court literally believed pebbles physically grew into boulders, the image works as a measure of time so long it’s barely usable. It’s not asking for a long reign in any countable sense. It’s asking for one that outlasts geology.

Taken together, the poem isn’t a statement of national pride or military strength. It’s a blessing, and a notably patient one. There’s no enemy in it, no victory, no declaration of identity. There’s just a wish that something good will go on for so long that stones will have time to turn into boulders.

That gentleness is exactly what makes its later history so striking, because almost none of that political weight existed when the poem was first written. The political weight came later, and it came from outside the poem entirely.

A Thousand Years, Several Different Songs

For most of its existence, Kimigayo had nothing to do with politics. During the Heian, medieval, and Edo periods, it functioned as a celebratory verse, recited at birthdays, banquets, and New Year’s festivities to wish prosperity on whoever was being honored that day. “Kimi” was simply the person being celebrated, whether that was a lord, a guest, or a family member. The poem traveled through Japanese culture for roughly nine centuries this way, as a piece of festive language rather than a statement about the state.

That changed with the Meiji Restoration in 1868.

When the Meiji government consolidated power and began rapidly modernizing Japan along Western lines, it ran into a problem that had nothing to do with culture and everything to do with diplomacy. Western nations arriving in Japan expected the trappings of a modern nation-state: a flag, a national anthem, military bands that could perform it during official ceremonies. Japan, despite over a thousand years of recorded history, had no anthem in the way Western diplomats understood the term.

The first attempt to fix this came in 1869, when Irish bandmaster John William Fenton, serving with the British military in Japan, set the old poem to a new melody. Fenton’s version was a fairly conventional European-style brass fanfare, the kind of thing you’d expect at a state ceremony in London or Berlin. It didn’t last. It sounded foreign, and it failed to register with Japanese audiences as something genuinely their own, so officials shelved it and went looking for something else.

That search led, in 1880, to an imperial committee turning toward Gagaku (雅楽), the same court music tradition that had shaped Japanese ceremonial life for centuries. Court musicians Yoshiisa Oku and Akimori Hayashi developed a new melody under the supervision of imperial music director Hayashi Hiromori, drawing on that older musical language instead of European convention. German military musician Franz Eckert later arranged the piece with Western harmonies so it could be performed by an orchestra, and that arrangement is essentially the one still used today. The result was something neither fully traditional nor fully Western, but distinctly Japanese in its slowness and restraint, which is part of why it still sounds so unlike everything else played before international matches.

With a melody in place, the Meiji government set about giving the song political meaning it had never previously carried. The 1890 Imperial Rescript on Education pushed Kimigayo into Japanese classrooms as a fixture of civic ritual, and the word “kimi,” once a flexible term of address for anyone deserving of respect, was narrowed to mean one person specifically: the Emperor, now positioned at the center of a rapidly centralizing, modernizing state. A poem that had spent nine centuries wishing well on whoever happened to be listening was now wishing, specifically and exclusively, for the endless reign of imperial Japan.

That shift deepened considerably over the following decades. As Japan moved into the militarism of the early Showa era and the Second World War, Kimigayo stopped being merely associated with the Emperor and became an active instrument of wartime mobilisation. Schoolchildren in Japan, and in territories Japan had colonised, including Korea and Taiwan, were required to sing it daily while bowing toward the Imperial Palace. Soldiers sang it heading into combat, framed as an offering of loyalty to an Emperor the state had elevated to the status of a living god. The same gentle wish about pebbles and boulders had become, in this context, a hymn of total submission to the state.

Japan’s defeat in 1945 didn’t erase that history, but it did force a recalculation. The Allied Occupation stripped the Emperor of political authority and reduced his constitutional role to that of a symbol — a representation of the state and the unity of the Japanese people, with no governing power attached. Occupation authorities restricted the use of the Hinomaru flag during this period but never formally banned Kimigayo. It kept functioning as Japan’s anthem in practice, sung at events without much fanfare, but the government deliberately avoided giving it official legal status. After everything the song had been used for during the war, making it official again carried obvious risk, and for more than five decades, Japan simply let the matter sit unresolved.

It might have stayed unresolved indefinitely, if not for a tragedy in 1999. A high school principal in Hiroshima took his own life after becoming caught between a conservative school board insisting the anthem be sung at graduation and a group of pacifist teachers refusing to comply. The case drew national attention to a dispute that had been quietly simmering in Japanese schools for years, and the government responded by finally passing the Act on National Flag and Anthem later that same year, formally designating Kimigayo as Japan’s national anthem for the first time in its history.

To make that designation politically workable, Prime Minister Keizo Obuchi offered an updated interpretation of the lyrics. “Kimi,” the government now said, referred to the Emperor strictly in his constitutional role as a symbol of the state, within a democracy where sovereignty belonged to the people, not to him. The text of the poem hadn’t changed at all in over a thousand years. What changed, again, was what the country agreed the words meant.

A Song Still Being Argued Over

That 1999 law settled the legal question of whether Kimigayo was Japan’s national anthem, but it didn’t settle the disagreement over what the song represents, and in some ways it sharpened it.

For many Japanese citizens, the anthem today is simply a quiet, dignified piece of national tradition, no more loaded than any other country’s anthem, and arguably less so given its gentle subject matter. For others, particularly pacifists, some on the political left, and many people across countries that suffered under Japanese wartime occupation, Kimigayo remains tightly bound to memories of empire, forced loyalty, and militarism that the song was actively used to encourage. Both readings draw on real history. Neither is simply wrong.

That tension has played out most visibly in Japanese schools. Following the 1999 law, regional boards of education, most prominently in Tokyo, issued directives requiring teachers and students to stand, face the Hinomaru flag, and sing the anthem at graduation and entrance ceremonies. A number of teachers, citing pacifist, anti-imperialist, or religious objections, refused to comply, and many were formally reprimanded as a result, facing salary cuts, suspensions, or difficulty being rehired.

These disputes eventually reached Japan’s Supreme Court, where teachers’ unions argued that compelling participation in the anthem violated Article 19 of the Japanese Constitution, which protects freedom of thought and conscience. The Court’s rulings landed in a genuinely split place: it found that requiring teachers to stand and sing was constitutional, treating it as a routine part of school ceremony rather than a violation of belief, but it also ruled that the harshest disciplinary measures taken against teachers who refused, including termination or extended suspension, went beyond what the law could reasonably justify. In other words, the song could be required. The punishment for refusing it could not be unlimited.

None of this makes Kimigayo unusually controversial by the standards of national symbols generally. Most countries have some version of this argument buried somewhere in their relationship to a flag, an anthem, or a founding myth. What makes Japan’s case distinctive is how directly the dispute traces back to the song’s own history, and how openly that history sits with the country today, neither fully resolved nor fully suppressed.

What the Players Are Actually Singing

Which brings the story back to the players, arms around each other, lining up on the pitch moments before kickoff, with cameras lingering on faces that looked like they were somewhere else entirely.

It’s tempting to read that moment as pure patriotism, the same uncomplicated pride you’d assume from any team standing for its anthem. But given everything the song has carried across the last thousand years, it seems more accurate to say that no two people in that lineup, or in the stands behind them, are necessarily hearing the same song. Some are hearing a wish for peace and continuity that predates the country’s entire modern history, a piece of language passed down from court poets a thousand years before anyone wore a national team jersey. Some are hearing tradition in a simpler sense, a connection to family, ceremony, and home. And some, including people watching from countries that remember Japan’s twentieth century very differently, are hearing an anthem that was once sung by soldiers on their way into a war built on conquest, sung in classrooms as an instrument of obedience, sung as an offering to an emperor treated as divine.

All of those readings are legitimate, because all of them are true of the song at different points in its history. Kimigayo never really resolved which one it is. It just kept being sung, generation after generation, carrying every one of those meanings forward at once rather than settling on a single one.

Maybe that’s actually the more interesting explanation for why the moment lands the way it does, more interesting than any claim about the melody being unusually beautiful or the players being unusually emotional. A thirty-two-syllable poem about pebbles slowly becoming boulders has now outlived nine separate eras of Japanese history, absorbing something from each one without ever losing the others. It’s been a birthday wish, an imperial decree, a wartime hymn, a quiet and unofficial tradition, and finally, in 1999, a piece of settled law that still hasn’t settled the argument underneath it.

Standing on that pitch in Houston, a century removed from the war that gave the song its darkest associations and a millennium removed from the Heian court that gave it its first ones, the players singing it are connected to all of that whether they’re thinking about it or not. It’s a strange kind of weight for a song that, on its surface, is just a wish for something good to last a very long time.

Leave a comment

The numbers themselves are not entirely surprising. English-speaking countries have traditionally directed their students who study abroad toward the United States, the United Kingdom, and increasingly other English-speaking destinations. Japan’s language barrier is real, and postgraduate opportunities there are often less visible to Australian students than equivalent programmes in Europe or North America.

But these numbers point to something beyond logistics. They raise a harder question: why do some students look at a country like Japan and see an exciting challenge, while others look at the same country and see an environment that feels fundamentally inaccessible? And for those who do go — what happens when they arrive?

Who Goes to Japan to Study?

Before asking what happens after arrival, it is worth understanding who is actually there.

According to the Japan Student Services Organization (JASSO), Japan hosted 336,708 international students across all institution types as of May 2024 — a record high, up more than 20% from the previous year. Of those enrolled in higher education institutions specifically, 229,467 students were studying at universities, graduate schools, junior colleges, and professional training colleges.

The composition of that student body reveals everything about how Japan fits into the global flows of international education. The five largest source countries for higher education were China (94,635 students), Nepal (37,203), Vietnam (22,633), South Korea (13,566), and Myanmar (6,996). Together, these five countries account for nearly three-quarters of all international students in Japanese higher education.

Now consider the English-speaking world. The United States sent 3,032 students. The United Kingdom, 769. Canada, 482. And Australia — despite being Japan’s closest major English-speaking neighbour, despite decades of trade and diplomatic ties, despite Japan being one of Australia’s most significant economic partners — sent just 383 students to Japanese higher education institutions in total.

The Gap Between Arrival and Belonging

The decision not to apply to study in Japan is made long before anyone fills out a form. It is made when a prospective student asks themselves: can I imagine building a life there? Can they form friendships? Find employment after graduation? Participate in society beyond the international student bubble?

These are not irrational questions. They reflect a calculation that many international students — and particularly those from English-speaking countries — face with unusual acuity in Japan. The prospect of learning a new language and adapting to a radically different culture can be exciting. It can also create an underlying fear that meaningful participation requires not just learning new skills, but becoming someone else.

This is where the research becomes relevant in ways that go well beyond economics.

When international students who do go to Japan struggle to integrate — when friendships stay surface-level, when the language keeps them at arm’s length from the culture around them, when the social world contracts to other foreigners — the experience is often interpreted as evidence that Japanese society is closed to outsiders. The research literature describes what can happen next as negative assimilation: a process by which people who become more familiar with a society simultaneously become more aware of the barriers preventing them from fully participating in it. Rather than finding their footing, they develop a growing sense of distance. And when they leave, they carry that sense of distance with them, sometimes passing it on to others who might have considered going.

In this way, the low participation of Australian students in Japanese higher education is not just a data point. It is, in part, the accumulated result of experiences that did not go the way people hoped.

What Negative Assimilation Actually Means

The conventional story of migration goes like this: you arrive, you struggle, you adapt, and over time — through language acquisition, cultural familiarity, social connection — you rise. Time in a new country is the engine of improvement. The longer you stay, the better you do.

This is what researchers call positive assimilation, and for a long time it was treated almost as a law of migration. But a growing body of research has been quietly complicating the picture, particularly in Japan.

A landmark study by Takenaka, Nakamuro and Ishida — drawing on survey data from Chinese, Korean, and Western immigrants living in Japan — found that for migrants from English-speaking Western countries, earnings actually declined the longer they stayed. The foreign skills they arrived with — English fluency, internationally transferable expertise, credentials from prestigious foreign institutions — were highly valued at the point of entry. But those same skills did not compound over time the way conventional wisdom would predict. As immigrants stayed longer, the premium attached to their foreign skills diminished. Yet the locally acquired human capital that assimilation theory predicts should improve outcomes often failed to deliver comparable rewards. Japanese education, rather than boosting earnings, was associated with lower wages than foreign-earned credentials. For many immigrants, the skills and qualifications most valued at entry proved more economically valuable than those acquired through deeper integration into Japanese society.

What makes this finding so unsettling is what it implies. If locally acquired human capital — Japanese education, professional socialisation, and other forms of host-country integration — does not consistently translate into higher earnings, then the entire logic of “just integrate harder” begins to unravel. Immigrants were not necessarily failing to assimilate. In many cases, the rewards attached to foreign capital remained stronger than those attached to becoming more like native workers.

The Employment System Nobody Warned You About

To understand why, it is necessary to understand something distinctive about how Japan’s labour market is structured — something that catches many skilled immigrants entirely off guard.

Research by economist Hashimoto Yuki lays this out with unusual clarity. The Japanese Employment System (JES) is built around a single, pivotal moment: the mass simultaneous recruitment of new graduates, every spring, into companies that then train and promote them internally for the rest of their careers. Miss that window — arrive a year late, come from a foreign university, graduate at the wrong time of year — and you are effectively locked out of the main pathway. The door does not close gently. It simply was never designed to open for you.

Hashimoto distinguishes between two types of highly skilled immigrant workers that emerge from this reality. Type I workers end up in professional and technical roles — hired for specific external skills that Japanese companies cannot produce internally: native English instruction, international IT expertise, country-specific knowledge. They remain, in a sense, complementary to the Japanese workforce rather than integrated into it. Type II workers, by contrast, are those who have navigated the Japanese system from within — often educated in Japan, fluent in the cultural expectations of the workplace, embedded in the internal labour market. They are treated more like Japanese employees.

The historical parallel Hashimoto draws is striking. Type I immigrants today, he suggests, resemble the oyatoi-gaikokujin — the foreign specialists Japan hired during the Meiji era to import Western knowledge. Highly paid. Respected for their specific utility. And never quite expected to become Japanese.

This distinction matters because it reveals how institutional structures can produce a kind of permanent foreignness regardless of how hard someone tries to belong. You can be exceptional at your job, deeply committed to your adopted country, and still find yourself positioned by the system as a temporary specialist rather than a permanent member — not through any act of discrimination you could point to, but through the accumulated logic of a hiring structure that simply was not designed for you.

The Persistence of the Gap

The economic data makes the structural problem concrete. Liu Yang’s econometric analysis of Japan’s employment data — drawing on the 2010 national census — found that across all immigrant groups, immigrants face significantly higher rates of unemployment and significantly lower rates of permanent employment than Japanese natives, even after controlling for education, age, and household characteristics.

The gap is not small. Native Japanese men held permanent employment at a rate of 87.1%. Immigrants from Asian countries: 70.9%. Immigrants from the US and UK: 72.6%. Immigrants from South America: just 34.9% — a gap of more than 52 percentage points compared with Japanese natives.

And the gap persists. For immigrants from English-speaking Western countries in particular, Liu describes what he calls a possible “failure in assimilation”: their unemployment rates approach those of natives over time, but this appears to be because they are accepting more impermanent jobs — not because they are being welcomed into the stable, lifetime employment relationships that define economic security in Japan.

So what is actually happening? Immigrants are adapting. They are making themselves more acceptable to the system. But the system is not reciprocating. People who have spent years — sometimes decades — reshaping themselves to fit find themselves looking at the numbers and wondering what all that effort was actually for.

The Language Nobody Talks About

Language sits at the centre of all of this, but not in the way most people think.

The conversation around language and migration almost always focuses on practical functionality: can you navigate bureaucracy, participate in a meeting, read a form? These things obviously matter. But language is not just a tool for getting things done. Language is the medium through which personality travels. It is how you make jokes, express subtlety, signal that you are someone worth knowing.

When you lose access to that — when you are stuck at a level of expression that makes you sound simpler, less interesting, less fully yourself than you actually are — the social cost is immense. You start to experience your host country through a kind of glass. You can see the life happening around you, but you cannot quite reach it. Acquaintances remain acquaintances. Conversations stay functional. The deeper layers of a society — the ones that are worth moving across the world for — stay just out of reach.

Many international students arrive expecting immersion. But immersion requires access, and access requires language competence that most people significantly underestimate how long it takes to develop. Without it, people gravitate toward others who share their background. There is nothing wrong with that instinct — community among people who understand your experience is genuinely necessary. But if that becomes your only social world, you can spend years inside a country while remaining entirely outside its culture. When that happens, disappointment can be misread — by the international student and sometimes by the host society alike — as evidence that belonging was never really possible.

The research bears this out. Liu found that Japanese language skills function as something even more significant than a communication tool. Japan has been described in the academic literature as “the most high-context country in the world,” meaning that effective communication requires not just vocabulary but a deep understanding of cultural codes, hierarchy, and implication. Immigrants who lack this are disadvantaged in the labour market in ways that cannot simply be fixed by working harder or learning more grammar. The language barrier is also a cultural barrier, and it operates in hiring rooms and promotion decisions long after someone can order food or read train signs.

The Difference Between Growing and Disappearing

None of this is an argument against adaptation. Living between cultures changes people, and that change can be one of the most valuable parts of the experience. Exposure to different societies forces people to question assumptions, develop empathy, and understand that their own way of seeing the world is not universal. The goal was never to remain unchanged.

The question is not whether to adapt. It is what kind of adaptation you’re being asked to do — and whether it is building something or eroding something.

There is a meaningful difference between expanding your identity to include new ways of being and replacing your identity with performances of fitting in. One of them leaves you more whole. The other leaves you, over time, with less of yourself to draw on — and with a growing unease that you cannot quite name.

Takenaka and colleagues capture something of this in their economic findings. What they found, counterintuitively, is that the foreign skills immigrants brought with them to Japan were worth more in the labour market than the locally acquired credentials they accumulated after arrival. The things that made someone distinctively themselves — their international education, English fluency, external professional networks — these retained value. The attempts to become more Japanese, more local, more embedded in the system, frequently did not.

This is economically specific, and should not be generalised too far. But it points toward something philosophically significant: that what you were before you arrived is not simply a liability to be overcome.

What Institutions Owe the People They Ask to Adapt

There is a version of the integration conversation that places all the responsibility on the migrant or the international student. Learn the language. Understand the culture. Work harder. Be patient. Give it time.

There is genuine truth in this. But the research on Japan makes clear that institutional structures carry enormous weight in determining whether integration is actually achievable — not just attempted. The Japanese employment system’s reliance on simultaneous graduate recruitment is not a natural law. It is a design choice. And it is a design choice that systematically disadvantages people who did not grow up inside it, regardless of their talent, their commitment, or the effort they put in.

When immigrants from English-speaking countries fail to achieve positive assimilation in permanent employment — not through lack of trying, but because the doors were structured to admit people who came through specific pathways they had no access to — that is a question about institutional design, not individual character.

Similarly, when Liu’s data reveal that ethnic penalties in Japan’s labour market persist even after controlling for education and experience, the conclusion is not that immigrants need to try harder. It is that Japan — like most countries — has not yet built institutions that make the promise of meritocracy real for people who look and sound different from the majority.

The most powerful forms of assimilation pressure are rarely overt. They do not arrive as demands. They arrive as the accumulated friction of a hundred interactions in which your difference is, subtly, treated as a problem. You learn to minimise it. You learn to present yourself in ways that make others more comfortable. And eventually, the minimisation becomes habit, and you stop noticing what you have put down.

Belonging Without Disappearing

Return, for a moment, to those numbers. Eighty-five Australian postgraduate students in Japan. Two hundred and seventy-three undergraduates. In a country of 125 million people, with one of the world’s largest economies, one of the most distinctive intellectual traditions, and research universities that rank among the finest in Asia.

Why so few? Part of the answer is practical — language barriers, unfamiliar application processes, misaligned academic calendars. These are real and should be addressed. But underneath the logistics is a more fundamental question that prospective students are implicitly asking: can I imagine belonging there?

That question deserves a serious answer. And the honest answer, based on the research, is complicated. Japan offers a genuinely rich intellectual and cultural environment. It also has institutional structures — in employment, in social life, in language — that make full participation difficult for people who arrive from outside. That is not a reason not to go. It is a reason to go with clear eyes.

The question worth asking — for Japan, for Australia, for anywhere that describes itself as genuinely multicultural — is not how quickly newcomers can be made to resemble insiders. It is whether the society is flexible enough to let different people fully participate while remaining, recognisably, themselves.

That is a harder goal. It requires institutions to design for diversity rather than assimilation. It requires host societies to examine the assumptions built into their systems. It requires honesty about the difference between asking newcomers to share a society and asking them to become a different kind of person.

And it requires those who live between cultures — in whatever form that takes — to hold onto the distinction between the changes that make us more whole and the changes that make us less visible.

Because the best version of cultural exchange is not one side finally becoming acceptable to the other. It is both sides arriving somewhere neither of them expected, each carrying something the other did not have before.

The goal was never assimilation. It was always belonging — and those are not the same thing.

Data on international student enrolments in Japan sourced from the Japan Student Services Organization (JASSO), Result of International Student Survey in Japan, 2024 (published April 2025).

The three academic sources informing this piece are: Takenaka, Nakamuro & Ishida, "Negative Assimilation: How Immigrants Experience Economic Mobility in Japan" (International Migration Review, 2016); Liu Yang, "Immigrant-Native Gaps of Unemployment and Permanent Employment in Japan" (Economic Analysis and Policy, 2024); and Hashimoto Yuki, "Highly Skilled Immigrants' Occupational Choices and the Japanese Employment System" (RIETI Discussion Paper, 2017).

Leave a comment

The 2026 World Cup is less than a week away. Japan’s squad has already departed for North America, coach Hajime Moriyasu declaring openly at Narita Airport that their goal is to win the tournament. The team’s official slogan, unveiled by the Japan Football Association in March, is 最高の景色を — “For Our Greatest Stage.”

It is a slogan full of ambition and poetry. But before we talk about where Japan is going, it’s worth remembering where they’ve already been — and what actually happened the last time the world came to their shores.

Because the story of the 2002 World Cup is not just a football story. It’s a story about institutional power, political survival, financial corruption, and a governing body willing to bend the integrity of sport to serve its own expansion ambitions. And Japan was caught right in the middle of it.

Two Nations, One World Cup

The controversy of 2002 began not in June of that year but in the summer of 1996, six years before the tournament kicked off, when FIFA announced that Japan and South Korea would co-host the World Cup. It was the first, and to this day only, time two nations have shared hosting duties for the tournament.

The decision papered over what had been a bitter, acrimonious rivalry. Both nations had been bidding independently, each determined to outdo the other in securing football’s biggest prize. Neither had the facilities to host alone, so FIFA forced them together — a decision framed as historic and progressive, but driven largely by logistical necessity and FIFA’s ambitions to crack open the Asian market.

There was an obvious irony here: at the time of bidding, Japan had never qualified for a World Cup. The logistics were also widely criticised. Fans travelling between venues would have to cross seas, European broadcasters would face morning kick-off times, and the monsoon season would hang over the entire tournament. As the first World Cup staged in Asia, many accused FIFA of prioritising political expedience and market access over sporting merit. Those accusations would only grow louder as the tournament unfolded.

The Man With Everything to Lose

To understand what happened at the 2002 World Cup, you need to understand who Sepp Blatter was and what position he was in when the tournament began. Blatter was elected as the eighth President of FIFA on June 8, 1998 — just two days before the tournament kicked off in France — but the circumstances of that election were murky from the start.

FIFA’s presidential voting gives every member nation an equal vote, regardless of size or footballing stature. It was a system Blatter understood better than anyone, and he used it to his advantage. His 1998 campaign was accused of bribing African member nations through promises to bring the World Cup to Africa, and of securing CONCACAF delegates through the influential — and later disgraced — Jack Warner. His opponent, Lennart Johansson of UEFA, had strong backing from the Confederation of African Football, the largest voting bloc in FIFA, but that backing, it seemed, could be bought. Blatter’s team reportedly reached out to African football officials through intermediaries, offering financial inducements in exchange for votes. Blatter won. Johansson lost. The full picture of what had been paid to whom would only emerge years later.

By the time the 2002 World Cup opened, Blatter was already fighting for his political survival. In May 2002, just weeks before the tournament began, FIFA’s secretary-general Michel Zen-Ruffinen produced a 21-page dossier, supported by 300 pages of internal documents, accusing Blatter of financial mismanagement, conflicts of interest and abuse of power. Eleven of the 24 executive committee members filed a criminal complaint to a Zurich court on the basis of Zen-Ruffinen’s evidence. Blatter won the presidential election at the end of May anyway, then turned on his accusers with characteristic contempt: “Friday, I throw Zen-Ruffinen out of the door.” In December 2002, the Zurich public prosecutor closed the investigation, remarkably describing the complainants themselves as “reprehensible.” Blatter remained in power for a further 13 years.

Japan Builds Ten Stadiums — Then Loses in the Round of 16

Japan spent enormously to host. The Nissan Stadium in Yokohama, which hosted the final between Brazil and Germany, cost $621 million to build — at the time the most expensive World Cup stadium ever constructed. Across Japan, ten new or heavily renovated stadiums were built or upgraded specifically for the tournament.

One member of Japan’s World Cup bidding committee later admitted: “They did not have a post-World Cup strategy” — and six out of ten stadiums went on to lose millions in the years that followed. Some found second lives: the Sapporo Dome, for instance, became a hybrid football-baseball venue. Others struggled badly. Sendai’s stadium — built for the tournament but never becoming the home of the local J-League side — was used mainly for athletics and occasional events.

The Samurai Blue performed respectably, topping their group and advancing to the Round of 16 before losing 1-0 to Turkey. It was a solid debut for a country appearing in only its second World Cup. But as Japan exited, their co-hosts were just getting started — and what followed made many question whether sport had anything to do with it.

The Referees, the Red Cards, and the Conspiracy That Still Burns

South Korea had a good team in 2002, managed brilliantly by Guus Hiddink in a disciplined 3-4-3 system built around high-pressing, energetic football. Their group stage — topping ahead of Portugal and Poland — was widely considered legitimate. But from the Round of 16 onwards, the results became increasingly difficult to explain.

Italy, Round of 16

The match against Italy in Daejeon became one of the most controversial in World Cup history. Ecuador’s Byron Moreno was the referee. Hiddink’s South Korea pressed aggressively and Italy — despite boasting Maldini, Cannavaro, Totti, Nesta, Inzaghi, Materazzi and Vieri — were rattled. Christian Vieri scored to make it 1-0, but Italy’s lead wouldn’t last.

What followed was a litany of decisions that left Italian players, fans and journalists in disbelief. Francesco Totti was sent off for diving in a decision widely considered unjust. Damiano Tommasi had a goal wrongly disallowed for offside. Two South Korean players committed serious fouls — including a two-footed challenge and an elbow — that went completely unpunished. Ahn Jung-hwan headed a golden goal three minutes before the match went to penalties, sending South Korea through.

Back in Italy, the reaction was volcanic. Legendary sports journalist Giorgio Tosatti wrote in Corriere della Sera: “Italy have been thrown out of a dirty World Cup where referees and linesmen are used as hitmen.” Jung-hwan, who was on loan from Serie A club Perugia, had his contract cancelled by the club’s president in an act of petty vengeance. Moreno was later investigated by FIFA over “a number of controversies” and in 2010 was sentenced to 30 months in prison in the United States on drug trafficking charges.

Spain, Quarter-Final

Italy had already had four perfectly good goals disallowed in the group stages — against Croatia and Mexico — before the South Korea match. Against Moreno, Tommasi’s golden goal became the fifth Italian goal chalked off in the tournament. If the Italy match left a bitter taste, the Spain quarter-final turned the bitterness into fury. Egyptian referee Gamal Al-Ghandour disallowed two perfectly legitimate Spanish goals — Ivan Helguera’s header and Fernando Morientes’ header from a Joaquin cross. Paul Hayward wrote in The Telegraph: “The records say that the Koreans knocked out Spain in a penalty shoot-out in Gwangju on Saturday. The records are a lie and this tournament has descended into farce.”

The response from the public was unprecedented. Following the Spain match, 400,000 emails were sent to FIFA complaining about the officiating decisions — so many that they crashed FIFA’s servers. Blatter was forced to acknowledge there had been problems with officials. Italian and Spanish television threatened to sue FIFA, arguing the skewed refereeing had cost them millions in broadcast revenue.

The Referee Selection Theory

Here is where it gets interesting from an institutional perspective. The referees appointed to South Korea’s knockout matches were not from major footballing nations. They came from Ecuador, Egypt, and Trinidad — countries with minimal international football pedigree, whose officials were arguably more susceptible to the pressure of 50,000 screaming Korean supporters.

Blatter’s motive was clear.

Asia represented half the world’s population and football’s largest untapped commercial market. Getting an Asian team to the semi-finals gave the entire project — the first World Cup on Asian soil, the controversial co-hosting decision, the years of political manoeuvring — its justification. A South Korean run deep into the tournament was worth billions in broadcast rights, sponsorship deals, and long-term fan development across the continent. South Korea were eventually knocked out by Germany in the semi-finals.

The 2015 Reckoning

For years, the conspiracy theories around 2002 remained exactly that — theories. Then came 2015.

In May of that year, US and Swiss authorities arrested seven FIFA officials in Zurich on corruption charges. The indictments alleged decades of bribery, kickbacks and financial fraud at the highest levels of world football. Blatter — who had somehow survived every previous scandal, winning re-election as FIFA president four times, including once while nine of his officials were being indicted — finally resigned, though not before winning yet another election days after the arrests.

In the wake of the 2015 scandal, Italy’s Corriere dello Sport reopened the 2002 files, directly alleging that FIFA and referee Byron Moreno had been involved in match-fixing to keep South Korea in the tournament. The paper argued that the 2015 corruption arrests provided the institutional context that had always been missing: here, finally, was proof that FIFA operated as a corrupt organisation willing to manipulate outcomes to serve commercial interests.

By Blatter’s third term, accusations of bribery, financial mismanagement and World Cup bid-rigging had become alarmingly commonplace. In 2006, FIFA vice president Jack Warner was accused of massive fraud involving the resale of World Cup tickets in Germany through a travel company controlled by his family — FIFA cleared him and merely expressed “disapproval.” By 2010, vote-buying allegations surrounded the awarding of the South Africa World Cup. Then in May 2015 came the reckoning: the US Department of Justice disclosed a 47-count criminal indictment charging seven FIFA executives with having received $150 million in bribes over more than two decades. They were arrested at the Baur au Lac hotel in Zürich, where FIFA was based.

Two days after the indictment, Blatter was installed as FIFA president for a fifth term. Three days later he resigned. The ethics committee subsequently imposed an eight-year ban on both Blatter and UEFA president Michel Platini over a $2 million payment made to Platini in 2011. Blatter received a further six-year ban in 2021.

His reign — which began with the 1998 election won through alleged bribes to African officials, ran through the 2002 refereeing controversy, the 2006 ticket fraud, the 2010 vote-buying allegations, and ended in the 2015 mass arrests — was the most comprehensively corrupt era in the history of international sport governance. He never accepted moral responsibility, and was never criminally convicted. He simply left football in disgrace, blaming everyone but himself.

FIFA Today: Same Structure, Different Faces?

The question worth asking is whether anything has fundamentally changed. The 2026 World Cup is to be hosted across the United States, Canada and Mexico and will be the largest in history, expanded to 48 teams. The 2030 edition will be spread across six countries on three continents. The 2034 edition has been awarded to Saudi Arabia, a decision that drew fierce criticism from human rights organisations and football governance reformers.

FIFA under current president Gianni Infantino has pursued the same expansion logic that Blatter championed: take football to new markets, grow the commercial footprint, manage the politics. The structure that allowed 2002 to happen — one country, one vote; small federations holding disproportionate influence; commercial interests shaping tournament outcomes — remains largely intact.

In 2026, FIFA is still far from a clean, non-political organisation. Under Infantino, the governing body that once bent tournament results to serve Blatter’s Asian ambitions has found new ways to blur the line between football and politics. At the 2026 World Cup draw ceremony at the Kennedy Center in Washington D.C. in December 2025, Infantino presented Trump with the inaugural FIFA Peace Prize — an award suddenly announced just three weeks after Trump had been overlooked for the Nobel Peace Prize. There was no explanation for the process behind the award, no candidates named, and no selection criteria given. Senior officials within FIFA were surprised by the announcement, which had not been discussed with the FIFA Council. Trump called it “truly one of the great honours of my life.”

The politics don’t stop there. With US-Iran tensions running high, Iranian players and fans faced genuine uncertainty over whether they could even enter the United States to participate. Iranian fans are permitted to attend matches, but entry remains complicated by travel restrictions, and FIFA has banned the display of the pre-revolutionary Lion and Sun flag inside stadiums. For a tournament billed as football uniting the world, the geopolitical fault lines running beneath it are hard to ignore.

Japan in 2026: 最高の景色を

Which brings us back to Japan, preparing to chase their most beautiful scenery on American soil.

Japan have never gone beyond the last 16 at a World Cup, but they were the first team to qualify for 2026 and their soaring confidence has been boosted by impressive recent results. They beat Brazil 3-2 in October, then defeated Ghana, Bolivia, Scotland, England and Iceland in succession — six consecutive victories heading into the tournament. Japan have mapped out a plan to win the World Cup by 2050. Whether 2026 is too soon for that plan to bear fruit is the question.

They have been drawn in Group F with the Netherlands, Sweden and Tunisia, opening against the Dutch in Dallas on June 14. The squad has genuine European club quality throughout, though the loss of Brighton winger Kaoru Mitoma to a hamstring injury is a significant blow. Addressing the crowd at the National Stadium after Japan’s send-off match, Moriyasu was unambiguous about the target. “It’s not easy to become world champion,” he said. “I think everyone on the team and all of our supporters understand that. But we want even more people across Japan to support us. … The players will give everything they have in every match, fighting bravely and with toughness until the end.” He framed the campaign in terms of Yamato spirit and the pride of the Japanese people — language that signals this is more than a football tournament for Japan. It is a statement of national ambition.

最高の景色を — the most beautiful scenery. Whether Japan will find it at this year's World Cup, only time will tell. Their journey begins against FIFA-ranked number 7 Netherlands on June 14 at Dallas Stadium — kick-off 4 p.m. ET (5 a.m. JST, June 15 for fans watching back home in Japan).

In May last year, the shop was paying ¥327 per kilogram for Brazilian chicken thigh. By March this year, that had climbed to ¥418 — a jump of nearly 30%. When they eventually passed some of that cost on to customers, raising prices by ¥30 across their menu, it wasn’t enough. The numbers stopped working. They shut the doors.

This story is playing out at restaurants, supermarkets, and food businesses across Japan right now. And to understand why, you have to follow a supply chain that stretches from the Strait of Hormuz to the American Midwest to the chicken farms of Brazil — and eventually lands onto the dinner tables of Japan.

Japan’s Quiet Dependence on Imported Chicken

Chicken has become Japan's most consumed meat in recent years, overtaking pork. Of that, 64% is domestically produced — but the remaining 36% is imported, and 70% of that comes from Brazil.

That dependency looks even more fragile when you dig deeper. Even “domestic” chicken is mostly raised on feed — corn and grain — that is imported. When you factor in feed dependency, only around 8% of chicken consumed in Japan is truly produced here from start to finish. Japan’s food self-sufficiency ratio has collapsed from 73% in 1973 to 38% in 2024. For chicken, the real number is far worse than the headline figure suggests.

For years, this system worked because it was efficient. Japan sourced from wherever was cheapest, kept inventories lean, and passed the savings to consumers. In a deflationary economy where food was expected to be cheap, that made sense. But efficiency, it turns out, is a terrible substitute for resilience — and the cracks are now very visible.

Three Shocks, One Supply Chain

Shock 1: The Currency Double Bind

The yen’s long slide against the dollar is well-documented, but for chicken imports, the more relevant number is the yen against the Brazilian real. Brazil’s central bank has kept interest rates at 15%, drawing global capital and pushing the real to its highest level against the dollar since mid-2024. The result: Japan is paying more in yen and paying more in real terms for the product it’s buying. A single yen movement in the exchange rate costs one major Japanese chicken importer an additional ¥17 million per month.

Shock 2: The Bird Flu That Shook the World’s Biggest Exporter

On May 16, 2025, Brazil announced something that had never happened before: high pathogen avian influenza confirmed at a commercial poultry farm in Rio Grande do Sul, the country’s main chicken-producing state. Brazil supplies more than a third of the world’s chicken exports. When that farm went under, the ripple effect was immediate and global.

China, Europe, Mexico, South Korea — major importers halted Brazilian chicken. Brazil’s own government data put the value of affected exports at over $4 billion USD annually — roughly 40% of its total chicken export market, effectively shut overnight.

Japan suspended imports from the affected region. According to data from the Agriculture and Livestock Industries Corporation (ALIC), Japan’s import chicken inventory as of end-January 2026 had fallen to 80.6% of the prior year level, and total chicken inventory (domestic and imported combined) had dropped nearly 10%. ALIC projected further declines through February and March.

Brazil declared itself avian flu-free in June 2025 after 28 consecutive days without new infections at commercial farms — but supply doesn’t snap back instantly. Chicken travels by ship, and the cargo that couldn’t leave Brazil during the outbreak wouldn’t begin clearing Japanese customs until September at the earliest. The effects of the October shipping delays were still being felt months later.

The price tells the story. In January 2026, Brazilian chicken thigh was selling at ¥600–650 per kilogram. A year earlier, it had been ¥390.

Shock 3: The Strait of Hormuz

Here is where the Iran situation enters — and where things get complicated in ways most news coverage misses.

The Strait of Hormuz is the narrowest chokepoint in global energy shipping: over 25% of the world’s seaborne oil passes through it. Japan sources approximately 94% of its crude oil from the Middle East, and roughly 90% of those tankers use this route. When the strait comes under threat, the effects fan out far beyond oil prices.

The chain reaction goes like this:

Rising oil and gas prices — NHK reported oil up approximately 1.5x and gas up 1.8x from baseline — flow directly into fertilizer costs. Nitrogen-based fertilizers are produced using natural gas as a primary input. When gas prices spike, fertilizer becomes more expensive. When fertilizer gets more expensive, the cost of growing corn and feed grain in the American Midwest goes up. That corn and feed grain is what Brazilian chicken farmers use to feed their birds. So before a single chicken even reaches a processing plant, its cost of production has already risen.

Then the production facility itself faces higher energy costs — the gas-powered heating systems that keep chickens at the right temperature for healthy growth, the factory machinery, the domestic trucks moving birds to processing plants. All of it moves with oil.

Then there’s the shipping. The cost of moving a container from Brazil to Japan has multiple fuel-surcharge components. Major container shipping lines have already introduced emergency bunker surcharges (EBS) in response to Middle East tensions, running at around $150–200 per TEU (twenty-foot equivalent unit) on long-haul routes. Add to this the “container shortage” problem flagged by industry insiders: when ships are forced onto longer detour routes, containers take longer to return, tightening supply and pushing freight rates even higher. The CIF price — Cost + Insurance + Freight — gets hit on the freight side from two directions at once.

Once the chicken arrives in Japan, it still has to reach restaurants and supermarkets via domestic distribution. Trucking costs are also tied to fuel prices. Every stage of the multi-layer Japanese distribution system adds a little more to the final number.

Even the PE bags (ポリエチレン袋) used to package and ship chicken have risen in cost, adding another layer of pressure for importers already squeezed by freight rates and the weak yen.

NHK estimated the cumulative effect of these pressures could add around ¥20,000–¥30,000 to annual household chicken expenditure.

The “Triple Punch” at Retail

TV Asahi’s reporting from a supermarket in Saitama captured what this looks like on the shop floor. The meat manager, with nearly 30 years in the business, described something he had never seen before: imported chicken priced higher than domestic pork. Brazilian chicken prices have risen more than 1.5x in a single year and the wholesale price for the next month is set to climb another ¥50 per kilogram.

“Whatever happens with Iran, if it continues, fuel costs go up, and that gets passed through to us,” the manager said. He was preparing for more.

A town-style Chinese restaurant in Tokyo famous for its towering karaage sets — the kind where customers joke they couldn’t finish it alone — had been going through around 50 kilograms of chicken a day, using frozen Brazilian thigh. This year, they switched to domestic breast meat after the Brazilian thigh hit ¥850 per kilogram. “We changed to survive,” the second-generation owner said plainly. He adapted his preparation method for the new breast meat to replicate the texture of thigh meat, and customers approved. But he remained anxious about what comes next.

At Kunichan’s Kitchen in Chuo, Tokyo, Brazilian chicken is the backbone of their karaage — prized for its fat content and juiciness. But between March and April alone, the price climbed ¥160 per 2kg bag. The shop goes through 5–6 boxes of 10kg per week, so that adds up fast. Owner Maruyama Kunio described the impossible position many small operators now find themselves in:

The same pressure is reshaping what supermarkets even stock. At Super Celcio in Wada-machi, the Brazilian chicken section has simply disappeared. “We couldn’t meet customer needs anymore,” said Tsuruta Satoshi, the store’s general manager. “We were handling Brazilian, but with prices rising continuously and the gap with domestic narrowing, we stopped.” The store switched to domestic-only — but domestic is rising too. Now they’re considering bringing in Thai chicken as the more affordable alternative. “We think we can offer it at a lower price point,” Tsuruta said. It’s a telling snapshot of how the substitution chain works in practice: Brazilian gets expensive, demand shifts to domestic, domestic rises, retailers scramble for the next option.

Karaage specialist Maruyoshoten in Showa Ward was using Thai chicken rather than Brazilian — but found no escape. “Brazilian supply is getting tight, so demand for Thai has increased, and Thai is going up too,” store manager Inano Masayoshi explained. Imported chicken that cost around ¥350 per kilogram in January last year had climbed past ¥400 by early this year — and the shop’s wholesale price for foreign chicken is now approaching double what it was in 2019.

To manage costs, the shop stopped buying pre-cut Thai chicken and switched to purchasing 2kg whole blocks, with staff cutting them in-house. Even so, they raised prices by ¥30 across part of the menu on April 1st. “It was a painful choice,” Inano said. “Chicken has gotten too expensive — we had no option but to do this.”

Shoppers are feeling it too. “If it’s not on sale, I can’t really bring myself to pick it up. I buy it when it’s cheap and freeze it,” one customer said. Another: “When I’m tired I buy ready-made — karaage, yakitori — but food costs really are piling up.”

A supermarket in nearby Seto city, Isseya, sells domestic chicken at the fresh meat counter and uses Brazilian chicken for its prepared foods. Even on a special sale day in late April, prices were broadly up. Meat section manager Matsura Masanori, with nearly 30 years in the business, described something he’d never seen before. Domestic chicken thigh was on the floor at ¥117 per 100g that day — normally ¥150 per 100g. “Our current regular price is about the same as what we’d normally see at the peak of Christmas and year-end, our busiest season,” he said.

The traditional assumption — imported chicken is cheap, domestic is premium — has flipped. “In the past two or three months, it’s essentially reversed. Oil is up, so shipping costs and feed costs have all gone up. In 30 years of doing this, it’s the first time. Our price increases can’t keep pace.”

Matsura also pointed to a Japan-specific dimension that is quietly making the situation worse. Japanese buyers require unusually precise handling: packages must be weighed to exact tolerances (240g to 260g, for instance), bone fragments near joints must be removed, trimming must meet meticulous standards that other markets don’t demand. “I’m starting to hear from overseas that processors are struggling to meet that level of precision”, he noted.

With global labor shortages at processing plants and competing buyers who don’t impose the same requirements, overseas suppliers are starting to route their product to easier customers. Japan’s high standards, once a mark of quality, are becoming a competitive disadvantage in a seller’s market.

The Japan Problem: Being “Outbid” on the World Market

Something more structurally troubling is happening in Brazil that goes beyond short-term price movements.

Japan has historically been one of Brazil’s most important and loyal chicken buyers. Large Japanese trading companies built relationships over decades. Japanese-Brazilian immigrants helped establish trust that shaped long-term contracts. But that legacy advantage is eroding.

Brazil’s domestic borrowing rate sits near 30% — so high that companies can earn better returns leaving cash in the bank than investing in expanded production capacity. New processing lines aren’t being built. Supply isn’t growing fast enough to meet global demand.

Into that constrained supply, new buyers are arriving — China, South Korea, India, Europe — prioritising supply security over price and buying in volumes of 100 containers at a time, at prices Japan won't match. They are buying in volumes of 100 containers at a time, at prices Japan won’t match. Japan, shaped by decades of deflationary consumer expectations, has a structurally low price tolerance. Restaurants can’t raise prices easily. Supermarkets are pressured to keep food cheap. The result is that Japan consistently bids lower than competitors, and is consistently being deprioritised by suppliers.

This is what 買い負け (kaimake) looks like in practice: Japan being outcompeted not because it lacks demand, but because its decades-long deflationary psychology has left it unable to compete in a world where food security is increasingly expensive.

The Amplifier: How Domestic Panic Makes It Worse

Global fundamentals alone would have pushed prices up. But Japan’s distribution structure has an amplification mechanism built in.

When supply risks emerge — a bird flu announcement, an escalation in the Strait of Hormuz — large domestic food companies and retail chains start building inventory defensively. They buy ahead of the key demand seasons: the May holiday period, the December Christmas chicken rush. When major wholesalers hoard stock “just in case,” the available supply in the open market shrinks rapidly.

This triggers secondary panic buying by other businesses, who fear being left without product. They’re not buying because real demand has increased — they’re buying because they expect scarcity. This speculative demand, or 仮需 (kariju), drives prices sharply higher without any change in underlying production costs.

The effect radiates outward. Imported chicken gets scarce, so buyers switch to domestic chicken, pushing domestic prices up too. Thai chicken becomes the alternative to Brazilian, so Thai prices rise. By December, restaurant owners were receiving sudden wholesale price increase notices they couldn’t explain — because the reasons had cascaded through three or four layers of the distribution system before reaching them.

The Structural Picture

None of the forces driving this situation are going away quickly. The yen’s structural weakness relative to other major currencies is a long-term feature, not a temporary anomaly. Brazil’s high-interest-rate, low-investment environment constrains supply expansion. Global demand for protein is growing. The geopolitical uncertainty around the Strait of Hormuz is unlikely to resolve on a schedule that benefits Japanese food importers.

The one variable that could bring meaningful cost relief — a substantial appreciation of the yen — is outside anyone in the food industry’s control.

What this means in practice is that the price of chicken has undergone a structural reset, not a temporary spike. The era of sub-¥400 chicken is not coming back.

What Needs to Change

NHK’s reporting pointed toward two levels of response.

At the industry level, importers are beginning to diversify their supplier base, investing directly in Brazilian grain research, funding production improvements, and deepening relationships at the farm and processing level rather than simply buying at spot market prices. The goal is to position Japan as a preferred customer — one worth supplying — rather than a price-sensitive buyer that can be deprioritised when better offers arrive. Building genuine, sustained partnerships with producers overseas, including equity investment in farmland and processing facilities, is increasingly seen as the only way to ensure reliable access.

At the policy level, the concept of food security is being reframed. Japan’s pursuit of supply-chain efficiency — always buying from the cheapest available source, maintaining minimal inventory — maximised cost competitiveness but created fragility. The shift being proposed is from 効率性 (efficiency) to 耐久性 (durability): accepting somewhat higher baseline costs in exchange for supply stability.

At the consumer level, NHK suggested practical adaptations: shifting toward more vegetables, buying domestic and seasonal produce where possible, and choosing products with fewer imported inputs. These are small individual adjustments to a structural problem — but they reflect the reality that the supply chain running beneath everyday meals is now visibly under strain.

The Bigger Picture

Japan’s food self-sufficiency rate of 38% is among the lowest of any major developed nation. The country has spent 50 years trading away supply security for price efficiency, and is now confronting the consequences…

A karaage shop in Nagoya closes. A restaurant in Tokyo quietly swaps thigh for breast. A supermarket manager in Seto, after 30 years in the trade, sees imported chicken priced higher than domestic pork for the first time in his life. These aren't isolated incidents. They're the visible surface of a structural shift that runs from the Strait of Hormuz, through fertilizer plants and grain farms in the US, to chicken farms in Brazil, across container ships on the Pacific, and onto the dinner tables of Japan.

The question isn't whether chicken prices will stay elevated. The question is whether Japan has the institutional appetite — and the consumer tolerance for higher prices — to rebuild the relationships and domestic capacity that a more resilient food system requires.

Buy Me A Coffee

Easy enough — English words reshaped in Japanese sounds. But here is the thing that surprises people: not all of them are English.

Japanese has a whole category called 外来語 — gairaigo, literally “words coming from outside” — and when you trace where they actually come from, you find something much stranger and more interesting than borrowed English. You find Portuguese. You find Dutch. German, French, Spanish, Italian, even Russian. You find words that crossed the Pacific and the Indian Ocean centuries before English did, in the hands of merchants and missionaries and the Dutch East India Company.1

The Portuguese came with Christianity

Portuguese missionaries arrived in 1549. Merchants followed. And with them came words that embedded so deeply into Japanese that many speakers today have no idea they are standing on Portuguese soil.

• パン pan — bread (from pão)

• タバコ tabako — tobacco (from tabaco)

• カルタ karuta — playing cards (from carta)

• テンプラ tenpura — tempura (from têmporas)

• ボタン botan — button (from botão)

These are not obscure loanwords kept alive by historians. パン is the everyday word for bread in Japan — children learn it before they learn to read. テンプラ is a national dish. The Portuguese fingerprints on Japanese cuisine and daily life are so old they have become invisible.

Then, in 1639, the shogunate expelled the Portuguese. When envoys sailed from Macau the following year to try to reverse the expulsion, the shogunate executed sixty-one members of the mission and burned their ship. The message was clear. The Portuguese chapter was over. But their words remained.

Sakoku: not a locked door, but a guarded gate

鎖国 — sakoku — is usually translated as “closed country” or “isolationism.” That framing, while convenient, is deeply misleading. Japan during the Edo period (roughly 1603–1868) did not close, it curated.

Interestingly, the word sakoku itself is a foreign import of sorts. It was coined by Japanese astronomer and translator Shizuki Tadao in 1801, while he was translating the work of 17th-century German naturalist, physician, explorer, and writer, Engelbert Kaempfer — whose book The History of Japan was one of the few Western accounts of the country to reach European readers. The very term we use to describe Japan’s isolation was invented by a Japanese scholar translating a German text. The irony is almost too neat.

The policy hardened in stages. It was never simply about keeping foreigners out — the full picture is considerably more political. The Tokugawa bakufu had a domestic agenda: to control Japan’s foreign trade so tightly that no rival daimyō could use it to build wealth and military power. Those daimyō, particularly the tozama (outside lords), had profited enormously from East Asian trade networks during the preceding Sengoku period. Locking down maritime access was, in part, a way of locking down internal competition.

Religion was the other major driver — or at least the most visible one. The number of Catholic converts in southern Japan, particularly Kyūshū, had been rising steadily for decades. Missionaries like Francis Xavier had been remarkably effective. The direct trigger most historians point to was the Shimabara Rebellion of 1637–38: an uprising of around 40,000 predominantly Christian peasants, which the shogunate crushed and then used as justification to expel missionaries entirely and ban Christianity on pain of death. The remaining Japanese Christians, mostly in Nagasaki, went underground — they became known as 隠れキリシタン, the “hidden Christians.”

The Protestant Dutch and English, crucially, were seen as able to separate religion from trade — unlike their Iberian counterparts. The Dutch, who were actively at war with Spain and eager to displace both Spanish and Portuguese merchants, had every incentive to reinforce this perception. It was a distinction that would shape the next two centuries.

So: Japan did not seal itself shut. It chose, deliberately and violently, who was allowed in. Trade continued through four carefully managed gateways — Nagasaki, Matsumae in Hokkaido (for trade with the Ainu), Tsushima (for relations with Joseon Korea), and Satsuma (as a conduit to the Ryūkyū Kingdom). Historian Tashiro Kazui draws a useful distinction between Group A — China and the Dutch, under direct shogunate control at Nagasaki — and Group B — Korea and Ryūkyū, dealt with through domain intermediaries. The Group A relationship mattered most for language, because it placed foreign merchants in direct, sustained contact with Japanese society.

The physical mechanism of that contact was a small fan-shaped island in Nagasaki harbour called Dejima — barely 120 metres wide. Chinese and Dutch traders lived there, separated from the city by a narrow strait. They could not enter Nagasaki without special permission. Japanese civilians could not enter Dejima at all. A guard watched the bridge to the mainland around the clock. And yet: trade happened. Words crossed that bridge even when people largely couldn’t.

The Dutch Stayed — by Agreeing to Everything

The Dutch East India Company survived the sakoku purge for one reason: they were willing to be subordinate in a way the Portuguese were not. The Dutch East India Company (VOC) had no missionaries. They wanted trade, not souls. They agreed to confine themselves to Dejima, to submit to humiliating inspections, to trade entirely on Japan’s terms. And so while British, Russian, French, and American ships all attempted to force open a relationship with Japan and were turned away — the Dutch stayed on their small island for over two hundred years.

This extraordinary arrangement produced an extraordinary intellectual exchange. Rangaku — 蘭学, “Dutch learning” — became the umbrella term for all Western science and medicine that filtered into Japan through Nagasaki. Japanese scholars, permitted limited contact with VOC staff, translated Dutch medical texts, studied Dutch anatomy, and learned Dutch cartography.

The word ランドセル (randoseru) — the hard-backed school bag that every Japanese child carries today — comes directly from the Dutch ransel, meaning knapsack.

• ビール bīru — beer (from bier)

• ゴム gomu — rubber (from gom)

• ランドセル randoseru — school backpack (from ransel)

• ピンセット pinsetto — tweezers (from pincet)

Every one of these words is a record of two centuries of enforced proximity on a 120-metre island. They are linguistic sediment.

The Gates Open and Everyone Rushes in

On July 8, 1853, Commodore Matthew Perry of the United States Navy steamed four warships into Edo Bay — ships that would become known as the kurofune, the Black Ships. He demanded Japan open to trade. The following year, at the Convention of Kanagawa, the Shogun signed. Within five years, Japan had signed similar treaties with Britain, Russia, France, and the Netherlands — treaties that Japanese intellectuals almost unanimously regarded as unequal, forced through the same gunboat diplomacy being used across Asia.

The floodgates opened, and with them came a new wave of foreign words — this time overwhelmingly from English, reflecting British and American commercial and cultural dominance through the nineteenth and twentieth centuries. Gairaigo from this era is what most people today assume is the whole story: テレビ (terebi). コンピューター (konpyūtā). インターネット (intānetto).

But the Portuguese and Dutch words were already there, embedded beneath the surface like an older stratum of rock. Sakoku hadn’t prevented borrowing — it had simply determined, with great precision, who Japan would borrow from.

The English Problem

The Portuguese and Dutch loanwords slipped into Japanese and settled. They filled gaps — new objects, new concepts, new foods that had no existing Japanese equivalent. パン named something that simply didn’t exist in Japan before the Portuguese arrived. The language absorbed the word and moved on, undisturbed.

Modern English borrowings are doing something different. They are not filling gaps. They are replacing perfectly functional Japanese words that already exist.

クレーム (kurēmu) from "claim" — used specifically to mean a complaint to a business, replacing 苦情 (kujō). シェア (shea), from “share,” is increasingly used where the native 共有 (kyōyū) once stood unchallenged. Some will argue that the english counterpart in Japanese has some different nuances, and they may be right, but the trend is overwhelming. Some English words arrive so reshaped that their origins become almost unrecognisable — a whole phenomenon the Japanese call 和製英語, wasei-eigo: “Japanese-made English.” Words like サラリーマン (sarariiman) — “salaryman,” a term that has actually looped back into English with its Japanese meaning intact.2

Japanese linguists have a term for this anxiety: 国語の乱れ, kokugo no midare — “the disordering of the national language.” It surfaces periodically in public discourse. The National Institute for Japanese Language and Linguistics has, at various points, issued official guidelines recommending native alternatives to fashionable English loanwords in bureaucratic and media contexts.

But language doesn’t take orders. Young Japanese speakers are not consulting official guidelines when they text each other. And the pressure is not easing — it is intensifying, driven by global media, social platforms, English-language media, and an economy that has made English proficiency valuable in a way Dutch never was.

What Sakoku Actually Tells Us

There is a temptation to look back at the sakoku period and see only restriction — a paranoid feudal state sealing itself off from progress. But the linguistic record tells a more interesting story. Japan during that period was not a static, sealed container. It was a system of deliberate, asymmetric openness. It chose its interlocutors. It set the terms. And the words that survived that system — the Portuguese bread, the Dutch backpack, the Chinese characters that form the backbone of the writing system — are evidence not of isolation, but of selection.

Modern Japan has no such gatekeeping. English words arrive faster than any commission can evaluate them, through screens rather than ships — fused with native Japanese words, twisted into verbs, their meanings stretched until they become hard to follow even for native speakers.

The question of what a language loses when it stops selecting is one that Japanese, and perhaps every non-English language, is quietly beginning to answer. Loanwords were never just words. They were always a record of power — of who was allowed in, on whose terms, and what they left behind when they came.

Subscribe now

Leave a comment