On July 17, 2026, the “Act on the Punishment of Flag Desecration and Related Acts” was enacted in Japan. On August 4, a group of constitutional law scholars published a statement protesting the law, gathering 201 signatories as of August 5. Among the initiating signatories are some of the most prominent names in the field, including Yasuo Hasebe, Kenji Ishikawa, and Asaho Mizushima.
The statement’s core arguments — that the law constitutes content-based regulation of expression, that it amounts to viewpoint discrimination, and that its scope is impermissibly vague — are all faithful to, and solidly grounded in, the standard framework of Japanese constitutional doctrine. But there is one passage — the passage carrying the statement’s single heaviest claim — that turns out to be argued the most thinly.
The Step the Statement Skips
Here is the passage in question. The statement first observes that acts of burning or tearing a national flag can serve as an extremely effective form of expression — a way of protesting the current government and its policies, or of conveying a strong negative judgment about prewar and wartime militarism and colonial rule — and then continues:
Such acts are referred to, in constitutional scholarship, as symbolic expression, and are understood to fall within the protection of Article 21(1) of the Constitution.
“Are understood to.” By whom, on the basis of which precedent, and by what standard are they “understood” this way? The statement never says. That flag burning and flag tearing have historically served as a form of protest around the world, and that the Hinomaru has not always been evaluated positively in Japan, are both true statements of fact. But they are facts — not a legal argument for why such an act qualifies as “expression” under Article 21. The statement leaps directly from a historical description to a normative constitutional conclusion.
This is, to some extent, an understandable feature of the genre. A statement of this kind is not an academic paper; it is a public advocacy document whose conclusions are meant to be underwritten by the authority of its signatories rather than by a chain of citations. But if this law’s constitutionality is ever actually litigated, this is precisely the thin spot the government side will press hardest.
The Reference Point: Texas v. Johnson
The case that should have anchored this argument is the U.S. Supreme Court’s decision in Texas v. Johnson, 491 U.S. 397 (1989). In 1984, at the close of a demonstration coinciding with the Republican National Convention, Gregory Johnson burned an American flag and was convicted under a Texas statute prohibiting the desecration of a “venerated object.”
What’s striking is the structural resemblance between that statute and Japan’s new law. The Texas statute punished flag desecration when the actor knew it would “seriously offend one or more persons.” Japan’s new law’s language — punishing acts done “in a manner likely to cause a person marked displeasure or disgust” — is nearly identical in structure. In both cases, the trigger for punishment is not the physical act itself, but the emotional reaction it provokes in an observer.
Justice Brennan’s majority opinion (5–4) first assessed whether Johnson’s conduct qualified as “expressive conduct,” applying the framework from Spence v. Washington (1974): whether the actor intended to convey a particularized message, and whether, given the surrounding circumstances, it was likely that those who viewed it would understand that message. The Court then considered whether the United States v. O’Brien test applied, found that Texas’s asserted interest in preventing breaches of the peace was not implicated on the facts, and held that the state’s interest in preserving the flag as a symbol of national unity was inseparable from the suppression of expression — and therefore could not qualify as the kind of compelling interest needed to justify a content-based restriction. On that basis, the statute was struck down.
The animating principle of the opinion can be summarized in a single line: government may not prohibit an idea simply because society finds it offensive or disagreeable. This is a doctrine aimed precisely at legislation that builds “causing offense” into the elements of the crime.
The Follow-Up: United States v. Eichman
Even more important is United States v. Eichman (1990). The year after Johnson, Congress removed the subjective element — “intent to offend” — and re-enacted a federal flag-protection statute framed instead around the ostensibly content-neutral goal of preserving the flag’s “physical integrity.” The Supreme Court struck this version down too. No matter how neutrally the text was drafted, as long as the government’s stated purpose was to preserve the flag, that interest remained inseparable from the flag’s symbolic value — and so the law could never truly be content-neutral.
This carries an important implication for Japan’s new law.
A brief digression is useful here. Constitutional doctrine generally sorts restrictions on expression into two categories. The first is “content-neutral regulation” — rules concerned not with what is being said, but with when, where, or how it is said (a ban on late-night sound-truck campaigning, for example). The second is “content-based regulation” — rules that target the substance of what is being expressed, which are treated as a more serious intrusion on free expression and therefore subject to stricter judicial scrutiny. Within content-based regulation, the most disfavored category is “viewpoint discrimination” — regulation that permits one side of an issue to be expressed while singling out the other side for suppression (or vice versa). The government’s act of sorting speech into the permitted and the forbidden strikes at the foundation of a free society.
With that distinction in hand, the earlier point can be restated more precisely. Suppose that at some future date, the law is amended to remove the plainly subjective phrase “in a manner likely to cause marked displeasure or disgust” and replace it with something more objective and ostensibly content-neutral — say, simply “a person who burns or destroys a national flag.” Even then, as long as the law retains its stated purpose in the supplementary provisions — “protecting the sentiment of citizens who cherish the national flag” — the Johnson–Eichman logic dictates that the law remains, in substance, content-based regulation. That is because the sentiment being protected is wounded precisely because flag destruction conveys the message of a negative evaluation of the state or the current government; protecting that sentiment and suppressing that message are two sides of the same coin. No amount of drafting neutrality can sever the law’s content-based character so long as its stated purpose is “protecting national sentiment.”
In other words, the vagueness the statement identifies in Article 2(2) — the “manner likely to cause marked displeasure or disgust” standard — is a technical flaw, fixable through better drafting. The viewpoint-discrimination problem, by contrast, is rooted in the purpose clause itself, and cannot be cured by any amount of drafting technique. It is a structural defect built into the law’s skeleton.
Why “Therefore Unconstitutional” Isn’t So Simple
There is a point worth pausing on here. In the United States, there is an accumulated line of precedent — Street v. New York (1969), Spence (1974), Johnson (1989), Eichman (1990) — supporting the proposition that flag desecration is symbolic speech, and that accumulation itself functions as a robust body of doctrine.
Japan, by contrast, has no precedent in which its Supreme Court has directly addressed this question. Nor is that the only gap: Japan’s Supreme Court has repeatedly been criticized for applying, in practice, a far less rigorous standard of review to restrictions on intellectual and expressive freedom than scholarly doctrine calls for. The Sarufutsu case, which upheld a blanket ban on political activity by civil servants; the customs inspection case, which upheld import restrictions on expressive materials; the Osaka City Outdoor Advertising Ordinance case, which upheld restrictions on outdoor advertising expression under a loose balancing test — in none of these did the Court apply anything close to the rigor that the textbook “double standard” theory calls for.
In short: as a matter of doctrine, applying the Johnson–Eichman logic to Japanese law makes unconstitutionality a relatively easy conclusion to reach. But there is no guarantee that if this law were actually reviewed by Japan’s Supreme Court, it would reach the same conclusion the American cases did.
The Statement’s Real Weakness
The statement’s deepest weakness is that it never does the work of bridging this gap — between “the conclusion doctrine points toward” and “how Japan’s Supreme Court actually tends to review these cases” — the work of normatively arguing why Japanese courts, too, should adopt this more rigorous logic. Instead, it simply states the conclusion as if it were already settled fact. The claim that such acts are “referred to as symbolic expression, and understood to fall within Article 21(1) protection” is a position that can, in fact, be well supported by the accumulated comparative-law record. But by writing it in the passive voice — as though it were already established, majority doctrine — the statement’s single most important claim ends up looking like its single least substantiated one.
Filling in the Missing Step, in My Own Words
So what would fill that gap? There are two threads worth pulling on.
The first is the text of Article 21(1) itself. What it protects is “freedom of assembly and association as well as speech, press and all other forms of expression.” Having named “speech” and “press” as linguistic examples, the provision then adds the unqualified phrase “all other” forms. Reading this to limit protection to linguistic expression arguably runs against the text itself, rather than being compelled by it. If flag burning functions, for an observer, as a vehicle for a clear political message, then the burden of explaining why it should be excluded from “all… expression” properly falls on whoever seeks to regulate it — not on the person claiming protection.
The second thread is the analytical framework the American cases have built up since Spence v. Washington: whether a given act is protected as expression turns on two concrete questions — did the actor intend to convey a particular message, and, given the surrounding circumstances, was it likely that an observer would understand that message. This framework isn’t a peculiarity of U.S. constitutional law; it’s a general method for determining the outer boundary of “expression,” applicable in principle to any constitutional order that protects freedom of expression. Japan’s Supreme Court is under no obligation to adopt this framework wholesale — but absent some alternative test of its own, there is likewise no principled basis for retreating to the narrow position that “conduct without language is not expression.”
Building on this: the fact that Japan’s Supreme Court has historically applied lenient review to restrictions on intellectual freedom is not, by itself, a reason to extend that same leniency to viewpoint discrimination — the single most dangerous category of restriction that exists. Sarufutsu and the customs inspection case involved content-neutral regulation, or regulation justified by a separate interest (border control). A law that specifically targets and punishes a particular political viewpoint — criticism of the current government, or a particular reading of history — belongs to a different regulatory category altogether. The burden of explaining why the leniency of those earlier precedents should simply carry over here falls on whoever argues the law is constitutional — not the other way around.
What the statement lacked was precisely this: connecting these two threads in its own words, and explaining — by reference to both the structure of Japan’s constitutional text and the way Japan’s own Supreme Court has actually differentiated its standard of review by regulatory category — why this case cannot simply be absorbed into the Court’s familiar pattern of deference.
The weight of 201 signatures is not a substitute for precision of argument. For the statement to function as an actual weapon in litigation, this is the step that needed to be filled in — grounded in the accumulated comparative-law record and in the way Japan’s own Supreme Court has itself distinguished between categories of precedent — in the scholars’ own words.