The Words, and What They Mean in Japan

Five words do most of the work in this field, and they do not mean the same thing: massage, bodywork, relaxation, treatment, therapy. In a shop window they look like synonyms with different amounts of polish. They are not. One of them is the name of a nationally licensed occupation in Japan. One of them is an industrial classification the state created in 2013. One of them is a claim about what happened to your body afterwards. A reader who cannot tell them apart cannot evaluate a single sentence anyone writes about touch, including the sentences on this page.

This page is the anchor of the chamber, so it takes the slow route: what each word denotes, where commercial usage blurs them, and then the part that matters most in this country — the licensing statute, what it actually says, and where its edges are genuinely unsettled. It is not legal advice, it does not diagnose anything, and it makes no claim about what any particular business may or may not do. Where something could not be established from a primary source, the prose says so instead of filling the gap.

"Massage" in ordinary English names a manual technique: pressure, gliding, kneading applied to soft tissue. "Bodywork" is broader and vaguer — an umbrella coined in the twentieth-century Western somatic scene to cover manual work, movement, breath and assisted stretch together, precisely because no older word covered all of it. "Relaxation" is not a technique at all; it names an intended state, which is why it can be attached to almost anything. "Treatment" and "therapy" are the two that quietly change register: both imply an intervention aimed at a condition, and therefore imply that a condition has been identified. That is the step where description becomes a claim.

Japanese complicates this in its own way, because several of these arrived as loanwords and then settled into different places from their originals. マッサージ entered the language as the name of a specific European technique and was later written into national law as part of a licensed occupation, so it now carries a legal weight that the English word does not. 施術 sounds clinical but is used across the whole field. ボディケア, リラクゼーション and 整体 are commercial category names rather than defined techniques, and a shop can adopt or drop them without changing what the hands do. The vocabulary is not a ladder from casual to serious. It is four or five different classification systems laid on top of each other.

In Japan, あん摩マッサージ指圧師 is a national qualification. The governing statute is あん摩マツサージ指圧師、はり師、きゆう師等に関する法律, Act No. 217 of 1947, promulgated on 20 December 1947 and commonly abbreviated あはき法. The odd orthography of the title — マツサージ and きゆう with full-size ツ and ゆ — is the statute's own spelling and has never been modernised, which is one reason the official name looks slightly wrong to anyone who has only seen it in a brochure. Article 1 is short: a person other than a physician who intends to make a business of anma, massage or shiatsu, of acupuncture, or of moxibustion must hold the corresponding licence.

Article 2 sets out what the licence costs in time. A candidate must be eligible for university entry, must have spent three years or more at a school approved by the Minister of Education or at a training facility approved by the Minister of Health, Labour and Welfare (or, for the acupuncture and moxibustion qualifications, by a prefectural governor), studying — the statute names them — anatomy, physiology, pathology, hygiene and the other knowledge and skills required, and must then pass the national examination. The statute also constrains what a licensed practitioner may do: Article 4 forbids surgery and the administration or prescription of drugs, and Article 5 bars an あん摩マッサージ指圧師 from working on the site of a dislocation or fracture without a physician's consent. Article 7 restricts advertising for these businesses to a closed list of items and adds that even those may not extend to the practitioner's skill, method or career history.

The name of the Act is itself a small history lesson. As promulgated in 1947 it was called あん摩、はり、きゆう、柔道整復等営業法 — a business law. It became a practitioners' law in 1951. マッサージ and 指圧 entered the title only in 1964, by Act No. 120 of that year; and 柔道整復 was moved out into its own statute, 柔道整復師法 (Act No. 19 of 1970), leaving the title we have now. So the very word マッサージ has been inside the licensing title for a little over sixty years, not since the beginning.

Article 1 is not the whole regime. Article 12 adds a general prohibition: no person, other than as listed in Article 1, may make a business of 医業類似行為 — a category usually rendered "quasi-medical acts" — with a proviso carving out 柔道整復, which has its own statute. Article 12-2 then grandfathers a narrow group: people who had already been practising such acts as a business for three months or more when the Act was promulgated, and who filed the notification required at the time, may continue. Penalties sit in Article 13-7, which as the statute now reads provides a fine of up to ¥500,000 for breach of either Article 1 or Article 12.

What does 医業類似行為 actually cover? The Ministry of Health, Labour and Welfare stated its own reading in a formal answer approved by the Cabinet on 31 May 2019 (内閣参質一九八第六二号): it is an act that is not 医行為 — not something that endangers or risks endangering the body unless performed with a physician's medical judgement and technique — but that nonetheless risks endangering the body unless performed by a person holding a certain qualification. The same answer states that the category as used in this Act includes anma, massage and shiatsu, acupuncture, moxibustion and 柔道整復 performed by licence holders, plus other manual and thermal 療術 acts carrying such risk. In other words, the licensed practices are inside the category rather than outside it — which is exactly why Article 12's phrase "other than as listed in Article 1" has to do so much work.

One detail in the same answer is worth keeping. Asked how many people remain covered by the Article 12-2 grandfather clause, the government replied that it does not know the current number. A statutory category that has been in force since 1947 has an exempted population whose size the state does not track.

This is the contested ground, and it should be reported as contested. On 27 January 1960 the Grand Bench of the Supreme Court decided case 昭和29(あ)2990. Its reasoning, quoted verbatim in a 2020 government administrative-evaluation report, runs: Article 22 of the Constitution guarantees freedom to choose an occupation so far as it does not conflict with the public welfare; the statute prohibits and punishes making a business of 医業類似行為 only because doing so was held to conflict with the public welfare; and it conflicts with the public welfare because such business conduct carries a risk of harm to human health. Therefore — 人の健康に害を及ぼす虞のある業務行為に限局する趣旨と解しなければならない — the prohibition must be read as confined to business conduct that carries such a risk.

Two months later the Ministry told the prefectures what it thought this meant (昭和35年3月30日医発第二四七号の一). Three points from that notice matter. First, the judgment concerned 療術 businesses using manual technique, heat, electricity, light and stimulation — it did not rule on anma, acupuncture, moxibustion or 柔道整復, so unlicensed practice of those four remained punishable on the fact of the practice itself. Second, for the acts the judgment did cover, punishment requires a finding not merely that a person performed treatment as a business but that the treatment carried a risk of harm to health. Third — and this is where the Ministry's reading is broad — if the act, viewed medically, poses even the slightest risk of harm to the body, that is enough to count.

So the question a reader actually wants answered — does the licensing requirement reach manual work that claims no therapeutic effect? — sits in the gap between those two readings, and the gap has not been closed. An older definition often cited in this area, from the Sendai High Court on 29 June 1954 (昭28(う)275) and quoted by Japan's National Consumer Affairs Center, builds purpose into the test: 医業類似行為 is 施術 performed for the purpose of treating disease or of maintaining health. On that reading purpose matters. On the Ministry's reading, medical risk matters. Asked directly in 2019 how the presence or absence of 人の健康に害を及ぼす虞 should be determined, the government answered that it is judged comprehensively from the concrete manner of the act, and that a single general answer is therefore difficult. That is the state declining, in writing, to draw the line in the abstract.

The administrative picture matches the doctrinal one. The 2020 report by the Administrative Evaluation Bureau found that most surveyed public health centres considered themselves to have no supervisory authority over health harm arising in this area, and that causation between a session and a complaint is often too difficult to establish for administrative guidance to proceed; it also noted that at least one prefecture's own operational manual states in terms that 整体 and カイロプラクティック fall outside the Act. None of this tells a reader whether any given service is lawful. It tells them that the question is decided case by case, on the concrete facts, by bodies that do not always agree on who is supposed to decide it. Anyone who needs an answer for their own situation should ask a qualified professional; this page cannot give one.

Meanwhile the relaxation business exists, in large numbers, and the state counts it. In the October 2013 revision of the Japan Standard Industrial Classification, a new detailed category was created: 7893 リラクゼーション業（手技を用いるもの）, defined as establishments that perform 施術 using manual technique in order to relax physical and mental tension. The classification explicitly routes aesthetic businesses to 7892 and establishments run by people who make a business of 医業類似行為 to 835 under the medical and welfare division. A statistical taxonomy does not confer or withhold legal permission; it records that a distinguishable kind of business is there.

The scale is not trivial. The same 2020 administrative report, drawing on the Economic Census, recorded 65,647 establishments in the 療術業 minor category as of June 2016, up 25.0% on 2012, with sales of ¥646.0 billion in 2015, up 26.1% on 2011. In the 2019 government answer, the Ministry of Economy, Trade and Industry is described as holding the portfolio for the development and improvement of industries providing health-related goods and services including リラクゼーション業 — expressly on the premise of compliance with the relevant laws, あはき法 included. That premise is doing a great deal of quiet work, and it is the same premise this essay has just shown to be unsettled at its edges.

The gap shows up in ordinary language before it shows up in litigation. In a 2012 report by the National Consumer Affairs Center, a customer who had a painful experience at a facility asked the operator about credentials and was told that the practitioner held no licence — because this was ボディケア, not マッサージ, and so no licence was needed. Whatever the legal merits, that exchange is a precise picture of how the vocabulary is actually used: the name on the menu is doing the regulatory argument.

The commonest error in this field is to collapse three independent questions into one. Whether a practice is regulated is a question about statutes and licensing bodies. Whether it is evidence-based is a question about trials and their quality. Whether it is traditional is a question about transmission and lineage. A practice can score high on any one of these and low on the other two, and marketing copy routinely borrows the prestige of whichever axis is most flattering.

Chiropractic in Japan is the clean illustration. It is not a licensed occupation here. A 1970 ministry interpretation held that because it aims at adjustment of the spine it is distinguished from anma, massage and shiatsu and is not included in them. A 1991 ministry notice (医事第五八号), reporting a research group's findings, stated plainly that scientific evaluation of its medical effect had not been settled and required further examination — and, in the same document, set out conditions for which it is not appropriate and prohibited a specific high-risk technique. So: not licensed, effect not established, and yet formally addressed by the state on safety grounds. Three answers, three axes, one practice.

On the evidence axis, the most careful available summary for the single best-studied application is sobering. The 2015 Cochrane review of massage for low-back pain (Furlan and colleagues, CD001929) pooled 25 trials with 3,096 participants. It found massage better than inactive controls for pain in the short term, and better than active controls for pain in both short and long term, with no difference for function against active controls. It graded the quality of evidence for every comparison as low or very low, on account of small samples and methodological flaws, and the authors' own summary sentence is that they have very little confidence that massage is an effective treatment for low-back pain. Adverse events were minor; increased pain intensity was the most commonly reported, in between 1.5% and 25% of participants across trials.

Note what that finding is and is not. "Low certainty" is a statement about the state of the research, not a verdict that nothing happens. It means the studies are too small and too loosely controlled to support confident claims either way. A reader who wants to know whether an hour of attentive touch will feel good to them is asking a different question from the one the trials were built to answer, and it is honest to say that the trials were not designed to settle it.

The practices named on this house's Experience pages come with accounts of themselves, and those accounts deserve to be reported as what they are: a tradition's own explanation, not a finding. Nuad Thai was inscribed on UNESCO's Representative List of the Intangible Cultural Heritage of Humanity in 2019. UNESCO's own entry describes it as regarded as part of the art, science and culture of traditional Thai healthcare, and as a manual practice in which the practitioner helps rebalance the recipient's body, energy and structure to address illnesses believed to be caused by obstruction of energy flow along sen — lines understood to cross the body. The careful wording there is UNESCO's, not this page's: "believed to be". A heritage inscription records cultural significance and transmission. It is not a clinical endorsement, and UNESCO does not present it as one.

Hilot is a Filipino manual tradition, and it sits on the regulation axis in a way that has no Japanese equivalent: the Philippines' Technical Education and Skills Development Authority maintains a national qualification, Hilot (Wellness Massage) NC II, with published Training Regulations covering core competencies and a defined training duration. That is occupational regulation in another country, not in Japan, and it says nothing about the legal position of anything here. It does say that "traditional" and "regulated" are separable: a practice can be both, in one jurisdiction and not another, without changing what the hands do.

Anma is the oldest of the names in the Japanese licence, and it is worth noticing how far it has travelled from a technique to a legal term. The statute that licenses it does not define it. Neither the Act nor its subordinate rules set out in concrete terms what anma, massage or shiatsu consist of — a point the National Consumer Affairs Center made in its own report, observing that the content of these practices is not clearly and specifically prescribed in law. So the licensed name and the described technique are held apart: what the state regulates is who may make a business under the name, and what the tradition transmits is how the hands are trained. Those two things are related, but they are not the same object.

Fascia is a real anatomical structure: sheets and layers of connective tissue that wrap and separate muscle and organ. The word became commercially irresistible around the same time it became scientifically interesting, and the commercial literature now runs a long way ahead of what has been shown. Claims that a pair of hands "releases" fascia, in the sense of permanently deforming it, are the part that has not held up well.

The most instructive paper here is not a hostile critique but one co-authored by Robert Schleip, among the field's most prominent researchers. Chaudhry and colleagues (Journal of the American Osteopathic Association, 2008) built a three-dimensional model of fascial deformation under manual force and compared the stresses needed for plastic deformation with what hands actually apply. Their conclusion was that the palpable sensation of tissue release reported by manual practitioners cannot be caused by deformation of firm tissues such as the plantar fascia or fascia lata, though it could arise in much softer tissue such as superficial nasal fascia. That is a finding from inside the field, and it is a useful discipline: the sensation is real and well attested; the mechanism usually offered for it is not the one the mechanics support.

What follows from that is not "fascia work does nothing." It is narrower and more honest: a description of what your hands are doing is not the same as an explanation of why something felt different afterwards, and the second one is much harder to establish than the first. This chamber will use fascia as an anatomical noun and as a description of where pressure is directed. It will not use it as a mechanism of action.

Manual work is not risk-free, and the most authoritative Japanese statements on this are worth knowing in general terms. The 1991 ministry notice on chiropractic listed conditions for which that therapy is not appropriate — among them tumour-related, haemorrhagic and infectious conditions, and a set of spinal conditions where manual adjustment is known to be capable of worsening matters — and singled out thrust manoeuvres involving rapid rotational extension of the cervical spine as carrying a large risk of injury. It also advised that where symptoms worsen, or fail to lessen over repeated sessions, the sessions should stop and the person should be examined promptly at a medical institution. Those cautions were written for a specific therapy, but the general shape of the reasoning — some conditions make manual force a bad idea, and the neck is the highest-stakes region — is worth carrying into any room.

On how often things go wrong, the honest answer is that nobody knows precisely, and the available figures come with a caveat that should be read as carefully as the numbers. Japan's National Consumer Affairs Center reported in August 2012 that PIO-NET had logged 825 consultations over roughly five years from FY2007 in which harm was said to have followed manual 施術 without instruments; of those consultations that reported a severity, about 80% involved a visit to a medical institution, and about 30% of those needed three weeks or more. The 2020 administrative report counted 1,534 accident records concerning 医業類似行為 in the government accident-information databank over FY2014–2017. Both sources state that these records rest on what the consulting person reported, and that the facts and the causal link have not been confirmed. They are a record of complaints, not of established injuries.

Nothing in this section is a statement about your body. It does not assume you have anything, and it is not capable of finding out. If something about your body is on your mind — anything at all, including things that seem too small to mention — the person to ask is a qualified medical professional, who can actually examine you. That is not a formality. It is the only route to an answer that a page of prose structurally cannot provide.

Two of Moonlight's Experiences are, in concept, massage and relaxation Experiences, built as hybrids rather than as one school: shiatsu pressure, hilot and Swedish technique, assisted stretching in the yoga and Thai tradition, slow work through fascia and the deeper layers, with aromatic oil if wanted. The clause attached to both is the same in substance: offered as relaxation and personal care, never as treatment, rehabilitation, correction or diagnosis, with no claim to cure and no guaranteed physical outcome. This chamber exists partly so that a reader can see what that clause is actually doing.

It is a constraint on writing and on conduct, not a line of small print at the bottom. Read as a constraint, it removes an entire register of sentence from everything this house publishes and from what is said in the room. It removes symptom talk, because naming a symptom is the first half of a diagnosis. It removes outcome promises, because the research above does not support them and it would be dishonest to imply otherwise. It removes the vocabulary of correction and realignment, because that vocabulary asserts that something was wrong and has been made right. It removes the language of release in the mechanistic sense, for the reason the fascia section gave. What is left is narrower, and it is the part that can actually be said truthfully: where attention goes, how long it lasts, what the pressure is like, what the room is like, and that you can stop it at any point.

One further thing this house does not do: it does not tell you what is or is not permitted in Japan for anybody, itself included. This page has set out the statute, the ministry's reading of it, the Supreme Court's limiting construction, and the places where those three do not meet cleanly. It has stopped there deliberately. A resource library can describe a landscape; it cannot adjudicate a case, and a reader who wanted it to would be better served by someone qualified to.

Several things a reader might reasonably expect to find here are missing on purpose. There is no statutory definition of anma, massage or shiatsu, because the Act does not contain one; and no binding judicial definition of them is asserted here, because none was located in a primary court record during the work behind this page. There is therefore no answer to the question of whether a particular manual technique is "massage" in the statutory sense — not a withheld answer, an absent one.

Nor is there case law resolving whether the licensing requirement reaches manual work offered without any claim of therapeutic effect. That is the question this chamber would most like to answer, and it could not be answered from the sources consulted. The 1960 Grand Bench judgment limits the Article 12 prohibition to conduct carrying a risk of harm to health; the ministry notice that followed it says the judgment did not decide the position of anma, massage, shiatsu, acupuncture, moxibustion or 柔道整復 at all; the older Sendai High Court formulation puts purpose into the definition; and in 2019 the government declined to give a general answer, saying the judgment is comprehensive and case-specific. Presenting any one of those as the settled position would be picking the convenient reading, so this page has not.

Two smaller gaps, for completeness. Article 19 of the Act permits the ministers, for the time being, to withhold approval from new あん摩マッサージ指圧師 schools and training facilities for students who are not visually impaired, where this is needed so that the livelihoods of visually impaired practitioners do not become severely difficult — that provision was read directly from the statute. Reporting on the litigation that challenged it says the Supreme Court upheld it in February 2022; that judgment itself could not be retrieved from the court's own records here, so the outcome is passed on second-hand and should be checked before being relied on. And on the research side, no attempt has been made to summarise the evidence for shiatsu, hilot or Thai-tradition assisted stretching specifically, because a summary worth printing would require a literature search this page has not done. Absence of a summary here is not a verdict about those practices.