Sources & Further Reading

This is the research layer behind the public pages: where the claims come from, and what each kind of source is allowed to support.

Almost nobody in this field publishes one, and the reason is not modesty. A sources page is a commitment: it lets a reader go and check, which means it lets a reader find you wrong. Most of the writing about Tantra, about touch and about Japanese women’s lives is published without any way of establishing where a particular sentence came from — and a sentence with no provenance is not a weak claim, it is an unfalsifiable one, which is the whole of its commercial appeal.

So what follows is specific enough to be used against this library, which is the only kind worth publishing.

For the history, this library relies on university presses, reference works and specialist scholarship rather than wellness writing: the Oxford Research Encyclopedia and Britannica for orientation, and monographs by David Gordon White, Hugh Urban and André Padoux for the contested detail. Where scholars disagree, the essays name who holds which view.

The preference for that kind of source is not snobbery about where a book was printed. It is that each of those has a named author with a professional reputation exposed to the claim, a citation trail you can follow backwards, and an editorial process that at least attempted to catch an error. None of that makes a scholar right. It makes the scholar checkable, which is a different and more useful property.

A hazard specific to this subject is worth naming, because it inverts the usual reading instinct. In Tantra writing, confidence and sourcing tend to run in opposite directions. The most assured histories — an unbroken five-thousand-year lineage, a single coherent teaching, a clean line from ancient India to a studio in a European city — are typically the least footnoted, because the confidence is only available to somebody who did not consult the material. The scholarship is full of hedging for the honest reason that the evidence is fragmentary and the traditions were several rather than one.

This is what lets the History essay say we do not know where the evidence runs out, instead of filling the gap with a good story. A good story is exactly what fills that gap everywhere else.

For rest and body awareness, the restrained language on these pages leans on general summaries from the U.S. National Center for Complementary and Integrative Health on relaxation techniques and massage, together with peer-reviewed work on slow-breathing physiology and interoception.

These sources support careful general statements such as slow breathing tends to calm the nervous system. They do not support, and this library does not make, claims about treating any condition.

The limits of that literature are worth knowing too, and stating them is not a way of quietly discounting them. Studies in this area often run on small samples, frequently rely on what people report about themselves, are usually conducted in laboratory conditions that resemble nobody’s evening, and sit inside a publication culture that prints findings more readily than non-findings. So a real effect can be genuine, modest, and considerably less reliable outside the room it was measured in. Anybody citing this literature to promise you an outcome is overreading it — including, if it ever happens, this house.

Behind the essays there is an apparatus, and describing it concretely is more use than any statement of principle.

Across this corpus there are 2,559 evidence-ledger entries. Each one carries four things: the claim, stated precisely enough to be wrong; a source class, saying what kind of thing is being relied on; the citation, with the date it was read; and — the field that does the most work — a stated limit, describing what that source does not license. A ledger entry that only recorded where a fact came from would be a bibliography. The limit is what makes it a control.

The read date matters more than it looks. A survey page changes, an organisation revises a definition, a government series is refielded, a URL dies. Recording when a source was consulted is the difference between a citation you can audit and one you have to trust.

The classes, counted rather than estimated: Moonlight interpretation, 902. Primary text, 846. Cultural signal, 310. Official statistics, 226. Peer-reviewed research, 213. Professional guidance, 34. Clinical model, 28.

The largest class being our own interpretation is the most useful fact on this page, and it would be easy to hide by not counting. It says plainly that most of what this library offers is a reading rather than a finding: an argument about what a film is doing, about what a statistic leaves out, about what a woman is describing when she says she is tired. Those are the essays’ real contribution and they are not evidence, so they are labelled as interpretation and you are entitled to disagree with every one of them without disputing a single fact.

The distinction between the other classes is what stops a category error in either direction. A clinical model is a way of organising observations, not a proof that the observations were right. A professional body’s definition establishes what a term formally counts, and establishes nothing about the health of any particular relationship. Official statistics answer the question the state asked, which is rarely the question you have. A cultural signal — a phrase in wide use, a magazine’s recurring headline — is evidence about what is being said, never evidence that it is true.

A source supporting a general mechanism does not prove a Moonlight-specific outcome, and claims have to stay proportionate to the evidence. When a page says an Experience may do something, that word is doing real work and it is there on purpose.

So here is the decoder, because hedges that a reader cannot decode are just texture. Tends to means a direction observed across a group, which says nothing certain about you on Thursday. Is associated with means two things travel together and the essay is not claiming to know which causes which. May means the mechanism is plausible and the specific outcome has not been established. Often means we have not got a number and are not going to invent one. And we do not know means exactly that — it is the strongest sentence in the library, because it is the only one nobody has a commercial reason to write.

The Glossary applies the same discipline term by term. Every entry is labelled with the lens it is written from — historical, Japanese religious, modern Neo-Tantric, Moonlight’s own, or general well-being — so that a reader always knows which kind of claim she is reading. The same word can be honest under one label and false under another, and the label is how you tell which one is in front of you.

A large share of what this library cites about Japan comes from national surveys, and they need a particular kind of reading — not scepticism, which is cheap, but attention to the question the instrument was built to ask.

Much of the best data on intimacy here was collected in service of fertility policy. That is not a scandal; a state facing a falling birth rate has a duty to ask. But an instrument built to find out whether married people are having sex will see exactly that, and will be structurally incapable of seeing whether anybody wanted anything — and it will not sample the unmarried, who are a growing share of the adults this library is written for. The mission essay on the sexlessness count works that argument through in full.

The practice that follows is simple enough to use on anybody’s statistics, including ours. Ask who commissioned it and what they needed to find out. Ask who was sampled and who therefore does not appear. Ask whether the answer was self-reported, and about what — people under-report what carries shame and over-report what carries credit, reliably and in both directions. A number that survives those three questions is worth citing. A number that does not is still worth citing, with the limits said out loud, which is what the ledger’s limit field is for.

The chamber’s essay on mikkyō sends readers here for the wider list, so here it is — short on purpose, and annotated, because a bare list of titles tells you nothing about which question a book answers.

For the history, in the order a curious reader would actually want them. André Padoux, The Hindu Tantric World: An Overview, is the shortest responsible way in — a scholar of the field summarising it without a thesis to sell. David Gordon White, Kiss of the Yogini, is the book to read on what the sexual material in the South Asian sources actually was and was for; it is unsentimental and will disappoint anybody hoping for continuity with a modern workshop. White’s edited anthology Tantra in Practice puts translated primary material in front of you, which is the fastest cure for any single confident summary, this library’s included.

Hugh Urban, Tantra: Sex, Secrecy, Politics, and Power in the Study of Religion, is the one to read if the question that actually interests you is how the Western idea of Tantra was manufactured — which, for most readers arriving from a wellness page, is the more useful question. For the Japanese thread, Ryūichi Abé’s The Weaving of Mantra, on Kūkai and the introduction of esoteric Buddhism, is the standard work, and the Japanese Journal of Religious Studies is where the contested questions are argued out by people who read the sources in the original.

For the body, two directions. Emily Nagoski’s Come As You Are is the accessible route into responsive desire and the dual control model, and Come Together applies the same material to long relationships; both are popular science written by somebody who cites her sources, which is rarer in this aisle than it should be. Masters and Johnson’s Human Sexual Inadequacy is worth knowing as the origin of sensate focus rather than as current practice — it is a historical document, and reading it that way is the point. And for anything about relaxation physiology, the U.S. National Center for Complementary and Integrative Health summaries are duller than any book on this list and considerably more honest about what is not known.

One omission is deliberate and should be explained. There is no list here of popular Tantra books, and not out of snobbery — it is that in this particular genre the reliable signal is whether a book tells you where it got something, and most of the popular titles do not, which leaves a reader no way to separate the careful ones from the confident ones. If you already have one you love, the useful exercise is not to abandon it. It is to pick its three strongest claims and try to find out where they came from.

Three honest gaps, because a methods page that listed only its strengths would be advertising.

The first is this house’s own room. Anything said about what happens in a Moonlight Experience — that a person tends to settle in the second hour, that a reaction can surface two days later — is the account of a practice, given by the person who runs it. It is not research, it has no control group, and the person reporting it has an interest in the answer. It is marked as interpretation for that reason, and it should be read the way you would read any practitioner describing their own work: worth hearing, and not evidence.

The second is that the Japanese-language and English-language literatures on several of these subjects do not agree, and are not translations of each other. They ask different questions, cite different work, and sometimes describe the same phenomenon with terms that do not map. Where this library writes both halves of a page, a choice has been made about which framing travels — and a translated term always carries an argument inside it. The Japanese is written rather than machine-converted for exactly this reason, and it is still a choice.

The third is the oldest material. For the early centuries the evidence is genuinely thin, and no amount of careful citation manufactures a fact that the surviving texts do not contain. Where that is the situation, the pages say so.

Every essay names who holds which view where scholars disagree. Every glossary entry carries the lens it was written from. Every cited source carries the date it was read. Nothing above requires you to take a sentence on trust, and the point of publishing all of it is that you should not have to.

If you find something wrong — a misread study, a statistic quoted past what it supports, a date that does not survive checking, a Japanese sentence that carries a claim its English pair does not — that is worth writing in about, and it will be corrected rather than defended. A library that publishes its sources and then argues with anybody who uses them has published an ornament.

The short version of everything above: a claim is only as good as what it is allowed to support, and most of the damage done by writing in this field is done by true statements used past their range. A real study about slow breathing becomes a promise about your anxiety. A real survey about married couples becomes a fact about you. A genuine old tradition becomes a guarantee about an evening.

So the ledger, the classes, the limit field and the hedges all exist to do one unglamorous thing: keep each sentence inside its range. It makes for more cautious prose than this field is used to. It also means that when a page here does say something plainly, you can find out where it came from — and that is the only reason a reader should believe any of it.