A librarian stamps a manuscript "authentic." A churchgoer hears a sermon called "authentic." They mean opposite things. The librarian means the name on the spine is the name of the person who actually wrote it — a fact about paperwork. The churchgoer means the words are true, divinely backed, not to be questioned — a fact about reality. Dukes spends a footnote pointing out that when Buddhists call a scripture "authentic," they mean the librarian's thing, never the churchgoer's. The whole quarrel about which sutras are "real" turns out to be an argument about correct filing, not about cosmic truth.1
This is Dukes's account of how Buddhism decides what counts as scripture — and the surprising news that the decision is mostly clerical. Across its history, many different canons of teaching were drawn up by various sects, and in most cases this was done "simply to accredit works to the correct authors — which is what is literally meant by 'authentic.'"2 The word does not certify that a teaching is right or wrong. It certifies attribution. A sutra is "authentic" if the tradition agrees the Buddha (or a recognized teacher) is the one who said it — the way a painting is "an authentic Rembrandt" without that telling you whether the painting is any good.
Behind this sits a test that has nothing to do with attribution at all. Dukes points back to the Kalama Sutta, where Shakyamuni told followers to accept no written word, saying, or tradition simply because it is so — but instead to test a teaching against their own personal experience.3 What survives the test gets kept. What can't be understood gets replaced by what can.4 So there are two separate questions running in parallel: who said it (authenticity, a matter of record) and does it work in your own practice (realization, a matter of experience). The first is about the label. The second is about the medicine inside the bottle.
The third leg is the door this leaves open. It was commonly held that "anything which was well-spoken was the word of the Buddha."5 A teaching does not need a pedigree if it is true and useful — being well-spoken is itself the credential. This is why Buddhism tolerates "newer" scriptures with an ease that bewilders anyone raised on Semitic ideas of heresy.6 There is no master sealed list that closed centuries ago and damns everything after it. There is a living conversation that keeps absorbing whatever rings true.
Walk the logic and the missing piece becomes obvious. Heresy needs a fixed, closed, divinely guaranteed canon to deviate from — a line in the sand drawn by an authority, with eternal stakes for crossing it. Buddhism never drew that line in that way. Each sect "developed and used its own canon, and this altered over the years to accord with the various practices which grew current within the sect itself."7 The canon is downstream of practice, not upstream commanding it. When the practice changes, the canon quietly follows. A list that bends to fit what practitioners actually do cannot also function as a tripwire for damnation.
Dukes is careful that this tolerance is not gullibility. "To those who know, the Sutra's teachings that have no place in Buddhism are easily discernible, and history has shown over and over again that Buddhists are seldom hoodwinked."8 The filter is not an official stamp; it is the trained eye of practitioners who can taste whether a teaching belongs. Some works were no doubt produced exclusively within China, and "this does not mean that they were 'wrong' or had no value."9 Origin does not decide worth. The teaching's effect on the practitioner decides worth — which loops straight back to the Kalama test.
He grounds the clerical reading in scholarship, not just assertion. He cites Buswell's Chinese Buddhist Apocrypha, which he says shows "the sometimes ad hoc, or political reasons, for the classification of a Sutra as official or not," and which suggests many texts in the Taisho Tripitaka collection are wrongly attributed.10 The official/unofficial split, in other words, was often decided by committee politics and convenience — exactly what you'd expect if "authentic" means correctly filed rather than true. The mess in the catalog is not a scandal. It is the predictable residue of a tradition that never needed the catalog to be sacred.
This page is a load-bearing epistemics page for the whole Dukes corpus, and it hands several other pages a tool they otherwise lack. The vault's running quarantine flags around Dukes — the Bodhidharma Revisionist Biography anachronisms, the Hsieh Mai Lun attribution question — can now be read in Dukes's own native frame: he would not be troubled by a disputed attribution, because for him attribution is the librarian's problem, not the truth's. That does not dissolve the vault's critical-historical concern; it sharpens the disagreement into something precise. The vault asks did this person really write this. Dukes mostly asks does this teaching work. Naming that gap is more useful than papering over it.
It is the sibling and complement to Forgery Has No Meaning Doctrine. That page argues that within Buddhism the concept of forgery barely applies; this page supplies the underlying machinery — if "authentic" means correctly attributed and "well-spoken is the word of the Buddha," then a "forged" sutra that teaches truly is simply scripture under a different name. Read together they form one position: the value of a teaching floats free of its paperwork.
Look closely at how Dukes deploys Buswell, because the move is quietly recursive.10 He is making a claim about Buddhism — that "authentic" is a clerical category — and he proves it not with doctrine but with a modern academic study of how the catalogs were actually built. Buswell's finding is that classification happened for "ad hoc, or political reasons," and that the canonical Taisho collection contains misattributions. Dukes treats this not as an embarrassment to be hidden but as confirmation of his thesis. Of course the canon is politically assembled and full of wrong names — that is what a canon is once you stop expecting it to be a register of cosmic truth.
The self-demonstration is this: Dukes is using a piece of well-spoken modern scholarship to settle a point about scripture, exactly as his own principle prescribes. The Buswell book has no place in any traditional canon. By Semitic standards it is the most "external" of sources — a 1990 university press volume. But it is well-spoken and it survives the test of experience against the texts, so Dukes admits it as authoritative on the question. He is practicing the doctrine in the act of describing it. The footnote performs the very tolerance it is explaining.
You're holding a sutra someone has just told you is "fake" — composed in China, not India, wrongly stamped with the Buddha's name. The old reflex tightens your chest: then it's worthless, throw it out. Notice that reflex. It is borrowed from a different religion's nervous system, the one with heresy in it. Set the attribution question down on the table to your left — it is a real question, but it is the librarian's, and it can wait. Now read the thing on its own terms. Does this passage describe your mind as you have actually found it in practice? Try the instruction. Sit with it for a week. If it opens something — if the medicine works — the bottle's label was never the point. If it teaches something that plainly has no place in the path, you'll feel that too, the way a trained tongue catches a wrong note. Either way you decided by tasting, not by checking the seal. The catalog can argue about the name afterward.
The source is a single dense footnote (Note 87) plus its bibliography.1 It is Dukes the practitioner-scholar laying out a position, and he leans on solid external scholarship to do it — Buswell, Sangharakshita, Mizuno, Conze — which is more corroboration than many of his lineage claims carry. The Kalama Sutta is genuinely canonical and the "well-spoken is the word of the Buddha" principle (subhāṣita) is a real and widely attested attitude, so the core claim is on firm ground.
One tension worth preserving: the position is more comfortable for Dukes's own enterprise than he acknowledges. A practitioner-scholar whose decisive lineage sources are his temple's privately-held manuscripts has an obvious interest in a doctrine that says attribution is secondary to whether a teaching works. The doctrine is real and ancient; it is also convenient for him. Both things are true and the second does not cancel the first. [PLAUSIBLE — needs corroboration] for the strongest version of "Buddhists are seldom hoodwinked," which is an assertion, not a demonstration.
Open: where exactly is the line between "well-spoken, therefore the word of the Buddha" and "has no place in Buddhism, easily discernible"? Dukes locates the line in the trained eye of "those who know," but never specifies the criteria — which is either a frank admission that the judgment is tacit and earned, or a convenient way to keep the gatekeeping inside the lineage.
Dukes against the vault's critical-historical corpus is the live friction here. Samuel and White spend their careers establishing when and by whom a text was actually composed — attribution as the central scholarly act. Dukes, in this footnote, all but says that question, while legitimate, is not the important one. He is not anti-historical; he cites the historians approvingly. But he reframes their findings: where they read a misattribution as an error to be corrected, he reads it as evidence that attribution was always clerical and political, never the seat of a teaching's value. The two camps agree on the facts of the catalog and disagree on what those facts mean. That is the cleanest possible statement of the vault's standing Dukes-versus-Samuel tension — not a clash over data, but over which question is the real one. Preserved, not resolved. Where Dukes converges with the historians: both insist the canon is a human, contingent, politically-assembled artifact. Neither treats it as dropped from the sky. They part only on whether that contingency is a problem.
Source Credibility vs Content Quality — Knowledge-work and research practice runs on a heuristic Dukes would recognize and partly reject: judge a claim by the credibility of its source. Provenance is a cheap proxy for truth — cite the peer-reviewed journal, distrust the anonymous blog. Dukes's canon-doctrine is the same proxy run in reverse and then abandoned: provenance ("authentic" = correctly attributed) is explicitly decoupled from value ("well-spoken is the word of the Buddha"). The insight neither domain produces alone: a credibility-by-source heuristic is an efficiency hack for when you cannot test the content yourself, and it silently fails the moment you can. Buddhism's Kalama test is a domain where the practitioner can run the experiment directly on their own mind — so the source-credibility shortcut becomes not just unnecessary but a category error. The handshake tells you when to use provenance (you can't verify) and when to drop it (you can). The research worker who treats every claim as source-rankable is committing the same mistake as the Buddhist who treats canonicity as truth — outsourcing a judgment they were equipped to make firsthand.
Bodhidharma Revisionist Biography — The vault flags Dukes for having Bodhidharma quote a sutra composed after his death — a clean anachronism that, by ordinary historical standards, discredits the attribution. Read this page back onto that flag and a structural tension surfaces that neither page resolves alone. Dukes's own doctrine predicts and even licenses exactly this kind of "anachronism": if well-spoken teaching is the word of the Buddha regardless of who literally voiced it, then putting a later, truer formulation into Bodhidharma's mouth is not fraud within his frame — it is attribution-as-honoring, the teaching credited to the lineage-figure who embodies it. The insight: the vault and Dukes are not making the same kind of claim and grading it differently; they are playing two different games with the word "true." History asks whether the sentence was uttered. Dukes asks whether the sentence is well-spoken. The anachronism is a scandal in one game and a non-event in the other — and seeing that is more honest than pretending either game is simply right.
Sharpest implication: a tradition with no concept of heresy cannot be refuted by attribution scholarship — proving the Buddha didn't write something leaves its claim to be "the word of the Buddha" entirely intact, because that phrase never meant what the critic thinks it means.
Generative questions: