A British karate magazine prints a confident line: because there were only sixteen Arahants, the whole "eighteen Arahants" tradition must be a Chinese forgery. It reads like scholarship. It is the kind of sentence that ends arguments in a dojo changing-room. And it is, Dukes says, simply wrong — wrong in a way that exposes how a martial subculture polices its own history with a half-read book. The writer had opened the Theravada scriptures, found sixteen Arahants there, and concluded that any larger number must be invention. He had never opened the Chinese histories, where eighteen — and more — show up in the earliest records, and where court documents casually list "eighteen Nobles" because round-numbered groupings were just how ancient China counted important things.1
This page is about a nomenclature dispute and the bad reasoning underneath it. The "18 Arahants of China" are a named set — a roster of eighteen perfected disciples, each with a Chinese phonetic name, that anchors one of the core nata forms of the Lion-art tradition (the "18 Subduings").2 A current in Western martial-arts writing dismisses this set as a Chinese fabrication on the grounds that the canonical Arahant count is sixteen. Dukes's response is three-layered. First, he shows the dismissal rests on reading one canon (Theravada) and ignoring another (the Chinese Buddhist histories). Second, he marshals positive evidence that the eighteen-count is ancient: a 217 B.C. account of Arahants arriving in China, and Han-dynasty court records listing "18 Nobles." Third, he points to Tucci's observation that the same dismissal was tried centuries ago against the Tibetan eighteen-Arahant tradition, by Theravada partisans trying to push the Dhyana school out of Tibet — so the "forgery" accusation is itself an old sectarian weapon, not a fresh discovery.3
The forgery charge has a tidy shape, and the tidiness is the trap. It runs: canonical number is sixteen → this tradition says eighteen → eighteen ≠ sixteen → therefore eighteen is forged. Each step looks rigorous. The hidden flaw sits at step one. "Canonical" smuggles in which canon — and the writer silently means the Theravada scriptures, where sixteen is indeed the common count.4 But the eighteen-tradition was never claiming to be Theravada. It belongs to the Chinese Buddhist world, where, Dukes says flatly, "the 18 Arahants (and more) appear in the earliest Chinese histories of Buddhism."5 Judging a Chinese count by a Theravada yardstick is like declaring a French word misspelled because it breaks English rules.
Dukes then sets down the corroborating texture. In the chronology, under 217 B.C., "Li Fong and 17 other Arahants arrive in China. They are imprisoned but by means of a 'miracle' escape."6 Seventeen companions plus Li Fong is eighteen — the number is load-bearing in the oldest entry. He cites Fa Yuan Chu Lin for the fuller account. And he reaches for the most deflating evidence of all: the ordinary court records of the Han Shu and Shi Chi, around A.D. 90, "list groups of '18 Nobles.'" Eighteen was simply a stock grouping in ancient China — "such numerical nomenclatures were common."7 The number is not a Buddhist invention smuggled in; it is a Chinese counting habit the tradition sat comfortably inside.
This page arms the Dukes corpus against a specific kind of debunking. Bodhidharma: The Revisionist Biography makes the general case that "the sources are late and contradictory" is a fact about archives, not a proof of fabrication; this page is that argument applied to a single number, with the counter-evidence laid out. It also pairs tightly with The Lion Becomes a Lap-Dog — both pages catch a martial subculture mishandling its own names, one by reading a word's worn meaning as original, the other by reading one canon's count as universal. And it underwrites Chuan Fa Transmission Lineage, since the "18 Subduings" nata is one of the named forms the transmission claim carries forward.
Dukes quotes the offending opinion almost verbatim and dissects its self-incrimination: it is "blissfully ignorant of Chinese history" and "obviously develops from reading the Theravada scriptures."8 Watch what he does with it. He does not just say the writer is wrong; he reconstructs how the writer got there, and the reconstruction is the indictment. The error is not a slip of fact — it is a procedure. Read one scripture. Treat its number as the only number. Meet a different number. Cry forgery.
The deeper point is who gets hurt by the procedure. The forgery charge does real work: it lets a publication dismiss an entire transmission lineage with a single confident sentence, sparing the writer the labour of learning the Chinese sources. And Dukes's Tucci card turns the tables completely. The same accusation, Tucci notes, was deployed centuries ago against the Tibetan eighteen-Arahant tradition — by Theravada proponents specifically trying "to oust the Dhyana School teaching from Tibet."9 So the modern karate magazine is unknowingly repeating an ancient sectarian attack. "It seems the misunderstanding is a very old one," Dukes writes — and the dryness is the knife. The amateur debunker thinks he is doing critical history; he is actually rerunning a thousand-year-old turf war he has never heard of.
A practitioner is mid-draft on a history of his style and reaches the "18 Subduings" form. A nagging memory surfaces — didn't someone prove the eighteen-Arahant thing was fake? He almost cuts the passage. Then he does the thing the magazine writer never did: he checks which number lives in which canon. Sixteen, he finds, is the Theravada count. His tradition is not Theravada. He pulls the Chinese chronology and lands on the 217 B.C. line — Li Fong and seventeen others, eighteen in total, arriving and escaping by miracle. He pulls the Han court records and finds "18 Nobles" listed as a routine grouping with no Buddhist content at all. The "forgery" dissolves into a counting custom. He writes the passage back in, but now with a footnote that names the canon-mismatch directly — because the next reader will have heard the same rumour, and the only durable defence against a tidy false argument is to show exactly where its first step cheats.
The base is Dukes's Notes 38a–38c. The roster of eighteen Chinese Arahant names (Pin to lo pa lo to, Chia no chia fa ts'o, and so on) is reproduced in full, and the canon-mismatch point — sixteen in Theravada, eighteen-plus in early Chinese histories — is the kind of claim a Buddhologist can check against the actual texts; it does not depend on Kongoryuji's private manuscripts.10 The Han Shu / Shi Chi "18 Nobles" datum and the Tucci citation (his Minor Buddhist Texts) are likewise checkable against published scholarship. That makes the defensive core of the page relatively solid.
The weak link is the offensive evidence: the 217 B.C. "Li Fong and 17 other Arahants" chronology entry. It is single-sourced to Fa Yuan Chu Lin via Dukes's own chronology, contains a "miracle" escape, and dates a foreign-monk arrival in China more than two centuries before the conventionally accepted introduction of Buddhism. [PLAUSIBLE — needs corroboration] is generous; on its own this is a [SINGLE SOURCE] claim carrying a legendary element, and it should anchor nothing by itself — its job here is corroborative texture for the number eighteen, not proof of a 3rd-century-B.C. Buddhist mission. The eighteen-count survives without it; the early-arrival story does not survive on it alone.
Open: Dukes uses the 217 B.C. account to show the eighteen-count is ancient in China. But the same account, taken as history rather than as evidence-of-a-number, would push Buddhism's arrival in China dramatically earlier than the standard 1st-century-A.D. dating. Dukes does not flag that consequence. The page preserves the tension: the entry is being asked to do light work (a number was in circulation) while quietly implying heavy work (Buddhism reached China in the Qin) that the source cannot bear.
Dukes and the vault's critical historians line up on the same side of this fight and split on the next one over. Against the karate-magazine debunker, Samuel and White would be Dukes's allies without reservation — judging a Chinese tradition by a Theravada count is exactly the canon-confusion they spend careers correcting, and the recognition that "forgery!" can be a sectarian weapon rather than a finding is pure critical-historical instinct. They converge with Dukes completely on the demolition of the bad argument. The split opens on the 217 B.C. chronology. Samuel would accept the eighteen-count's antiquity and immediately quarantine the Li Fong entry as a legendary single-source datum that cannot date a mission; Dukes lets it stand in his chronology with lighter scepticism. So the page records a genuine asymmetry in Dukes himself: ruthless about other people's credulity (the magazine), more permissive about his own tradition's oldest stories. That asymmetry is the live edge of reading him — sharp blade, occasionally pointed the wrong way.
Manas Self-Looping and Pudgalabhava — the forgery charge is a cognitive loop made social. The Manas page describes a mind that takes its own prior output as input and mistakes the feedback for solid ground. The magazine writer does the institutional version: he takes one canon he has already read as his fixed frame, then processes all new data (an eighteen-count) only as confirmation or threat to that frame — never as a signal that the frame itself is partial. The error is not lack of information; it is a closed loop treating its own contents as the whole world. Neither domain alone gives the insight: the consciousness theory supplies the mechanism (auto-reference mistaking itself for reality), and this history supplies a clean public specimen of that mechanism producing a confident false claim — and a cure, which is structurally identical to the meditative one. You break the loop by feeding it genuinely external data (the Chinese canon, the Han court records), the way the meditator breaks the self-loop by feeding attention non-self objects. Debunking and ego-dissolution turn out to share a method: starve the closed loop with something it did not generate.
Bodhidharma: The Revisionist Biography — the biography page argues that late, contradictory sources are what real contested institutions produce, not proof of nonexistence; this page argues the mirror-image case, that a number's absence from one canon is not proof of forgery. Put together they form a single methodological rule the history domain needs in both directions: the gaps and mismatches in a tradition's records are evidence about the records, not automatically evidence against the tradition. The biography defends a person against dissolution-by-late-sources; this page defends a number against forgery-by-wrong-canon. The combined insight neither case produces alone: amateur "critical history" tends to run one move — find a discrepancy, declare the thing fake — and that single move is wrong by the same logic whether the target is a man or a number. Real source criticism has to ask what the discrepancy is evidence of, and the answer is rarely "fabrication."
Sharpest implication: a confident debunking can be a sectarian attack with the sect filed off. The karate writer thought he was doing history; he was rerunning a Tibetan turf war he had never heard of. The lesson generalizes: when a "this is fake" claim feels suspiciously clean, check whether someone, somewhere, once needed it to be fake.
Generative questions: