Halfway through Faxian's chronicle, at Ch. XXVI line 1216, Faxian records the death of Ananda — Buddha's beloved cousin, the disciple whose memory preserved most of the Sutras. The text describes Ananda's miraculous mid-river-Samadhi-fire body-division: he sets his body alight via meditative absorption in the middle of the Ganges-confluence-of-five-rivers, divides his body in two, leaves half on each bank, so that king Ajatasatru (pursuing from Magadha) and the Lichchhavis (coming from Vaisali) each get half a body for relic-veneration.
Legge writes a footnote. The footnote runs five paragraphs. It surveys the Buddhist-scholarly literature on Samadhi (Eitel, Hardy, Turnour, Burnouf, Edkins). It quotes Eitel at length on Samadhi as "the highest pitch of abstract, ecstatic meditation; a state of absolute indifference to all influences from within or without; a state of torpor of both the material and spiritual forces of vitality; a sort of terrestrial nirvana, consistently culminating in total destruction of life." Then Legge says:
"All this is simply 'a darkening of counsel by words without knowledge.' Some facts concerning the death of Ananda are hidden beneath the darkness of the phraseology, which it is impossible for us to ascertain."1
He continues: "The account of Ananda's death in Nien-ch'ang's 'History of Buddha and the Patriarchs' is much more extravagant. Crowds of men and devas are brought together to witness it. The body is divided into four parts. One is conveyed to the Tushita heaven; a second, to the palace of a certain Naga king; a third is given to Ajatasatru; and the fourth to the Lichchhavis. What it all really means I cannot tell."
Legge has just admitted, in the apparatus of his own translation, that he does not understand what the text in front of him is saying. Not because the words are unclear — he translates them with precision. Because the events the words describe cannot be reconstructed as historical happenings from inside Legge's epistemic frame.
That admission is the discipline. Other 19th-century scholars in Legge's position would either (a) suppress the miraculous-detail because it embarrasses the scholarly project, or (b) embrace the miraculous-detail as devotional content the translator does not question, or (c) explain it away with elaborate symbolic-interpretation that converts the literal-miracle into a metaphor. Legge does none of these. He translates the text accurately. He preserves the scholarly tradition's explanations. He notes that the explanations don't help. He says he cannot tell what the text really means. He moves on.
The footnote is the translator's epistemic-discipline register at its sparest. The discipline is to mark what you don't understand rather than to fake comprehension.
Legge's apparatus contains multiple instances of the same register. Read them as a pattern.
Inter-translator disagreement preservation (Ch. XXV line 1184): Three translators rendered Buddha's last-walk-at-Vaisali phrase differently. Beal: "In this place I have performed the last religious act of my earthly career." Giles: "This is the last place I shall visit." Rémusat: "C'est un lieu ou je reviendrai bien longtemps après ceci." Legge preserves all three, offers his own conjecture ("Perhaps the 'walk' to which Buddha referred had been for meditation"), and lets the reader see the disagreement.2 No suppression of alternatives in favor of the translator's preferred reading. The reader gets the full disagreement.
Christian-missionary-terminology refusal (Ch. XXXII Legge note 3, line 1452): On the Chinese term for hell-realm, Legge writes: "I prefer to retain the Sanskrit term here [naraka], instead of translating the Chinese text by 'Earth's prison,' or 'a prison in the earth;' the name for which has been adopted generally by Christian missionaries in China for gehenna and hell."3 Legge is consciously refusing to use the missionary-default translation choice that would import Christian-hell connotations into the Buddhist material. The discipline: don't let your own tradition's vocabulary distort the translation.
Translator-honesty about gaps:
Each of these is the translator saying: the text references something I cannot locate in the broader scholarly literature. The default 19th-century move would be to invent context or fake confidence. Legge marks the gap.
Alternative-reading acknowledgment:
Self-correction:
The pattern is consistent: mark what you don't know, preserve what others have said even when you disagree, declare your translation-choices and the costs of those choices, refuse default vocabulary that would import alien categories. This is the translator-discipline register. It is distinct from Legge's missionary-apologetic register (see Legge's 1886 Buddhism Population Revisionism), which operates with different stakes and different commitments.
Read the two registers side by side and Legge is operating as two coexisting scholarly identities in the same book.
The missionary-apologetic Legge writes the Introduction part 3 (Buddhism is fifth place), the Ch. II note 3 (Buddhism has no God, no soul, no sacraments), the Christian-comparative parallels (the alms-bowl's widow-mite parallel, the Brahmins-converted-by-lamp-stealing-devas Dagon parallel, the Avalokitesvara Mary parallel). This register is theologically committed, deploys statistical revisionism, and positions Buddhism unfavorably against Christianity.
The translator-discipline Legge writes the footnotes that admit "I cannot tell," preserves inter-translator disagreements, refuses missionary-default translation choices, marks gaps in the scholarly record, declares interpretive uncertainty. This register is craft-disciplined, professionally honest, and refuses to fake comprehension.
Both registers are in the same book. The translation between them is fluid. Legge can pivot from missionary-apologetic to translator-disciplined within a few pages. Sometimes the registers operate in the same footnote — a Christian-comparative observation followed by a translator-honesty admission.
What this tells you: scholars are not unified identities. They operate in multiple registers depending on what specific task they are performing. The same person who wrote the Introduction part 3 wrote the Ch. XXVI Ananda-death footnote. The Introduction is identity-committed; the footnote is craft-honest. The two coexist.
Modern reception of Legge has tended to either (a) honor the translator-craft and elide the missionary-apologetics (treating the translation as if it were neutral), or (b) attack the missionary-apologetics and discount the translation (treating the translation as suspect because of the framing). Both are wrong. Legge is both registers, and the registers are different. The translation is excellent craft; the apologetics is theological commitment. Both have to be assessed on their own terms.
Legge's translator-discipline is structurally identical to what Diarmaid MacCulloch articulates explicitly as the historian's "may-have-discipline" (see May Have Discipline — Degrees of Confidence from the 2026-05-12 ingest). MacCulloch's discipline is to mark the confidence-tier of every claim ("we may say," "we can say," "we know") so the reader can adjust.
Legge, 130 years before MacCulloch, is doing the same thing in different vocabulary. "I have not met with" is Legge's "we cannot say." "I cannot tell" is Legge's "we may say at low confidence" (i.e., we can speculate but the speculation is weak). "Perhaps" is Legge's "we may say at medium confidence." The translator-discipline articulates the same epistemic-tier-marking that MacCulloch's historiographic method articulates.
What this implies: the discipline is cross-temporal and cross-domain. It appears in 19th-century English missionary-scholarship translation-craft AND in 21st-century Anglo-academic historiography AND in 5th-century Chinese Buddhist pilgrim-chronicle writing (see Faxian Witness vs Hearsay Epistemic Discipline). Wherever a scholar is operating in good faith on materials they cannot fully verify, the discipline emerges. The vocabulary varies. The structure is constant.
The discipline is not a 21st-century invention. It is the deep-time form of intellectual honesty in the face of incomplete information. Faxian had it in 405 CE. Legge had it in 1886. MacCulloch articulates it explicitly in the 21st century. The articulation has gotten more explicit over time but the underlying practice is much older.
Take Legge's discipline seriously and notice what it produces in the chronicle.
The reader can recover the original material's epistemic tier. Faxian recorded what he saw. Legge translates it accurately. When Faxian's text describes a miraculous event (the alms-bowl filling for the poor, the Ananda-body-division, the Buddha-shadow at Nagara), Legge translates the description and notes what scholarship can and cannot say about the underlying claim. The reader gets both the description and the epistemic context. The reader can adjust their belief-state appropriately.
The translation's load-bearing decisions are visible. Whenever Legge had to choose between possible renderings, he typically notes the alternatives. The reader can see which decision was made and can assess whether it was the right one. The translation does not hide its own choices.
The scholarly tradition's state of knowledge is preserved. The footnotes are also a snapshot of what 19th-century Buddhist scholarship knew (or thought it knew) about each topic Faxian addresses. A modern reader can use the footnotes as a baseline for the field's evolution. The translation thus serves as both translation and history-of-its-own-discipline.
The honest gaps are flagged. The "I have not met with" notations identify places where the scholarly tradition lacked information in 1886. Some of these gaps have since been filled by later scholarship. The notations let a modern reader identify where to look for updated material.
This is what good translator-discipline produces: a translation that is also a piece of meta-scholarship, marking the field's confidence-tier on each claim it transmits. The Faxian-as-Legge-translates-it text is not just Faxian's voice — it is Faxian's voice in dialogue with 1886 scholarship in dialogue with Legge's specific judgment about what each can and cannot tell.
The primary witness passages are at the line-references given throughout: Ch. XXVI Legge note 4 (line 1216) for the Ananda death; Ch. XXV Legge note 4 (line 1184) for the Buddha's last-walk three-translator disagreement; Ch. XXXII Legge note 3 (line 1452) for the missionary-vocabulary refusal; Ch. XX, XX, XXXII, XXVIII footnotes for the "I have not met with" pattern; Ch. III Legge note 17 (line 380) for the self-correction.
The Preface (lines 101-145) contains Legge's own articulation of his translator-method: "I have endeavoured to eschew those matters [translator-disagreement debates], and have seldom made remarks of a critical nature in defence of renderings of my own."6 This statement is itself an instance of the discipline — Legge declares his methodological choice and its costs upfront.
Cross-validation: 19th-century translation-craft discourse documents Legge as one of the more careful scholar-translators of the period. The specific footnote-discipline patterns are documented in comparative analyses of Beal, Giles, Rémusat, and Legge as Faxian-translators.
The most acute tension: the translator-discipline register and the missionary-apologetic register operate in the same book without explicit acknowledgment of their distinctness. Legge does not declare that he is operating in two registers. The reader has to identify them. A casual reader might encounter the missionary-apologetics in the Introduction and conclude that the whole book is missionary apologetics — missing the translator-discipline operating in the footnotes. Or might encounter the translator-discipline in the footnotes and conclude that the whole book is craft-disciplined — missing the missionary-apologetics in the framing.
The two registers operate at different stakes. The Introduction is directed at the general Anglo-Protestant reader who needs to be reassured that engaging Buddhism does not threaten Christian commitments. The footnotes are directed at the specialist scholar who needs to know what the translator does and doesn't understand. Different audiences, different registers. Legge does not consciously toggle between audiences; the toggle happens implicitly.
A second tension: the translator-discipline is applied selectively. Legge's "I cannot tell" appears in some places and not in others. He does not write "I cannot tell" about the Christian-comparative footnotes — those are presented as if the parallels are obvious and natural. The selective application means the discipline operates as a register he can choose to deploy or not deploy. When the topic is Buddhist mysticism (Samadhi-fire body-division), the discipline appears. When the topic is Christian-comparative parallel, the discipline disappears. The selective application reveals that the discipline is not unconditional methodological commitment but conditional rhetorical choice.
A third tension: the discipline preserves inter-translator disagreements but does not always preserve the original-text tradition's own disagreements. Faxian's chronicle preserves multiple variants from the Corean, Sung, Ming, and Japanese textual streams (line 180). Legge cites some of these but does not always document where his rendering chose one variant over another. The textual-criticism layer is more incomplete than the inter-translator-criticism layer.
Three figures across fifteen centuries are doing structurally the same craft-discipline. Faxian in 405 CE marks each claim as witnessed / heard-on-site / unreachable. James Legge in 1886 marks each translation choice as confident / uncertain / "I cannot tell." Diarmaid MacCulloch in the 2010s articulates the same move explicitly — "we may say," "we can say," "we know." The vocabulary changes across centuries. The discipline does not.
Three more figures from creative-practice handshake into this discipline alongside the historian-craft lineage: May Have Discipline — Degrees of Confidence (MacCulloch) and Manuscript Copying as Voice Training (Wang).
MacCulloch's may-have-discipline articulates degree-of-confidence marking in historical scholarship. Legge is doing structurally the same thing in 1886 translation-craft. They converge on the principle: scholars must mark their confidence-level for the reader. They diverge on register — MacCulloch articulates the principle as principle; Legge embeds the principle in footnote practice. The handshake reveals that the principle is older and more widely practiced than the explicit articulation suggests. MacCulloch's framing is an explicit recovery of what Legge was already doing implicitly in the 19th century, what Faxian was doing implicitly in the 5th century. The practice precedes the articulation by many centuries.
Wang's manuscript-copying-as-voice-training framing is about the developmental function of slow textual reproduction for the writer's own craft. Legge's hand-translation of Faxian (the work involved producing the Corean text typesetting plus the English rendering plus the footnote apparatus over multiple years of revision — Preface line 105: "I wrote out a translation in English for my own satisfaction of nearly half the narrative") is the structural parallel for translation-craft. The manuscript-copying mechanism produces voice; the translation-craft mechanism produces translator-discipline. Both are developmental outcomes of slow, hand-mediated engagement with an original text.
What the convergence reveals — that none of the three pages alone shows — is that slow-craft engagement with original texts produces both voice and epistemic discipline as joint outcomes. The two are not separable. The translator who has spent years moving slowly through a difficult text develops both the voice to render it and the epistemic discipline to mark its uncertainty. The writer who has spent years copying others' manuscripts develops both the voice to write authentically and the discipline to know when their own claims are uncertain. The same slow-engagement mechanism produces both.
The translator-discipline reaches into three adjacent vault domains.
Creative-practice: May Have Discipline — Degrees of Confidence (MacCulloch) — see Author Tensions section above. The handshake reveals the cross-temporal continuity of the discipline.
Creative-practice: Manuscript Copying as Voice Training (Wang) — see Author Tensions section above. The handshake reveals the joint-outcome of slow-craft engagement.
History: Faxian Witness vs Hearsay Epistemic Discipline — Faxian's three-tier epistemic register in the 5th century and Legge's translator-discipline in the 19th century are doing structurally identical work at different stages of the textual transmission chain. Faxian marks what he saw vs what he heard vs what he could not visit. Legge marks what scholarship knows vs what scholarship is uncertain about vs what scholarship cannot reach. The structural parallel: both authors are tier-marking their epistemic claims. They diverge on what they are tier-marking — Faxian on witnessed-vs-heard; Legge on understood-vs-not-understood. The handshake produces an insight: the epistemic-discipline necessary for reliable textual transmission must operate at every stage of the transmission chain. Faxian's discipline buys one stage of reliable transmission (5th century India to his chronicle). Legge's discipline buys another stage (1886 reception of the chronicle into English). The intervening stages (medieval Chinese transmission, Corean recension, Japanese textual stream) each had their own discipline-needs. Where the discipline was maintained, the text survives reliably. Where it was relaxed, the text degrades.
The Sharpest Implication
Take Legge's translator-discipline seriously and the standard model of "translation as transparent transmission" inverts. Translation is not transparent. Every translation involves decisions, and good translation makes those decisions visible. The "transparent translation" ideal (the reader gets the original author's meaning unmediated) is not achievable; the translator is always present in the choices. The third-wire reading: all translation is interpretation, and the question is whether the interpretation is honest about itself. Legge's translation is honest: it tells you what it doesn't understand, it preserves alternatives, it declares its choices. Translations that pretend to be transparent are hiding their interpretation; they are not actually more accurate than honest translations, just less honest about the unavoidable interpretation. The implication for any reader of any translation: ask which discipline-register the translation is operating in. Honest translations mark their uncertainty. Hidden-interpretation translations don't. The latter is more dangerous because the reader cannot adjust for what they cannot see.
Generative Questions