Terminology note: Perdue’s Course is a Geluk bsdus grwa textbook and uses several English renderings that are variants in our
raw/glossary.tsv, not the preferred terms. In this page (and in all wiki pages enriched from it) we use the preferred renderings and parenthesise Perdue’s variant with the Tibetan at first occurrence. The recurring substitutions are: investigative thought (Perdue: “correctly assuming consciousness”, ཡིད་དཔྱོད་) · direct perception (Perdue: “direct perceiver”, མངོན་སུམ་) · inference (Perdue: “inferential cognizer”, རྗེས་དཔག་) · conceptual cognition (Perdue: “thought / conceptual consciousness”, རྟོག་པ་) · wrong cognition (Perdue: “wrong consciousness”, ལོག་ཤེས་) · doubt (Perdue: “uncertain consciousness”, ཐེ་ཚོམ་) · conceptual generality (Perdue: “meaning generality”, དོན་སྤྱི་) · ascertaining cognition (Perdue: “determinative knower”, ངེས་ཤེས་) · self-awareness (Perdue: “self-knower” / “self-knowing direct perceiver”, རང་རིག་) · valid cognition (Perdue: “valid cognizer”, ཚད་མ་).
Main argument
Perdue’s Course is a stand-alone English-language pedagogical introduction to the Geluk monastic debate tradition, distilled from the bsdus grwa (Collected Topics, ཚད་མའི་བསྡུས་གྲྭ་) literature on valid cognition (ཚད་མ་). It is the practical-classroom companion to his earlier Debate in Tibetan Buddhism (1992), which Perdue characterises as a record of oral commentary that failed to actually train readers to debate (pp. 19–20). The thesis is that Buddhist reasoning and debate is not formal logic: validity is determined “only in relation to certain persons at certain times” because “the only value of an argument is in its efficacy in bringing forth new knowledge” (p. 142). Validity is “inextricably linked with the possibility of epistemological verification” — a Buddhist argument is sound only for someone who has ascertained its components. Logic is therefore subordinated to, and included within, the topic of valid cognition (Skt. pramāṇa, Tib. tshad ma), which Perdue’s teacher Kensur Yeshi Thupten glosses as “the measure of what exists and what does not exist” (p. 60).
The book operates explicitly from the standpoint of the Sūtra School Following Reasoning (Sautrāntika) as transmitted in the Collected Topics genre, and within that tradition follows the Geluk-pa order — chiefly through three teachers: Lati Rinpoche, Denma Lochö Rinpoche, and Kensur Yeshi Thupten (pp. 66–67). The Collected Topics texts draw their material and reasoning style from Dignāga, Dharmakīrti and Vasubandhu, with Pur-bu-jok Jam-ba-gya-tso’s Tutor’s Collected Topics (Yongs ‘dzin bsdus grva) cited as the source-text behind Debate in Tibetan Buddhism (pp. 65–66, 178). Pedagogically the Course foregrounds the standard tripartite analytical scheme: every topic is approached through definition, divisions, and illustrations (མཚན་ཉིད་ / མཚན་གཞི་ / མཚོན་བྱ་, with མཚན་གཞི་ functioning as the example), with the study of philosophy in Tibet itself called mtshan nyid — “definitions” (p. 378).
Chapter breakdown
1–3. Introduction, The Human Situation, Reasoning within the Buddhist Context — frames the South Asian wisdom-tradition rationale for reasoning. Ch. 3 explicitly positions the Course within the Sautrāntika tenet system as a beginner’s bsdus grwa (pp. 65–66). 4–5. Comparison of Phenomena; Two Kinds of Statements — the four-fold scheme of comparing any two phenomena (three possibilities / four possibilities / mutually exclusive / mutually inclusive) and the distinction between statements of qualities and statements of pervasion. 6. The Buddhist Syllogism — the Tibetan single-sentence enthymeme (“the subject S is P because of being R”), the three modes (property of the subject, forward pervasion, counterpervasion), the epistemological-verification requirement (pp. 142–158). 7. The Only Two Valid Syllogism Forms — restricts valid forms to “is”-statements and “exists”-statements. 8. Three Types of Correct Signs — division of correct signs by entity into effect signs (smoke proving fire), nature signs (product proving impermanence — same-nature reason), and signs of nonobservation (set forth “for the sake of realising that all things are selfless,” p. 179, citing Manchester-Rogers’ Tibetan Logic). 9–10. Basic Buddhist Ontology 1 and 2 — the selfless, existents (with seven divisions), non-existents; functioning things divided into matter, consciousnesses, and nonassociated compositional factors. Sautrāntika ontology asserting truly existent external phenomena. 11. The Paradigms for Proving a Comparison of Phenomena — formal proof-templates for each of the four comparison-types. 12. Consciousnesses in Relation to What They Perceive — the seven-fold typology of cognitions ordered by reliability (pp. 293–305): wrong cognition (Perdue: “wrong consciousness”, ལོག་ཤེས་); three grades of doubt (Perdue: “uncertain consciousness”, ཐེ་ཚོམ་) — leaning-wrong, equivocating, leaning-right; investigative thought (Perdue: “correctly assuming consciousness”, ཡིད་དཔྱོད་); inference (Perdue: “inferential cognizer”, རྗེས་དཔག་); direct perception (Perdue: “direct perceiver”, མངོན་སུམ་). Includes the long discussion of appearing object vs. object of engagement, conceptual generality (Perdue: “meaning generality”, དོན་སྤྱི་) vs. sound generality (སྒྲ་སྤྱི་), and the four types of direct perception (sense, mental, self-awareness, yogic). 13. Valid Cognition — the Geluk argument that there are exactly two valid cognitions (PV III: “because objects of comprehension are two”); discussion of the term “valid cognition” (Perdue: “valid cognizer”) as a division from the point of view of the name applied to person (Buddha), scripture, and mind (pp. 316–317). 14–18. Three Spheres of Agent–Action–Object; choosing partners; Challengers and Defenders; the Clap; Mañjuśhrī — the ritual and interactive frame of debate. 19. The Three Purposes of Debate — refuting wrong views (‘khrul ba dgag pa), presenting one’s own system (rang lugs bzhag pa), and dispelling objections (rtsod pa spong ba); these structure the chapters of Collected Topics texts (pp. 371–377). 20. Definitions, Divisions, and Illustrations — the formal tripartite analytic structure; the eight approaches of pervasion between a definition and its definiendum (p. 381); definition as substantial existent (rdzas yod) vs. definiendum as imputed existent (btags yod); the four types of division (true partition, non-comprehensive, division in name only, division from the point of view of the name). 21. One and Different — the partition of established bases into “one” and “different” (singular and diverse). 22. Consequences — thal ‘gyur (prasaṅga) as the principal tool for refuting positions by drawing out unwanted implications; structurally similar to a syllogism but without a “promise” — the formulator is bound only to fairness of derivation, not validity of conclusion (pp. 411–419). 23–25. Procedures, Strategies, Bringing It All Together — defender’s answers, the standard challenger responses, debate strategy across the four comparison-types. 26–27. Efforts and Practices; Conclusions — practice regimens and a final stance against epistemological relativism.
Key claims
- Validity is not formal but epistemological-verification-relative: a syllogism is valid only for a person who has ascertained its property-of-the-subject and pervasion (p. 142, pp. 149–150).
- The “definition of a correct sign” is exactly “that which is the three modes” (tshul gsum) — property-of-the-subject, forward pervasion, counterpervasion (p. 146).
- The three types of correct signs are exhaustive by entity: effect, nature, and nonobservation. Signs of nonobservation “were set forth for the sake of realising that all things are selfless” (p. 179, citing Manchester-Rogers).
- A seven-fold gradation of cognitions by reliability: wrong cognition; three grades of doubt; investigative thought; inference; direct perception (pp. 293–305). Only the last two are valid cognition; the first five are mere beliefs or misperceptions.
- An investigative thought (Perdue: “correctly assuming consciousness”, ཡིད་དཔྱོད་) is “a knower that does not get at an object with respect to which superimpositions have been eliminated although it adheres one-pointedly to the phenomenon which is its principal object of engagement” (p. 298).
- Only two valid cognitions (perception, inference) because objects of comprehension are exactly two (specifically- and generally-characterised); citing Dharmakīrti, PV III (Perdue p. 314).
- The Buddha is called “valid cognition” (Perdue: “valid cognizer”) only by division from the point of view of the name — only consciousnesses are actual valid cognitions; persons and scriptures are merely called by that name (p. 317).
- A conceptual cognition (Perdue: “conceptual consciousness”, རྟོག་པ་) is mistaken with regard to its appearing object (a conceptual generality) but may be correct with regard to its object of engagement; this is the standard structure of inference (pp. 286–287).
- Conceptual generalities (Perdue: “meaning-generalities”, དོན་སྤྱི་) are permanent phenomena — eliminations of all that is not-thus — even when the things they represent are impermanent (pp. 278–279).
- Definitions are substantial existents (rdzas yod) — the actual referents themselves — while definienda are imputed existents (btags yod) — names ascribed to those referents (p. 379).
- A correct definition stands in eight approaches of pervasion with its definiendum (four pervasions of being plus four pervasions of existence, pp. 381–382). MI alone is insufficient: definition and definiendum must additionally stand in the priority-of-ascertainment relation — one must ascertain the definition before one can ascertain the definiendum with valid cognition (p. 382, citing Lati Rinpoche).
- Buddhist debate has three purposes: refuting wrong views, presenting one’s own system, dispelling objections to it (pp. 371–377). These structure the layout of every Collected Topics chapter.
- Geluk asserts that at the end of a continuum of sense direct perception there is generated one moment of mental direct perception, which then induces conceptual cognition (p. 283).
- The Collected Topics ontology is Sautrāntika, but explicitly held to be a stepping-stone, not the final view (pp. 65–66).
Notable quotes
- “Valid cognition is the measure of what exists and what does not exist.” — Kensur Yeshi Thupten, paraphrased by Perdue, p. 60.
- “Because objects of comprehension are two, valid cognizers are two.” — Dharmakīrti, PV III, quoted at p. 314 (Perdue’s wording; our preferred rendering: “…valid cognitions are two”).
- “The Buddha is called a ‘valid cognizer’ or a ‘valid person’ because Buddhists find that they can rely upon his sayings as validly established… However, from among the three — valid persons, valid scriptures, and valid minds — only the minds are actual valid cognizers.” — p. 317.
- “What we experience as sense perception is a continuum of moments of consciousness apprehending a continuum of moments of an object which is also disintegrating moment by moment.” — Lati Rinpoche, Mind in Tibetan Buddhism, quoted at p. 283.
- “Although direct perceivers may induce conceptuality, they themselves are totally free of conceptuality.” — p. 283.
- “The yogi first understands subtle impermanence conceptually. Then through continued and sustained familiarization… the meditator is able to bring the image appearing to that inferential cognizer into exceptionally clear focus… eventually passes beyond the need for a representative image and develops a direct perception of the object.” — p. 284.
- “It is suitable to have the same thing in parts 1 and 2 or to have the same thing in parts 1 and 3, but it does not work in parts 2 and 3.” — p. 150 (on why “S is R because of being R” cannot be a valid argument: ascertainment order would collapse).
- “Doubting consciousnesses are among the worst types of mind. If one is travelling along a road constantly wondering, ‘Is this the right road or not’, it is difficult to arrive at one’s destination.” — Lati Rinpoche, quoted at p. 294.
- “Monks and scholars should well analyse my words, like gold [to be tested through] melting, cutting and polishing, and then adopt them, but not for the sake of showing me respect.” — Buddha, quoted at p. 316.
(Direct quotes are preserved with Perdue’s wording inside quotation marks. Surrounding prose uses our preferred terminology — see terminology note above.)
Relevance to the three main texts
pramanasamuccaya
Perdue’s presentation is structurally downstream of Dignāga: the two-pramāṇa thesis, the perception/conception distinction by way of free-from-conceptuality, the four-fold typology of direct perception (sense, mental, self-awareness, yogic) all derive from PS. But Perdue’s Course engages PS only mediated through the Collected Topics tradition — he cites Dharmakīrti’s Commentary on (Dignāga’s) “Compilation of Valid Cognition,” not PS directly (e.g. p. 314). The book is most useful for PS-readers as a witness to how the bsdus grwa tradition flattens PS-PV into a single Geluk teaching framework.
pramanavartika
This is the operative root text for the Course. Perdue quotes Dharmakīrti directly at PV III to argue for exactly two pramāṇas (p. 314). The seven-fold gradation of cognitions in Ch. 12 — and the careful treatment of conceptual cognition as “mistaken w.r.t. its appearing object but correct w.r.t. its object of engagement” — is a Tibetan systematisation of PV’s distinction between specifically- and generally-characterised phenomena, the foundation of PV Ch. 3 (Perception) and Ch. 1 (Inference for Oneself). The Buddha-as-valid-cognition move (p. 317) is the soteriological frame of PV Ch. 2 (Pramāṇasiddhi). Where Perdue contributes new content beyond what is on pramanavartika is in his explicit pedagogical sequencing — the Course shows what the Geluk monastic uptake of PV actually looks like in the classroom.
pramanayuktanidhi
The most consequential cross-reference. Perdue’s seven-fold typology of cognitions in Ch. 12 is precisely the early-Tibetan (Cha-pa / Geluk) five-type non-valid-cognition list that sakya-pandita’s Treasury (Ch. 2) refutes — with the three grades of doubt telescoped to a single ཐེ་ཚོམ་ category. In particular, Perdue’s investigative thought (Perdue: “correctly assuming consciousness”, p. 298) is the very ཡིད་དཔྱོད་ that Sapan dismisses as a pure fabrication (non-valid-cognition). Perdue inherits this category from the bsdus grwa tradition without flagging the Sa-gya objection at all. The Course therefore provides the wiki with a clean, explicit, modern statement of the position the PYN argues against, in the words of a sympathetic contemporary expositor. It is also instructive on the wider tripartite definition-structure (definition, divisions, illustrations) that PYN Ch. 8 (mtshan nyid brtag pa) thematises — Perdue’s Ch. 20 is the most accessible English statement of that structure.
Critical assessment
The Course is a careful, classroom-tested introduction with high internal consistency, but it is single-tradition: the Geluk bsdus grwa position is presented as simply “the Buddhist position,” with Sa-gya and other dissenters generally absent. Where Perdue does mention divergence (e.g. that “most scholars agree that self-knowers are mental direct perceivers”, p. 304), the alternative views are not engaged. The book is invaluable for understanding the Geluk training pedagogy and for the careful articulation of the seven-fold cognition typology and the three modes of a correct sign — but readers approaching it as a guide to the whole Buddhist epistemological tradition would miss the actual debates and would be unable to reconstruct the PYN’s case from what is here. Most useful in tandem with dreyfus-recognizing-reality and sonam-introduction-treasury, which document the Sa-gya counter-position.