March 04, 2013


I am not going to discuss here the idea of tshad ma bzhi, which is said to be mentioned in the famous rDo rje tshig rkang of the Lam-’bras tradition. For details, see Sobisch 2008: 7, 99, where it is translated as “four authenticities.” Dan Martin translates the same expression as “four Truth-tests.” See Martin 1997: 269. The four are: bla ma tshad ma (in Sobisch 2008: bla ma’i tshad ma), nyams myong tshad mabstan bcos tshad ma, and lung tshad ma.

It is perhaps useful to assume that the term tshad ma can be employed (a) technically, that is, in the sense of gsar du mi bslu ba’i rig pa (how should one translate this actually?) or rang stobs kyis mi bslu ba’i rig pa (“a cognition that is independently/autonomously non-erroneous”), and (b) non-technically, sometimes called “metaphorically” as in the case of pramāam bhavatī “Your ladyship is the authority” (MW, s.v.) or according to some tshad ma in ston pa tshad mar sgrub pa (“establishing the Teacher to be an authority”). Possibly tshad ma bzhi above can only be understood in the second sense, that is, as “four authorities,” or, perhaps as “four means of attestation/validation.”

Actually I did not intend to discuss tshad ma bzhi, but just note down what seems to be a Terminus technicus, namely, slob dpon tshad ma, for fear that I will never find the reference if and when I need it. The expression slob dpon tshad ma occurs in a citation in gNubs-chen’s bSam gtan mig sgron (p. 32): slob dpon tshad mar ma phyin spre’u’i lung ||. But is it at all a technical term here? What does this line mean?

Some venues of considerations would be the (Buddhist) notions of (a) one just pramā(ultimate/finally), (2) two pramāas, (c) three, (d) four, and so on. 

(a) All pramāas consumate in svasavedana/svasavitti (which is by definition pratyaka). This is known in the Dharmakīrtian tradition, I think. The idea that ultimately only the omniscience of the Buddha is the pramāa, I think, has been proposed by Candrakīrti. (b) The concept of two pramāas (i.e. pratyaka and anumāna) is posited by also by the Dharmakīrtian tradition. (c) There seems to be two types of the three-pramāconcepts. First, the more popular one is the notion of pratyakaanumāna, and āgama. One of the earliest sources for this version of the three-pramāa scheme may well be the Vyākhyāyukti, where the expression tshad ma rna pa gsum po has been employed. See Verhagen 2008: 244, n. 34. Note that āgama is said (by the Tibetan traditions) to be subsumable under anumāna. But presumably for Dharmakīrti, āgama can be considered “authoritative” or “reliable” or “trustworthy” (pramāa), if and only if, its contents are verifiable by means of pratyaka and anumāna, not because of the mere fact that it is regarded as an āgama by a person or tradition “x.” One can theoretically throw out an āgama (even if Buddhistic) out of the window, if its doctrinal contents turn out to contradict pratyaka and anumāna. That is why I love Dharmakīrti! He would, of course, say that incidentally it turns out to be that the contents of the Buddha’s teachings turn out to conform pratyaka and anumāna and hence can be accepted as “authoritative” or “reliable” or “trustworthy.” There is yet another three-pramāa scheme, which may be historically linked with the four-pramāa scheme of the Lam-’bras tradition. Rong-zom-pa, in his Dam tshig mdo rgyas (p. 307), explains that three pramāas (tshad ma gsum), namely, man ngag tshad malung tshad ma, and rigs pa tshad ma, have been taught in the Vajrapāyabhiekatantra. He also employs the corresponding terms man ngag dam palung dam pa, and rigs pa dam pa (p. 308). He makes it clear that this three-pramāa scheme is Mantric or Tantric and it is said to excel “ordinary” (tha mal pa’i) scheme of pratyakaanumāna, and “distinct from or other than” (gzhan pa) other (Sūtric) āgama and yogipratyaka. Supposing the Vajrapāyabhiekatantra is relatively older than rDo rje tshig rkang, it is well-neigh possible that the four-pramāa scheme of the Lam-’bras tradition is an adoption-cum-extension of the three-pramāa scheme proposed in the Vajrapāyabhiekatantra. I hold this position of mine particularly significant (without intending to self-praise) for explaining the history of the four-pramāa scheme of the Lam-’bras tradition! This brings us to the four-pramāa scheme in Buddhist tradition, for which two versions prevail. Of the two, the four-pramāa (tshad ma bzhi) scheme—consisting of pratyakaanumānaupamāna, and āgama—attributed to Candrakīrti’s Prasannapadā by Tibetan scholars and occurring in a non-Tantric context, should be treated to be earlier than the four-pramāa scheme of the Lam-’bras tradition. 


In a citation from the rDo rje bkod pa (perhaps = Kun ’dus rig pa’i mdo—also titles dGongs ’dus and Kun ’dus occur there), an Anuyoga scripture, in gNubs-chen’s Sam gtan mig sgron (p. 235), the expression tshad ma gsum occurs. Here, too, the three pramāas seem to be lung (tshad ma), rigs pa (tshad ma), and man ngag (tshad ma). See ibid. (p. 241). The word tshad ma is added by me. By the way, one of the most common scribal errors seem to be the confusion between rigs pa and rig pa. Our text reads rig pa (p. 241), but I assume that that it should read rigs pa.

Cf. Davidson 1999 [= “Masquerading as pramāa”]: 33–34. I see that it also mentions Vajrapāyabhiekatantra but the point I made above does not seem to be made.

The Mahāparinirvāṇasūtra (T, fols. 165b6–168a7) calls the rten bzhi (= rton pa bzhi), that is, chos nyid, donnges pa’i don, and ye shes as tshad ma and their counterparts as tshad ma ma yin pa.  


  1. I really like your blog and the questions you put forward. They are very inspiring and make me think. I sometimes used "criteria" in the past, but seeing your new evidence, I feel like using "reference". This works perhaps better in French than in English. I think the translations of pramāṇa are often too heavy, and your examples make that even more clear.
    Generally the four "references" are pratyakṣa (T. mngon sum), anumāna (T. rjes dpag), upamāna (T. (dpe) nyer ‘jal), and śabda (T. sgra) or śruti or a witness that can be trusted (āptopadeśa). I guess a slob dpon and a lama would fall into the latter category. So: "Those who don't take the teacher as their reference are following a monkey tradition (āgama)."

    1. Dear Hridaya artha,

      Thank you for your kind remarks. Of course, we cannot afford to discuss the issue of pramāṇa in general. That various Buddhist and non-Buddhist schools have different notions about the number of pramāṇa is well-known. Tibetan scholars would usually attribute the concept of the “four [kinds of] pramāṇa” (tshad ma bzhi) to Candrakīrti’s Prasannapadā (i.e. pratyakṣa, anumāna, upamāna, and āgama). In my view, Dharmakīrti, who insisted only one two kinds of pramāṇa (i.e. pratyakṣa and anumāna), may not have approved this.

      So much for now.



  2. Hi guys! I think Janus doesn't like the word 'authority' just because most young people like him don't respect authorities, and are particularly irked by being asked to follow them. On the other hand I think too much weight is given to the word 'authorities' in academic scholarship and those who think they understand it. It really just means that you've found some grounding for what you're saying, and it's not just floating in from the blue, or getting tossing on the waves. Some people think they can just say what pops into their heads (often calling upon their imaginative or spiritually 'evolved' natures as their only authority, ironically) without paying attention to other types of support that may or may not be available for their ideas... Admittedly, absolute authority absolutely corrupts!
    So yes, moving back to my third or fourth hand, I believe tshad-ma means criteria or standard or backup... And yes, logic or some such method of 'establishing' (sorry, another of those words) truth. Perhaps you can tell I've been doing some earth and water contemplations these last few days. Especially the water.

    1. Obviously no two persons would agree on how to render the term pramāṇa. In the pure context of Buddhist logic and epistemology, it seems reasonable to understand pramāṇa (a) technically as “valid cognition (be it perceptual or conceptual)” or “valid means of cognition/attestation,” and (b) non-technically and metaphorically as “authority (i.e. with regard to cognition).”

  3. Cyrus Stearn translated in his Lam-'bras book tshad-ma as "authentic quality". I was chiefly concerned with the function of the tshad-ma-bzhi, which is, to my mind, establishing the authenticity of a teaching (such as Lam-'bras or the dGongs-gcig). BTW, I discussed some aspects of the tshad-ma-bzhi with H.H. Drikung Kyabgön Chetsang Rinpoche, and he thinks that it is quite obvious that Phagmodrupa brought the terms from Sa-chen's teachings into the Drikungpa tradition. For lack of any alternatives offering themselves, I think I agree.
    Thanks for the notes, Dorji!


  4. Dear Jan,

    Thanks. Please take a look again above for some updates, that may be some interests to you.



  5. Dear Dorji, Dan and Jan,

    Yes the number of pramāṇa or pamāṇa in Pāli varies. The Kālama-sutta gives quite a long list :
    "Do not go upon what has been acquired by repeated hearing; nor upon tradition; nor upon rumor; nor upon what is in a scripture; nor upon surmise; nor upon an axiom; nor upon specious reasoning; nor upon a bias toward a notion that has been pondered over; nor upon another's seeming ability; nor upon the consideration, 'The monk is our teacher.' Kalamas, when you yourselves know: 'These things are bad; these things are blamable; these things are censured by the wise; undertaken and observed, these things lead to harm and ill,' abandon them."

    The point seems to be not to go upon any pamāṇa...

    The Sutta Nipāta (Sn 5 :6) confirms this ultimate goal:
    "One who has reached the end has no criterion (pamāṇa) by which anyone would say that - for him it doesn't exist. When all phenomena are done away with, done away are all means of speaking as well."

    What a rebel that Buddha! :-) For Dan. We all (even young guys of 54) use authorities, criteria or references. The difficulty seems to be not to do so. It is the attitude towards the authority that a young rebel like me finds sometimes problematic. Genuine respect is not shown by kowtow, but by genuine discussion, debate, and testing of the authority. Using it, instead of bowing and submitting to it. A sparring partner :-)

  6. I sentence Janus to 100,000 instances of resisting the urge to prostrate in front of those high ideals Buddhists are aiming for!

    No difficulty not appealing to background, context, tradition, authority if all you want to do is not speak... OK I'll be quiet. Nothing I say makes any sense to me, nor should it to you.

    Well, a PS for Dorji. Please fix that spelling typo in the heading, will you?

    1. Dear D, Haha! Seriously, I am becoming blind to mistakes! Thanks. Promptly done. :-) D.

  7. @Dan Thank you, I will try to do so whilst simultaneously trying not to think of a pink elephant. BTW I expect the "speaking" to be along the lines of Nāgārjuna's view or position (dṛṣṭi). Otherwise what a sad sad world this would be. And we wouldn't have the pleasure of squabbling about authoritarianism and contrarianism.

  8. I'll bet you're still thinking about that pink elephant! Hah! Gotcha!
    Now whatever you do, don't think about that goose inside the bottle.