因明入正理論疏 1.47

1.47

又陳那以前古師，宗中復說宗因相違過，陳那《理門》自破之云：「諸有說言，宗因相違名宗違者，此非宗過。以於此中立聲為常，一切皆是無常故者，是喻方便惡立異法，由合喻顯，非一切故。」陳那意言，如聲論者立聲為常，一切皆是無常故因，是彼外道立宗之喻，方便矯智，惡立異法無常之義，非欲成宗所立聲常。釋所因云：「由合喻顯非一切故。」陳那正云：「立聲為常，正因應言非一切故。以外道說非常之法有多品類，種種差別，名為一切。故立聲常，非一切因。」合喻中云：「諸非一切故者，皆體是常，猶如虛空。」何得乃以一切皆是無常之因立常宗也？復云：「此因非有，以聲攝在一切中故。」陳那意言，此古所引一切皆是無常故因，於其所立常聲非有，以聲攝在一切皆是無常中故，便是因中兩俱不成，共不許因，有法有故，其立聲常，非一切因。陳那復云：「或是所立一分義故。」唯外道許非一切因，於宗中有；內道不許聲非一切因，於宗無，即是隨一不成因過。故此二徒皆非宗過，名因過失，亦是異喻倒離之過。一切離法先宗後因，既立常宗，非一切因，異喻離言：諸無常者皆是一切。而今說言一切皆是無常故，故先因後宗，故成倒離惡立異法之無常義。由此宗違，非是宗過，是因喻過。陳那既破，天主順從，故亦不立。若以因過，宗中不立，既是喻過，因應不立，且如相違及不共、不定，應非喻中能立不成。今釋不然，因有三相，體義最寬，但陳其因，有是相違、不共、不定。未舉其喻，過已彰訖。今陳其宗，猶未有過，舉因方過，何得推過乃在宗中？不同比量相違，彼但舉宗，已違因訖；今此不爾，是故但應如陳那說。外道因明四不成中，但說兩俱及隨一過，不說猶預、所依不成。此不成因，亦不成宗，立敵或偏所不成故。陳那說言，其理雖爾，因依於宗，或決或疑，宗或有無，既有差別，總合難知，故開為四。《理門論》中，古亦有說不定有五，除不共因，異品無故。陳那加之，由不共故。此如何等？

Furthermore, ancient teachers before Diṅnāga also spoke in the thesis (pratijñā) of the fallacy of thesis-reason contradiction (pratijñā-hetu-virodha). Diṅnāga himself refuted this in the Nyāyamukha, stating: 'As for those who say that contradiction between thesis and reason is called thesis contradiction (pratijñā-virodha), this is not a fallacy of the thesis. Because in this [case], when one establishes sound as permanent because everything is impermanent, this is a faulty formulation of a dissimilar instance (vaidharmya) through the device of the example (dṛṣṭānta); since it is shown through synthesis (upanaya) that [sound] is "not everything".' Diṅnāga's meaning is: when, for instance, the Śabdika establishes sound as permanent with the reason 'because everything is impermanent', this is an example used by that non-Buddhist to establish the thesis; through contrivance and deceitful cleverness, it is a faulty formulation of the dissimilar meaning of impermanence, and not that he intends to establish the sound's permanence that was asserted as the thesis. Explaining the reason, he says: 'Since it is shown through synthesis that [sound] is not everything.' Diṅnāga correctly says: 'In establishing sound as permanent, the correct reason ought to say "because it is not everything". For the non-Buddhist holds that impermanent phenomena have many types and diverse distinctions, which are called "everything". Therefore, establishing sound as permanent takes the reason "not being everything".' In the synthesis of the example, it says: 'Whatever is not everything is permanent in essence, like empty space.' How could one instead establish a thesis of permanence with the reason 'because everything is impermanent'? He further states: 'This reason does not exist [in the subject], because sound is subsumed under everything.' Diṅnāga's meaning is: this reason 'because everything is impermanent' cited by the ancients does not exist in the established permanent sound, because sound is subsumed under 'everything is impermanent'; hence it is simply the fallacy of unestablished for both (ubhaya-asiddha) within the reason, an unaccepted reason for both, since it does not exist in the subject (dharmin). In establishing sound as permanent, the reason is 'not being everything'. Diṅnāga further states: 'Or it is because it is a partial meaning of what is to be established (sādhya).' Only the non-Buddhist accepts that the reason 'not being everything' exists in the subject; the Buddhist does not accept that sound is not everything, so it does not exist in the subject, which constitutes the reason fallacy of unestablished for either party (anyatara-asiddha). Therefore, these two formulations are not fallacies of the thesis; they are called fallacies of the reason, and are also the fallacy of inverted contraposition (viparīta-vyatireka) in the dissimilar example. All contrapositive methods put the thesis first and the reason after; since one establishes the thesis of permanence with the reason of not being everything, the contraposition of the dissimilar example says: 'Whatever is impermanent is everything.' But now it says 'because everything is impermanent'; thus putting the reason first and the thesis after, which constitutes an inverted contraposition and a faulty formulation of the dissimilar meaning of impermanence. For this reason, thesis contradiction is not a fallacy of the thesis; it is a fallacy of the reason and the example. Since Diṅnāga refuted it, Śaṅkarasvāmin conformed to him, and so likewise does not establish it. If it is not established in the thesis because it is a fallacy of the reason, then since [certain reason fallacies] are fallacies of the example, they ought not to be established in the reason; for instance, the contradictory (viruddha), the uncommon (asādhāraṇa), and the inconclusive (anaikāntika) should not be the unestablished probans in an example. The present explanation is that this is not so: the reason has three characteristics (trirūpa) and its substantive meaning is the broadest; simply stating the reason, there are cases where it is contradictory, uncommon, or inconclusive. Before one has even brought up the example, the fallacy is already apparent. But now, in stating the thesis, there is as yet no fallacy; the fallacy only arises when the reason is advanced. How can one ascribe the fallacy to the thesis? It is not like contradiction with inference (anumāna-virodha), where merely stating the thesis has already contradicted the reason; this is not the case here, and therefore it should only be as Diṅnāga stated. Among the four kinds of unestablished reasons in non-Buddhist hetuvidyā, they only state the fallacies of ubhaya[-asiddha] and anyatara[-asiddha], and do not state doubtful (saṃdigdha-asiddha) or unestablished locus (āśraya-asiddha). This unestablished reason also does not establish the thesis, because it is unestablished by either both proponent and opponent, or by one side. Diṅnāga states that although the principle is thus, the reason relies on the thesis; it is either definite or doubtful, and the locus in the thesis either exists or does not exist. Since there are distinctions, combining them together makes them difficult to understand, so he opens them into four. In the Nyāyamukha, ancients also spoke of five kinds of inconclusive [reasons], excluding the uncommon reason because it does not exist in the dissimilar class. Diṅnāga added it because it is uncommon. What are these like?