唯識二十論述記 2.23

2.23

其正理師，恐違自宗眼等五識不緣假法，異於經部。若順於古，即有陳那，五識之上，無微相故，非所緣失。遂復說言：色等諸法各有多相，於中一分是現量境，故諸極微相資各有一和集相，此相實有，各能發生似己相識，故與五識作所緣緣，如多極微集成山等，相資各有山等量相。眼等五識，緣山等時，實有多極微相資山相，五識並得，故成所緣。不爾，即有非所緣失。許有實體，但為緣故，故論說言：實有眾多極微皆共和集。廣如陳那《觀所緣論》及《成唯識》第一卷說：然舊唯識，但有鞞世師及古薩婆多義，其和合、和集文但說一，謂隣虛集色，仍意難知，文亦難曉。於下破中，言非隣虛聚集成塵，披者自知，豈如今論？

The master of the Nyāyānusāra, fearing contradiction with his own school's [tenet] that the five consciousnesses such as the eye do not take conceptual designations as objects, differed from the Sautrāntikas. If he followed the ancients, Dignāga's objection would apply: because there is no appearance of atoms on the five consciousnesses, they would suffer the fault of not being objects. Thereupon he further said: 'Form and other phenomena each have many aspects, among which one portion is the sphere of direct perception. Therefore, atoms assist each other so that each possesses an aggregate aspect; this aspect exists substantially, and each can produce a consciousness resembling its own aspect, thereby serving as an ālambana-pratyaya for the five consciousnesses. Just as many atoms aggregate into a mountain or the like, and assisting each other, each possesses the dimension and aspect of the mountain or the like; when the five consciousnesses such as the eye take the mountain or the like as an object, there are substantially many atoms mutually assisting into the mountain aspect, and the five consciousnesses apprehend them together, so they become the object. Otherwise, there would be the fault of not being an object.' Admitting that they have substantial entities solely as conditions, the treatise says: 'They hold that substantially existing numerous atoms all come together in aggregation.' This is explained at length in Dignāga's Ālambanaparīkṣā and the first fascicle of the Cheng weishi lun. However, the older translation of the Viṃśikā had only the positions of the Vaiśeṣikas and the ancient Sarvāstivādins; regarding composition and aggregation, the text only mentioned one, calling it 'aggregate form of adjacent infinitesimals', yet its meaning was hard to discern and its phrasing hard to understand. In the refutation below, it says 'not that adjacent infinitesimals aggregate into a dust particle'; a reader can understand this on their own—how could it compare to the present treatise?