Dignāga

Pratyakṣa is a kind of awareness that acquires information about particulars, and is immediately present to one of the senses. This is the topic of the first chapter of the Pramāṇa-samuccaya.[9] For Dignāga, perception is pre-verbal, pre-conceptual and unstructured sense data. In chapter two of the Pramāṇa-samuccaya he writes:

Sensation is devoid of structure. That cognition in which there is no structure is sensation. What kind of thing is this so-called structure? Attaching a name, a universal and so forth.[10]

According to Dignāga our mind always takes raw sense data or particulars and interprets them or groups them together in more complex ways, compares them to past experiences, gives them names to classify them based on general attributes (samanyalaksana) and so forth. This process he terms kalpana (arranging, structuring).[11] This cognitive process is already different from sensation, which is a simple cognition based only on the immediately present. Thus pratyakṣa is only awareness of particular sense data such as a patch of green color and the sensation of hardness, never awareness of a macroscopic object such an apple which is always a higher level synthesis.[12] For Dignāga, sensation is also inerrant, it cannot "stray" because it is the most basic and simple phenomenon of experience or as he puts it:

"it is impossible too for the object of awareness itself to be errant, for errancy is only the content of misinterpretation by the mind."[13]

Also, for Dignaga, pratyakṣa is mostly phenomenalist and is not dependent on the existence of an external world. It is also inexpressible and private.[14]

Anumāṇa (inference or reasoning) for Dignāga is a type of cognition which is only aware of general attributes, and is constructed out of simpler sensations. Inference can also be communicated through linguistic conventions.[15]

A central issue which concerned Dignāga was the interpretation of signs (linga) or the evidence (hetu) which led one to an inference about states of affairs; such as how smoke can lead one to infer that there is a fire.[16] This topic of svārthānumāna (reasoning, literally "inference for oneself") is the subject of chapter two of the Pramāṇa-samuccaya while the topic of the third chapter is about demonstration (parārthānumāna, literally "inference for others"), that is, how one communicates one's inferences through proper argument.[17]

According to Richard Hayes, in Dignāga's system, to obtain knowledge that a property (the "inferable property", sadhya) is inherent in a "subject of inference" (paksa) it must be derived through an inferential sign (linga). For this to occur, the following must be true:[18]

1. The inferential sign must be a property of the subject of the inference. That is, there exists in the subject of inference a property, which is different from the inferable property and which is furthermore evident to the person drawing the inference; this second property may serve as an inferential sign in case it has two further characteristics.
2. The inferential sign must be known to occur in at least one locus, other than the subject of inference, in which the inferable property occurs.
3. The inferential sign must not be known to occur in any other loci in which the inferable property is absent.

Richard Hayes interprets these criteria as overly strict and this is because he sees Dignāga's system as one of rational skepticism. Dignāga's epistemology, argues Hayes, is a way to express and practice the traditional Buddhist injunction to not become attached to views and opinions.[19] According to Hayes, for Dignāga, the role of logic is:

to counter dogmatism and prejudice. As a weapon in the battle against prejudice that rages in every mind that seeks wisdom--in minds of the vast majority of people who do not seek wisdom, prejudice simply takes full control without a contest-there is nothing as powerful as the kind of reason that lies at the heart of Dignaga's system of logic. For it should be clear that very few of our judgments in ordinary life pass the standards set by the three characteristics of legitimate' evidence. Taken in its strictest interpretation, none of the judgments of any but a fully omniscient being passes. And, since there is no evidence that there exist any fully omniscient beings, the best available working hypothesis is that no one's thinking is immune from errors that require revision in the face of newly discovered realities.[20]

Dignāga considered the interpretation of conventional and symbolic signs such as the words and sentences of human language to be no more than special or conventional instances of the general principles of inference or anumana.[21] He takes up several issues relating to language and its relationship to inference in the fifth chapter of his Pramāṇa-samuccaya.[22]

During Dignāga's time, the orthodox Indian Nyaya school and also Hindu Sanskrit grammarians (such as Bhartṛhari) had discussed issues of epistemology and language respectively, but their theories generally accepted the concept of universals which was rejected by most Buddhist philosophers. Influenced by the work of these thinkers as well as by Buddhist philosophers of the Sautrantika school who rejected Hindu theories of universals in favor of nominalism (prajñapti), Dignāga developed his own Buddhist theory of language and meaning based on the concept of "apoha" (exclusion).[23] Hattori Masaaki explains the doctrine thus:

a word indicates an object merely through the exclusion of other objects (anyapoha, -vyavrtti). For example, the word "cow" simply means that the object is not a non-cow. As such, a word cannot denote anything real, whether it be an individual (vyakti), a universal (jati), or any other thing. The apprehension of an object by means of the exclusion of other objects is nothing but an inference.[24]