Perception and Inference are, According to Dinnāga, both Pramāņa. What is The Knowledge Gained Through Each?
Perception and inference are, according to Dinnāga, both Pramāņa.
What is the knowledge gained through each?
The traditional Buddhist epistemological stance on pramāņa is that only perception is a source of knowledge. The pramāņa of conceptual thought fails to grasp the elusive realm of particulars. ‘Concept’ includes language, arithmetic, induction, means-ends reasoning etc. The Buddhist believes that it is futile to apply truth values to our perceptions, since the essence of raw-sense data cannot be comprehended. We can apprehend them through the senses, but beyond this awareness, our knowledge must remain in a realm of concepts. Conceptual thought can be used for communication, sciences, or promoting understanding (but it is at this point that suffering begins). If knowledge gained through perception is fruitful but incommunicable, while knowledge gained through concepts is communicable but pointless, then is there any point in seeking or acquiring knowledge?
The Buddhist philosopher Dinnāga brings the two pramāņas closer together without compromising the integrity of either. He defends the distinctions marked by the Buddhist school, and provides an alternative epistemology to the realism of the Vaiśesika. Dinnāga established the first formal logic in Indian philosophy in the ‘The Wheel of Reason’ (Hetucakra). In the ‘The Collection on Knowing’ (Pramānsamuccaya) he developed a theory of meaning based on the exclusion theory of apoha of the Vaiśesika realist school.
This essay will discuss theories in Pramānsamuccaya, dealing first with inference and the apoha theory, and in the second half with perception. Within each I shall attempt to provide some examples to help explain the concepts and also discuss the two pramāņas.
In the Pramānsamuccaya, Dinnāga distinguishes between only two sources of knowledge. Knowledge through inference is gained through universals which are concepts contained exclusively within the mind, while perception is only of particulars. In the Pramānsamuccaya these are called fields of operation. The following quote refers to two distinct means of acquiring knowledge: inference, and perception:
1.3.0 “Now if both [sensation and inference] are characterised as cognitions, what is the difference between them? ‘Their fields of operation and essential natures are dissimilar.’ Sensation and inference have distinct fields of operation, and their essential natures are also distinct in accordance with their having different cognitive images.”1
When someone perceives an object, our inferential mind immediately applies a label to that object. This would have to operate instantaneously, such that our mind is not capable of perceiving without simultaneously perceiving the object. The label is not present in reality. One might reproach Dinnāga for asserting that we project reality onto the world, or even suggesting idealism, but as long as the faculty of perception exists, than it must mean there is a reality, except that we distort it upon its arrival.
We cannot deny sense-data ie. We cannot reject our sensation of the present, what we can reject are the concepts we attempt to classify them under. In conclusion, Dinnāga argues that it is impossible to conceive of a particular and impossible to perceive a concept. Two questions arise. How is communication through language possible? How can we explain our ability to make sense of the world?
To begin with the first question, we shall refer to the theory of exclusion which Dinnāga employs in the realm of concepts. What makes us even able to apply concepts, since we cannot comprehend their relation to the particulars? Whereas the Buddhist would simply concede that the realm of concepts is fictitious, Dinnāga produces a reformative theory of inference which maintains both ontological realms.
An example cited is of a lotus, the process of exclusion which takes place when communicating the word ‘lotus’ is as follows:
14.1.0 “Possession of qualities rules out non-substances and possession of aroma rules out non-earthen substances, and possession of a sweet fragrance rules out things
that stink, and a particular sweet fragrance rules out what is not a lotus. Each of these eliminations makes [the lotus more clearly] known. Otherwise,”
15.0.0 “If the sign made the object known by a means of confirmation similar to direct sensation, then either the object would not be known at all, or it would be known in its entirety.”2
In 14.1.1, Dinnāga illustrates the process of inference. One might compare the example with another, fictitious, one. There is a common party game called 20 questions. The aim is for the audience to guess, in 20 questions what the person is thinking of. One version of the game is to keep the domain of objects to one room, so that all the objects are reduced to those which are perceivable to all present (more akin to eye spy with my little eye). The audience may attempt to guess the object by listing all the objects in sight, and with luck someone might guess the object the guesser has in mind. Alternatively, the audience could make the guessing easier by limiting the range of possibilities such as “is it on walls?”, “Is it on the ceiling”, “does it have legs?” or “is it moving?”. It must be noted that all of these questions make the assumption that all present can perceive the same object. The answer may be a dog, person or a chair. When a person employs concepts to communicate particulars, the process through which the concept is brought to the understanding, is one of exclusion.
Not-tableness, not on-the-floorness, not-animalness, are all subdivided into lotus-hood, but none of these, in total, or on their own have as their opposite, lotus-hood.
The obvious problem for the logician, is how not to read a double negation as
a tautology. Analytically speaking, it seems obvious that non-not lotus is the same as
lotus, just as obvious as the truth that every mountain must have a valley. The two truths arise together. To define something entails defining something by what it isn’t, which makes sense, but to define something by what it is, seems a tautology. The difference is a metaphorical one only. We are climbing up the same ladder except from the other side. This changes as soon as we introduce another perspective of the whole endeavour. An allegory might be drawn here: When I say to someone that I used reverse psychology on someone, it is a slang term for having tricked someone into doing something by making them believe something about myself. If I were to say I was using reverse-reverse psychology they might say that I am not using any tricks. But taking into account the distinction between belief and knowledge, it might imply me tricking someone into believing I am tricking them. The subsequent stages of assumption depend on my foreknowledge of the other person’s foreknowledge. I am not moving backwards and forwards on a plane but upwards in a zigzag manner.
Let us say I want someone to be grateful for an action I have done. I could act as though I expected gratefulness. This would be “psychology”. But it might not get me gratefulness, since some people believe in unconditional giving. But I could also act as thought I didn’t expect gratefulness “no, it was no problem at all, seriously, please don’t...etc.” knowing that this will get me more praise. I success rests on the assumption that the other believes that although I want gratefulness I am modest and do not want gratefulness. Let’s say that I myself dislike people acting in such a way (since I know how the game works I might get tired of it eventually). If I assume that the other is aware of what I’m aware, then I might be able to trick him into giving me praise by making him believe I am trying to act as though I am modest. So I trick him into believing I tricked him.
Now, the arbitrary A or B situation might represent the tautology. Me expecting gratefulness failing. Whereas the predicted train-of-thought relation beween A and B is akin to Dinnāga’s notion of a not-not lotus. It is both a lotus and not a lotus at the same time, and for this reason it can be a not-not lotus without being a tautology.
This process of inference is common to all rational beings, and may be viewed
as simple common sense. We perceive an object, and using previous knowledge (in
our memory), we draw inferences from the object. What is interesting about both of these forms of inferring, is that, whereas the employment of universals seem to multuiply infinitely, (ie. “The list of all lists”), whereas the process of exclusion seems to divide infinitely, each division never getting closer to the truth. Moreover, the epistemological problem of actually communicating has not been resolved, for if we accept that utterances are particulars, than how is it possible to recognise these words as concepts. What is required is a mechanical explanation of apoha, and not merely one grounded in language.
Another aspect of Dinnāga’s theory is the trairūpya theory. For an inference to be valid, according Dinnāga, it must fulfil three requirements.
The translation by B.K. Matilal is as follows
The evidential characteristic must belong to the intended object.
It must belong to (at least) one similar object.
It must not belong to any dissimilar object.
Dinnāga refers to these qualifications as a sign. We can infer from one awareness of an object the existence of an object and even the qualities of that object. One used example is that of smoke on the hill. After perceiving smoke, we infer that there must be a fire. As long as we can cite a previous example of fire and smoke present together, and cannot cite an example in which they are not together, then we can state that there is smoke on the hill. Dinnāga states that it is also possible to make inferences of the object not perceived.
“4.2.0 …In the above explanations of inference, attention has been focused on signs not connected with speech. But verbal communication also does not apply to a unique things. Therefore one should regard inference as being of two typoes according as [its object is] visible or invisible. In the case of a visible object, we may teach its name. with reference to an invisible object there is only a concept but there is no awareness of a particular object.”3
Dinnāga asserts that language deals with indeterminates as it does with things perceived., through the exclusion of contraries. With regards to the previous example therefore, if we exclude all the objects within the room, and expand the range of possibilities to any concept, language is even further removed than the situation in which (we assume) all present can see the same object.
On the other hand, what Dinnāga calls knowledge acquired through the senses consists of nothing except the raw sense-data. As soon as one attempts to classify this information it ceases to be a particular and becomes a concept. Perceptions have no truth values, they simply are. Perception is a mechanical process similar to the way instincts or our reflexes operate, interacting with our body without our ability to manipulate them. We cannot control our surroundings, but we can control our concepts, because the concepts are mind-only reflections of the world, but regardless how sharp or closely linked or identical they may seem, our interaction with the world will always be a step away from the way things really are. The disillusionment with our conceptual world is something we ascribe to the gradual rationalising and structuring o
f the world around us, far removed from the incomprehensible joys of an infant which perceives an ever changing world. Dinnāga accepts direct awareness as the fundamental source of gaining knowledge, but is not skeptical about the importance of knowledge gained through inference. Those accustomed to the idea of the subconscious may find it easier to accept the idea of other disguised mental processes, and non-conscious awareness, albeit with a less sharp division than Dinnāga’s epistemology.
But don’t memories seem to depict particulars? Dinnāga would reject the idea of particulars existing in the mind. These particulars would have been replaced by concepts the mind. The uniqueness we ascribe to them being the concept of uniqueness, or the feeling of uniqueness which we unite the various concepts under.
This would make sense when we compare it with dreaming. Dreams being subconscious manipulations of memory-data (mainly images), they can be bent to such a degree in the dream-state that whatever particular they once may have been accurate representations of, they soon lose their sharpness, and get lost in other mental activities. The concept of the subconscious would be an ideal mediator between particulars and concepts. One may see the futility of comparing particulars, just as much as we accept that memories, good or bad, beyond a certain limit, cannot be compared, or classified. A feeling arises out of them, but particular to that memory alone. We say that it is impossible to compare them, not only because they are two different sorts, but merely because. Or in the same we remember our youthful ignorance with a sense of amusement, we classify each one as a sui generis, something which as its own characteristics, but unexplainable, and beyond an awareness of them, and particular feelings they arouse in us we have no place to put them in.
To conclude, problems arise when one considers the possibility of concepts being particulars, and reject the idea of a reason as an independent source of knowledge from experience. Metaphysical questions such as free-will or the notion of a soul would arise from such a challenge. Moreover, if we cannot access true knowledge through introspection, since they are merely metaphors for our thought patterns; or through science, than how can we ever claim to know ourselves or the true motives behind our actions? Such questions and other theories within philosophy of the mind are linked to the ontological divide put forward by Dinnāga. Pertinent to this discussion however, is what thoughts consist of and to what degree all our mental activities (including means-ends reasoning, logical inference and other fixed things we associate with the reason gland) consist of experience. If one accepts the idea that experience provides us with metaphors for understanding, then to what extent is our mind merely dependent on the continuous interaction between particulars
The epistemological problems which arise out of Dinnāga’s theory are mainly how concepts and particulars interact. If we follow in Dinnāga’s argument they simply don’t. Even the exclusion theory merely assumes a connection between the two, but one can never be completely sure of the accuracy of this relationship. It would make more sense to claim that concepts are themselves particulars. One way to view the problem is such. ‘How is it possible to trust the philosophy of a being who is unable, through introspection to get at the concepts which makes his thoughts, or convince him of his rationality’?
1 Pramānsamuccaya (Dignaga on the Interpretation of signs. R. Hayes)
2 Pramānsamuccaya (Dignaga on the Interpretation of signs. R. Hayes)
3 Pramānsamuccaya (Dignaga on the Interpretation of signs. R. Hayes)