Friday, 1 July 2011


To begin with I think it is worth while quoting the entire treatment of the valid means of knowledge (pramana) knows as upamana as given in the locus classicus of Vedanta Paribhasa by Dharmaraja Adhvarindra who seems to have flourished in the 17th. Century. It will be noted just how short a treatment is given of what I hope to show is an extremely important pramana.

Now comparison is being described. The instrument of the valid knowledge of similarity is comparison. For instance a man who has seen a cow's form in cities and has gone to a forest, where his eyes have come in contact with a gayal (gavaya - bos gaurus) has the cognition, "This thing is like a cow". Then he has the conviction, "My cow is like this." Here by a process of agreement and difference, the knowledge of the likeness of a cow which exists in a gayal is the instrument, and the knowledge of that likeness of a gayal which exists in a cow is the result.

this is not possible through perception, for then the cow's form is not in contact with the eyes. Nor is it possible through inference, for that likeness of a cow which exists in a gayal cannot be the sign (reason) for inferring the likeness of a gayal in a cow. Nor can it be urged that this is possible through the following inference:

My cow is like this gayal.
Because it corresponds to its likeness existing in a gayal.
That which corresponds to its likeness existing in a thing is like the latter.

As Caitra, who corresponds to his likeness existing in Maitra, is like him.

For even without this sort of inference, the cognition, "My cow is like this," is a matter of common experience, and has also the apperception, "I am comparing the two,". Hence comparison is a distinct means of knowledge.

The first thing that must be remarked is the insistence that upamana is a distinct means of knowledge which is to say that it cannot be reduced to anything more basic or primitive. It thus takes its place along with the others such as Perception and Inference. There are 6 in all accepted within the philosophical system known as Advaitic Vedanta. It would take us too far out of our way to go into any detail about the rest of them though later I may offer a note on anupalabadhi or non-apprehension of existence.

Let me first make the point that perhaps 'comparison' may not be quite the apposite word in relation to this means of knowledge. I will be staying with the canonical example of the gayal for the moment. When the cow is seen in the gayal where is there what we would strictly in English call a comparison? When one speaks of comparison one generally requires that there be two things, a eye-witness description and a known individual. The one is compared to the other and a match is declared or a partial match or a mismatch. This seeing of the cow in the gayal seems to me to more like the successful deployment of the concept 'cow'. Our having the concept 'cow' allows us to 'see' the cow in the gayal. There is no comparison as such, having the concept is what enables us to do this, to have this recognition so to speak. For this basic reason I would stay with the term upamana to avoid what I think is an erroneous identification with the notion of comparison.

In Mental Acts Peter Geach speaks of concepts as capacities exercised in acts of judgement. Does exercising the concept just involve 'seeing' the cow in the gayal? Is the notion of comparison a misunderstanding that the translation of upamana underlines? We can construe what we have done as comparison only after the fact. That is what we must be doing we say to ourselves presenting an analysis that makes sense to us. It makes sense to us because of the inherent hankering after empiricist explanations. Evidence is required even if we have to make it up to have a likely story.

Why do we find it hard to accept that there things that we can do, that there are powers we just have and concept acquisition and concept utilisation are those sort of powers or they are perhaps a single power of double aspect. Having the concept 'white' means among other things seeing it in milk and chalk. No comparisons needed.

I am I have to admit ignorant of later speculation about pramanas and specifically this one amongst that school known as the New Logicians (Navya Nyaya) in its later phase. The perennial focus on the gavaya (bos gaurus) by commentators in general leads me to think that the general understanding of what was at issue was not clear. The canonical example is often a lifebuoy to cling to in a sea of incomprehension. My own speculations may be a matter of my own fancy and no better.


elisa freschi said...

Dear Michael,
I agree with your last point. The fact that commentators only mention the case of the cow/gayal means that they had probably lost track of the purpose of that pramāṇa. I tend to think that upamāna used to be a useful way to gather new information, through analogy. In rituals, if the Soma is not available, one has to substitute it with something else and one chooses the substitute out of its being similar to the Soma. Hence, upamāna might have had a ritual origin. It might have been very much in use for a while (because, in fact, we learn always by means of comparisons). Later, when the pramāṇa theory gor formalised, one could not admit but means of knowledge yielding JUST valid knowledge. Analogy was no longer welcome, due to its inner risks. But one did not want to contradict a well-established common sense idea about its being a distinct pramāṇa. Hence, one defined in a way as to make it always valid…and almost useless.
This being said, I agree with you about the fact that in the late tratments of the cow/gayal case, what is at stake is just the recognition of the form of a cow in a gayal. But is the Indian concept of artha or sāmānya or ākāra broad enough to allow for this? If I am not wrong, the only thing one could say is that cow and gayal are recognised as lower members of the broader sāmānya "bovines". But this is not the same as saying that the cow is seen in the gayal…

ombhurbhuva said...

Thanks for that intriguing and stimulating view. Yes it probably is connected with rite, sign and symbol and the foundations of language. Viewed from a purely epistemological point of view I think upamana as an account of concept acquisition is not fully worked out. All concepts whatever involve upamana in their discovery and use. However the offering of the basic elements which are used in upamana, i.e. the comparands (? such a latinism), is dependent on something that is innate. Sankara has this notion of ‘vedic words’ which I’ve mentioned before (BSB. 1.iii.28). In this commentary he says that the word ‘cow’ as genus is an eternal vedic word which allows us to recognise the individual of the species.

Ironically then even though the treatment of upamana seems to stick with the gavaya it is in fact universal and almost any use of concepts involves it.

The connection between artha, samanya and akara is not something I know anything about to be honest, if you have a chance a word or two on that subject would be welcome.

elisa freschi said...

Dear Michael,
I agree that all concepts need some sort of analogical assesment in order to be used in a new context. But this does not seem to be what Indian authors had in mind while describing upamāna as a pramāṇa. It is rather your original reinterpretation of it (which is very welcome, Sanskrit philosophy would be dead without such enterprises).
As for innate concepts and Śaṅkara's commentary, it seems to me that the Veda plays the role of the innate in Western philosophy, i.e., an apriori we cannot make without.
I am not an expert of sāmānya/ākāra etc. either. What I meant is:
suppose we agree that the artha of gauḥ is the sāmānya (universal) gotva. Then we will probably say that it is because of that that we can use gauḥ in order to designate each single cow. But this does not bring us any closer to the gayal. At this point one is reminded of the Vaiśeṣika idea that there are higher universals (the highest being sattā, being). Both the gayal and the cow could belong to such an higher universal (i.e., that of bovines). In this case, it would be relatively easy to recognise them as 'bovines'. But this has nothing to do, as far as I can see, with recognising the similarity of specifically the cow and the gayal (as against that of cows and buffalos).
Hope this does not make things fuzzier.

ombhurbhuva said...

The Vedas are as you say the ultimate source. Together with arthapatthi, which is a sort of inference to the best explanation, a truth expressed in that source of reliable truth can give a clue as to how it must be true in the world so to speak. If something is true by the Vedas, if it is accord with the Vedas then it is as they say of products, warranted genuine. This is the acid test. An intimation of this is that in some recitations of the gavaya example the discoverer is primed by a reliable authority on both cow and gavaya. This reliable authority is a relic maybe of the sabda pramana/sruti pramana.

The problem of similarity was one that engaged Wittgenstein. How do you know which factor it is that you are to pick out for a judgement of similarity? How could you learn how to do this? It's quite mysterious. If you look at a familiar object long enough it becomes strange.