Friday, March 18, 2011

Memory and experience

If we had an intellectual intuition (that is, a direct perception non mediated by any sense organ) of another person's memory's contents, would these contents appear as memory to us? No, answers the kaśmīri philosopher Utpaladeva, because the character of being past is not enough to identify a memory. On top of that, one needs the connection with an experience. We only identify as memory what is connected with our previous experience. A sheer object would not count as one. See below for a possible objection.

The Vivṛti (long gloss) of Utpaladeva on his ĪPK (see below for references):

It is not the case that in memory an object previously experienced is manifested in its sheer own form (kevala), like in an intellectual intuition (yogijñāna) having a past object as its content. For, in this way (if the object devoid of any connotation were manifested), it would not have its character of memory. It is said that in memory, the quality of being past only [occurs] through the experience, although the object is in itself changeless. Hence, the manifestation of the experience is here useful. Therefore [the author of the short gloss] says: "And a previously experienced object, together with the experience". And memory has the form of a cognition, therefore he speaks of "light of memory" (my translation).

(pūrvānubhūtasyaivārthasya kevalasya na smṛtau yogijñāna ivātītārthaviṣaye prakāśaḥ. tathā hi smṛtitvaṃ na syāt. anubhavamukhenaiva cārthasya svayaṃ sthiratve ’pi smṛtāv atītatvatvam ucyata ity anubhavaprakāśa evātropayogī. tad āha “pūrvānubhūtaś cārtho ’nubhavena saha” iti | smṛtiś ca jñānarūpaivety āha “smṛtiprakāśa” iti. The text has been edited by R. Torella, Festschrift Brunner, 2002).

Well, one could object that some psychological experiments have shown that in case one is provided with a single content, unconnected with experience and only characterised as past (for instance, an old photograph), one tends to make up a fake memory around it. Hence, the distinction may be less clear-cut… However, the same experiments show that the "trap" works by far better if one includes in the photograph the person one is analysing, thus strongly suggesting that that content is linked with her experience.
What do readers think?

On the metaphor of light as referring to cognition, see here.
I have dealt with memory and subjectivity in other posts. See here.


David Dubois said...


Even assuming that memory as a past experience had by one self is an illusion, still it remains to be explained how such an illusion is possible.

Even if the sens of personnal connection with the experience is a delusion, how is it possible ? Can we explain dreams on the basis of mere samskaras ?

If memory is a purely mechanical process as described by Dharmakirti, how is it that an I appears at some point ? and from where does the feeling of "deja vu" come ?

Given that, according to Utpaladeva, a cognition cannot take another cognition as its object, how can we explain memory with just ephemeral cognitions ?

And can a mere unconscious samskara "remember" another one ? Or even perceive its similarity with the original experience ?
And even accepting such a description of daily life, how are those illusory synthesis possible ?
The crux is : Can a cognition know another cognition ?

If yes, then the buddhist solution is possible.

If no, then Utpaladeva is right, and perhaps philosophers like Daniel dennett could be challenged ?

I'm not sure. But it seems to me that the "illusion" theory doesn't solve the mystery. It only postpone the difficulties.

Thanks for the blog

elisa freschi said...

Cher David,
thanks for this comment. I am not an expert in Pratyabhijñā and I am very glad to read your views.
As far as I am concerned, I am inclined to believe in a personal subject, constant through time, (although my philosophical arguments in favour of that are different than Utpaladeva's ones). Hence, I share your dissatisfaction with the Buddhist view of memory. However, I think that it is theoretically sound. Why should not saṃskāras be credited with a 'personal outlook'?
As for your final dilemma (Can a cognition know another cognition?), I would add:
1) Śaiva thinkers such as Utpaladeva (or Rāmakaṇṭha, see Watson 2006) have themselves somehow been influenced by this Buddhist approach, insofar as they directly identify the subject with cognition.
2) One could also imagine that a cognition cannot know another cognition, but that cognition A (the actual one) arises because of conditions S (saṃskāras), which bestow on it the apperance of "being the memory of … (a past event)".
What do you think about the two points above?

David Dubois said...

Chère Elisa,

Though I'm quite far from any erpertise on Pratyabhijna, I'm fascinated by this debate on memory.

Can a sense of being a person be produced by samskaras ? If they are objects, how can a subject, an "inside" be produced by any amount of such objects ? This is similar to the zombie argument by western philosophers like Dennett, Chalmers, etc. A robot could be loaded with tons of informations. Would it ever become a person ? From the outside, perhaps. But from the inside ?

Regarding 1), yes, Utpaladeva integrates the samskara explanation, but only to show its limits. The Self is a cognition... or rather a free synthesis of cognitions. They depend on him, not he on them.
2)How can an insentient object be a cause or a condition ? Causing is an instance of "becoming aware of oneself as this and that". Or, if a samsKARa is really an action, then it samskara is another name for consciousness, because consciousness is the only action, all other actions being only metaphors.

I'm not sure of any of this, still trying to undertsand Utpaladeva, etc.

But do you think a feelin of being a person could be produced by impersonnal interactions between objects ? Is it possible to account for the first-person experience from a third person point of view ?


David Dubois said...

PS : an excellent study on those issues, to be released soon

elisa freschi said...

Cher David,
I am also still struggling with Utpaladeva, etc., and I will surely benefit from Isabelle's work, as soon as it is published, thanks for letting me know.
As for 1), if you speak of 'free synthesis of congitions' you seem to presuppose a sort of adhyavasāya, as in Nyāya. On the contrary, Utp. seem rather to speak of prakāśa as the only reality. Yes, this prakāśa synthesises cognitions, but only insofar as it is a broader cognition, isn't it?
As I said, I am myself not convinced by the non-subject hypothesis, but I think that non-personal aspects of a subject can be explained away through saṃskāras, understood as automatisms, carrying with them the illusion of being part of a subjective experience. For instance, one could say that there is no subject A remembering herself while doing x, but just the event of a cognitive act having the content "A did x".

David Dubois said...

Chère Elisa,

Yes, consciousness is not reducable to any particular cognition.
But it is not an altogether different entity than cognitions : it wholly pervades cognitions, and it is not just some separate synthetiser, but rahter the very activity fo synthesising (anusamdhâna, rather than adhyavasâya)cognitions. Otherwise, one would fall into an infinite regress. This is one point where Utpaladeva departs from Nyâya. No dharmin separate from dharmas, but the dharmin is not just a sum of dharmas either.


elisa freschi said...

Cher David,
very good point. I have been dealing a little bit with dharma/dharmin in Rāmakaṇṭha Bhāṭṭa (see my Śaivasiddhānta theories about ātman here) and I guess that Utpaladeva's position is made smoother by his non-dualism. I hope I can post more on this in the next days.

Licenza Creative Commons
Quest' opera è distribuita con licenza Creative Commons Attribuzione - Non commerciale - Non opere derivate 2.5 Italia.