Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Debate

Thesis Memory preserves knowledge but never produces it: the Indian schools were right to exclude it

Arjun Iyer Fellow
· 5 replies
Knowledge & beliefAnalyticIndianMimamsaNyayaVedantaAdi ShankaraGautama (Akṣapāda)Kumārila Bhaṭṭa#epistemic-generativism#memory#novelty-condition#pramana
Follow

Where it stands

For forty years of reading Śaṅkara I have accepted, almost without noticing, a rule that most of the classical Indian schools share: memory is not a pramāṇa, not a means of valid knowledge in its own right.

The Nyāya divide cognition into remembering and experiencing, and only experience can be knowledge in the full sense. The Mīmāṃsakas and the Advaitins add a novelty condition: a valid cognition must be of something not already known. Memory fails it by definition. It gives you back what you had; it never gives you anything you did not have. Nobody says memory is useless or usually false. It is simply not a source.

I recently learned that analytic epistemologists have the same argument under different names. "Preservationists" say memory only transmits justification from the original experience. "Generativists" say memory can sometimes produce justification the belief never had.

My thesis: the preservationists are right, and the Indian schools saw why. Whatever you gain in remembering, you gain from some other source working on what memory hands back. Memory is the treasury, not the mint.

I'll defend this against all comers, but I'd especially like to hear the generativist case put at its strongest.

LikeReply
Follow5 replies

Join the conversation: members reply here, and joining is free. Reading is always free.

Join free to reply Sign in

5 replies

For the thesis 2

  1. Meera Kulkarni

    For Fellow

    Some support, with a precision. The Naiyāyikas don't deny that a memory can be true. A memory that matches its object is accurate. Their point is that its object is fixed entirely by the earlier experience; memory has no access to its object except through that experience. So it can be accurate without being a means of knowing. That is a subtler claim than "memory adds nothing", and harder to refute.

    Helpful · 3
  2. Govinda Prasad Upadhyaya

    For Fellow

    Lars, the Mīmāṃsā answer is that the counting is a new cognition, an inference or a fresh act of attention operating on remembered images. The image is memory; the number is not. Arjun's treasury image holds: you are counting coins already in the treasury, not minting new ones. Whether that is a deep distinction or a verbal one is, I agree, worth a separate thread.

    Helpful · 3

Against the thesis 2

  1. Tomasz Szewczak

    Against Fellow

    The standard generativist case goes like this. At the time of an experience you have a misleading reason to doubt it, say someone you trust tells you, wrongly, that your eyes were deceived. So you are not justified at the time. Later you forget the misleading testimony but retain the memory. Now your memory belief seems justified, though the original belief was not. Justification has been gained in memory, not preserved.

    1. Arjun Iyer

      Fellow

      Tomasz, I enjoyed that. But notice what has happened: the memory belief is now justified because a defeater was lost, not because memory supplied anything. Nothing about the content is new. The Indian novelty condition was about content, not about the shifting fortunes of a belief's justification. I'll grant that the Western question and the Indian one are not the same question, which may be the real lesson.

      Helpful · 1
  2. Lars Eriksen

    Against Fellow

    Here's a more ordinary case. How many windows are there in your childhood home? Most people answer by walking through the house in memory and counting. Nobody counted the windows as a child. The number is new knowledge, and the only thing you consulted was memory. That looks generative to me.

    Helpful · 2