Is there a seeming contradiction in Kumārila’s account of I-cognition?

Suguru Ishimura and I are currently preparing a heavily annotated English translation of Ślokavārttika ātmavāda, verses 107–139, the section concerned with I-cognition (ahampratyaya).  The translation is to appear in a Kumārila Reader, edited by Elisa Freschi and Nilanjan Das.  After looking at our first draft, Elisa recently published two posts on this blog on the topic of Kumārila and I-cognition.  I’d like to respond to both of these: to post two responses to her first and two responses to her second.  Below is my first response to Elisa’s first post:
https://indianphilosophyblog.org/2026/02/16/does-kumarila-accept-i-cognition-as-a-kind-of-perception/

Elisa’s post begins:

Kumārila is an extremely systematic thinker. Thus, if there is a seeming contradiction in Kumārila’s thought, it is likely the case that the contradiction is only a seeming one and that it can be solved.

I completely agree that seeming contradictions in Kumārila’s oeuvre have the habit of dissolving when familiarity with other relevant parts of his system is achieved.  In fact in this case I question whether we even have a seeming contradiction in the first place.

What is the seeming contradiction that Elisa claims to have identified?  She sees evidence that makes her question whether Kumārila accepted I-cognition as a kind of perception.  She claims that the “I” cannot be “known through perception”, because of Kumārila’s “stating in the pratyakṣa chapter within the ŚV that perception is sense-perception”.

No reference is given there for where in the pratyakṣa chapter of the Ślokavārttika Kumārila states this.  Elisa’s formulation (“perception is sense-perception”) allows of two interpretations:

1. All perception involves one of the five external sense-faculties
2. All perception involves one of the six sense-faculties, i.e. the five external sense-faculties and the inner sense, the manas.

Only if Kumārila says somewhere that all perception involves one of the five external sense-faculties (claim 1) do we have any “seeming contradiction”.  If his claim is rather that perception involves one of the six sense-faculties (claim 2) then there is no difficulty at all in I-cognition being perception: it can be mānasa-pratyakṣa.  So I think the very foundation of the blog post is shaky.  For surely Kumārila nowhere says that all perception involves one of the five external sense-faculties?

That Elisa is assuming that Kumārila does somewhere claim that all perception involves one of the five external sense-faculties (claim 1) is suggested not only by the fact that she’s seeing a “seeming contradiction”, but also by the following.  She writes “Could it be that Kumārila accepts perception as sense-perception AND ahampratyaya?”, wondering here whether Kumārila included within the domain of perception two different things: sense-perception and I-cognition.  But these are only two different things if by sense-perception Kumārila means what is produced by one of the five senses.  If he includes within perception some cognitions produced by the manas, i.e. allows for the validity of mānasa-pratyakṣa, then I-cognition will not exist over and above the other kinds of perception that Kumārila accepts; it will just be one kind of mānasa-pratyakṣa – one kind of “sense-perception” in interpretation 2.  And it is surely more likely that Kumārila includes mānasa-pratyakṣa as a kind of perception than that he excludes it.  If he excludes it, what are we to make of all his discussions of perception of pleasure and pain in the pratyakṣa chapter?

So this is the situation as I see it:

• Elisa sees a seeming contradiction between I-cognition being perception and an unnamed place in the pratyakṣa chapter of Ślokavārttika where Kumārila claims that all perception is sense-perception.
• There is certainly no seeming contradiction if “sense-perception” there includes mānasa-pratyakṣa.
• Therefore Elisa thinks that somewhere in the pratyakṣa chapter of Ślokavārttika, Kumārila denies that mānasa-pratyakṣa is pratyakṣa.

Now I don’t think Elisa can really think that Kumārila denies that mānasa-pratyakṣa is pratyakṣa.  So what has gone wrong?  How has she arrived at the surely mistaken view that Kumārila does not accept mānasa-pratyakṣa?  (If she hadn’t arrived at this view then she would not have seen any seeming contradiction with I-cognition’s being perception.)

These are the possibilities I have come up with. 

• She finds Kumārila arguing against svasaṃvedana in the pratyakṣa chapter of Ślokavārttika.  She thinks svasaṃvedana is included in the Buddhist category of mānasa-pratyakṣa.  In fact it is not: the two are separate for the Buddhists.

• She finds Kumārila arguing against svasaṃvedana in the pratyakṣa chapter of Ślokavārttika.  She sees some of these arguments as hingeing on the need for perception to have some instrument/faculty/organ.  She thinks the instrument must be one of the five external sense-faculties – forgetting that it can be the manas.

• She finds Kumārila arguing against the validity of the Buddhist’s mānasa-pratyakṣa, and takes that to mean that Kumārila is against mānasa-pratyakṣa more widely.  But being against Buddhist mānasa-pratyakṣa does not mean one is against mānasa-pratyakṣa more widely.

In order to further investigate the seeming contradiction between

• I-cognition being perception
  
• what Elisa refers to as Kumārila’s claim “in the pratyakṣa chapter within the ŚV that perception is sense-perception”

I would have to know exactly what verse or verses in that section of the Ślokavārttika Elisa sees as making that claim.  If there is a verse claiming that the instrument of all perception has to be one of the five external sense-faculties, that would indeed imply that the “I” cannot be grasped perceptually, because it is indeed beyond the reach of any of the five external sense-faculties.  But such a verse would present us with a much bigger problem than that it contradicts the evidence that Kumārila held the self to be perceived by I-cognition; it would contradict all of the places where Kumārila accepts perception of pleasure and pain by means of the manas.

So I suspect that any Kumārila verse claiming that all “perception is sense-perception” is best interpreted as claiming that all perception involves one of the six sense-faculties.  If so, then the instrument of I-cognition can be the manas, so we have no seeming contradiction.

***

Here are a few comments on other parts of Elisa’s post.

Could it be that ahampratyaya is just a case of recognition? In favour of this view run a few passages in Pārthasārathi’s commentary, where he says pratyabhijñārūpeṇa ahampratyayena (ad ŚV ātmavāda v. 109) and ekasantānasambandhino ‘hampratyayā ekajñātṛviṣayā iti pratyabhijñeti (ad ŚV ātmavāda v. 139).

The second statement has nothing to do with recognition.  Elisa has mistyped pratijñeti as pratyabhijñēti, taking this sentence to be about recognition, whereas it is actually about the “assertion” or “thesis statement”, pratijñā, of the inference given in the verse under comment (Ślokavārttika ātmavāda 139).  Not only does this make much better sense, it is the reading of all witnesses (the printed editions, the Asiatic Society manuscript and the Varanasi manuscript).* It makes better sense because (1) the verse being commented on does not concern recognition, (2) what precedes the iti in Pārthasārathi’s sentence better expresses a pratijñā than a pratyabhijñā, (3) the point is that the pratijñā of the inference in the verse was not articulated there, so Pārthasārathi articulates it here.

The meaning of Pārthasārathi’s statement is: “The thesis statement [of the inference expressed in 139] is: All I-cognitions connected with a certain stream of consciousness have the same cognizer as their object.”  The sentence speaks of a plurality of separate I-cognitions having the same object, not of a single recognition identifying a past and a present object.  Since it does not concern recognition, Elisa’s use of it to give evidence concerning recognition is not fitting.

* Note that in order to facilitate our translation of Ślokavārttika ātmavāda 107–139, Ishimura produced a first edition of Sucarita’s commentary (Kāśikāṭīkā), and better editions of the verses and Pārthasārathi’s commentary (Nyāyaratnākara):
https://independent.academia.edu/SIshimura
His edition of the Nyāyaratnākara not only makes many improvements, it also supplies new text where the printed editions turned out to be lacunose.

The other phrase of Pārthasārathi that Elisa cites is pratyabhijñārūpeṇa ahampratyayena (ad Ślokavārttika ātmavāda v. 109).  This is the context of the phrase: jñātṛpratyabhijñā tv iyam ucyate bhāṣyakāreṇa “vayam eva hyaḥ, vayam evādya” iti. tena jñātṛviṣayeṇa pratyabhijñārūpeṇa ahampratyayena jñānād anyasya sthirasya jñātuḥ siddhir iti.  So he cites a formulation from the Śābarabhāṣya of recognition of the cognizer (jñātṛpratyabhijñā; elsewhere – ad 136 – he terms it ātmapratyabhijñā) and then describes it as an “I-cognition, having the nature of recognition, focussed on the cognizer”.  This is indeed significant – in that it shows that Pārthasārathi regards self-recognition as a kind of I-cognition.  But it certainly doesn’t show that he regards all I-cognition as recognition.  For a detailed treatment of the relationship between I-cognition and self-recognition in Kumārila, see our (Watson and Ishimura’s) forthcoming translation of Ślokavārttika ātmavāda 107–139.

Elisa’s sentence continues in this way:

Could it be that ahampratyaya is just a case of recognition? In favour of this view run a few passages … as well as the fact that the self is proven to exist because of phenomena like desire, which need some extension through time (perception of X, memory that X in the past produced pleasure, desire of X, see vv. 104–105).

Here Elisa is running together things accepted by Kumārila with things not accepted by him. She’s seeing something he does not actually accept as evidence for his holding that “ahampratyaya is just a case of recognition”.  The Nyāya and Vaiśeṣika inferences of the self recounted by Kumārila in verses 92cd–106 – including those she refers to in 104–105 – are not accepted by him.  They are recounted in order to show how they can be refuted by Buddhist arguments – in order to show the value of the Mīmāṃsā idea that the self can be perceived.

What is the evidence for my claim that Kumārila does not accept these inferences?  

• See verse 92cd (at the beginning of the passage) where Kumārila mentions that the “inferences of the self brought forward formerly by others” are refuted by the Bhāṣyakāra.
• Kumārila refutes them with Buddhist arguments in 101–106.
• Then in 107 he states that: “The inferences taught by others (i.e. Naiyāyikas and Vaiśeṣikas) having been repudiated in this way, the self is now ascertained as that which is known by itself through I- cognition.”
• Kumārila here reflects accurately what is going on in the Bhāṣya (pp. 50,13–56,6), where the Buddhist interlocutor is given the final word in the discussion of all of the following Nyāya arguments for the self:
  • that from the requirement for a possessor of pleasure and such like (sukhādi)
  • that based on the need to postulate an agent of cognition in order to account for the linguistic usage “he/she cognizes” (jānāti)
  • that from desire (icchā)
  • that from memory (smṛti)

So the verses Elisa mentions are certainly not concerned with the kind of recognition that Kumārila accepts of the self and discusses in verses 115ff.

One might wonder whether ahampratyaya could ever count as knowledge, given that it is unfalsifiable. However,
a. Kumārila has defeated the alternatives already (mainly bhūtacaitanyavāda, or physicalism and the non-physicalist illusionism of Buddhist epistemologists)
b. Kumārila showed at least in one case that he is comfortable with an unfalsifiable knowledge, namely the Veda.

Relevant here is ātmavāda 125cd, where Kumārila explicitly says there is no falsifier (bādhaka) of I-cognition.  That is compatible both with it being falsifiable (but as it happens unfalsified) or unfalsifiable.  Elisa’s position is the second, but there is some evidence for the first.

(1) Those who claim that I-cognition lacks a falsifier often articulate what would count as a falsifier.  For Sucarita (see Kāśikāṭīkā ad 125cd) an example of a falsifying cognition would be a memory that presents the subject of an earlier experience as not me: I.e. one that suggests that we are wrong to say “It was me who saw that”. 

For Śaṅkara (Brahmasūtrabhāṣya ad 2.2.25) an example would be: “I am now remembering a visual experience in which it was not me but someone else who was seeing” (ahaṃ smarāmi, adrākṣīd anyaḥ;smarāmy aham, asāv ado ’drākṣīt).  Later in the passage he writes, na “nāham”ity ātmano darśanaṃ nirvṛttaṃ nihnute: “There is no cognition “It wasn’t me [who had the earlier experience]”, which contradicts the perception of the self that has occurred”.  “Perception of the self” refers back to an experience Śaṅkara has mentioned of oneself as identical with the subject of an earlier experience.

So we’re certainly not dealing with the kind of unfalsifiability where it’s not even possible to give the verbal formulation of a cognition that would count as a falsifier.  Elisa could say that cognitions having this kind of verbal formulation never occur.

But even if that were true, the evidence such as we find in Sucarita and Śaṅkara certainly places I-cognitions in a different category from cognitions of the meaning of Vedic sentences.  The latter, being for Mīmāṃsakas the only authoritative source for matters of what we are to do, cannot in principle be falsified by anything else.  Empirical investigation is restricted to establishing facts (things which are siddha, not sādhya), so there is no possibility of any empirical discovery falsifying what we have learnt about our duties from the Veda (no possibility of deriving an “ought” from an “is”).  But we can name an empirical discovery that would overturn the validity of I-cognition: the occurrence of a cognition such as “It was not I who experienced that headache yesterday”.

(2) These kinds of cognition, furthermore, do actually arise in certain kinds of so-called “mental illness”.  The empirical psychology literature on their existence has been drawn on in the context of this Indian self-debate by Tom Tillemans, Roy Perrett and Monima Chadha.

(3) The Buddhists claim we have this kind of cognition in yogic perception (yogipratyakṣa), where the subject of the present cognition appears as utterly distinct from the subjects of all previous cognitions in my stream (santāna).

(4) I-cognition is frequently described as abādhita; I have never seen it described as abādhya.

Unfortunately, Uṃveka’s commentary is not available for this chapter and Pārthsārathi does not add much to this short section.

As I mentioned above, Ishimura has produced an edition of Sucarita’s commentary (Kāśikāṭīkā) on verses 107–148, this part of his commentary never having been published before:

https://independent.academia.edu/SIshimura

So we now have not only Pārthasārathi but also Sucarita to help with the interpretation of this part of the ātmavāda chapter.  For what Elisa refers to as “this short section” (142–145), Sucarita’s commentary is more than twice as long as Pārthasārathi’s.