CF-08: Mahiṣāsura: Two Slayings

A passage in the Skanda Purāṇa narrates that Hari, perceiving the Dānava's intent, filled His mouth with divine weapons; that Mahiṣa, struck, fell; and that Hari then revived him, telling him he was not destined to die at His hand but at a woman's, as Brahmā had declared, and dismissing him from the field.

Locator as circulated: Skanda Purāṇa 1.2.20.66–70. Grade C. The Skanda is the largest of the Purāṇas, exists in markedly divergent recensions, and its internal numbering is not stable across editions. We have not collated this locator. Register entry D-17.

It establishes nothing, and the Śākta reply writes itself: the Devī-Māhātmya narrates the same event with the Goddess as the sole agent, formed from the combined potencies of the devas — including Viṣṇu's — because none of them individually could accomplish it. Two Purāṇas, two narratives, opposite implications. Whoever produces his preferred Purāṇa and declares the matter settled has demonstrated only which Purāṇa he prefers.

And the Vaiṣṇava version has an evident apologetic shape: it concedes the famous event to the Goddess while inserting a prior, unfamous victory for Viṣṇu, which is what a harmonising interpolation looks like.

The pūrvapakṣa is essentially correct about the evidential situation, and the honest adjudication follows from admitting it rather than resisting it.

1. Narrative is tier T5 and rank six territory. Under §8.3, conclusions drawn from narrative report sit below the logical instruments, which sit below direct scriptural statement. Under §10.3, Purāṇic narrative is T5. A contest between two T5 narratives is structurally unresolvable at T5. This is not a weakness in the method; it is the method reporting, correctly, that the question has been asked at the wrong tier.

2. What the ranking actually does. The tattva question — whether the ultimate cause is Viṣṇu or the Devī — is settled, if at all, at T1–T3 by direct statement and samanvaya. That is where CF-07 argued it. Once settled there, a T5 narrative that agrees is admissible under rule C-01 — accepted because it agrees with the higher tiers, not adduced to establish them. Reversing that order is failure mode M13 and is the most common error in popular apologetics on every side.

3. What the narrative does contribute, if genuine. Not a ranking, but a mechanism. Its actual subject is the integrity of a boon: Brahmā had declared that Mahiṣa would fall to a woman, and the revival exists so that Brahmā's word should not be falsified. Read that way, the passage is about the reliability of the created order's guarantees — a theme with substantial support at higher tiers — rather than about who is stronger.

4. On the interpolation charge. We note it and decline to adjudicate it. Text-critical claims about Purāṇic recensions require manuscript work we have not done, and the tradition's counter-position — that the criteria by which a passage is judged interpolated are themselves theory-laden — is a real methodological point, not an evasion. Neither side's assertion is evidence.

5. On kalpa-bheda. Appeal to distinct cosmic cycles is available in the tradition and is not, in itself, illegitimate. But invoking it here, to dissolve this particular tension, would manufacture a resolution the evidence does not support: no source we have located assigns these two narratives to different kalpas, so the appeal would be supplying the very datum in dispute. Under step 8 of the conflict protocol it is refused. Earlier drafts of this file resolved the case by kalpa-bheda; that resolution has been withdrawn as failure mode M18.

Admissible, non-probative. The narrative is consistent with the conclusion established at T1–T3 and is therefore admitted under C-01. It does not establish that conclusion and is not cited anywhere in this document as establishing anything. Grade C pending collation.