Uddhava is listing what Kṛṣṇa did as a child, and he arrives at Pūtanā in two sentences:
She had virulent poison in her breasts. The wicked one wished to kill him and nurtured him with this. However, she obtained a state that was appropriate for a proper nurse. Who is as compassionate as him?1
Intent: infanticide. Method: poisoned breasts. Outcome: the destination of a woman who nursed a child.
The passage does not say she repented, was forgiven, or was shown mercy on some ground extraneous to what she did. It says she got what a nurse gets, and the reason given is what she performed — she nurtured him — not what she meant by it.
The wicked one wished to kill him and nurtured him with this.1
Nurtured him with this. The poison is the instrument of nursing rather than an alternative to it. She did not attack him; she fed him, and the food was poison.
That is the hinge, and it is a matter of what act was actually performed. To poison an infant that way you have to hold it, offer the breast, and let it feed. Whatever she intended, the thing she did was nurse him.
And the accounting is done on the act. The state she obtains is appropriate for a proper nurse — appropriate to the office, filled, however badly meant.
The clause that follows is not decoration:
Who is as compassionate as him?1
Uddhava reads the outcome as generosity, and it is worth noticing that this is an interpretation, not a restatement. The passage could have been read as strict accounting — she performed a nurse's act, she received a nurse's wage, no compassion involved.
He does not read it that way, and the question is rhetorical, which means he expects Vidura to agree.
The reading he offers is that ignoring the intent is the mercy. A just accounting would have weighed what she meant. Declining to weigh it is the favour, and the favour is available to the worst-intentioned person in the story.
The sentence immediately after the Pūtanā case is the one that reveals what the example is for:
Who is as compassionate as him? Which refuge will we go to?1
That is not a rhetorical flourish. Uddhava is a survivor whose people have just annihilated themselves, talking to an exile whose family destroyed itself, and the question is literal: where do we go now.
And the Pūtanā example is the answer's premise. If the woman who came to poison him received a nurse's state, then the accounting is not done on worthiness, and two men with nothing left to offer are not thereby disqualified.
So the case is being deployed as a floor. Not look how kind he is in the abstract, but look at the worst applicant on record and what she received — which is the only kind of precedent that is any use to someone who has run out of merit.
Register: this is a reading of why the example appears where it does, in Uddhava's speech, immediately before the refuge question. The text does not state the connection. What can be shown is the adjacency and that the two sentences are consecutive.
Uddhava does not stop there. The next claim extends the principle past its plausible limit:
I think that the asuras are devotees of the illustrious lord of the three. They focused their intelligence along the path of enmity towards him, though they could see him descend with the weapon with the excellent nave, on the shoulders of Tarkshya's son.2
Enmity as a form of devotion, on the ground that it is sustained attention correctly aimed.
That is the same move as Pūtanā's, run on a whole class rather than an individual. In both cases the intent is hostile, the orientation is toward him, and the intent is discounted. The nurse gets a nurse's state; the enemies get counted among the devotees.
The two sentences are adjacent and the second is offered as I think, which marks it as Uddhava's inference rather than received doctrine. He is generalising, in real time, from the Pūtanā case to a conclusion he needs.
The principle is not free, and the volume does not itemise the charge.
If offices are credited by performance rather than by intent, then intent has been removed from a great deal of moral accounting. The volume elsewhere cares enormously about intent — Duryodhana's motives, Vidura's, the difference between the wicked and the virtuous runs through every chapter.
Here it does not count. And nothing in the passage marks the exception, or limits it to nurses, or says under what conditions performance overrides intent.
Which leaves the principle stated and unbounded. Read strictly it would license a great deal; read narrowly it is a single miracle about a single demoness. The text gives no basis for choosing, and Uddhava's immediate generalisation to the asuras suggests he read it broadly.
One detail worth pressing: the state she obtains is specifically a nurse's, not a devotee's or a liberated soul's.
That is a modest reward and an exactly fitted one. She is not elevated past what she did. She performed the office of nurse, badly and murderously, and receives the office's due — no more.
So the principle is narrower than he forgives everyone. It is the office you actually filled is the office you are paid for, and it happens that the worst possible motive did not prevent the office being filled.
Which is a rule about categories of act rather than about mercy at all — and makes Uddhava's who is as compassionate as him a generous gloss on something more like accounting. The two readings sit in the same two sentences and the passage does not separate them.
You are assessing something where the intent was bad and the effect was not — a colleague whose self-serving move happened to fix a real problem, a policy passed for the wrong reasons that works, help given for the sake of being seen giving it.
The passage runs the accounting one way and it is worth being deliberate about when to copy it.
Establish what office was actually filled. Not what was intended, not what was said afterwards — what act was performed, described plainly. She held an infant and fed it. That is the fact the credit attaches to.
Then pay the office's rate and no more. A nurse's state for a nurse's act. The failure mode here is inflation: crediting the act and then treating the person as having been well-motivated all along, which is a second and unearned payment. Pūtanā gets nothing beyond the office.
And keep the intent on the record separately. The wicked one wished to kill him is not deleted by the outcome; the passage says both, in order, in one sentence. You can credit an act and continue to hold that the motive was murderous, and a system that cannot hold both will end up either refusing to credit anything or laundering everyone.
The reason to run it this way is Uddhava's reason. If credit is only ever available to the well-motivated, then anyone whose record is bad has no route back, and which refuge will we go to has no answer.
The strongest evidence that the accounting is done on the act is the specificity of the reward. Not liberation, not devotion — a state appropriate for a proper nurse. The office performed determines the payment exactly.
Tension — the volume weighs intent heavily throughout and does not weigh it here, and marks no exception. The principle is stated without limits and immediately generalised by Uddhava to the asuras. Preserved unresolved.
Open question. Is the passage a rule or a miracle? Who is as compassionate as him reads it as an exercise of mercy; the fitted reward reads it as accounting. The two produce different scopes and the text supplies both without choosing.
This vault's Gauḍīya material treats the Pūtanā episode as the tradition's standard proof of Kṛṣṇa's disregard for the offerer's qualification — cited to show that the offering, however corrupt, establishes a relation.
What this passage adds is the frame. It is not cited here as doctrine. It is cited by a survivor, in a conversation about where to go now, one sentence before he asks which refuge is left and one sentence before he decides that the demons are devotees too.
That placement makes it a piece of reasoning under pressure rather than a settled teaching. The Asuras Are Devotees Too documents where the reasoning goes next, and it goes further than the tradition usually follows it. Whether the Pūtanā case supports that extension is the question the volume leaves open by putting them side by side and marking only the second as I think.
With Judge by Effects, Not Intentions. That page works the strategic discipline of assessing people and moves by what they produce rather than by what they meant — intent being unknowable, self-reported, and easily counterfeited, while effects are observable.
Pūtanā's case is that discipline applied at the extreme, with the intent not merely unknowable but known, stated, and murderous — and discounted anyway.
What the pairing produces: the strategic version discounts intent on epistemic grounds — you cannot see it, so do not price it. The Bhāgavata's version discounts it where the intent is fully visible and undisputed, which removes the epistemic justification entirely and exposes what is left. What remains is a claim about what kind of thing an act is: the office of nursing was filled, and offices are the units the ledger is kept in. That is a substantially stronger position than the strategic one, and it has a consequence the strategic framing avoids. If effects are used because intent is opaque, then better information about intent should change the judgement. If offices are the unit, better information changes nothing — and the volume, which everywhere else cares intensely about motive, here does not adjust. Neither account alone shows that the two justifications for the same practice come apart precisely when the intent becomes visible.
Sharpest implication. The reward is fitted to the office, not to the person: a nurse's state for the act of nursing, with the murder attached to it left on the record and unpaid. Which means this is not forgiveness — nothing is remitted — but a claim that acts are counted by what they are rather than by what they were for, and that a person can therefore be simultaneously credited and condemned for one motion of the body.
Generative questions.