Across four groups this vault has recorded the same finding in different words. The Kalyāṇ Hanumān Aṅka prints the costs of Hanuman's actions and has no category in which they weigh anything.
Rākṣasa wives miscarrying indoors from the sound of a battle-cry, recorded twice by different contributors and treated by both as escalation. A garden destroyed and the destruction read as reconnaissance. A ritual sabotaged and the sabotage filed under wit.34
⚑ The finding was that the corpus is not concealing these. It is printing them and lacking a place to put them.
The comparative section shows the place existed the whole time.
Skanda Purāṇa, Avanti-khaṇḍa, chapter 83: Hanuman installs the Hanumanteśvara-liṅga on the bank of the Narmadā — in expiation for killing a Brahmin.1
⚑ A named act. A named sin. A named remedy. A shrine that is still there.
The item is four words long in the essay and every one of them does work.
Killing a Brahmin is not a generic wrong in this literature. ⚑ Brahmahatyā is the paradigm case of the gravest class of offence — the standard example, the one the expiation literature is organised around. It is not a regrettable side-effect. It is the thing the category was built for.
And the remedy is not remorse. ⚠ It is an installation. ⚑ He builds something, at a specific place, and the thing he builds outlasts him and can be visited.
Chapter 84 continues it: Kapitīrtha is established nearby, and Rāma and Lakṣmaṇa install two further liṅgas on the way back from Laṅkā.1
⚑ So the expiation is not private and it is not singular. It becomes a site, the site acquires companions, and the companions are installed by the two people the whole campaign was for. ⚠ A shrine cluster is the most durable form of record a pre-modern tradition has, and this one exists because somebody did something wrong.
There is a second one, in a different Purāṇa, for a different killing, and the two must be kept apart.
Brahma Vaivarta Purāṇa, Brahma-khaṇḍa 46: Hanuman is sent to fetch a liṅga from Kailāsa in order to atone for killing Rāvaṇa.1
⚑ Killing Rāvaṇa requires atonement. ⚠ The volume's other four groups treat that killing as the campaign's purpose and its vindication — and here it is a debt.
The reason is available and the essay does not state it: Rāvaṇa was a Brahmin by descent. ⚑ Which would make both expiations instances of one category — and this build does not assert that, because the essay gives no reason at all and the inference, however natural, is not in the source.
⚠ What is in the source is that the two are separate. Different Purāṇas, different rivers, different sins, different shrines. ⚑ The tradition did not build one all-purpose apology. It kept a ledger.
The whole value of this material is that it is a different move from the one the volume otherwise makes, and the difference is structural.
The volume's standard handling of a cost is a justification. ⚑ The grove destruction had three calculated purposes.5 The one-syllable sabotage was wit. The battle-cry's effects were escalation. ⚠ Each converts the cost into evidence of something admirable, and each therefore removes the cost — after the justification there is nothing left over.
An expiation does the opposite. ⚑ It concedes the act was wrong, leaves it wrong, and adds an obligation. Nothing is reinterpreted. The killing is still a killing after the liṅga is installed — which is why the liṅga has to be installed.
⚠ And that is exactly what a category for cost is. ⚑ A justification is a way of not having one. A tradition that can only justify has no method of accounting; a tradition that can also expiate has a second column.
⚑ Which sharpens the build's standing finding rather than overturning it. ⚠ The volume does lack the category — in its devotional essays. The category exists in the Purāṇic literature those essays are drawing on, and the survey essay that reports it does so without noticing what it is reporting. ⚑ The failure is not that the tradition has no place to weigh a cost. It is that the essays defending Hanuman never reach for the one it has.
The distribution is not random and it follows from what each register can afford.
⚑ An essay arguing that Hanuman is the ideal devotee cannot use an expiation, because an expiation entails a sin, and a sin entails a fault that the essay's whole argument denies. ⚠ The instrument is unavailable to precisely the writers who most need a way of handling the costs they print.
A survey essay has no such problem. ⚑ Upādhyāy is not defending anything. He lists what the Purāṇas contain, and what they contain includes a Brahmin-killing with a shrine attached. He gives it the same treatment as the birth accounts and the boast — a locator and a sentence.
⚠ So the volume's two halves are not disagreeing. They are running incompatible accounting systems and never meet, and this build has now recorded the same split three ways: who is allowed to concede a lapse, who prints a boast without flinching, and who can name a sin. ⚑ In every case the survey and comparative registers carry the harder material and the devotional register carries the argument.
The find is real and its reach should be stated exactly, because the temptation is to over-claim it.
⚑ The expiations cover killings. A Brahmin, and Rāvaṇa. ⚠ They do not cover the costs this build has actually been recording.
The rākṣasa wives who miscarried from a sound were not killed by anybody. ⚑ No agent, no act aimed at them, and therefore nothing for an expiation to attach to. ⚠ The category is built around a deed with a victim and a doer, and collateral effects have neither.
This is the same shortfall the liberation-of-the-killed doctrine had.6 ⚑ That doctrine covers those killed by the incarnation or its agents and says nothing about the miscarriages. This one covers those killed by Hanuman and says nothing about them either. ⚠ Two independent categories for cost, from two different registers of the tradition, and the same material falls through both.
⚑ Which is now a considerably stronger version of the build's finding. It is no longer the volume has no category. It is that the volume has at least two, they were built for deeds rather than for effects, and the one cost this build keeps returning to is an effect.
The most transferable thing here is the form the acknowledgement takes.
⚑ The expiation is a building. Not a statement, not a confession, not a passage in a text — a physical installation at a named place, which persists, can be visited, and requires no one to remember anything.
⚠ And that solves a problem statements have. A confession is an event; it happens once, to whoever was present, and then depends on transmission. ⚑ A shrine is a standing claim. Anybody who arrives at that bend of the Narmadā encounters the fact that something needed expiating, whether or not the story travelled with them.
⚑ It is also non-defensive in a way words cannot be. ⚠ Every verbal acknowledgement carries its own framing — the tone, the extenuations, the ordering — and this build has recorded three contributors timing a recovery in the sentence after a lapse.7 A liṅga has no tone. It says only that this happened here and something was owed.
Which is the thing worth taking out of the cosmology. ⚑ An acknowledgement that survives its author's framing has to be structural rather than rhetorical — a fixture, a record, a standing item — and the tradition reached for stone because stone is the only medium that cannot be re-narrated.
When something has gone wrong, notice whether you are justifying or accounting. ⚑ A justification explains the cost away and leaves nothing outstanding; an accounting concedes it and adds an obligation. ⚠ The test is whether anything remains to be done after you have finished speaking. If not, it was a justification, however honest it felt.
Make the acknowledgement structural rather than verbal where you can. ⚑ A statement depends on transmission and carries your framing; a fixture does not — a standing item on a checklist, a permanent entry in a register, a named process that exists because of the failure. ⚠ The tradition's version was a shrine, and the property that mattered was that it could not be re-narrated by whoever tells the story next.
⚑ And check what your category is built around. A remedy designed for deeds will not reach effects, and effects are where most real harm sits — the sound that carried, the second-order consequence, the thing nobody aimed at. ⚠ Two separate remedial systems in this volume both miss the same case, which suggests the gap is a property of deed-shaped categories rather than an oversight by either.
One Aṅka essay for both expiations,1 a second for the group's method,2 and four of this build's pages for the standing finding and the comparisons.34567
Bounded. 🚩 SINGLE SOURCE. ⚑ Both expiations are reported by one contributor. The chapter locators are given — Skanda Avanti-khaṇḍa 83 and 84, Brahma Vaivarta Brahma-khaṇḍa 46 — and this build has checked none of them. [TRADITION-CLAIM — verified] as the essay's citations; not verified against the Purāṇas.
Bounded again, and this one matters. ⚠ The essay gives no circumstances for the Brahmin-killing — no name, no occasion, no account of how it happened. ⚑ The whole item is that an expiation was performed and a shrine installed. Nothing here claims to know what the act was.
⚠ And the connection between the two expiations is not made by the source. ⚑ The natural inference — that Rāvaṇa's Brahmin descent makes both instances of one category — is not in the essay, and is not asserted here.
🔴 A tension preserved. Killing Rāvaṇa requires atonement in the Purāṇic material and is the campaign's vindication in the devotional material. The volume prints both and no contributor notices.
Open question, and it is checkable. ⚑ Do the Character & Virtue or Devotion to Rāma essays mention either shrine? This build has read thirty-seven essays across those two groups and recorded no reference to Hanumanteśvara, Kapitīrtha or Hanumal-liṅga — ⚠ but it was not looking for them, and the absence is stated as not found in the range read rather than as an absence.
Upādhyāy is the only contributor in this build who reports a fault of Hanuman's without any evident awareness of doing so, and the reason is his register.
⚑ His essay is a survey. It lists what the Purāṇas contain, in order, with locators — birth accounts, tīrthas, liṅgas, boasts, a viśvarūpa — and it applies no hierarchy of authority and no dating whatever.1 ⚠ Three competing birth narratives are laid side by side with no judgement about which is earlier or weightier.
⚑ That is the volume's kalpa-plurality machinery, and this build has praised it as the compendium's most portable contribution.2 Here it has a cost nobody has recorded.
⚠ A survey that refuses to rank also refuses to notice. ⚑ The Brahmin-killing arrives between a curse-motive and a tīrtha, in the same tone, at the same weight — and the machinery that lets the volume hold three birth stories without collapsing is the same machinery that lets it hold a mortal sin without registering it. The device is a levelling one, and it levels in both directions.
⚑ Which is worth stating plainly because it is the first cost this build has found in that apparatus. Publishing your interpretive licences makes a tradition arguable. It also means nothing in it is ever foregrounded — and a reader who wants to know what the tradition thinks is serious will not learn it from a survey.
Never Apologize, Never Explain. That page holds Greene's advice for the moment after something you did has gone wrong in front of people: a true, good, ninety-second sentence is forming — the constraint, the missing information, the reasonable call that turned out badly — and the counsel is to not open your mouth. ⚑ The Purāṇic response is the opposite in method and, oddly, the same in one respect: it does not explain either.
What the pairing produces is that Greene and the Purāṇa agree the explanation is worthless and disagree about what to do instead. ⚠ Greene's alternative is silence, which leaves the fault unmarked and lets the audience's attention move on. ⚑ The tradition's alternative is a fixture, which leaves the fault marked permanently and takes the framing out of everybody's hands. Both refuse the ninety seconds; only one refuses because the ninety seconds are too weak rather than too costly.
⚑ And the pairing names the variable that decides between them: whether you need the fault forgotten or need it settled. ⚠ Silence works when the goal is to survive the moment, and Greene is explicit that the sentence loses status even when true. It fails completely when something is genuinely owed, because an unmarked debt does not stop being a debt. ⚑ The strategic literature has no instrument for a wrong that must be conceded and cannot be talked away, and the expiation is exactly that instrument — which is why it is stone and not speech.
Tishya-Day Enforced Public Reading as Accountability. That page holds Aśoka's mechanism for making an edict bind: it is read aloud on the days the moon enters Tishya — roughly three times a year — by designated officials to assembled subjects, with the schedule written into the edict itself, plus solo readings by officials in between. ⚑ A commitment given a calendar so that it cannot quietly lapse.
What the pairing yields is that both solve the same problem with different coordinates. ⚠ A commitment or an admission decays because nothing forces re-encounter. ⚑ Aśoka fixes it in time — three occasions a year, mandated. The expiation fixes it in space — a bend of the Narmadā, permanently. Neither depends on anybody remembering, and both remove the discretion of whoever would rather not bring it up.
⚑ And the difference is instructive. ⚠ The calendar needs an administration to keep it; when the officials stop reading, the edict stops binding, which is why Aśoka had to specify who and when. The shrine needs nobody. ⚑ A spatial fixture is the lower-maintenance instrument and reaches only whoever comes to the place; a temporal one reaches everybody assembled and requires an institution to survive. Which is a real design choice available outside either cosmology: an admission can be made unavoidable in time or unavoidable in place, and the two fail in different directions.
The uncomfortable possibility is that the expiations are etiologies rather than acknowledgements.
⚑ Shrines generate stories. A liṅga on the Narmadā called Hanumanteśvara needs an explanation for its name, and Hanuman installed it to expiate a killing is exactly the kind of account that gets attached to an existing site rather than an account of why the site exists. ⚠ On that reading the sin is downstream of the shrine, not upstream.
Two things point that way. ⚑ The essay gives no circumstances for the Brahmin-killing at all — no name, no occasion, no narrative — which is what a back-formed cause looks like. ⚠ And the Rāmeśvaram material in the same chapter is unmistakably etiological, explaining why there are two liṅgas at a site rather than one.
⚑ But it does not dissolve the finding, and this is the important part. ⚠ Even if the sin was invented to explain the stone, the category was not. Somebody reaching for a story to explain a Hanuman shrine reached for expiation for killing a Brahmin — which means that in the imaginative world of the people telling it, Hanuman was the kind of figure who could incur that debt. ⚑ The etiology reading changes what the shrine is evidence of and leaves intact the thing this page is actually about: the tradition had a category, and was willing to put Hanuman in it.