Menā has just seen her daughter's fiancé for the first time and he is worse than she feared — matted hair, ash, snakes, a bull. She loses control completely. She screams at the sages, at her sons, at Nārada, at her husband. Then she turns on her daughter and says this:
"You have yourself wickedly given away gold and brought a piece of glass instead. Having given up sandalwood paste, you have smeared yourself in mud... You have cast aside the sun and have carefully collected a firefly."1
It's a good line. Three images, escalating, the kind of thing a furious parent produces under pressure.
Except it isn't hers.
Seventeen chapters earlier, an old brahmana came to Pārvatī in the forest and said, almost word for word:
"Handing over a golden coin, you wish to accept a piece of glass. Giving up pure sandalwood paste, you desire to smear yourself with mud... Giving up the energy of the sun, you are wishing for a firefly."2
Same three figures. Same order. And the old brahmana was Śiva in disguise, disparaging himself.
Menā was not there. Nobody who was there is in this room. She has reproduced, under genuine grief, a speech her daughter's husband wrote about himself.
Set them side by side, because the correspondence is what carries the page and it needs to be checkable rather than asserted.
| Ch 122 — Śiva, disguised, to Pārvatī | Ch 139 — Menā, in grief, to Pārvatī | |
|---|---|---|
| 1 | "Handing over a golden coin, you wish to accept a piece of glass." | "You have yourself wickedly given away gold and brought a piece of glass instead." |
| 2 | "Giving up pure sandalwood paste, you desire to smear yourself with mud." | "Having given up sandalwood paste, you have smeared yourself in mud." |
| 3 | "Giving up the energy of the sun, you are wishing for a firefly." | "You have cast aside the sun and have carefully collected a firefly." |
Three pairs, same sequence, same grammatical shape — give up the valuable, take the worthless.
Debroy does not flag it. Across an endnote layer that supplies a missing negative, announces "we have deviated from the text," declares that certain sentences don't belong where they are, and pre-emptively defends the text against a seam-reading nobody made, this goes past unmarked.3
Both passages sit inside longer lists, and the surrounding items rhyme too.
Chapter 122's version keeps going: give up silk from China for a hide, give up a home for the forest, give up treasures for a lump of iron, give up Indra and the world-guardians for Śiva. Then it moves into the paired-attributes catalogue — your eyes are lotus petals and he has three, your face is the moon and he has five faces, you have sandalwood and he has ashes, you have a girdle and he has an elephant hide, you have ornaments and he has serpents.
Chapter 139's version keeps going too: you made the swan fly away and clutched a crow, you threw away sanctified water and drank from a well, you threw away rice and are eating husk, you gave up serving a lion and are serving a jackal.
So it isn't one borrowed sentence. It's a set — an anti-Śiva disparagement kit, with a fixed core and a variable tail.
And once you know the kit exists you find it four times.
Chapter 120. The Saptarṣis, commissioned by Śiva to lie: he's a naked yogi, no home, no lineage, keeps company with pretas and bhūtas, "crooked and will use his own maya to destroy your vijnana."4
Chapter 122. Śiva as the brahmacārin Jaṭila, in person, with the gold/sandalwood/sun core.2
Chapter 126. Śiva as a Vaiṣṇava brahmana claiming to be a professional matchmaker: "He is malformed and possesses no qualities... His name and lineage are not known. His conduct is vile... He is deceitful and has abandoned the path of the Vedas."5
Chapter 139. Menā, who was present at none of these, reproducing the core in order.1
[n≥3 PATTERN] on the block.
Here is what turns a compositional observation into something with teeth.
The Saptarṣis' version was commissioned by Śiva — "use deceitful words and be harsh." The brahmacārin was Śiva. The Vaiṣṇava matchmaker was Śiva, and he confesses it four chapters later: "Urged by devas, to diminish their devotion, I assumed the form of Vishnu's devotee and criticized myself."6
Three of the four deployments are authored by the man being slandered.
And then the fourth comes out of a woman in real distress, at the worst moment of her life, using his figures.
The chapter 139 instance is different in kind from the other three and the difference is the whole finding.
The first three are performances. Everyone speaking knows the content is a tactic. The Saptarṣis are described as loving Pārvatī while they say it. Śiva is describing himself. There is no sincerity anywhere in the first three deployments, and the text says so.
Menā means every word.
Look at what surrounds her version. She has just asked why she wasn't born barren, why her foetus didn't miscarry, why her daughter didn't die. She says "I will slice off your head now." She tells her husband to bind the girl by the throat and throw her down, or drown her in the ocean. She asks Brahmā — to his face — "Why don't you kill this Shivaa yourself?" And she physically attacks her daughter: "she gnashed her teeth and seized her body, beating her with her fists and knees," until the assembled sages pull the girl out of her hands.7
That is not a rhetorical performance. And in the middle of it, the gold and the glass and the firefly.
The volume also tells you, in the same chapter, what is running her: "Mena did not accept the idea because she was confounded by Shiva's maya" and "The blessed one was goaded by Shiva's wishes, which confound everyone."8
So the fourth deployment of Śiva's self-slander comes out of a woman Śiva's own māyā is driving, in a rage Śiva's own māyā produced, while she tries to kill the daughter he intends to marry.
You think you've found a repeated block in a long text and you want to know whether it's real.
Quote both instances in full and lay them in a table before you interpret anything. Not paraphrase — the actual sentences, aligned. A repeated block is a claim about wording, and wording is the only evidence for it. If the correspondence survives being set out plainly, you have something; if it needs your summary to hold together, you don't.
Then check the order. Shared imagery is weak — gold-versus-glass is a common enough figure that two authors could reach it independently. Shared imagery in sequence is much stronger, because sequence is arbitrary. Three figures in matching order is the load-bearing fact here.
Then look for the tail and see if it varies. A block with a fixed core and a variable tail is behaving like a formula rather than a quotation. Chapter 122 continues into silk and iron; chapter 139 continues into swans and jackals. Same spine, different limbs.
Then count the deployments and identify the speakers. This is where a stylistic observation becomes an argument about the text. Four uses is a pattern; four uses of which three belong to the person being described is a structure.
Then check whether the translator caught it, and record the answer either way. He didn't. That is a fact about the instrument this build has been using, and it belongs in the finding.
Be honest about what this supports, because it is the kind of find that invites more than it can carry.
Reading one: it's stock. Purāṇic composition reuses set-pieces constantly. An anti-ascetic disparagement kit would be exactly the sort of thing a compiler keeps to hand and deploys whenever a character needs to run Śiva down. On this reading the ch 122/139 correspondence is a compositional fingerprint and means nothing about Menā.
Reading two: it propagated. The volume shows Śiva manufacturing a slander against himself and putting it into circulation, and then shows it arriving, intact, in the mouth of someone who believes it and acts on it violently. On this reading the block is doing narrative work: it's the receipt for what a manufactured slander does once it's loose.
The text supports both. It never connects the two passages, never has anyone notice, and never comments.
What it does not support is any claim about compilation history, or about how many hands are involved, or about which passage came first. A repeated block establishes repetition. It does not date anything.
What would sharpen it: a fifth deployment, or the same core turning up attached to a figure other than Śiva. Volume 1 is now fully read and contains neither. Volumes 2 and 3 are out of scope.
One handshake. A second candidate — reading the kit against propaganda-seeding in the vault's influence corpus — kept arriving at the same insight through different vocabulary, so it's cut rather than padded.
The Manufactured Overt Act — same architecture with the payload changed, and the change is what the pairing is worth.
The manufactured overt act is an influence technique in which an operator engineers a situation that produces a real, attributable action by the target — and the action is then genuine evidence, because it really happened. Its power comes from the gap between authorship and attribution: the operator wrote the script, the target performed it, and everything downstream treats the performance as the target's own. Provocation cases turn on exactly this, and the reason they're hard to unwind is that nothing in the record is false.
Chapter 139 runs the same split with speech instead of action. Śiva authors the disparagement; Menā delivers it; and by the time it reaches Pārvatī it is unmistakably Menā's — spoken in her voice, from her grief, backed by her violence. No one in the scene could reconstruct the authorship, because the two deployments are seventeen chapters and one costume apart.
The insight the pairing produces is a distinction the influence material tends to blur: a manufactured act damages the person who performs it, but manufactured speech damages the person who believes it, and the two require different remedies. In the provocation case the target's problem is the record — they did the thing, and clearing it means proving authorship. Menā's problem is not the record; nobody accuses her of anything. Her problem is that the content is now load-bearing in her own reasoning, and she will defend it against eight separate authorities because it feels like something she worked out. Read practically: authored speech, once believed, becomes indistinguishable from conviction, and the usual counter — showing the target the authorship — is unavailable when the author has an interest in never disclosing it. Śiva confesses the operation to the Saptarṣis and never to her.
Strongest support. Three figures in matching order across two passages seventeen chapters apart, quotable in full.12 Four deployments of the wider kit in twenty chapters.1245 Śiva's authorship of three of the four is stated in his own mouth or the narrator's.6 The translator does not flag any of it, in an endnote range where he flags a great deal.3
Tension, preserved. Stock formula versus narrative propagation. Both readings survive every sentence on the page and the volume never adjudicates.
Tension, second. The kit's content is accurate. Śiva is naked, houseless and without lineage — the volume asserts all three constantly and treats them as credentials. So what circulates is not a lie but a valuation, which is harder to correct and which chapter 143 turns into a procedural failure at the wedding.
Open. Does a fifth deployment exist? Not in Volume 1, now fully read. Volumes 2–3 out of scope.
Open. Does the core ever attach to a target other than Śiva? Not found in Volume 1.
Read this straight after The Test That Śiva Commissioned and the two pages want to merge, which is worth resisting because they are doing different work.
That page is about a single commissioned lie and the one true sentence buried in it. The finding there was local: seven sages, one brief, one damning admission about Satī. What made it uncomfortable was the frame — the truest thing in the volume, spoken by mouths the text had pre-discredited.
This page finds the same lie four times, and the fourth time it is spoken sincerely by someone with no access to the first three. That converts a scene into a mechanism. The commissioned slander of chapter 120 was not an isolated tactic; it was one deployment of a kit that keeps being used, and its final user believes it.
And it changes what Disabled by Māyā, Then Held Liable is describing. That page had four instances of an agent being impaired and then charged. Menā is a ninth instance, and she is the first where the impairment supplies not just the act but the script. The others were made to do things. She is made to say something specific — something already on the record, in the same book, in the mouth of the god whose māyā is driving her.
Sharpest implication. Śiva manufactures a disparagement of himself, deploys it three times through commissioned or disguised mouths, and it arrives seventeen chapters later — three figures, same order — in the sincere grief of a woman his own māyā is driving to attack his intended bride. The volume shows a slander being authored, circulated, and believed, and never once has anyone notice.
Generative questions.
The content of the kit is true. Every charge in it is something the volume elsewhere calls a credential. Does that make it slander at all — or is the thing circulating not a falsehood but a valuation, and is a valuation correctable?
Śiva confesses the operation to the Saptarṣis and never to Menā. Is there any point in the volume at which someone is told what was done to them?
If the block is stock rather than propagated, the compiler chose to give a grieving mother the same words as a disguised god. Is that indifference, or is it the only way the scene could have been written by someone who had the kit and needed it twice?