Chapter 28 prices things in Parthiva lingams. A thousand for learning. Five hundred for wealth. A crore for emancipation. Seven hundred to kill a man.1
Chapter 46 prices the same things in Parthiva lingams. Ten crore for a kingdom. A hundred thousand for freedom from imprisonment. Five crore for emancipation. A hundred thousand to kill a man.2
Same goods, same currency, same volume.
And not one overlapping price agrees.
Put the overlaps side by side, because the sizes are the finding.
| Outcome | Ch 28 (Vidyeshvara) | Ch 46 (Rudra) | Factor |
|---|---|---|---|
| learning | 1,000 | 12,500 | ×12.5 |
| emancipation | 1 crore | 5 crore | ×5 |
| freedom from prison | 10,000 | 100,000 | ×10 |
| confound someone | 800 | 50,000 | ×62.5 |
| kill someone | 700 | 100,000 | ×143 |
Seven hundred lingams to kill a man in one chapter. A hundred thousand in another.
That is not a variant reading or a copyist's zero. It is two orders of magnitude, in the same volume, for the same act, denominated in the same unit.
Here is what makes this more than a list of discrepancies.
The disagreement has a direction. Every single overlapping price is higher in the Rudra Samhita. Not one goes down.
Five kinds of thing, five increases, ranging from fivefold to a hundred and forty-threefold. If these were independent traditions with independent numbers, you would expect roughly half to fall.
And this is not the only place the Rudra material inflates a Vidyeshvara specification. The morning protocol has the mouth rinsed eight times in chapter 20 and sixteen times in chapter 45 — a clean doubling.3 The ketaki flower's sentence in chapter 15 is partial, with the flower permitted to Shiva's followers and in canopies above him; by chapter 46 the ban is flat, with the concession gone.4
Where the Rudra Samhita restates a Vidyeshvara number, the number goes up. Where it restates a Vidyeshvara penalty, the mercy comes off.
That is a systematic drift rather than noise, and directional drift is much better evidence of two layers than disagreement alone would be. Noise says the numbers were never fixed. A ratchet says somebody was competing.
Before that gets too tidy, the same chapter contradicts itself internally, and badly.
Chapter 28 prices sons twice: "a person seeking sons must use one thousand and five hundred" — and, a few lines later, "a person who has no sons will devise fifty-five thousand."1
Those are the same customer. Fifteen hundred and fifty-five thousand.
Worse, the chapter also says "ten thousand yield every object of desire" and "if an intelligent person fashions ten thousand Parthiva lingams, he has no reason for any fear" — while pricing emancipation at a crore, a thousand times more.
If ten thousand yields every object of desire, the crore is unnecessary. If the crore is necessary, ten thousand does not yield every object of desire.
The vault already holds a version of this: three lingams accomplishing every wish, four sentences after a table pricing every wish at ten thousand.5
So the incoherence is not only between the Samhitas. It is inside a single chapter, between adjacent sentences.
Take the numbers seriously for one section and ask what following them would involve.
A Parthiva lingam is made of clay, by hand, with the correct proportions, and then worshipped.6 Call it a few minutes each if you are quick and not careful.
Seven hundred is a long weekend's work.
A hundred thousand, at three minutes each, is five thousand hours — roughly two and a half years of full-time work.
A crore is ten million. At three minutes each, without sleeping, that is fifty-seven years.
Five crore, chapter 46's price for emancipation, is two hundred and eighty-five years of continuous unbroken work.
Nobody has ever done this. Nobody was ever expected to. And the text shows no sign of thinking anyone would — there is no discussion of how to fit it into a life, no dispensation for the elderly, no arrangement for doing it in parts, which is the sort of thing the volume supplies constantly elsewhere.7
So the numbers are not a schedule. They are a magnitude vocabulary.
Which explains the contradictions completely. If a crore means unimaginably many rather than ten million, then a crore for emancipation and ten thousand for every-object-of-desire do not conflict, because neither is a quantity. They are both saying a great deal, at different volumes.
And it explains the ratchet. If numbers are rhetoric, there is no cost to raising them, and there is a benefit — a later text competing with an earlier one has every reason to say a bigger number and none to say a smaller one.
You are comparing two consultancy proposals for the same project.
One says the work will take four hundred hours. The other says it will take fifty thousand.
You do not conclude that one of them has miscounted. You conclude that at least one of them is not really talking about hours — that "fifty thousand" is doing the job of this is an enormous undertaking and you should be impressed, and any resemblance to a duration is incidental.
And you know how to tell which. You ask what happens next in the document. Does anyone schedule against the number? Is there a milestone at hour ten thousand? Does the invoice reference it?
If the number never appears again in an operational context, it was never an estimate.
Run that test on chapter 28 and chapter 46 and both fail it completely. No chapter anywhere in this volume takes a lingam total and does anything further with it — no schedule, no division across a lifetime, no proportional partial credit, no discussion of what happens if you die at four hundred thousand.
The numbers appear once, in a list, and are never used again.
Which is the same tell, and it is available to any reader, and it means the two price lists were never in competition as specifications. They were in competition as claims about how much this matters.
One detail deserves separating out, because it is the ugliest thing in either chapter and the drift is largest exactly there.
Both chapters price harm. Chapter 28 sells killing at seven hundred, confounding at eight hundred, kindling hatred at five hundred, stupefying at a thousand.1 Chapter 46 sells killing enemies at a hundred thousand, confounding at fifty thousand, and adds japa flowers, rajika flowers, mustard oil, pepper and sarshapa seeds, all for the death of enemies.2
And the largest single disagreement in the whole comparison is the price of murder: a factor of a hundred and forty-three.
Two readings, and they point in opposite directions.
Charitable: the later chapter is pricing harm out of reach. A hundred thousand lingams is two and a half years of full-time work, which is a genuine deterrent in a way that seven hundred is not. On this reading the inflation is a moral correction performed in the only vocabulary available.
Uncharitable: it is the same inflation that hit every other price, applied without any moral thought at all — and the fact that the murder price rose further than emancipation's did is an artefact of starting from a smaller number, not a judgement.
The text gives no way to choose, because neither chapter comments on the abhichara prices at all. They sit in the middle of the lists, between garments and land, in exactly the same tone.
That silence is the thing to record. The volume prices killing a man twice, disagrees with itself by two orders of magnitude, and never once indicates that this item is different from the others.8
The general point, and it reaches past this volume.
A number in a devotional text is doing one of two jobs, and they look identical on the page.
Operationally, it specifies a quantity someone is expected to produce, and it can be wrong.
Rhetorically, it conveys magnitude and seriousness, and it cannot be wrong, because it is not a claim about quantity at all.
The two are told apart by a single test: is the number ever used again? An operational number gets referenced, scheduled against, divided, or checked. A rhetorical one appears once and is never revisited.
Almost every number in this volume fails that test. The crores of japa, the lingam totals, the flower counts, the gayatri quota — they appear in lists and are never operated on.
Except one class, and the exception is exactly where you would expect it. The mouth is rinsed eight times, or sixteen. Earth is used five times, or four, or three, or two. The twig is twelve angulas. Those numbers are small, checkable, and performed daily — and they are the only ones a person could actually comply with.
So the volume runs two numeric registers side by side: a rhetorical one for soteriology and a procedural one for the body. And the tell that separates them is not the size but the reachability.
Which means the contradictions in this page are contradictions only if you read the wrong register operationally. Seven hundred and a hundred thousand do not disagree, because neither of them is seven hundred or a hundred thousand.
Strongest support is the direction: five overlapping prices, five increases, from ×5 to ×143, with none falling. Independent traditions with independent numbers should disagree in both directions.
Tension one. Chapter 28 contradicts itself internally — sons at 1,500 and at 55,000; "ten thousand yield every object of desire" against a crore for emancipation. The inconsistency is not only between Samhitas.
Tension two. The largest disagreement in the corpus is the price of killing a man, ×143. Whether that is a moral correction or an artefact of a low starting number is not decidable, because neither chapter comments on the abhichara items at all.
Tension three. The numbers are unreachable by construction — five crore is two hundred and eighty-five years of continuous work — and the volume supplies dispensations for the elderly, the poor and the incapable everywhere else, and none here.7
Tension four. The reading that resolves everything — that these are magnitude vocabulary, not quantities — also dissolves the vault's ability to treat any soteriological number in this volume as content. That is a large concession and it should be made explicitly rather than by implication.
Open question, and it is testable. If the Rudra Samhita systematically inflates Vidyeshvara numbers, the pattern should hold on any further overlap in chapters 47–150. A single counter-example — a Rudra figure lower than its Vidyeshvara counterpart — would break the ratchet reading and return this to noise. Worth watching on the Units 4–5 read, and cheap.
The Arithmetic of Lingams holds the chapter 28 list and made the vault's sharpest observation about it: that abhichara sits uncommented inside a devotional manual, with killing priced in the same units and the same tone as emancipation.
That finding survives intact and gets worse. The volume prices murder twice, disagrees with itself by two orders of magnitude, and still never marks the item as different.
But that page treated the numbers as content — seven hundred against a crore as a meaningful ratio, a moral partition that was never drawn. This page argues the ratio may not be a ratio. Both readings should be held: if the numbers are rhetoric, the uncommented pricing of murder is if anything more damning, because it means killing was reached for casually as an illustration of magnitude.
Excess Is Never Punished found that deficiency is penalised and surplus is always free — a floor with no ceiling, producing permanent insufficiency.
This page supplies the mechanism. A number that is never operated on cannot have a ceiling, because a ceiling is an operational concept. And a ratchet between recensions is exactly what a floor-with-no-ceiling produces over time: each restatement can only go up.
Two Manuals That Do Not Agree established procedural contradiction as the build's fourth mechanical line of evidence. This is the same phenomenon in the soteriological register, and it is weaker evidence for exactly the reason set out above: north and south cannot both be followed, but a crore and five crore can both be meant, because neither is a quantity.
Disguising Numbers and Direction (behavioral-mechanics). The material there starts from the footprint as a confession — press your foot into wet ground and you have told anyone following how many of you there were, how heavy, which way you went — and then observes that the confession can be forged. You leave a track that says fewer men, a lighter body, the wrong direction.
The point that transfers is not deception. It is that a number is read as a residue of a process, and that this is what makes it persuasive and what makes it forgeable.
Seven hundred lingams reads as though someone counted. A hundred thousand reads as though someone counted a great deal more. Neither is a count, but both carry the authority of one, because a specific figure implies a procedure that produced it.
The insight neither gives alone: precision is a claim about method, not about magnitude, and it is claimed most freely exactly where no method exists. Seven hundred is more convincing than many for the same reason a forged footprint is more convincing than no footprint — it implies a process the reader then reconstructs on the text's behalf. Which gives a practical instruction with wide reach: when a figure is oddly specific and never used again, the specificity is the signal being sent, and the right question is not whether the number is right but what it would have taken to produce it, and whether anyone did that.
Fairness Norms in Pricing (psychology). The finding there is that a hardware store will not raise the price of snow shovels during a blizzard even though demand has spiked five hundred per cent, because customers experience the increase as gouging and the reputational cost exceeds the profit. Markets are constrained by what buyers will accept as legitimate.
Chapter 46 raises every price and nobody objects, and the contrast explains why.
Fairness norms bite when a real transaction happens. You hand over money; you can compare this price to last week's; you can go elsewhere. The constraint is enforced by the moment of payment.
Here there is no moment of payment. Nobody assembles five crore lingams, nobody compares chapter 28's rate to chapter 46's, and there is no competing vendor whose rate you could check. A price nobody pays is a price nobody can find unfair.
What the pair produces: fairness norms are not a feature of buyers' psychology in general — they are a feature of buyers who transact, and they vanish the moment a price becomes purely declarative. Which suggests a broad test for any domain where numbers inflate without resistance — devotional accounting, mission statements, corporate targets nobody is measured against, credentials nobody verifies: look for whether anyone ever settles up. Where nothing is ever tendered, the ratchet runs unopposed, and the numbers will keep rising until they stop meaning anything, at which point they will keep rising anyway.
Sharpest implication. The volume prices the same goods twice in the same currency and no overlapping price agrees: learning at 1,000 and 12,500, emancipation at 1 crore and 5 crore, freedom from prison at 10,000 and 100,000, and killing a man at 700 in chapter 28 and 100,000 in chapter 46 — a factor of a hundred and forty-three. And the disagreement has a direction: every overlapping price rises in the Rudra Saṃhitā, not one falls, which is the same drift that turns eight mouth-rinses into sixteen and strips the concession off the ketakī's sentence. A ratchet is not noise; it means somebody was competing. But the reading that explains it also costs something: at three minutes a lingam, five crore is two hundred and eighty-five years of continuous work, and the volume supplies dispensations for the poor and the elderly everywhere else and none here — so these were never quantities, they are magnitude vocabulary, and the test that separates the two registers is whether a number is ever used again. Almost none in this volume are. The exceptions are the small daily ones — eight rinses, five handfuls, twelve angulas — which are the only ones anybody could actually comply with.
Generative questions.