Do the rite badly and one of three things happens.
You get less than you hoped. You get nothing at all. Or you get hurt.
The Shiva Purana never says this. There is no chapter on ritual failure, no list, no vocabulary for the distinction. What there is, scattered across nine chapters, is a consistent grammar — the text uses different words for different failures, uses them the same way every time, and never once explains itself.
Collect the clauses and the system is visible. And the moment it is visible, it is startling, because the tier a mistake lands in has almost nothing to do with how serious the mistake looks.
The lightest tier is a straight discount, and the phrasing is always quantitative.
An excellent lingam is one which measures a span that equals twelve fingers of the devotee. If it is shorter, the fruits are less.1
Less. Not none.
The same register appears at the seat:
The seat can be circular, with four faces, or in the form of a triangle... If the seat is shaped like a couch in the middle, few fruits are obtained.2
And at the offering measure:
The naivedya, consisting of shali rice, must always be measured out in kudavas. For worship in a house, it should amount to one kudava... This is known to yield complete fruits.3
Complete fruits is the benchmark, which implies the incomplete kind as the ordinary failure state.
So tier one is dosage. Undershoot the quantity, undershoot the return. The rite worked. It worked proportionally.
Nothing moral is happening here and nothing is at risk. This is the tier a reasonable person would expect all ritual error to sit in.
The middle tier is different in kind, and the text has a different word for it. Not less. Futile.
If the sandhya rites are performed at the wrong time, they are futile.4
And, in the chapter on ash:
Without Tripundraka, donations, japa and other rites become futile.5
And, at the top of chapter 28:
Without fashioning a Parthiva lingam, if a person worships any other deva, that worship is futile. Self-control and devotions are in vain.6
Futile means the act does not convert. You performed it, it took the same time and the same materials, and nothing was produced.
Notice what puts an act in this tier. Timing. A missing mark on the forehead. Directing the worship to the wrong god.
None of those is a defect in the act itself. The sandhya said at the wrong hour is a perfectly executed sandhya. The japa without Tripundraka is correct japa. What has failed is a precondition — a frame around the act rather than anything inside it.
So tier two is about eligibility. The rite is fine; you were not in a state where it counts.
The third tier is the one that makes this page necessary, and it is triggered by something almost nobody would predict.
Chapter 26, on how a lingam must be constructed:
If one worships a Chara lingam that is made out of a single piece, this bestows complete fruits. If one worships a Chara lingam that is made out of two pieces, this is said to lead to great harm. If one worships a stationary lingam that is made out of a single piece, this does not yield the fruits one desires.7
Three outcomes in three sentences, and the text has just given all three tiers in one place:
Look at what earns "great harm."
Not blasphemy. Not neglect. Not worshipping the wrong god. Not doing it drunk, or angry, or for a bad purpose — chapter 28 will shortly price killing a man at seven hundred lingams without threatening anybody.8
Great harm is what you get for a portable stone made of two bits instead of one.
The chapter is emphatic about it, too. It repeats the rule four times in one paragraph, calls those who disagree "deluded in their knowledge," notes that "sages who know about the sacred texts do not agree with this conclusion," and finishes with "one must never act against what the sacred texts have said."9
That is the volume's most heated defence of any rule anywhere, and its subject is joinery.
There is one other clear tier-three case, and it is instructive because it is not about equipment at all. Chapter 6: a listener who sits at the same level as the narrator commits a sin the text equates with violating the guru's bed.10 Same structure — a physical arrangement, catastrophically penalized.
You are carrying a lingam in your bag because you travel, and this morning you are looking at it in the light of somebody else's courtyard.
It is one piece. You checked when you bought it and you have checked since, more than once, in the way that people check things they have been told carry great harm.
The man who sold it to you said one piece and you believed him, and you have no way of verifying that a repair was not made and smoothed. A good joint would not show.
So there is a small permanent uncertainty travelling with you, and you have made a private accommodation with it, which is that you do not look too closely any more.
The sandhya you say at the wrong hour, on the days the journey runs long — that one you know is futile. You say it anyway, and you know it does not count, and you would rather do the futile thing than nothing.
And the stone you got for your daughter is a shade smaller than twelve of her fingers. Fewer fruits. You decided that was acceptable and you did not agonize.
Three failures, three completely different weights, and only one of them keeps you up. The one you cannot inspect.
Line the tiers up against what triggers them and the pattern is unmistakable.
| Trigger | Tier |
|---|---|
| Lingam too short | reduced |
| Seat dished in the middle | reduced |
| Naivedya under-measured | reduced |
| Stationary lingam of one piece | reduced |
| Sandhya at the wrong hour | futile |
| Rites without Tripundraka | futile |
| Worshipping another deva instead | futile |
| Mobile lingam of two pieces | harm |
| Sitting level with the narrator | harm |
Timing errors are futile. Quantity errors are discounted. Equipment and posture errors are harmful.
That ordering is the reverse of what a devotional reading predicts. A tradition organized around inner states should punish inattention and reward sincerity. A tradition organized around correct performance should punish timing, because timing is the classic Vedic anxiety.
This one reserves its heaviest penalty for the physical configuration of objects and bodies.
And there is a second pattern inside it. The tier-three triggers are the two errors that are invisible in the moment. A missed sandhya announces itself — you know the sun has moved. An under-sized lingam can be measured against your own hand any time you like. But a concealed joint in a stone, and a seating arrangement that everyone in the room has already accepted as normal, are both errors that will not surface on their own.
The heaviest penalty attaches to the errors nobody will catch for you.
Which is a coherent design, even if the text never states it. Where a mistake is self-correcting, price it lightly. Where nothing in the situation will alert you, make the stakes high enough that you check.
The text says great harm and never says harm from whom, or of what kind.
That silence is worth holding rather than filling, because the volume is perfectly capable of specifying. It has hells with hot iron rods, a trident for the lax brahmana, rebirth as a python for the listener who reclines and as a donkey for the one who argues.11 When it wants to threaten graphically, it does.
Here it does not. Great harm, and nothing else.
Three readings are available.
Mechanical. The two-piece chara lingam is structurally wrong in the way a badly built thing is wrong, and it fails on you. Chapter 26's own reasoning supports this: the base is Mahavidya and the lingam proper is Maheshvara, so a fixed lingam needs two pieces and a portable one must not be split — the split is a division of something that should be whole.12
Jurisdictional. The god or his attendants take it badly. This fits Chanda's right over naivedya, where an attendant has claims and enforces them.13
Unspecified deterrent. The vagueness is the mechanism. A named penalty can be weighed against a benefit; an unnamed one cannot, and therefore cannot be traded off.
The third is the most likely and the least provable. Notice that the volume's graphic punishments all attach to social offences in a room full of witnesses — where the point is that the audience hears the threat. The unspecified one attaches to a private object nobody else will inspect.
Step back from the content and look at the fact of the system.
Three graded outcomes, applied consistently across nine chapters by a text that never names the grading, never lists the tiers, and never explains a single assignment.
That is not how a doctrine behaves. A doctrine gets stated.
It is exactly how craft knowledge behaves. A carpenter does not have a theory of tolerance; he knows which mistakes cost you a little, which waste the piece, and which will hurt somebody later. The knowledge lives in the specific cases and never gets abstracted, because in the workshop it never needs to be.
And that fits the other evidence about these chapters. The panchasutri proportions are a workshop specification stated without explanation.14 The lingam is measured against the devotee's own hand. The chapter on installation mentions artisans and commissioning patrons. The materials history — "at first, clay or rock; later, iron" — is a technical note nobody theologized.15
So the three tiers are probably not a theology of ritual efficacy at all. They are trade knowledge about what goes wrong, absorbed into a devotional text and preserved with its original grammar intact — which is why the assignments look arbitrary to a reader looking for doctrine and perfectly sensible to one looking for a workshop.
Strongest support is chapter 26's three-sentence sequence, which gives all three tiers in immediate succession with three different phrasings — complete fruits, great harm, does not yield the fruits one desires — attached to three configurations of the same object.
Tension one. The heaviest penalty in the ritual system attaches to a construction defect, in a volume that prices contract killing without any penalty at all.
Tension two. The tiers are used with complete consistency and named nowhere. A system that operates and is never stated is either very old or was never anyone's doctrine.
Tension three. Chapter 26 defends the one-piece rule harder than the volume defends anything else — four repetitions, an explicit dismissal of dissenters as "deluded" — while conceding that sages disagree.
Tension four. "Great harm" is unspecified, in a text with a rich vocabulary of specified punishment. The vagueness is either a gap in transmission or the point.
Open question. Do the tier-three triggers share the property of being uninspectable, as the two clear cases suggest, or is two cases too thin a base? Units 3–5 of this volume have not been read, and the Rudra Samhita's narrative material may or may not produce further instances. Flagged for the later clusters rather than settled here.
The vault's practitioner-tier Śaivism has no failure system at all, and the absence is the finding.
Do Whatever You Want: The Śivarātri Non-Prescription holds that the vows are tools rather than rules — that there is no wrong way to do it. Upavāsa: Fast Only If It Nourishes runs the same argument for austerity.
Chapter 26 says a stone made of two pieces leads to great harm and calls people who disagree deluded.
The obvious reading is medieval rigidity against modern permissiveness, and it does not survive contact with the rest of the volume — the thumb clause, the "no prohibitions," the waived caste assignments, the permission to offer two upacharas instead of sixteen.16 The Purana is more permissive than its reputation almost everywhere.
Except here. The one place this text will not bend is equipment, and that is a much stranger fact than general strictness would have been. It suggests the modern permissiveness and the ancient permissiveness are the same permissiveness — and that what the modern tradition actually dropped is a workshop's knowledge of what breaks.
The Blemishes Taxonomy: Failure Modes (creative-practice). The Sanskrit dramaturgical tradition catalogues doṣas — the specific ways a performance fails — as a named, graded, teachable list, with the grading explicit and the reasoning given.
The Shiva Purana has the identical apparatus and never names it. Same culture, same period-range, same three-fold instinct about degrees of failure — one tradition abstracts it into a taxonomy, the other leaves it distributed across nine chapters as case law.
What the pair produces: whether a body of practical knowledge gets abstracted depends on whether it has to be taught to strangers. The Nāṭyaśāstra is a teaching text for a profession with pupils who arrive knowing nothing; it must name its failure modes because there is no other channel. Ritual instruction passes down a lineage where the cases are demonstrated, so the tiers can stay implicit for a thousand years without anyone noticing they form a system. The abstraction is not a sign of sophistication. It is a sign of a transmission problem the other tradition did not have.
The Sacred Container as Operational Architecture (behavioral-mechanics). The ritual-design corpus treats the boundary conditions of a rite — its timing, its markings, its spatial arrangement — as the load-bearing structure, with the content secondary.
The Purana's middle tier is precisely a boundary-condition tier. Wrong hour, no forehead mark, wrong deity: futile. Nothing inside the act failed; the container did not close.
The insight neither gives alone: a three-tier failure system reveals what a tradition thinks a rite actually consists of — and this one says a rite is a correctly-configured object, inside a correctly-closed container, performed in sufficient quantity. Intention appears in none of the three tiers. Not once, in nine chapters. The behavioral pages design containers to produce psychological states; this text designs them to produce validity, and treats the worshipper's state as irrelevant to whether the thing worked. Which is the same commitment that lets ash purify involuntarily and lets seven hundred lingams kill a man — and it means the volume's much-quoted mercy and its much-avoided amorality are the same doctrine seen from two sides.
Sharpest implication. The Shiva Purana runs a consistent three-tier penalty system across nine chapters — reduced fruits for quantity errors, futile for timing and precondition errors, great harm for configuration errors — and never names it, lists it, or explains a single assignment. The tiering is backwards from what a devotional reading predicts: a mistimed sandhya is merely futile, an undersized liṅgam merely yields less, and "great harm" is reserved for a portable liṅgam made of two pieces instead of one, defended more fiercely than any doctrine in the volume. The two clear tier-three triggers share one property — they are the errors nothing in the situation will alert you to. And intention appears in none of the three tiers, which is the same commitment that lets ash purify involuntarily and lets the volume price a killing without a word of warning.
Generative questions.