The argument runs in two steps and takes about a sentence. Impermanence is the chief feature of the universe. A religion is a thing in the universe. So it goes.
in time every religion will disappear, every nation will fall, and every people will scatter. Impermanence is the chief feature of the universe, so this has to be so. Even Buddha predicted that Buddhism would one day disappear, as have a variety of religions already. Christianity, Islam and Judaism, despite their current prominence, cannot escape this fate either.1
Four traditions get named and one of them is the tradition that supplied the doctrine. Nobody carves out an exception for the true teaching, nobody promises a remnant, and nobody's forecasting a restoration. The scheme that generates the prediction is the first thing the prediction eats, which is unusual enough to notice.
Sit that inside a book of stages for a moment. The whole machinery — realms, dhyānas, rankings — gets offered as a description of how minds and bodies work rather than as the property of a school. If the description is right, it outlasts whatever words are currently carrying it. The practices are re-derivable, because they run on human equipment that keeps getting manufactured. The names for them aren't.
The impermanence claim would be a slogan if it stopped at religions, and it doesn't. It runs all the way up, and the mechanism is worth having because it's the only part of this that behaves like machinery rather than like an assertion.
Merit is the currency. Greater merit buys greater resources — better circumstances, longer life, a higher rebirth — and the highest heavens are populated by whoever accumulated the most of it.7 So far this reads like a reward scheme. Then comes the clause that turns it into something else entirely:
when this life is over or the merit you've earned for any situation has been used up, then your good karma is gone as well.7
So merit gets spent. You do not own the heaven — the merit is what is holding you there, and living in it is what uses it up. Nothing has to go wrong for the arrangement to end. That is why the top of this ladder is not a destination. It is a very long and very pleasant stay that ends when the account does.
There's a matching rule about how high you get in the first place, and it's stated flatly. The heavenly stage you can ultimately reach corresponds to the limits of your own cultivation achievement.7 Not to what you believe, not to which tradition issued the ticket. To how far you actually got. Put the two together and you have the whole doctrine in miniature: your depth sets the ceiling, and your merit sets how long you stay, and neither of them is permanent. If that's true of a heaven, a church isn't going to be the exception.
Now the part that does the most work here, because it explains why the world's cosmologies differ rather than just noting that they do. The strongest version of the claim is short:
many of the differences we find across the various world religions are due to the depth of samadhi attainments achieved by their sages.4
So a tradition's account of the heavens records how far its founders got. They went in, reached whatever level they reached, and wrote down what was there. What is on the account is real. What is missing from it is not missing from the world — it is missing from their reach. Two things follow from that, and the passage draws both.
Independent traditions converge where the terrain is the same. The Aztecs had no contact with Europe and held thirteen heavens full of celestial beings and nine hells full of devils and suffering souls — close enough to the Christian structure that the arriving Spaniards couldn't account for it.4 Ancient Egypt held six heavens immediately above the earth, matched here to the six Desire Realm heavens, with the realm where God lived above them corresponding to the Form and Formless Realms taken together. The Egyptians also held that beings on the higher levels come down to help those below, which is the bodhisattva pattern arrived at separately.4
And two accounts disagree where their founders stopped at different depths. The worked instance is oddly specific. Moses gets rated at roughly the second dhyāna and his fellow Egyptian initiates at about the first, which reaches the top of the Desire Realm heavens. That one-stage gap is offered as the entire explanation of the contest — why he could match their feats and go further, and why they could duplicate some of what he did but not all of it.4
Run the same test inside one tradition and it splits that tradition in half. The God of the Old Testament — punishing, vindictive, an authoritarian ruler — gets placed at the lowest of the three heavenly tiers, and the reading is that Moses took one of that tier's kings and dressed him in images his people could actually use.4 Wrapped is the source's own verb, and it describes founding a religion as a piece of work: pick which level you are pointing at, then pick the images and messages the people of your time can actually use. The God in St John's writings gets filed to the Form Realm, a higher attainment, and St John's own depth is offered as one reason he was the beloved disciple and one reason the two testaments read so differently.4
Here's the caveat the method needs and doesn't always get quoted with. Not every account is a record of somebody's own experience. Some are borrowed copies of visions belonging to another culture — the Roman imitation of Greek religion is named — and some may be nothing but fanciful imagination, which the Aztec example openly concedes as a possibility about its own case.4 So there are three slots, not one. That matters for what you can do with it, because the test only works if you can tell somebody's own experience from a copy of somebody else's, and the passage gives you no way to do that.
You don't have to buy the ratings to take the method. Treat an account of the heavens as testimony about the people who wrote it, keep the three possibilities in mind, and the differences between religions turn into data about attainment rather than about truth.
None of this is aimed at belief. It's aimed at a posture, and the target is named:
so you mustn't believe that any one particular group is supreme or "chosen," despite any propaganda to that effect. This attitude of supremacy is a big problem with a number of nations, races, cultures and religions who refuse to recognize the fact that in time every religion will disappear.1
It works as a deflation because chosenness is itself a claim about permanence. A people specially selected. A covenant that holds. A church the gates won't prevail against. Every version of it implies durability, and durability is exactly what the teaching removes. The instrument is unusually clean, too — you don't have to dispute anybody's history or refute anybody's theology. Nothing's lasted yet.
There's a subtler observation attached and it's the more useful one. A realised teacher born into a culture — usually by choice, on this account — will work to lift that culture and will say it's the best, because lifting it is the assignment. But in the deep mind no race or culture registers as supreme, since even moderate attainment is enough to scan and find countless realms further along in every respect.5
So the real line isn't between traditions that claim supremacy and traditions that don't. It's between supremacy as a working posture and supremacy as a belief about reality. The first is a teacher committed to the people in front of them, and it expires when the job does. The second is what the first hardens into once the teacher is gone and nobody remembers it was a posture.
Now the mechanism with the most reach, and it's a claim about irreversibility:
sages arriving later in a particular tradition would be hampered by what had been previously established (and by then inculcated into rigid dogma) by much lower adepts in the traditions. A greater sage arriving later might not be able to do anything to correct the mistaken, incomplete or wayward notions previously set forth by those of inferior cultivation attainments.2
We tend to assume traditions self-correct. A great teacher turns up and sets things right, and the record improves over time. This says no. Dogma laid down early by lesser figures ends up holding the roof, and a later and better figure inherits it as a constraint rather than as a draft.
The prediction that falls out is checkable, which is what makes it worth keeping. A tradition should end up looking like what the source calls a hodgepodge — teachings from very different levels of attainment sitting side by side, stacked by the date they arrived rather than by how good they are, with the earliest ones underneath everything else and therefore impossible to shift. And the named instance is offered without hedging: Brahmanism and Hinduism look like hodgepodge constructions of teachings from different levels of attainment, and they look that way because the observation is perfectly true.2
Put that together with the point above and it turns nasty. If an account records how far the founders got, and nobody who later goes further is allowed to change it, then the tradition is stuck forever at the level of whoever arrived first with enough authority to be listened to. Being early beats being right. And the two have nothing to do with each other.
The illustration was chosen for the size of the figure standing in it.
when Tsong Khapa taught in Tibet, he realized that so much damage had been done to true Buddhism in this culture that his best hope would be to just try and make sure that the situation didn't slide any further. In other words, he realized that there was no way to weed the culture of some of the tumors that had already taken root from incorrect teachings, so the best he could hope to do is to help prevent the culture from declining further rather than insure that it will ascend upwards.2
Tsong Khapa sits at 8 on this book's ten-point scale. That's above Socrates, above Wang Yang-ming, level with Milarepa. He founded the Gelugpa order. And the account here has him surveying his own position, concluding that reform wasn't available, and playing for damage limitation instead.
The teaching value is in the asymmetry. He had every advantage a reformer can have — top of the scale, founding an order, working inside his own tradition, with the standing to be heard. If that's what an 8 manages against established dogma, the ceiling isn't a function of the person standing under it. The constraint is structural, and realisation doesn't dissolve it. Which is a genuinely uncomfortable result for anyone who thinks the answer to a bad institution is a better person at the top of it.
Two things get recorded here rather than carried. The reconstruction concerns a historical figure's private strategic assessment and arrives with no source at all — no text, no biography, no attributed statement. And the damage gets named as the influence of yoga, Bon and Taoism, which is the wholesale-dismissal pattern this vault excludes. It's on the record and it isn't repeated.2
So what happens to the sediment once it sets? It gets tidied, and there's an image for the tidying that earns its place.
the haphazard construction of religious teachings and dogma is usually the rule for most religions, and it's only smoothed out over a long period of time after various intellectuals get involved in resolving inconsistencies through intellectual arguments and logic rather than through spiritual achievement.
Imagine that you have a stack of haphazardly oriented papers that you want to put through a copy machine. Before you can place them in the photocopy input hopper, first you pick up the stack of papers and bang them on a table trying to make all the edges line up nice and neat.
This is exactly the task that religious professionals perform — they take all the pieces of dogma which seem contradictory or out of place and try to either smooth them away or make them conform to the general outline of the dogmatic structural stack that has previously been built up.3
Banging a stack on a table doesn't change a single page. It lines up the edges. It makes the collection look like a document, and the contents stay whatever they were. That's what systematic theology does on a bad day: it removes the look of contradiction without settling anything, and what you're left with reads as a coherent body of doctrine and is a stack.
The detail that carries the weight is who does the aligning. Alongside the revelations, a tradition accumulates "intellectual" shapings performed by religious academics with no attainment at all.3 These people were selected for argumentative skill. Being best at making things fit isn't the same capacity as knowing which pieces are right, and nothing in the selection connects the two.
What the reader is then left holding is the practical problem:
you therefore have gold and pyrite, or fish eyes and pearls mixed together. The ordinary man is unable to distinguish between what's genuine and what just sounds good or turned out to be a practical compromise in order to settle an issue by satisfying a troublesome faction.6
The sting is in that last clause. Some of what sits in a tradition's doctrine got there as a negotiated settlement with a faction — it ended a dispute, and nobody checked it. Once it's smoothed into the stack you can't tell it from anything else, and that isn't bad luck — smoothing is exactly the operation that removes the texture difference.
The direction of travel isn't fixed, either. Intellectualisation can raise a stream or wreck it. Chu Hsi is named as the wreck, his interpretation judged to have destroyed the message of Confucius, and the Western path after Socrates is read as a slide into materiality and intellectualism.6 So the working test the whole passage has been building toward isn't a test for coherence, because coherence was manufactured. Ask instead whether anything in a position required attainment to produce. Gold and pyrite look identical in a tidy stack, and they behave differently the moment you try to use one for something.
You're inside something — a tradition, a company, a discipline — and you keep hitting the same wall on the same question. Before you decide the wall is a considered position, go and find out when it was built.
Start with the earliest statement of the thing, not the current one. The current one has been smoothed and reads well, which tells you nothing. Chase it back to the first formulation and then find out who wrote it and what they were actually good at. Sometimes it's a serious practitioner. More often it's an administrator, a compiler, a committee, or somebody who was solving a local problem that stopped existing two generations ago. Then hold the two apart, because they come apart more often than anyone admits: is this credited for who said it, or for when it was said? Most walls are holding weight because of when they went in, and nothing else.
Now run the revisability test, which is the one that tells you what kind of object you're dealing with. Has anyone since had both the standing and the reason to reopen it? A position nobody senior has ever reopened hasn't survived scrutiny — it hasn't had any, and those two states look identical from where you're standing. Lay the tradition's positions out by date rather than by topic while you're at it. If the quality tracks the date instead of the calibre of the person who contributed it, you're looking at sediment, and the bottom layer isn't going to move for you.
Then ask the survey question about each wall, the same one you'd ask of a cosmology. Whoever set this down went as far as they went — how far was that, and does the position report the terrain or the ceiling? And sort the two kinds of commitment while you have them laid out. A working posture is somebody doing their job for the people in front of them, and it's negotiable, because its author would have said so. A belief about reality that outlived its author isn't negotiable at all.
The last move is where the years go. Where a wall turns out to be early, incidental and built upon, stop arguing with it and start navigating it. People burn a decade trying to win an argument against a decision nobody is actually defending, because nobody remembers making it. Tsong Khapa's assessment was that he couldn't move his, and he was better placed than you are. That doesn't make the audit pointless — it changes what the audit is for. Knowing which walls are conclusions and which are sediment is how you find out where effort turns into movement.
[POPULAR SOURCE] throughout. [SINGLE SOURCE] on all the comparative material. The attainment ratings attached to Moses, St John, the Aztec priesthood and Tsong Khapa are the source's own assignments on its own scale, not independent findings.
The tension the material doesn't close is that the impermanence teaching is weaker than its rhetorical force. That nothing has lasted is compatible with a tradition being true while it lasts, so the deflation bites only against versions that assert permanence — and a tradition holding we were given this and it will end like everything else is untouched.
A second tension runs internally, between the two theses on the same two pages. If a cosmology records the depth its founders reached, the map should improve whenever a deeper practitioner surveys again. If early codification can't be revised, it never does. Both are asserted and neither is reconciled.
⚠ On the record and not carried: the Tsong Khapa reconstruction is unsourced and concerns a historical figure's private strategy; the sedimentary prediction rests on one asserted case graded perfectly true; the Aztec rating is flagged as speculation by the passage itself, which also warns that two heavens of the same level could be miscounted as two levels.24 Bon, Taoism and yoga named as the damage in Tibetan Buddhism is the wholesale-dismissal pattern this vault excludes, as is the surrounding treatment of Islamic paradise doctrine.
Open questions. Does comparative religious history actually show quality strata tracking date rather than contributor attainment, hardest to revise at the bottom? And if these accounts are testimony about what their writers reached, do independently developed ones agree more closely at the levels their traditions had practices for?
Something unusual happens in this passage and it's worth marking before anything else. Every other ranking in the book spares Buddhism, either explicitly or by using it as the measuring stick. Here Buddhism disappears alongside the rest, on the Buddha's own prediction, and the teaching that removes chosenness removes it from everybody who might have wanted it.
Then follow what the first writing-down actually does, because it does two separate things and nobody notices either at the time. It fixes how fine a distinction the tradition can ever draw — write the base coarsely and no later refinement will sit on it, which is what A Tradition's Ceiling Is Set by Its Codifiers works through. And it fixes what anyone is allowed to go back and change. Somebody choosing the wording of a founding text is setting both the resolution and the permanence of everything downstream, and thinking about neither.
The sharpest pull is against Buddhism as Comparative Reference Grid, which uses the five schemes as the measuring instrument for everything else. If Buddhism disappears, the instrument goes with it, and what's left standing is that the practices can be found again and the vocabulary can't. A future rediscovery would produce the methods again and wouldn't produce the eight consciousnesses.
The extinction mode a tradition can't see from inside — Parsi Zoroastrianism: When Ecology Triggers Religious Extinction. That page follows a living tradition into an ecological trap. Its central funerary practice depends on vulture populations, and the vultures collapsed. Membership rules make replacement impossible. Nobody stopped believing anything; the conditions the practice needed simply went away.
That's the concrete instance of the abstract teaching, and it corrects the teaching in a specific place. The impermanence account says how everything is — impermanence is the chief feature of the universe, therefore religions end. True, and it explains nothing about how or when, because it applies uniformly and so predicts no differences at all. Every decline mechanism named on this page is internal: bad codification, dogma set by lesser adepts, academics aligning edges. The Parsi case is entirely external, and nothing here would have caught it.
Put them together and you get what neither gives alone. A tradition can die two ways, and it can only see one of them coming. It can rot, and rot is something it can argue about — every reform movement in history is a tradition noticing its own rot. Or the world can quietly stop supplying something the tradition needed and never thought to write down, and that one gives no warning at all, because nobody audits a thing that was never doctrine in the first place. It was just there, like the vultures.
Which yields a diagnostic neither page asks for on its own. What does this tradition require from the world that it has never had to write down? Ordination needs a population willing to be ordained. Pilgrimage needs safe roads. Monasticism needs a surplus that somebody is prepared to donate. Textual transmission needed a scribal class and the leisure to feed it. None of that is doctrine, all of it is holding the tradition up, and a tradition auditing itself for decay won't find a single item of it, because it isn't in the doctrine to be audited.
Run it the other way and the irreversibility mechanism gives the historical case something it lacks. The Parsi analysis treats the vulture dependency as a fact about that tradition. Irreversibility suggests the dependency is itself an early codification — somebody fixed what the practice required at a moment when the requirement was cheap and abundant, and then it locked. The vultures weren't chosen because they were essential. They were chosen because they were there.
(One handshake. The codifier-ceiling material has its own page in this corpus and a second bridge would restate it rather than extend it.)
Sharpest implication. Two accounts of disappearance are on offer and only the weaker one gets developed. Impermanence tells you religions end and predicts nothing about which ones or when, because it applies to everything equally. Irreversible early codification by mediocre hands explains why traditions accumulate strata they can't revise, why later greatness makes so little difference, and why the people who tidy a tradition were selected for tidying rather than for knowing. The second is about how institutions work. It would still hold if the big cosmic claims turned out wrong, which is more than the first one can say.
And there's a sharper edge underneath it that the passage doesn't reach for. The merit rule and the codification rule are the same rule at two scales. A heaven ends because living in it spends the thing that bought it. A tradition ends because being used spends the thing that made it worth using — the early decisions get built upon, the builders get replaced by tidiers, and by the time anyone notices, the layer nobody can move is holding up everything above it. Neither collapse needs an enemy. Both are what running costs, and that's a far more unsettling account of decline than any story about corruption, because there's nothing in it to resist.
Generative questions.