Kāśī has five circuits around it, one inside the next, and each one is a walk you can actually do. Going round a sacred place on foot, keeping it on your right, is a parikramā — it's what the crowds are doing when they circle a temple rather than going into it. A krośa is an old measure of distance, the way a mile or a kilometre is now, and the outermost circuit of Kāśī is eighty-four of them — a ring big enough to take several days on foot. ⚠️ No modern equivalent for the unit is given anywhere in the cited range, so none is asserted here.
That's the Chaurāsī Krośī, and the first thing said about it is how many shrines are on it:
"number of shrines covering in 84 krośa parikramā is 144. 144 shrines are covered in 84 krośa parikramā."1
Then the same figure for each ring inward, as the speaker works down through them:
| Circuit | Shrines | |
|---|---|---|
| 1 | Chaurāsī Krośī Parikramā | 144 |
| 2 | Pañcakrośī Parikramā | 108 |
| 3 | Nagar Pradakṣiṇā | 72 |
| 4 | Avimukta | 72 |
| 5 | Antargṛhī Yātrā | no count given |
The wording throughout is shrines covered in a parikramā.2 That's a count of what a walker passes, not a census of what stands in the city. These are properties of a route rather than of a place, which is why they can be paced out, and why somebody who walked the eighty-four krośa ring and counted a hundred and thirty-nine would have found something.
The rings shrink, and the shrinking is stated in the same breath as the counts:
"after this 84 krośa circuit it is the, from outer side, second-most. So this is the 25 krośa. Then again it has reduced from 84 to the 25 — mean 60 krośa has been reduced. Second circle."3
Eighty-four out, then twenty-five, and sixty gone between them. The arithmetic is done aloud, which is unusual. It makes the point that this is a nested set, not five separate pilgrimages that happen to share a city.
⚠️ The second ring's name and its figure don't match. Pañcakrośī means five krośa. The number given for it is twenty-five, said twice.3 Both the name and the figure are in the range and the passage never puts them together, so nothing is reconciled here either — the name is recorded as spoken and the figure as given.
And the shrine-counts fall with the extents. A hundred and forty-four on the eighty-four-krośa ring, a hundred and eight on the twenty-five. ⚠️ The two sequences are given about four minutes apart and are never put side by side. Nothing in the range says the shrine-count tracks the distance, though the numbers do.
The reason to expect them to is the overlay this material runs across all five rings. Each circuit is matched to an element and to a part of the body at once — sky, then earth, then air, then water, then fire on the outside; the head, then the leg, then the face, then the blood on the inside; and the five koṣas, the sheaths from food up through breath and mind to intellect, running alongside.4
That scheme moves from the largest thing to the smallest, and the rings do the same. A finer ring covers less ground and needs fewer stations on it, and the walker arriving there has the outer ones behind them already. Sixty krośa dropped between the first ring and the second; thirty-six shrines dropped with them. ⚠️ Nothing in the range puts those two facts in the same sentence, and the passage that gives the extents and the passage that gives the counts are four minutes apart.
Only one of the four numbers gets explained, and the explanation is astrological:
"Again this 144 — I will show the pictorial diagrams. It is like nine planets, and with all these celestials, it comes [to] the 144."5
⚠️ The arithmetic isn't given. Nine planets don't produce 144 by any operation stated anywhere in the range, and the speaker hands the job to a diagram the transcript can't see.
But the gloss does settle what kind of number 144 is being claimed to be. Not a tally of buildings that happen to have accumulated along a road over fifteen centuries. A figure worked out from the navagraha and the celestials, and then found on the ground. The outer ring's shrine-count is being offered as a consequence of a sky-scheme rather than an accident of history.
⛔ Nothing derives 108, 72 or the second 72. Those three arrive as bare figures, and 108 is a number this tradition uses so often that its absence of derivation here isn't remarked on.
The guardians are given per ring, and the outer one has a count of its own that's easy to miss:
"whereas this 56 Vināyakas and three ⚠️[?] that are in the 84 krośa outer parikramā — outer circuits covering 56 Vināyakas and all three of … now it is reduced into this Nagar Pradakṣiṇā parikramā, [and] this circuit is guarded by the … Vināyaka."6
Fifty-six Vināyakas on the outer ring, and something else numbered three alongside them. ⚠️ What the three are isn't recoverable from the caption. Then the same guardian class reappears, reduced, on the third ring.
So each ring has both a shrine-count and a guardian-count, and they're different numbers. A hundred and forty-four shrines and fifty-six Vināyakas on the same eighty-four krośa. The fourth ring is guarded by Baṭuka Bhairava instead.7 ⚑ The outer ring's Bhairava guardians are carried elsewhere.8
⚑ The repeated 72 is the odd datum. Rings three and four have the same shrine-count and different guardians — Vināyakas on one, Baṭuka on the other — so the number isn't indexing the guardian class, and nothing in the range offers anything else it might be indexing.
The fourth ring is the one the passage stops on, and it gets an etymology and a text:
"Lord Śiva himself tells Pārvatī in Liṅga Purāṇa and various scriptures that: I will never abandon Kāśī, this area, Avimukta. That's why it is called as Avimukta — never forsaken."9
A-vimukta. Not-abandoned. The ring is named for a promise, and the promise is quoted as direct speech from Śiva to Pārvatī with a text attached to it — the only named scripture anywhere in this range.
That's why the speaker breaks off to say it's "very very important."7 The other rings are described by what they cover and who guards them. This one is described by an undertaking not to leave, which is a different sort of fact about a place, and it's the only ring whose name is a sentence.
The fourth ring is also the only one given a centre. Straight after its count: "72 shrines are in the15 — and centre of this is Viśveśvara, Kāśī Viśveśvara temple."15 So this ring isn't described as a loop with a length, the way the outer three are. It's described as a boundary drawn around one named building, and the building is the one the fifth circuit will turn out to be.
And Baṭuka's work on it is stated in the same breath as a single word repeated until it lands: "it dissolves — dissolving, dissolution. Dissolution means what? Dissolution, your ahaṅkāra, ego. That is called dissolution."10 What the fourth ring dissolves is carried in full elsewhere.7 What matters here is the pairing: the ring that will never be abandoned is also the one where the walker's I is taken apart.
Then, almost in passing, the rings get described the way somebody who lives there would describe them:
"Nagar Pradakṣiṇā is circumambulating the urban city area. Third one out is very far, far away. Then Pañcakrośī is also very far away — 25 krośa … and Nagar Pradakṣiṇā is urban city area, and this Avimukta mārg is the never-forsaken."11
That's the scheme translated out of cosmology and into a map. One of these rings goes through streets with people living on them. The Nagar Pradakṣiṇā is the city, the working town, houses and shops. The Pañcakrośī is out past that, a serious journey. The eighty-four krośa is further still.
And the fourth ring is described not by distance at all, but by the promise. Its size gets given in landmarks instead — it "covered our12 temple area … this Vināyaka, and up to the12 …" — with the caption losing both place names.
Four circuits get counts. The fifth doesn't, and that's the most interesting fact on the page.
"the last one, foremost, comes the Antargṛhī yātrā … It is the innermost sanctum place of the city — that is the Kāśī Viśvanātha temple, or this ⚠️[Maṇikarṇikā], or this Jñānavāpī. These are the innermost sanctum temples."13
The last one, foremost. The word order isn't a slip. Five rings have been walked through outermost-first, so the fifth comes last in the telling — and it's the first in rank, which is why the sentence has to carry both.
Three named places, no figure attached. And that's structurally right rather than an oversight. The other four circuits are rings, and a ring has a length you walk along and stations you pass. The Antargṛhī isn't a ring. It's the point at the centre — the bindu — and a point has no route, so there's nothing to count.
The sequence runs 144, 108, 72, 72 — and then nothing. Not zero, and not a smaller number, but the end of counting as an activity. Each inward ring has fewer stations because it covers less ground and the walker needs less, and at the centre both go to nothing at once: no distance left, and nothing left to pass.
⚠️ The summary at the end of the passage doesn't match the itemisation. It runs the whole set together as "number of shrines also 144, and this 108, and remaining all three koṣas: 72, 72, 72 shrines are there"14 — three seventy-twos, where the walk-through gave two. Neither figure is corrected, and the third 72 would have to belong to the Antargṛhī, which has just been described as a point. ⛔ Not reconstructed.
Everything here is walkable, which makes this one of the few pages in this material where the practice is literally the subject.
Start at the outside and go in. The rings are described outermost-first and the shrinking is stated in krośa, so the sequence is the practice: eighty-four, then twenty-five, then the city, then the Avimukta, then the sanctum.3 Somebody who begins at Viśvanātha and never leaves the centre has done the last step of a five-step walk.
Count what you pass. Shrines covered in is the phrasing every time.2 The number isn't decoration and it isn't a target — it's the description of the route, and knowing it is what tells you whether you've walked the whole ring or a section of it.
Know which ring you're on by where the houses are. The Nagar Pradakṣiṇā goes through the urban city; the Pañcakrośī is well out; the eighty-four krośa is further again.11 That's a check you can run with your eyes rather than a map.
And on the fourth ring, expect the work to be subtraction. Baṭuka guards it, and what he does there is named three times in one sentence as dissolution of the ego.10 It's also the ring named for a promise that the place will never be left. ⛔ No duration, no count of circuits, no season and no marker is given for any of the five — the numbers here are stations passed, not repetitions performed.
K2 19:38–27:33, given ring by ring as the overlay is built. [PARAPHRASED] — auto-generated captions, guest episode.
⚠️ The whole passage is delivered against slides the transcript can't see. The speaker refers to "pictorial diagrams" at 20:14 and to pictures he means to show. Everything here is recovered from the spoken track.
⚠️ Unreconstructed: Maṇikarṇikā at 26:42, as mananka; both place names in the Avimukta extent at 23:27; the item numbered three alongside the 56 Vināyakas at 21:04; and the 144 derivation, which is deferred to a diagram.
Trust calibration. This is the corpus's guest speaker, and his register is different from the channel's usual: he reasons from named scripture — the Liṅga Purāṇa quotation at 25:08 — and from a claim he says is "mentioned and proven" in the purāṇas, and he gives figures that could be checked by walking. The weakest link is the 144 derivation, which is the one place he asserts an arithmetic result and then declines to show it, handing it to a diagram. Everything else on the page could be verified or falsified on the ground; that single number rests on a picture nobody watching the transcript can see.
⚑ On the counts themselves. These four figures aren't found anywhere else in the corpus, so nothing contradicts them and nothing supports them. The Kāśī Khaṇḍa is the episode's stated authority for the circuits; ⚠️ the shrine-counts aren't separately attributed to it, and the only text named in this range is the Liṅga Purāṇa, cited for the promise rather than for the numbers. 🚩 SINGLE SOURCE.
Rebuilt 2026-08-25. The previous version was built around the numbers' standing relative to the rest of the corpus, which put an assessment of the source where the walk should have been. Tier held at high. Recovered from inside the cited range and previously absent: the fifty-six Vināyakas at 21:04, an anchored count the earlier page missed entirely; the Avimukta etymology and the Liṅga Purāṇa quotation at 25:08, the only named scripture in the range; Baṭuka's dissolution of ahaṅkāra at 24:47; the relative distances at 25:42, which put one ring inside the inhabited city; and the three named innermost temples at 26:42.
Open. How nine planets yield 144. Whether there are two seventy-twos or three. What the item numbered three beside the Vināyakas is. Why the ring called five krośa is given as twenty-five. And whether the counts are in the Kāśī Khaṇḍa at all, since the circuits are attributed to it and the numbers aren't.
The counts and the summary don't agree, and the disagreement is left standing. Four rings are itemised with numbers and the fifth is described as a point. Then one sentence rolls the whole thing up as 144, 108, 72, 72, 72. Either a fifth number exists and wasn't given in the walk-through, or the summary miscounted, and the passage moves on without noticing.
And the fourth ring's description doesn't work like the others'. Rings one, two and three are given by extent — eighty-four krośa, twenty-five krośa, the urban city. The fourth is given by a promise and a temple area, with no distance attached at all.7 It's the one the speaker calls very very important, twice, and it's the one whose size the passage never states.
Sacred Containers: Designing Rituals That Reorganize — a bounded space and time in which the ordinary rules are suspended, and which works because it strips out the competing demands a person is normally holding at once. A military unit, a startup team, a dojo. The claim is that the boundary itself does the work: inside it, attention stops being divided.
What that account needs and rarely gets is a boundary somebody can actually find. A container defined by intention alone tends to leak, because nobody can say precisely when they're in it. Here the boundary is a road with a hundred and forty-four shrines on it, and the count is what makes it legible — you know you're inside because you passed the stations, and you know you completed it because the number came out.
And this material supplies something the container literature doesn't have at all: nesting. In the behavioural account there's one container, with an inside and an outside. Here there are five, each one inside the last, each finer than the one before it, and a practitioner moves inward through them in sequence. That turns a boundary into a gradient — and it answers a question the single-container model leaves open, which is what happens to somebody for whom the container has stopped doing anything. On this scheme they haven't outgrown it. They've finished a ring and there's a smaller one.
One handshake. The tension with the vault's other pilgrimage-geography material is internal to the subject and sits above.
Sharpest implication. Four rings have counts and the fifth doesn't, and the missing number is the most informative thing in the passage. Counting is what you do while you're still going round something. At the centre there's no route left to pass stations on — which means the whole numbered scheme is scaffolding for people who are still on the outside, and the last step is the one where it stops applying.
Generative questions.