Somebody at work says a phrase in a stupid voice. It's a bad joke about a bad slogan and everyone laughs once.
Three weeks later the office cannot say the phrase any other way. The stupid voice has replaced the real one. And the slogan — which nobody believed, which was the whole point of the joke — is now said forty times a day by people who despise it.
Then somebody uses it sincerely, in a meeting, without irony, and doesn't notice.
That's the ordinary version. Ādi 17 has the theological one, and the men it happens to are frightened.
The complaint is brought by Muslim meat-eaters, and it is a complaint about themselves.
They had chanted "Hari, Hari" — to mock the Hindus. Not as an experiment, not half-seriously. As derision, aimed at neighbours, using the neighbours' own sacred words as the instrument.
And now:
"I have no desire to say it, but still my tongue says it. I do not know what to do."1
Read the grammar. I have no desire. My tongue says it. The speaker has separated himself from his own mouth, and he is reporting the mouth as a third party that has gone on without him.
There is no piety in this passage. Nobody has converted. What is described is a man who finds his own body producing a sound he disowns and cannot stop, and who is telling the authorities about it because he does not know what else to do.
This is the corpus's phonetic-efficacy claim with the most hostile possible input.
The general principle, stated elsewhere in the same corpus: the letters of the name "have so much spiritual potency that they act even when uttered improperly."2 Potency in the syllables, not in the heart.
Every other case the corpus offers for that claim is neutral — a dying man's cry, a father calling a son, a habitual exclamation with no religious content.3 Those establish that intention is not required.
This case establishes something stronger: intention can be actively hostile and the mechanism still runs.
Which is why the extraction record kept this page separate from the nāmābhāsa material rather than merging them. Same doctrine, different input, and the corpus's own boundary — no intent works, wrong intent does not — passes precisely between them.4
The boundary needs care here, because mockery looks like it should be the worst case and the corpus puts it on the safe side.
The corpus's disqualifying category is instrumentalising intention: chanting for wealth, for followers, for bodily benefit, to neutralise one's own conduct. Each of those interposes a purpose between the chanter and the sound. The person is using the name to get something.
The mockers are not using the name to get anything. They are making a noise at somebody. The sound is not being pointed at an outcome; it is being pointed at a neighbour, as an insult.
So nothing is interposed, and the syllables arrive unobstructed.
The scoffer is in a better position than the devotee with a motive. The corpus does not draw that conclusion in as many words and it follows directly from the two passages, and this build states it because leaving it implicit is what produces the flattened version of this doctrine that circulates everywhere.
Worth being careful about what the passage actually contains, because the theological reading and the ordinary reading are both fully available and the text does not adjudicate.
The theological reading. The name is potent independently of the speaker; hostile use is still use; the men have been caught by something real and are experiencing its early stages.
The ordinary reading. Involuntary verbal repetition after heavy exposure is one of the best-attested phenomena in ordinary experience. The jingle, the catchphrase, the parody that installs what it parodies. Mockery requires rehearsal — you have to produce the sound repeatedly and with attention to get the imitation right — and rehearsal is exactly the input that produces intrusion. On this reading the men have earworms and a religious framework for interpreting them.
The two readings agree on every observable and differ entirely on the cause. Nothing in the text distinguishes them, and the corpus is not trying to — it is not aware that a second reading exists.
What is worth noticing is that the ordinary reading is not a debunking. It explains the mechanism and leaves the corpus's practical conclusion standing: if you want a sound to lodge in a hostile population, having them repeat it in mockery is a highly effective delivery method, and the corpus has observed this correctly whatever it thinks is happening.
And the two readings differ on one testable point, which neither the corpus nor a sceptic usually bothers to name. The theological account predicts the repetition is good for them — it is deliverance, arriving unrecognised. The ordinary account predicts it is merely persistent, and will fade with disuse like any other intrusion. Those come apart over time, and the corpus, having recorded the complaint, never checks.
A tradition selects which true things it preserves, and the selection tells you what it needed.
This story does three jobs at once.
It proves the doctrine at its limit. Anyone can believe the name works for a sincere chanter. Producing a case where it works on someone attacking you is the strongest available demonstration.
It converts an insult into an asset. The community was being mocked. This story means the mockery was the mechanism, which is a considerable improvement on being mocked.
And it removes the sting from opposition generally. A movement whose enemies' derision is itself evangelism has made hostility survivable — and this corpus needed that, because its early history is full of hostility, including a petition from its own Hindu neighbours to have Caitanya expelled.5
None of which requires the story to be untrue. It requires only that we notice why this one, out of everything that happened in Navadvīpa, is the one that got written down.
There is a fourth job, and it is the one that would have mattered most at the time.
The men are not punished. They come and report a distressing condition, caused by an act of contempt against the community they are reporting to, and nothing is done to them. In a chapter that also contains a magistrate breaking a drum, threats of confiscation and forcible conversion, and a hundred thousand people in the street, that is a conspicuous omission.
Which makes the story usable as an assurance rather than only as a proof. A Muslim in Navadvīpa hearing it learns that the people running this movement treat an offender as somebody something happened to. That is a de-escalating story about a community in the middle of a communal dispute, told by the side with the crowd.
And it stands against the same chapter's worst sentence — the line the text puts in Caitanya's mouth about killing all the Muslims.8 One chapter, one community, a threat of massacre and a lenient handling of actual offenders. The conduct and the rhetoric point in opposite directions again, and again the conduct is the better of the two.
Layer A reports the complaint in the men's own voices and lets it stand as testimony.
Layer B is warm about it and uses it to argue that anyone who comes into contact with the name benefits regardless of attitude.6 That is exactly the verse's logic and it is applied outward, to a twentieth-century readership, as an argument for public chanting in places where it will not be welcome.
Worth marking one thing the commentary does not do: it does not turn the mockers into villains. Both layers treat them as people something happened to. In a corpus that can be extremely harsh about offence, the men who used the holy name as an insult are handled gently, which is a real datum about where this tradition thinks the danger actually lies.
You are mocking something by repeating it. Notice you are rehearsing it. Parody is the highest-repetition form of engagement there is, and it installs the thing at a rate sincerity never matches. This is the whole reason bad slogans outlive their campaigns.
You want something to spread in a room that is hostile to it. The corpus's observation is that opposition which requires quoting you is a delivery mechanism. Design the phrase so that attacking it requires saying it.
You find yourself producing a phrase you disown. Don't treat it as a failure of will. It arrived by exposure and it will leave by disuse, and the effort of suppressing it is itself a form of rehearsal.
Somebody in your group is delighted that your critics keep quoting you. They are right about the mechanism and it cuts both ways — everything you repeat in order to attack it is being installed in you at the same rate.
The evidence is a short first-person complaint with a stated cause — chanting in derision — and a stated present condition.7 It is testimony from hostile witnesses, which is a strong evidential form and is also, here, testimony collected and recorded by the party it flatters.
The tension the source doesn't close: this passage and the intention-based doctrine of devotion are both absolute. The corpus's central distinction between love and lust rests entirely on the actor's aim; the name's efficacy here rests on nothing at all. The corpus's own boundary reconciles them, and the reconciliation is never applied in the disciplinary chapters where it would have changed rulings.
A second one, and it belongs to the reading rather than the text. The theological account and the ordinary account of involuntary repetition agree on every observable. The corpus does not know it is making a claim that has a competitor.
Open: what happened to these men. The text records the complaint and moves on. Nothing in the range read says whether the tongues stopped, whether they converted, or whether anyone followed up — which is a recurring silence in this corpus wherever an ordinary person has served a demonstration.
This page and namabhasa-and-the-yavanas were assessed for merger during the consolidation pass and the merger was declined, which is upheld here. They carry the same doctrine with opposite inputs — habit and hostility — and the corpus's boundary between absence of intent and instrumentalising intent runs exactly between them. Merged, the page could state the doctrine but could not exhibit the line.
Against Kāma and Prema: The Direction Test this is the corpus's sharpest internal collision and it is worth holding open. That page makes the actor's aim the entire criterion, with no content test at all. This one has an act performed with the worst available aim producing the intended effect. Both are absolute; both are load-bearing; and the corpus's reconciliation covers the name and not the conduct.
And it supplies the mechanism for The Inundation and Those Who Fled. That chapter worries about people expert at avoiding the flood. This one shows the one route that expertise cannot block: a sound you have to produce in order to attack.
Mental Contagion and Mass Delusion — that page opens with a nine-year-old who invents a headache to avoid a maths test and discovers, through her mother's anxious tenderness, a set of responses that do not switch off when the test does. Its subject is how a state that begins as something a person is doing becomes something that is happening to them, and then spreads.
The mockers are the same transition, compressed and reported by the people it happened to.
Chanting in derision starts as an act with an author. It requires effort, aim, and an audience. And at some point in the repetition the authorship drops out and what remains is the behaviour, running on its own — I have no desire to say it, but still my tongue says it. The man is describing the exact moment the contagion page is about, and he is describing it about himself, which is the position from which it is almost never reported.
What the pair produces is a distinction neither draws alone. Contagion literature is mostly about states spreading between people — the schoolgirl's family, the hysteria that runs through a factory floor. This case is a state spreading within one person, from voluntary to involuntary, with no second party required. The mockers did not catch it from the Hindus. They caught it from themselves, by rehearsal.
Which reframes what a hostile audience actually is. The contagion model treats exposure to believers as the risk. This case says the risk is exposure to your own production of the material, and that the most exposed person in any argument is the one doing the quoting. Attack requires reproduction, reproduction is rehearsal, and rehearsal is the input. A movement that is being mocked in its own words is being propagated by its opponents' mouths at a rate its own members cannot match, because its members say the thing sincerely and occasionally while its opponents say it derisively and constantly.
That is a real and slightly alarming general finding about how ideas move, and it does not require anybody to believe anything about the syllables.
Sharpest implication. Opposition that has to quote you is not opposition; it is distribution with a hostile tone, and the people at greatest risk of absorbing a phrase are the ones repeating it to attack it. This corpus noticed that in the sixteenth century, kept the story because it was useful, and built a public practice on the premise that being audible matters more than being welcome.
Generative questions.