An object that mattered to a lot of people is no longer where it was.
There is always an account of how that happened, and the account almost always has one of two shapes. Either it is a story about somebody taking it — an army, a collector, a government — or it is a story about the object arriving somewhere, told from the destination, in which the taking has quietly become a coming.
The second version is not a lie in any sentence. It simply assigns the verb differently. Nobody carried it; it came. And once the verb has moved, so has the question anybody would otherwise ask, which is who moved it and what they wanted.
The Caitanya-caritāmṛta does this three times, in three chapters, with three different kinds of force behind it — and it does it so consistently that the pattern is easier to see here than in most places it operates.1
One — concealment. The Gopāla of Govardhana is found buried in a bush. He was put there by a priest who fled a Muslim raid, and the corpus states that cause once and never returns to it.2 In the dream that leads to the recovery, the Deity's own account of his situation is "I reside in this bush, and because of this I suffer very much from severe cold, rain showers, winds and scorching heat."
Two — flight, and a stated alibi that the text itself corrects. A Deity is transferred to Gāṅṭhuli-grāma "on the plea of being attacked by Muslims." And the text says plainly that this was not the reason. The real reason was to spare Caitanya the climb up Govardhana, which he had vowed not to make.3
Three — conquest. King Puruṣottama-deva takes Vidyānagara, carries off the jewelled Māṇikya throne to Jagannātha Purī, and takes the Gopāla Deity to Kaṭaka. This is narrated as a request, graciously accepted: "Please come to my kingdom."4
Three chapters. Three forms of force — concealment, flight and conquest. And in every one of the three, the god is given the agency and the circumstance is given the alibi.
The second instance is the important one, and it is important because the corpus catches itself.
The stated reason for the transfer is a Muslim attack. The text then tells the reader, without embarrassment, that the stated reason was a plea — a pretext — and that the operative reason was the convenience of one man who had made a vow about a hill.
So the corpus knows exactly what a cover story is. It can identify one, it can name the real motive underneath it, and it does so here in the ordinary course of narration.
Which makes the other two instances impossible to read as innocence. A text that can see the mechanism in the middle case is not failing to see it in the outer two; it is applying a different standard, and the difference is that the outer two have nothing awkward to disclose. The burial and the conquest can be told as the god's own arrangement all the way through, because nobody in the story needs an excuse.
The pretext gets named only when there is a second, smaller truth that the tradition would rather have on record — that this happened for Caitanya's convenience. The larger truths, the raid and the annexation, are absorbed without comment, because the devotional account is the only account and there is no competing version to reconcile.
The device is entirely grammatical and that is why it is hard to notice.
Every one of these sentences is true. The Deity was in a bush. The Deity was transferred. The Deity went to Kaṭaka. Nothing has been fabricated, and no reader has been told anything false.
What has been done is the removal of the agent from the sentence and its replacement with the object. A priest hid it becomes he resides in this bush. A king took it becomes please come to my kingdom. The event survives intact; the actor evaporates.
And the effect is not primarily on belief. It is on what remains askable. If the god came, there is no further question — a god's movements are his own business. If a king took it, there is immediately a second question about the king, the campaign, and what else was carried out of Vidyānagara that year, and the corpus has no interest in any of that.
A story with the agent removed is a story with no follow-up. That is the whole return on the device, and it is available to anybody who can write a sentence.
The most revealing detail is the one that sounds least significant, and it is the buried Deity's own account of his condition.
He does not say he was hidden. He does not say a raid happened, that his priest was frightened, or that his temple is gone. He says he is cold, and wet, and too hot, and that the wind gets at him.5
That is a domestic complaint. It is the register of somebody in an uncomfortable room, and it is the least political sentence available about a statue buried in an emergency.
And it is the sentence that produces the action. A raid is a historical event and produces reflection; discomfort is a present condition and produces a spade. The corpus's mobilisation follows immediately and at enormous scale — villagers cutting a path, nine hundred pots, an entire village's grain — and none of it is framed as a restoration or a reversal.6
Which is the deeper version of the finding. The grammar does not merely remove the agent; it re-times the event. A concealment is something that happened, in the past, done by somebody, with a cause. A cold god is something happening now, to a person, that anybody can fix.
Political facts are in the past tense and can only be resented. Discomfort is in the present tense and can be relieved — and a tradition that consistently converts the first into the second will get a great deal built and will remember almost nothing.
It is worth being exact about the cost, because it is not that the tradition is deceived.
What is lost is the priest. Somebody buried that Deity, under threat, at speed, and by doing so preserved the object the entire subsequent institution is built around. He fled and he is not named, not thanked, and not present in any part of the celebration.
What is also lost is the other end of the third case. The Gopāla that arrives at Kaṭaka arrived from Vidyānagara, which had just been conquered. There were people there for whom that Deity was the local one, and the corpus's account — a gracious acceptance of an invitation — has no room in it for their side, because in that account nothing was taken from anybody.
And what is lost in the middle case is nearly nothing, which is the tell. There, where the concealed motive was merely a vow about a hill, the corpus discloses. The disclosure rate tracks the cost of disclosing, and that is a property of accounts in general rather than of this one.
Layer A supplies all three relocations, the burial and its cause, the "plea of being attacked by Muslims" with its correction, and the king's carrying of the Deity to Kaṭaka.7
Layer B expands the material into the doctrine of the Deity's non-different status and the reciprocity between the Lord and his devotees, and treats each movement as the Lord's own arrangement for a devotee's benefit.8
Worth marking what the commentary does with the middle case. Layer A's correction — that the Muslim attack was a plea and the real reason was the climb — is the most historiographically interesting sentence in the set, and it is a Layer A sentence. The commentary has no need to press it, because on Layer B's account both reasons are subordinate anyway: the Lord arranges his own location, and the human motives are surface.
Which means the corpus's own most sceptical moment belongs to the seventeenth century and not to the twentieth, and a reader who works only from the purports will not meet it.
You are reading an account of how something ended up somewhere. Find the verb and ask who is doing it. It came, it was transferred, it found its way are all agentless constructions, and the agent has usually been removed rather than never existed. This costs nothing to check and it works on museum labels, company histories and family stories alike.
You want something done about a bad situation. State it as a present discomfort, not a past injustice. The raid produced nothing for two centuries; I am cold in this bush produced a temple in a day. That is not a recommendation to suppress history — it is an observation that the two registers recruit completely different faculties, and only one of them recruits hands.
You are writing an account and you have something awkward to disclose. Notice that your disclosure rate is tracking your cost. This corpus discloses the small pretext and absorbs the large ones, and it is not being strategic about it — the small one just felt worth the sentence. Everybody's record has this shape, and the reliable read of any account is not what it admits but what it did not think needed admitting.
Somebody's story leaves you with no further questions. That is a property of the telling, not of the events. A complete-feeling account is usually one from which an agent has been removed, and the fastest way back in is to ask who benefited from the movement rather than why it happened.
The evidence is three separate relocations in three chapters, each independently narrated, with the pattern verified at [n≥3] in the extraction record.9
The tension the source doesn't close: the corpus can identify a pretext and does so, once. It names the Muslim-attack plea as a plea and supplies the real motive — and applies no such scrutiny to the burial or the conquest, which are the two cases with actual force behind them.
A second one. The theology holds that this party lacks no capacity whatever, and all three narratives depend on him being somewhere he did not choose to be and needing human beings to move him. The corpus resolves this by making each move his idea after the fact, which preserves the theology and dissolves the history.
Open: what happened to the priest who buried the Gopāla. He is the reason the institution exists, and he is named nowhere in the corpus and thanked by nobody in it.
This page and The Excavation of Gopāla are the same burial at two scales. That page reads the recovery as the founding of an institution and follows the pots, the cows and the priesthood; this one reads the burial as the first of three forced movements and follows the grammar instead.
Against Sākṣi-gopāla the Witness it is the third relocation seen from inside. There the god walks to Kaṭaka's predecessor dispute under conditions he sets himself and after losing an argument; here he is carried off by a king who had just annexed the territory — and the corpus narrates both as his own decision, which is the point.
And it belongs beside The Epilepsy Cover Story, which is the same corpus recording a cover story in its own protagonist's mouth. There the alibi is disclosed to the reader and its subject supplies it knowingly; here the alibi is disclosed once in three instances and the other two are absorbed. This text is capable of naming what it is doing and does so unevenly.
Divine Attribution — that page's mechanism is the claiming of a supernatural mandate for one's own authority: the commander who is the son of Zeus-Ammon, operating in a register where ordinary standards of evidence do not apply, so that what would need justifying instead needs only announcing.
The three relocations are the same instrument pointed the other way, and the pairing is what makes the mechanism legible.
Standard divine attribution transfers authority toward the speaker. I am chosen, therefore my decision stands. The speaker gains licence, and the cost is that the claim is about them and can be disputed as such.
Here the attribution transfers agency away from the speaker. He wished to come, therefore nobody decided anything. The king gains something better than licence — he gains absence from the sentence. A claim of divine mandate invites the question of whether you have one. A claim that the god acted invites no question at all, because the actor has no interests anybody can interrogate.
Which gives the finding neither page reaches. Attributing agency to a god is more powerful as an exculpation than as a credential, because a credential is a claim about a person and can be tested, while an exculpation removes the person from the account entirely — and there is no procedure for testing a sentence that has no defendant in it.
That is also why the device survives secularisation intact. The market decided. The system moved. Circumstances required it. These are the same sentence with the deity removed, and they do the same work: an event with a cause, no agent, and nothing further to ask.
Sharpest implication. The most effective use of divine agency is not to claim a mandate but to vacate the sentence — an event attributed to a god has a cause and no defendant, and therefore no follow-up question. This corpus proves it knows the device, because it names the pretext exactly once, in the one case where the concealed motive was trivial. Disclosure rate tracks the cost of disclosing, and the reliable read of any account is what it did not think needed admitting.
Generative questions.