Three le south of Pataliputra in 405 CE, Faxian stands in front of a stone pillar. The pillar is fourteen or fifteen cubits in circumference and more than thirty cubits high. There is an inscription engraved on it. Faxian cannot read the inscription — the script is Brahmi, which has been illegible to Chinese travelers for centuries and will remain so until James Prinsep deciphers it in 1837, more than fourteen hundred years after Faxian's visit.
But somebody at the site reads it to him. Or paraphrases it. Or tells him what tradition holds about it. And Faxian writes down what he is told: "Asoka gave the jambudvipa to the general body of all the monks, and then redeemed it from them with money. This he did three times."1
That sentence is one of the most important single lines in the entire chronicle. It is a 5th-century-CE intermediate witness to the content of an Asokan inscription, transmitted via on-site oral tradition, recorded by a Chinese monk who could not himself read the script, preserved in his book that survives via Korean recension to a James Legge translation in 1886. Every link in the chain except Faxian and the modern reader is dead. And yet the sentence is still readable, and what it claims is administratively specific enough to test against the surviving epigraphic record.
Pull it apart. The sentence makes three distinct claims, each load-bearing.
Claim 1: Asoka gave the jambudvipa to the monks. "Jambudvipa" is the Buddhist cosmological name for India — the inhabited landmass. The gift is a sovereignty-transfer. Asoka conveys legal title to the territory of India to the Buddhist Sangha. This is not a metaphor. The verb is "gave."
Claim 2: He then redeemed it from them with money. Asoka bought it back. After the gift, he paid the monks an amount sufficient to repurchase the title. The territory returned to royal sovereignty, but the monetary transaction completed a closed loop.
Claim 3: This he did three times. The give-and-redeem cycle was performed not once but three times. The repetition is part of the claim.
Legge's footnote on this is unusually pointed: "We wish that we had more particulars of this great transaction, and that we knew what value in money Asoka set on the whole world. It is to be observed that he gave it to the monks, and did not receive it from them. Their right was from him, and he bought it back. He was the only 'Power' that was."2
Legge's reading is precise. The direction of the transaction matters. Asoka is not asking the monks for permission. He is giving them sovereign title, and then exercising the only-sovereign's-right to repurchase what he gave. The Sangha is the legal beneficiary, but the sovereignty Asoka displays through the cycle is his alone. He is the actor; the monks are recipients-then-sellers. The three-times-repetition emphasizes that the cycle is willed, deliberate, and exercised as a sovereign prerogative.
The standard reading of the give-and-redeem cycle in Buddhist tradition is that it is the Asokan founding-gesture for state-to-sangha property transfer. Asoka performs the symbolic transfer of the entire territory. The monks receive sovereignty. Asoka pays them back. The Sangha now holds a paid-up sovereignty title that has been honored by purchase. The mechanism establishes the precedent: the state acknowledges the Sangha as a legitimate property-holder and pays for the right to retain sovereignty itself.
In administrative-symbolic terms, this is the founding move for the entire post-Asokan property regime. The metal-plate land grants Faxian witnesses six centuries later (see Metal-Plate Land Grants as Monastic Economic Foundation) are descendants of this gesture. Each grant is a small re-enactment of the Asokan founding cycle: the king gives, the Sangha receives, the title is honored multi-generationally. Without the founding precedent, the grant system has no symbolic grounding. With it, every subsequent king is operating within an Asokan-established framework where state-to-sangha property transfer is the recognized normal mode.
The three-times repetition is plausibly a ritual-numerical emphasis (three being the symbolic Buddhist number for the Three Jewels — Buddha, Dharma, Sangha). It may also be reading what Asoka actually did — a multi-stage symbolic transaction performed at multiple sites or multiple times in his reign, each repetition deepening the precedent.
Here's the load-bearing thing: no Asokan inscription with this specific content has been preserved. The surviving Asokan epigraphic corpus (Major Rock Edicts, Minor Rock Edicts, Pillar Edicts, Schism Edicts, Cave Inscriptions) does not contain a passage that paraphrases as "I gave the jambudvipa to the monks and bought it back three times." We do not have the original inscription Faxian was standing in front of.
This means Faxian's paraphrase is the only surviving witness to the substance of an Asokan inscription that has otherwise been lost. The pillar he saw at Pataliputra either no longer exists or has been damaged beyond legibility. The inscription Legge note 10 commented on so suggestively is preserved only in Faxian's secondhand report.
That is what makes the line important. It is primary-source-of-the-content even though Faxian could not read the script. The content was preserved in on-site oral tradition at the pillar, communicated to Faxian by whoever was acting as informant at the site (a local monk, an attendant, a translator — Faxian doesn't say). Faxian then recorded what he was told. Six centuries of intervening events have erased the pillar. Faxian's chronicle preserved the substance.
The implication for how Asokan history reaches us: the surviving Asokan corpus is not the totality of what Asoka inscribed. It is the stuff that survived the medieval period to be read by Prinsep. Inscriptions that were destroyed, lost, or repurposed before the 19th century are gone — except for the tiny fraction whose content was preserved in non-epigraphic witness records like Faxian's. The Asokan documentary base is therefore the surviving inscriptions + Faxian's reported paraphrases + a small number of later Buddhist-text references. The base is wider than the surviving stones.
Faxian could not read Brahmi. The existing vault page on The Emperor's Afterlife — Buddhist vs Political Memory (from the Lahiri ingest) flags this directly: Brahmi was unreadable to Faxian because Brahmi script had fallen out of common literacy by the 5th century CE — even most Indians of the period could not read it, the script having been replaced by later derivative scripts (Gupta Brahmi → Siddham → various regional descendants). Prinsep's 1837 decipherment recovered Brahmi after fourteen centuries of dormancy.
So how did Faxian get the inscription's content? Four possibilities, in decreasing likelihood:
Possibility 1: On-site Buddhist oral tradition. The most likely. The pillar was a known Buddhist pilgrimage site (it was three le south of Pataliputra at the largest of Asoka's 84,000 topes, with a footprint of Buddha at the base). Local monks would have known what the inscription said because the inscription's content was part of the site's instructional landscape. They told Faxian what tradition said the pillar declared. Faxian wrote it down.
Possibility 2: A literate informant who could still read the script. Possible but less likely by 405 CE. Brahmi-literacy may have persisted as a specialist competency among a few learned individuals — religious scholars, court historians — even as it had vanished from common use. One of them could have read the inscription for Faxian directly.
Possibility 3: A later Sanskrit or Prakrit translation visible at the site. Some Asokan inscription sites were re-inscribed in later scripts as the original Brahmi became illegible. A bilingual or re-translated marker at the pillar would have been readable to Faxian's interpreters.
Possibility 4: Faxian invented or embellished the content from generic Buddhist-political-tradition framings. This is the least likely but worth noting as a sanity check. Faxian's epistemic discipline (see Faxian Witness vs Hearsay Epistemic Discipline) speaks against this — he marked his tier-shifts when claims came from hearsay or speculation. The paraphrase is recorded as inscription-content, not as legend. Faxian himself plausibly believed he was reporting what the inscription said.
The most likely chain is Possibility 1: site-specific Buddhist oral tradition preserving the substance of the inscription across the period when its script became illegible. This is exactly the same mechanism by which the on-site oral traditions at jataka-marker topes preserved the legendary narratives of those sites — local monks held the content; Faxian collected it.
The vault page The Emperor's Afterlife states that "Brahmi unreadable by Faxian" is a defining fact of the post-Mauryan political memory — Asoka's inscriptions became illegible monuments during the long period between his death and Prinsep's 1837 decipherment. The Faxian-paraphrase finding does not contradict this. It enriches it.
The reconciliation: Faxian saw the pillars and the inscriptions but could not read them; he received the inscription content through on-site oral interpretation; the content he records is therefore primary-witness to the substance but not direct-witness to the script. The Lahiri-corpus claim (Faxian could not read Brahmi) and the Faxian-paraphrase claim (Faxian recorded the inscription's substance) are both true, with no conflict, once the interpretive chain is named.
This matters for how the entire post-Mauryan Asoka-reception history reads. Asoka's apparatus did not become fully illegible. It became scriptually illegible while the substance was preserved orally at major sites. The on-site Buddhist oral tradition was the bridge across the centuries of script-dormancy. Faxian's chronicle is one of the few documents that preserved a piece of that oral bridge in writing before it too was lost.
The primary witness is Ch. XXVII line 1240: "On the pillar there is an inscription, saying, 'Asoka gave the jambudvipa to the general body of all the monks, and then redeemed it from them with money. This he did three times.'"3 Legge note 10 (line 1264): "We wish that we had more particulars of this great transaction, and that we knew what value in money Asoka set on the whole world. It is to be observed that he gave it to the monks, and did not receive it from them. Their right was from him, and he bought it back. He was the only 'Power' that was."4
Cross-validation from outside the source: the Asokan Schism Edict (preserved at Sarnath, Sanchi, Kausambi) does contain references to property-and-jurisdiction conveyances between the king and the Sangha, but not in the give-and-redeem-three-times format Faxian reports. The surviving epigraphic corpus and Faxian's paraphrase point to overlapping but not identical content — the surviving inscriptions show Asoka regulating sangha-internal affairs through inscriptional law; Faxian's paraphrase shows Asoka transacting territorial sovereignty with the sangha. The two registers are complementary, not contradictory. The give-and-redeem cycle Faxian records may be from a now-lost inscription in a different register from the surviving Schism Edicts.
Tibetan Buddhist tradition (preserved in later canonical Buddhist literature) records that Asoka performed a "great gift" of the territory to the Sangha at his consecration and at subsequent stages of his reign. The motif of state-territory-given-to-sangha-and-then-repurchased is therefore not unique to Faxian's chronicle. It appears across Buddhist tradition in various forms. This is evidence that the underlying narrative — whether it corresponds to an actual Asokan inscriptional act or to a post-Asokan symbolic re-reading — was widespread Buddhist memory by the 5th century CE.
The most acute tension: was there an actual Asokan transactional event corresponding to Faxian's paraphrase, or is Faxian recording a Buddhist symbolic re-reading of more conventional Asokan property donations?
The evidence is mixed. If there was an actual three-times-give-and-redeem event in Asoka's reign, it would be among the most striking sovereign acts in pre-modern history — a sitting emperor of one of the largest empires of the ancient world literally transferring legal title of the whole territory to a religious order, three times. There is no other corroborating evidence for this event happening at the scale Faxian's paraphrase implies. The surviving Asokan inscriptions document many specific Sangha-property donations (rest houses, water tanks, monastery endowments) but not a wholesale territorial-sovereignty transfer.
The alternative reading: Faxian's paraphrase records a Buddhist-tradition reframing of Asoka's pattern of repeated Sangha donations as a single mythic give-and-redeem cycle. The cumulative effect of many donations is read back as a singular symbolic act — Asoka "gave the jambudvipa" via the cumulative donation-pattern, and "redeemed it" via the continued exercise of royal sovereignty over the territory. The three-times-repetition is a ritual amplification of the cumulative pattern.
A second tension: even if the paraphrase records actual content, Faxian's interpretive chain (on-site oral tradition → Faxian's interpreter → Faxian's writing) introduces transmission uncertainty. The phrase "did this three times" is exactly the kind of mythicizing detail that tends to enter as oral tradition layers a numerical pattern onto a more diffuse historical reality.
A third tension: the existing vault corpus on Asoka (Lahiri 2026-05-13) reads Asoka primarily through his confessional-public-genre inscriptions (the Rock Edicts where Asoka talks about his moral evolution after Kalinga). Faxian's paraphrase reads Asoka through a transactional-property-conveyance inscription. These are different Asokan registers. The same emperor was capable of multiple epigraphic genres — moral confession in one inscription, property transaction in another. The Lahiri-foregrounded confessional register and the Faxian-witnessed transactional register both belong to Asoka's actual epigraphic toolkit. Neither alone is the complete Asoka.
Lahiri stands at Maski with a surviving Asokan inscription in front of her and reads it in its physical context. Faxian stands at Pataliputra with an Asokan inscription in front of him and cannot read it — Brahmi has gone illegible — but a local monk tells him what it says. Both scholars are doing landscape-mediated work on Asokan epigraphy. Lahiri has the text; Faxian has the on-site interpretation. Together they document layers of a transmission chain neither sees alone.
Lahiri's method is to read each Asokan inscription in its specific physical and political context — landscape, audience, immediate political circumstances. Faxian's paraphrase is the inverted situation: an inscription Lahiri cannot read because it does not survive, only the on-site Buddhist-oral preservation reaches us. They converge on the principle: understanding an Asokan inscription requires understanding the local interpretive chain around it. Lahiri does this for the surviving inscriptions. Faxian's witness is the rare case where the interpretive chain (on-site Buddhist oral tradition) is the only thing that survives — the inscription itself is gone.
They diverge on what they can recover. Lahiri can recover the inscription-as-text plus its surrounding interpretive context. Faxian can recover only the interpretive-chain-output, with the original text inferable but not directly accessible. The split reveals an important asymmetry: when the inscription survives, modern scholarship can question the local interpretation; when only the local interpretation survives, modern scholarship has to take it at the weight of its on-site preservation, with no ability to back-check against the inscriptional language.
What the convergence implies — that neither alone shows — is that the full Asokan documentary record is a layered transmission, not just a set of inscriptions. The inscriptions are one layer. The on-site Buddhist oral interpretations are another. Pilgrim chronicles that record fragments of the oral layer are a third. Later Buddhist textual references are a fourth. Modern decipherment recovers Layer 1 (inscriptions) but the inferences about Layers 2-4 are necessary to read Layer 1 in context. Lahiri does this work for the surviving inscriptions. Faxian is one of the few sources that lets us see Layer 2 directly.
The Faxian-paraphrase reaches into three adjacent vault domains in productive ways.
History: Ashoka as Confessional Public Genre Inventor (Lahiri corpus) — Lahiri's reading positions Asoka as the inventor of the confessional-public genre (the Rock Edicts where Asoka admits his Kalinga-era violence and announces his moral conversion). The Faxian-paraphrase records a transactional-property-conveyance genre that is a different epigraphic register. Both genres are part of Asoka's epigraphic toolkit. The structural relationship: confessional and transactional are two distinct uses of the same physical-medium (inscribed stone) for different communicative purposes. Together they show that Asoka's epigraphic apparatus was multi-register, not unified. He used inscriptions for moral self-disclosure, for property donations, for jurisdictional regulation, for genealogical statements (or strategically for genealogical silence — see Ashoka's Deliberate Genealogical Silence). The handshake produces an insight neither domain generates alone: the most consequential pre-modern epigraphic apparatus was genre-promiscuous. Modern reception has tended to flatten Asoka into the confessional register because the Rock Edicts survived and are dramatic. Faxian's witness shows the transactional register operating at scale too. The flattening is a modern artifact of survival-bias.
Eastern-spirituality: Dana — Religious Charity as Cosmic Practice (concept implicit in sadhana-practice-hub) — Dana is the first of the six Buddhist paramitas, the means of liberation. The Faxian-paraphrase records Asoka performing dana at the cosmic-territorial scale — giving the whole jambudvipa to the Sangha. The structural parallel: dana at the individual level is alms-giving to monks. Dana at the sovereign level is sovereign-gift of territory to the Sangha. The mechanism is the same — gift to the Sangha is meritorious — but the scale produces qualitatively different consequences. The Faxian paraphrase shows what dana looks like when the practitioner is an emperor. The repeated cycle (give-redeem-give-redeem-give-redeem) is dana performed at maximum amplitude. The handshake produces an insight: the Buddhist doctrinal framework of dana scales unboundedly with the giver's resources. There is no upper limit on the merit-generating capacity of dana. The cosmic-scale version Asoka allegedly performed is the inscriptional template for every subsequent royal donation across the Buddhist political world.
Behavioral-mechanics: sovereignty as public performance (concept implicit in propaganda-mass-persuasion-hub) — Asoka's three-times give-and-redeem is sovereignty performed as public spectacle. The state's possession of territory is enacted through the public cycle of giving and repurchasing. This is a high-cost public-performance mechanism — the king must actually transfer title, pay the repurchase, and do it three times — but the cost is exactly what makes the sovereignty-claim credible. The structural parallel: behavioral-mechanics frames performance-of-power as a category of influence operation. The Asokan give-and-redeem is the pre-modern instance of sovereignty-credible-via-conspicuous-renunciation-and-recovery. The handshake produces an insight: the highest-credibility sovereignty performances involve temporary voluntary renunciation of the sovereign claim, because only a real sovereign can afford to give it up and take it back. The mechanism is structurally identical to potlatch-style status displays, but at imperial scale. The give-and-redeem cycle is potlatch as state-craft.
The Sharpest Implication
If Faxian's paraphrase is accurate even approximately, the post-Mauryan property regime in India ran on an Asokan founding precedent that explicitly recognized the Sangha as a sovereignty-comparable institution. The state and the Sangha are not nested (state on top, religious institution within it). They are peers, with the state acknowledging the Sangha's right to hold territorial sovereignty if the state chose to transfer it. The state retains sovereignty because it pays for the right to retain it, not because the alternative is unthinkable. The third-wire reading: modern state-religion separation assumes religion is a sphere-within-the-state. The Faxian-Asoka structure assumes religion is a sphere-coordinate-with-the-state — different in function, equal in standing. The modern arrangement is structurally weaker because it forecloses the option Asoka's three-times-cycle keeps open: that sovereignty can pass between institutions through paid transactions. We have made that motion impossible by treating state-sovereignty as constitutive rather than as one institutional possession among others.
Generative Questions