In 405 CE at Sravasti, Faxian writes a single sentence that should be more famous than it is: "There are also companies of the followers of Devadatta still existing. They regularly make offerings to the three previous Buddhas, but not to Sakyamuni Buddha."1
Read what just happened. Devadatta was the Buddha's cousin, his rival within the early Sangha, the figure who tried to murder him at least three times (the rock at Gridhra-kuta that hurt Buddha's toes, the poisoned-claw attack, the intoxicated black elephant). Devadatta led a schismatic faction that broke from the Sangha during the Buddha's lifetime. The Buddhist tradition canonized Devadatta as the archetypal enemy of the Dharma — Eitel's note describes him as Buddha's "deadly enemy" whose hatred persisted across lifetimes. The Pali canon's standard story is that Devadatta died being engulfed by fire, called out to Buddha for salvation in his last moments, and is destined eventually to become a Buddha himself (Devaraja in the Deva-soppana universe) after extended hellish purification.
That's the canonical narrative. Faxian reports a different historical fact: the Devadatta-faction did not die out. Nine centuries after the original schism, organized communities are still practicing a Buddhism that explicitly refuses to venerate Sakyamuni and instead venerates the three Buddhas of the present kalpa who preceded him (Krakuchanda, Kanakamuni, Kasyapa).
This is a ~900-year-surviving Buddhist sectarian split. The mainstream Sangha has tried to write Devadatta out of the tradition as the cautionary villain. But Devadatta's own followers, in 405 CE, are still there. They have an institutional identity. They have a veneration practice. They are organized enough that Faxian, walking past, can identify them as a distinct group.
The Devadatta-faction's practice is doctrinally precise. They venerate the three previous Buddhas of the present kalpa: Krakuchanda ("he who readily solves all doubts"), Kanakamuni ("body radiant with the colour of pure gold"), and Kasyapa ("swallower of light"). These are the three Buddhas who appeared in the present cosmic age before Sakyamuni. In standard Buddhist cosmology, Sakyamuni is the fourth Buddha of this kalpa, and Maitreya will be the fifth and last.
By venerating the three previous Buddhas but not Sakyamuni, the Devadatta-faction is making a specific theological claim: Sakyamuni is not a true Buddha. The three before him are recognized as authentic. The fourth, who was Devadatta's cousin and rival, is not. The veneration pattern operationalizes a counter-claim to Sakyamuni's legitimacy: the Devadatta-faction accepts the Buddha-lineage cosmology but excludes Sakyamuni from the lineage.
This is structural anti-Buddhism within the Buddhist conceptual framework. The Devadatta-faction is not non-Buddhist (they accept the Buddha-cosmology, the four-Buddhas-of-this-kalpa schema, the veneration practice). They are anti-Sakyamuni Buddhist — Buddhist by all the structural features except the specific founder-figure of the mainstream tradition.
The doctrinal mechanism for this: Devadatta during his lifetime attempted to claim leadership of the Sangha and proposed five additional ascetic rules (the "five points") that Buddha rejected as excessive. Devadatta then schismed and led a separate community. The Pali canon presents Devadatta's schism as motivated by personal ambition; the Devadatta-faction's own self-understanding (insofar as it can be reconstructed) was probably that Devadatta's additional rules represented a stricter, more authentic asceticism than Sakyamuni allowed, and that Sakyamuni's leadership was therefore lax or compromised. The veneration of the three previous Buddhas (who in tradition had stricter ascetic requirements, given the longer human lifespans of their epochs) coheres with this self-positioning.
Faxian could have left this out. The Devadatta-faction was the canonical enemy. Recording their persistence at Sravasti, the city where Buddha spent twenty-five teaching years, the site of the Jetavana vihara, makes the chronicle's portrayal of the mainstream Sangha less triumphal. Mainstream Buddhism has not eliminated its rival. The schism continues.
But Faxian records it anyway. The witness-discipline (see Faxian Witness vs Hearsay Epistemic Discipline) requires him to report what he sees. He sees Devadatta-followers at Sravasti. The chronicle records them.
The brevity of the recording is itself worth noting. Faxian gives the fact one sentence. He does not editorialize. He does not describe their numbers, their leadership, their texts, their daily practices, or their relations with the mainstream Sangha. He just notes that they exist. This is the discipline of factual reportage at its sparest. The presence of the rival is acknowledged; the chronicle does not amplify it.
What we lose by Faxian's brevity: we do not know how many Devadatta-followers there were at Sravasti, what their institutional infrastructure looked like (did they have monasteries? lay communities? specific patrons?), what their textual canon was, or whether the same factions existed elsewhere on Faxian's route but went unnoticed or unmentioned. The single sentence is a marker of an underground history that the chronicle preserves only in trace.
How did this faction survive nine centuries?
The mainstream Sangha had every institutional advantage. Mainstream Buddhism had royal patronage from Asoka onwards, metal-plate-grant economic substrate (see Metal-Plate Land Grants), textual transmission infrastructure, the international hospitality protocol (see Monastic Hospitality Protocol), and the relic-distribution geography (see Faxian's Buddha-Relic Inventory). The Devadatta-faction presumably had none of these — they were a sectarian remnant operating outside the mainstream sangha's institutional ecology.
Yet they survived. For nine centuries. Across the rise and fall of dynasties. Through the consolidation of Buddhism under Asoka, the Kushan patronage period, the Gupta efflorescence. Whatever they were doing, they were doing it well enough to retain organized institutional presence at Sravasti at the start of the 5th century CE.
Several possible mechanisms for the persistence:
Lay-community-based rather than monastery-based. If the Devadatta-faction operated primarily through lay practitioners with minimal monastic infrastructure, they would have been less dependent on royal patronage and metal-plate grants. The mainstream Sangha's institutional advantages would not be relevant to them.
Geographic dispersion below the mainstream radar. Faxian only notes the Devadatta-faction at Sravasti. They may have existed in many places without attracting Faxian's attention. The sparseness of the chronicle's mention may reflect a sparse-but-widespread distribution rather than a concentrated Sravasti-only presence.
Stricter-ascetic appeal. Devadatta's proposed five additional rules (no eating fish or meat, dwelling only in forests, wearing only refuse-rags, etc.) would have appealed to ascetics seeking stricter practice than the mainstream Sangha required. The faction may have continuously recruited from the strictest end of the lay-religious spectrum.
Counter-orthodox identity-stability. Sectarian groups often have unusually stable identities precisely because their oppositional stance is clearly defined. The Devadatta-faction's identity claim ("Sakyamuni is not the true fourth Buddha") is sharp and durable. A faction with this kind of clarity can persist across centuries even with small numbers.
None of these can be confirmed from Faxian's one sentence. The mechanisms are inferential.
The primary witness is the single sentence at Ch. XX line 950.2 The Devadatta-Buddha relationship is documented across the chronicle at Ch. XX line 944 (poisoned-claw attack, descending alive into hell — but Faxian does not specify that the descent is the conclusion of the rival's career), Ch. XXVIII line 1276 (Ajatasatru and Devadatta's collaboration to send the intoxicated black elephant against Buddha), Ch. XXIX line 1310 (Devadatta throwing the rock at Buddha's toes on Gridhra-kuta), Ch. XXX line 1342 (Devadatta's rock dwelling near Rajagriha).
Cross-validation: the Devadatta-faction's continued existence is mentioned in other Buddhist sources from later periods — Xuanzang in the 7th century records similar observations in some monasteries. The faction's exact later history is poorly documented; whether they survived into the medieval period or merged into other sectarian streams is unclear.
The three-previous-Buddhas identification (Krakuchanda, Kanakamuni, Kasyapa) is documented at Ch. XXI line 1016 (Faxian's chapter on the three predecessors of Sakyamuni) and Ch. XVII Legge note 16 (line 870) — "These three predecessors of Sakyamuni were the three Buddhas of the present or Maha-bhadra Kalpa."
The most acute tension: Faxian's chronicle preserves the Devadatta-faction's existence but does not preserve their voice. We have no Devadatta-faction texts. We do not know how the faction described itself. The mainstream Sangha's narrative is the only narrative we have, and that narrative is hostile.
The historical reality of Devadatta — was he a personal-ambition schismatic as the Pali canon presents him, or was he a serious ascetic-reformer with legitimate doctrinal differences from Sakyamuni? — cannot be reconstructed from the surviving evidence. Both readings are consistent with the textual record. The Devadatta-faction's continued existence in 405 CE suggests there was substance to their position beyond personal-ambition narratives; people did not maintain a schismatic identity for nine centuries on the strength of a founder's personality alone.
A second tension: Faxian's chronicle records the Devadatta-faction at Sravasti but does not record any institutional friction between them and the mainstream Sangha. Were the two communities operating in peaceful parallel? Were there active disputes that Faxian missed? Did the cross-school hospitality protocol extend to Devadatta-followers or stop at the boundary? Faxian's brevity leaves all of this unanswered.
A third tension: the Devadatta-faction's veneration of the three previous Buddhas is unusual. Other heterodox Buddhist movements typically claim direct lineage from Sakyamuni or from a specific bodhisattva. The Devadatta-faction's veneration-pattern is explicitly retrospective — they reach back to the three Buddhas before Sakyamuni rather than forward to Maitreya or laterally to alternative bodhisattvas. This is a structurally unusual move in Buddhist tradition.
Faxian's brief note sits in productive tension with the dominant Pali-canonical narrative about Devadatta.
The Pali canon presents Devadatta as the archetypal corrupted disciple — motivated by ambition, attempting to usurp Buddha's authority, performing increasingly desperate acts of aggression, ultimately destroyed by his own evil. The narrative function is moral cautionary tale: this is what happens when ego corrupts religious life.
Faxian's witness presents Devadatta as the founder of a still-living tradition. The brief sentence tells you Devadatta was institutionally serious enough to leave a faction that survived nine centuries. The Pali narrative's framing of Devadatta as a failed individual sits awkwardly next to the existence of an organized Devadatta-tradition six hundred years later.
They converge on the principle: Devadatta was a major historical figure within early Buddhism whose claims for institutional authority were rejected by the mainstream. They diverge on whether the rejection succeeded in eliminating his influence. The Pali narrative suggests the rejection was total. Faxian's witness shows the rejection was institutional-mainstream-level only — at the sectarian-margin level, Devadatta's claims persisted.
What the convergence reveals — that neither alone states — is that the mainstream Sangha's narrative-victory over Devadatta did not produce his institutional-elimination. The Pali canon won the textual battle. The Devadatta-faction won the long survival game by operating outside the textual battlefield. The mainstream tradition controls the historical record; the rival tradition operates as a parallel stream that the mainstream's textual control does not penetrate.
This is a structurally important fact about religious schisms: textual dominance does not necessarily produce institutional elimination of the rival. The two can run on separate channels for very long periods.
The Devadatta-faction's persistence reaches into three adjacent vault domains.
Psychology: Mass Movement Mechanics — Hoffer's analysis of mass-movement persistence depends partly on whether the movement has a clearly-defined oppositional identity. The Devadatta-faction's identity is exactly this kind of clear opposition: "the Buddhists who reject Sakyamuni." Whatever else changes, the opposition-marker is stable. The structural parallel: mass-movements with sharp oppositional identity-claims have unusual longevity even when small. The handshake produces an insight: counter-orthodox sectarian movements often outlive their majority-tradition opponents in narrative-identity stability, even if they are dwarfed institutionally. The Devadatta-faction was tiny compared to mainstream Buddhism. But the faction's narrative identity (refusal of Sakyamuni) was sharper than mainstream Buddhism's identity (which was constantly evolving across the eighteen schools and the Mahayana / Hinayana split). Identity-sharpness compensates for institutional-smallness in long-term survival.
Behavioral-mechanics: the manipulation and influence hub — the standard behavioral-mechanics analysis of sectarian movements focuses on charismatic-leader mechanics and recruitment dynamics. The Devadatta-faction has no surviving charismatic leadership (Devadatta himself died nine centuries before Faxian) yet maintains organized identity. The structural parallel: long-surviving sectarian movements often operate post-charismatic, with the founding-figure as historical reference rather than living presence. The mechanism shifts from charismatic-recruitment to inherited-identity-maintenance. The handshake produces an insight: sectarian-identity-maintenance mechanisms after the founder's death are different from charismatic-recruitment mechanisms during the founder's life, and the long-term survival depends on the post-charismatic mechanism. The Devadatta-faction at 900 years post-founder is operating purely on inherited-identity-maintenance. The original recruitment dynamics are irrelevant.
History: The Emperor's Afterlife — Buddhist vs Political Memory (Lahiri ingest) — Lahiri's analysis of Asoka's two-afterlives (Buddhist vs political memory) shows how dominant institutional traditions can preserve very different versions of the same historical figure than the non-institutional tradition does. The Devadatta-faction is the analogous phenomenon for Devadatta himself — the mainstream Buddhist memory preserves Devadatta as villain, while the Devadatta-faction (insofar as it survived) preserved a different memory the chronicle does not record. The structural parallel: historical figures have multiple parallel afterlives in different institutional streams, and the mainstream-stream's version is not the only one. The handshake produces an insight: for any historically significant religious figure, expect multiple parallel reception-streams with different valuations, even if one stream dominates the surviving textual record. Devadatta's parallel reception-stream survived institutionally but did not survive textually.
The Sharpest Implication
Take the Devadatta-faction's nine-century persistence seriously and the standard assumption that "religious schisms are usually resolved within a few generations" looks wrong. The Devadatta schism began c. 480 BCE during the Buddha's lifetime. Faxian witnesses its continued existence c. 405 CE. That is approximately 900 years of institutional persistence for a faction that was textually marginalized from its origin. The third-wire reading: religious-sectarian persistence is typically much longer than the mainstream tradition acknowledges, because the mainstream tradition's narrative tools cannot represent the rival's continued existence without weakening the mainstream's institutional victory claim. So mainstream narratives systematically under-document long-surviving rivals. Faxian's chronicle is a rare case where the mainstream-tradition author's witness-discipline preserves the rival's existence anyway. Most mainstream traditions did not have a Faxian. The implications for current religious-historical scholarship: there are probably many more long-surviving sectarian movements in the historical record than the dominant traditions' textual archives acknowledge. The Devadatta-faction was visible enough that Faxian noted it. Many others may have been less visible without being less persistent.
Generative Questions