Walk up to Ashoka with only his own edicts in hand and you get a man who appears, fully formed, at the moment of his Buddhist conversion. He has no past. He never names his parents. He killed a hundred thousand people at Kalinga and barely lingers on it. He sounds like a person who has decided what kind of person to be remembered as.
Walk up with only the archaeology and you get cities — Pataliputra's wooden palisade, Taxila's mud-coated stone houses, Maski's gold-belt megaliths — but no protagonist. You can tell what kind of road his army marched on, what kind of soil their bullock-carts struggled through after the monsoon. You cannot tell what the army was thinking.
Walk up with only the legend traditions — Ashokavadana from the 2nd century CE, Dipavamsa from the 4th, Mahavamsa from the 5th — and you get the brother-killer who burned 500 concubines alive and then built 84,000 stupas. The pre-conversion sin is maximized. The post-conversion virtue is maximized. The gap between them is the point of the narrative. You're reading hagiography, and you're far enough downstream from Ashoka that the figure has been working as a religious icon for centuries.
Each witness alone is unreliable. The first is too curated. The second has no voice. The third is too theological. Lahiri's method is to read all three at once, against each other, and let the cross-corrections produce a figure none of them produces alone.
The edicts tell you what Ashoka wanted future readers to know. The archaeology tells you what kind of landscape that message landed in. The legends preserve circumstantial details the hagiographers wouldn't have invented — geography, dates, awkward facts.
Take the Bindusara-disliked-Ashoka detail. The edicts don't mention it; Ashoka never names his father. The archaeology can't tell you about family relations. But the legend traditions all agree on it. Hagiography normally wants beloved-sons converting (a beloved son's transformation is dramatic enough); a hated son's conversion is less so. The unanimous awkward detail across three independent traditions reads as preserved memory rather than later invention. Lahiri uses this kind of triangulation systematically.1
Take the four-year interregnum between Bindusara's death (~273 BCE) and Ashoka's formal consecration (~269/268 BCE). The Dipavamsa records the gap explicitly. The edicts confirm the consecration year via Hellenistic synchronisms. The archaeology doesn't speak to it. Two of three streams converge; the gap is real. What happened in those four years is then the question, and the answer comes mostly from the Ashokavadana's court-faction kingmaker narrative — corroborated by the Mahavamsa's "caused his eldest brother to be slain" but not by it. The reading is interpretive, but built from independent confirmations.
Take Megasthenes. The Indica is lost; what survives are fragments quoted in later writers. Lahiri accepts the Pataliputra descriptions (the wooden palisade, the king's security arrangements) because archaeology has independently turned up sal-wood ramparts in Patna excavations matching Megasthenes's description. She discounts the "race of men whose legs are contorted like snakes" passage because no archaeology supports it and Megasthenes was "prone to speculative grandeur" when describing parts of India he hadn't seen. Same author, different epistemic weight on different passages, based on cross-stream confirmation.1
Watch how the method behaves under pressure. Two cases.
The Olivelle case. Patrick Olivelle is a major scholar of ancient Indian texts. His 2012 translation and commentary on the Arthashastra argued that the text has two layers — a "Kautilya recension" and a later "Shastric redaction" — and that even the Kautilya recension is post-Mauryan. If Olivelle is right, the Arthashastra was composed centuries after Chandragupta and Bindusara, and the prescriptive scribal preamble Ashoka allegedly violated did not exist in Ashoka's lifetime. Lahiri's reading of Ashoka's deliberate genealogical silence depends on the Arthashastra being Mauryan-vintage. She doesn't dodge the disagreement. She writes: "the cultural data on the basis of which he has suggested this is flawed and some of that data actually underlines that the original recension was likely to be of Mauryan vintage."1 One paragraph. Names the senior scholar. States the disagreement. Specifies the grounds (cultural data flawed). Moves on. This is what scholarly disagreement looks like when the historian is willing to make it.
The ASI case. The Archaeological Survey of India's official annual publication, Indian Archaeology — A Review, dated the inscribed stone umbrella at Panguraria to the second century BCE — making it post-Mauryan. Lahiri examines the plates published in the same volume and notes: the Brahmi characters could well be Mauryan, and the shaft and umbrella show the mirror-polish characteristic of Ashokan-era stoneworking. She overrules the ASI's own dating using the ASI's own plates. The override appears in a footnote with no fanfare — just a citation and the contrary reading.1
The Coral Case. The Olivelle disagreement has a specific evidentiary anchor that shows the method's texture at the level of a single object-class. Part of Olivelle's post-Mauryan dating argument rests on the Arthashastra's mentions of coral, which he reads as Mediterranean coral — meaning the passage could not have been written before Mediterranean trade routes were flourishing, no earlier than the first century BCE. Lahiri does not answer this with a counter-text. She answers it with the same kind of evidence her PhD thesis was built on: where coral actually turns up in the ground. Coral appears in India from the Indus civilization onward (Harappa), through the second-millennium-BCE neolithic-chalcolithic sites (Navdatoli, Prakash, Maski), into early historic sites well before 200 BCE (Taxila, Rajghat, Ganwaria-Piprahwa, Vaishali, Nevasa, Paunar, T. Kallupati). Indian coral did not need to wait for Mediterranean trade; it had been moving through the subcontinent for two thousand years before Olivelle's proposed date. Lahiri's own footnote is candid about the limit of this rebuttal: "there is no clarity on whether these were from the Mediterranean or Indian seas" — she is not proving her case, she is showing that Olivelle's is not proven either. The coral case is Lahiri's artefact-route-tracking method deployed defensively, against a textual-dating argument, using exactly the kind of distributional archaeological evidence her Ch 11 PhD disclosure names as her home method.2
This is what MacCulloch calls "may have" discipline in operation. It's not gentle hedging. It's the willingness to publicly contradict consensus when the evidence requires it, while clearly marking the evidence-strength of one's own counter-claim. Lahiri does it twice in a single book against senior players in the field.
In a Chapter 11 footnote, Lahiri makes a methodological disclosure that's unusual in scholarly biography. "Understanding the pattern of routes on the basis of artefacts of non-local origin formed the core methodology of my PhD thesis. Specifically for lapis lazuli in early historic India, Lahiri (1992): 371."1
Her PhD work tracked routes by which non-local materials moved across ancient India — lapis lazuli from Afghan mines to Bengal funerary deposits, sea-shell from the eastern coast to the middle Gangetic plain, carnelian and agate beads moving along the Uttarapatha. Where you find a non-local artefact, you can reconstruct the route by which it reached the find-site. Aggregate enough finds and the routes become readable. Routes, reconstructed this way, reveal the geography of ancient commerce more reliably than any single textual source.
This is the foundation of her landscape-anchored reading of Ashoka's edicts. The Ashokan pillar at Lauriya-Nandangarh sits at the intersection of two principal communication routes (N.G. Majumdar identified them). The Topra pillar sits on the line between Sirsawa and Ambala. The Maski minor rock edict sits in the heart of the Dharwar-band gold deposits, with a provincial capital named Suvarnagiri — "golden mountain" — probably for the gold trade it controlled. The landscape that hosts an edict is not random. It's the same landscape Lahiri's PhD method showed produces ancient routes. Her reading of the edicts is a continuation, into a new domain, of her own pre-existing method.1
That disclosure matters because it gives the method a name. Lahiri isn't just doing biography. She's applying artefact-route-tracking historical archaeology to the corpus of Ashokan inscriptions. The biography is a worked case of a method.
The protocol generalizes past Ashoka. Any ancient figure with at least two of the three evidentiary streams — primary contemporary source, archaeology of landscapes the figure operated in, later legend traditions — can be approached this way.
The protocol:
Map the streams you have. Inventory: do you have anything contemporary in the figure's own voice? Do you have archaeology of the places they operated in? Do you have later traditions that narrate them? Mark each stream's source-tier (primary vs scholarly vs hagiographic).
Read each stream alone first, then suspend belief. Read the contemporary source for what it says. Read the archaeology for what it shows. Read the legends for what they narrate. Form an initial picture from each. Then put each picture aside.
Identify what one stream tells you that the others contradict. This is the most informative case. If the contemporary source says one thing and the archaeology says another, somebody is wrong — and which one is wrong is itself a historical fact about the period.
Identify what two streams agree on independently. Two-stream convergence is stronger than three-stream consensus (which can be downstream-influence). When the contemporary edicts and the archaeology both say the empire reached Karnataka, you can be confident. When all three streams say the king killed his brother, check whether the legend traditions might have a shared source.
Identify the awkward unanimous details. Detail that all streams agree on, that none of them needed to invent, is preserved historical memory. The Bindusara-dislike detail is the textbook case for Ashoka. Use these to anchor the biographical reconstruction.
Read the silences. What the contemporary source refuses to say, against the prescriptive baseline of what its genre normally said, is data. Ashoka's omitted family is silence-as-evidence.
When you disagree with consensus, name the senior scholar and specify the grounds. Don't hedge. State the disagreement. The discipline lives in the willingness to do this in print.
The method works when you have rich evidentiary triangulation. It fails when you have only one stream — the figure remains in whatever distortion that stream produces. Cleopatra, Hannibal, the Buddha, Confucius — each is reconstructible only to the limit of their available triangulation.
The three-stream method is laid out programmatically in the Prelude (lines 264-326) and reiterated as a frame throughout the book.1 The Olivelle and ASI disagreements are at lines 670-674 and Ch 10 footnote 47 respectively. The PhD-thesis disclosure is Ch 11 footnote 15.1
The scholarly apparatus Lahiri triangulates: Hultzsch (1925) for the edicts; Falk (2006) Asokan Sites and Artefacts for the foundational bibliography; Strong (1989) The Legend of King Ashoka for the Ashokavadana arc (especially Ch 12 of Lahiri); Olivelle (2009) Asoka in History and Historical Memory for the afterlife reception material; Geiger (1912) for the Mahavamsa translation; Beal (1880/1884) for Faxian and Xuanzang.
The three-stream method presumes the streams correct each other's distortions. But shared distortions are possible. If the legend traditions invented the fratricide narrative, and the edicts independently happen to omit Ashoka's family, the agreement is not corroboration — it could be two outputs of the same selection bias (later Buddhist preference for the Chandashoka → Dhammashoka arc may have shaped both the legend's content and what the inscriptions were thought to mean). Lahiri's reading mostly avoids this trap by anchoring claims in archaeological independent confirmation, but the trap is real.
The landscape-anchored reading is generative but cannot be falsified. We cannot know what the audience at Maski actually heard. We only know what Lahiri infers from the landscape's properties. The method produces readings that are richer than the edicts alone allow, but the readings are not testable in the way archaeology of dated stratigraphy is.
The disagreements with Olivelle and ASI cut both ways. Lahiri might be wrong. If Olivelle is right about the Arthashastra's post-Mauryan dating, the deliberate-genealogical-silence reading collapses. Lahiri argues against Olivelle but cannot prove against him.
Lahiri and Diarmaid MacCulloch — the historian whose Perell interview anchors the vault's historiography-hub — agree on the discipline's core: skepticism first, then sympathy; "may have" as verbatim linguistic protocol; willingness to publicly disagree with senior scholarship; cross-discipline nerving (drawing on archaeology, anthropology, art history, philology in addition to textual sources). They diverge on the function of the discipline. MacCulloch frames it as sanity-protection against the lies of the powerful; the historian's job is to keep the reader sane against propaganda. Lahiri's framing is more recovery-focused: the historian's job is to bring a figure or period back into the realm of the knowable when the surviving evidence is sparse. Both framings can be true; they emphasize different stakes.
Where they share ground matters more. The two historians are operating the same discipline on different periods with different evidentiary bases. MacCulloch works the Reformation; Lahiri works the third century BCE. The discipline transfers. The "may have" protocol, the willingness to overrule official publications when the plates contradict them, the cross-discipline humility — these are general tools. The tension between sanity-protection and figure-recovery as the discipline's function is real but workable: each framing emphasizes part of what the discipline does. The method is the same. The frame is contextual.
Reading them together reveals something neither states alone: historiographic discipline is a transferable skill set, and a historian trained on one period (Lahiri on early historic India) can deploy the same discipline on a different period (her PhD on lapis-lazuli routes was Bronze-Age-to-Iron-Age; her Ashoka book is Mauryan; her method continues). The discipline is not period-specific. The training is. Once you have the training, the periods you can work on are bounded only by the evidentiary streams available.
A historian working with sparse evidence has to read multiple imperfect streams in parallel, accept that no single stream is reliable, and use cross-corrections to triangulate the past. The same logic shows up elsewhere — in clinical diagnosis, in deception detection, in any field where the truth must be reconstructed from systematically biased witnesses.
Creative Practice: Historiography Hub (MacCulloch) — Pick up MacCulloch's "may have" discipline and watch Lahiri operate it. MacCulloch teaches the historian to mark uncertainty in the prose itself — to write "Cromwell may have adored Wolsey" rather than "Cromwell adored Wolsey." The discipline becomes visible in the words. Lahiri operates the discipline at a different layer: she marks her disagreements with named senior scholars in the footnotes, with the specific grounds cited, and she does this in print at scale. The Olivelle and ASI cases are the verbatim "may have" structure applied to live scholarly disagreements rather than to historical claims. Historiographic discipline isn't only about reading sources humbly; it's about disagreeing with consensus humbly-but-clearly in print. The "may have" protocol scales from the historical claim to the scholarly disagreement. Both pages reveal a layer of historiographic practice the other doesn't fully address.
Behavioral Mechanics: Elicitation Framework (Hughes) — Hughes teaches that what a subject refuses to say is often more informative than what they say. The questions the subject deflects map the territory the subject is protecting. Lahiri reads sources the same way. Ashoka's edicts deflect from family, from his pre-conversion years, from the Kalinga slaughter in the early Minor Rock Edicts. The deflections, against the Arthashastra's baseline of what edicts normally said, map what Ashoka was protecting. The mechanism is identical across the two domains: identify the prescriptive baseline (what the subject / source typically would say), notice the deviation (what they refused to say), and read the deviation as evidence. Source-reading and subject-reading are the same mechanism applied to different domains. The historian's "reading of silences" and the elicitation specialist's "reading of deflections" are the same epistemic move.
Psychology: Projection as Information Channel — Buddhist hagiographers projected onto Ashoka the conversion-arc they wanted converts to have. Lahiri reads this projection critically: the legend material isn't invalid, but it's the Buddhist tradition's projection on top of the historical Ashoka. The Chandashoka → Dhammashoka arc tells you about Buddhist conversion-narratives. The residual circumstantial details (geographies, dates, awkward facts the hagiographers had no interest in inventing) tell you about the historical Ashoka. A structural principle: in any source that distorts its subject for ideological purposes, the historical residue is what the distortion did not need to invent. Discounting the projection requires identifying what the projector's interests would have required them to fabricate — and what they would have had no interest in fabricating. The fratricide is inflated by hagiographic interest; the four-year interregnum is preserved against hagiographic interest. Lahiri uses this distinction systematically. The same logic applies to reading patient narratives in therapy, reading interview testimony, reading any source where the speaker has selection-pressure on the content.
The Sharpest Implication
Historical biography of figures from before the medieval period is structurally impossible by modern criteria — no contemporary chronicles, no diaries, no photographs. What Lahiri demonstrates is that the impossibility can be partially overcome by reading multiple imperfect evidentiary streams against each other. The implication is methodologically generative: it suggests the impossibility-of-ancient-biography is overstated whenever sufficient secondary evidence (archaeology + legend + foreign accounts) survives alongside the primary source. The deeper implication is uncomfortable for fields that have given up on certain ancient figures — Cleopatra, Hannibal, Pericles, the Buddha himself — because their evidentiary triangulation is sparser than Ashoka's. Lahiri's method suggests the limit isn't at "no contemporary primary source" but at "no triangulation possible at all." Where triangulation is possible, the historical figure is knowable. Where it is not, the figure remains in legend. The methodological frontier of ancient biography is not "find new sources" — it's "develop better cross-stream reading protocols for the sources we have."
Generative Questions
Lahiri's method works with three streams — epigraphs, archaeology, critical legend. The Faxian ingest adds a fourth stream that operates inside the third one: 5th-century-CE pilgrim chronicles preserving on-site Buddhist oral traditions that recorded Asokan content in periods when Brahmi was illegible.
The case in point: Faxian stands at a major Asokan pillar three li south of Pataliputra in 405 CE. The pillar is fourteen-fifteen cubits in circumference, thirty cubits high. Brahmi has gone unreadable to him (and to most of his Indian-Buddhist hosts by this point). But a local monk tells him what the inscription says, and he writes it down: "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."
This inscription does not appear in the surviving Asokan epigraphic corpus. It is in a transactional-property genre distinct from the confessional-public Rock Edicts that survive most visibly. Faxian's paraphrase is the only surviving witness to the substance of an Asokan inscription that has otherwise been lost — see Faxian's Asokan Inscription Paraphrase.
What this adds to Lahiri's method:
Pilgrim chronicles preserve Asokan content the surviving epigraphic record does not contain. The on-site Buddhist oral tradition that transmitted the inscription's substance to Faxian no longer exists. The pilgrim chronicle is what survives of it. Without Faxian, this particular Asokan inscription would be entirely lost to modernity.
The transactional-property genre complicates the confessional-genre framing. Lahiri's reading positions Asoka primarily as the inventor of the confessional public genre. The Faxian witness shows Asoka also operating in transactional-property genre. Asoka's epigraphic toolkit was multi-register, and the modern reception has flattened the register-multiplicity by over-emphasizing the surviving confessional material.
The reception-stream Lahiri reconstructs has a missing layer. Lahiri triangulates epigraphs + archaeology + critical legend. The fourth layer — pilgrim chronicles preserving substance-of-lost-inscriptions via on-site oral tradition — operates between her second and third streams. Faxian's chronicle is one of the documented instances; Xuanzang's Si-Yu-Ki (645 CE) is another; Yijing (7th-8th c.) is a third. Adding this layer to the methodology would strengthen it.
The full collision stub LAB/Collisions/2026-05-13-faxian-asokan-inscription-vs-lahiri-method.md develops this further. The relationship is not opposition but layer-completion: Lahiri's three-stream method is correct as far as it goes, and the Faxian witness adds a fourth stream that operates inside the gaps the surviving epigraphic record leaves.