"What is the name of Ashoka?" is a trick question — the answer is baked into how you asked it. Ask it the honest way instead: what is the name of the Indian king who wrote the inscriptions and erected the inscribed pillars in the third century BCE? Asked that way, Olivelle says, "the answer is not obvious or clear."1
Here's the strange part. The name everyone uses — Ashoka — appears exactly four times in his entire surviving corpus. Four. Out of an empire's worth of inscriptions, spanning roughly twenty years, across thousands of kilometers, in multiple scripts and languages.
The name the book is titled after is nearly absent from the evidence the book is built on.
What's actually on the rocks, over and over, is a three-word title: "The Beloved of Gods, King Piyadasi" (devanapiye piyadasi laja), or some abbreviation of it. "Beloved of Gods" — used self-referentially — appears here for the first time in all of Indian literature.
It's an honorific, not a personal name; Ashoka himself uses it for former kings too, in Rock Edict VIII, treating it as a generic designation any legitimate king could claim. His grandson Dasharatha later used it as his own title in three separate inscriptions. It's a crown, not a signature.1
Curiously, that same honorific curdled over time. In later Sanskrit usage, devanapriya became a term of mockery — a word you'd use to call someone a fool.
One theory (Harmut Scharfe's) traces it to a Greek parallel, "friend of the king," a title Hellenistic monarchs granted to subordinate rulers who owed them allegiance. But the historian Grant Parker points out that phrase doesn't actually appear in Greek inscriptions, and it's implausible the Mauryas considered themselves subordinate to anyone in the Hellenistic west. Madhav Deshpande instead traces Vedic precedents for the expression.
Whatever its origin, "Beloved of Gods" didn't survive the Mauryas with its dignity intact — within a few generations it curdled from an emperor's self-designation into an insult, a trajectory almost nobody could have predicted while it was still being carved into rock as a mark of majesty.1
The second word, piyadasi (Sanskrit priyadarshin), is where Olivelle's argument does its real work. He treats this as Ashoka's actual personal name — the name he preferred, the name he wanted people to use.
The meaning isn't perfectly settled, but it's close: piya (Sanskrit priya) means dear, pleasant, beloved. dasi (Sanskrit darshin) means seeing, or — in a causative reading — "made to see." Sanskrit grammar offers two ways to combine them.
A tatpurusha reading treats both halves as substantives in a governing relationship: "seeing of what is dear," or "seeing the dear." A karmadharaya reading — the one Olivelle prefers — has the first word adjectivally qualify the second: a dear sight, something pleasant or lovely to see, often rendered "of benign countenance."1
The grammatical choice isn't pedantry. It decides whether the name describes the king as a person who has a pleasant gaze, or as a person who is pleasant to look upon — a subtle but real difference in what kind of self-presentation the name is doing.
Here's the evidence that actually settles whether "Piyadasi" is a title or a personal name, and it's a lovely piece of philological reasoning. Ashoka's inscriptions exist in translation at Kandahar — into Greek and Aramaic, for a population that read neither Prakrit nor Brahmi.
In those translations, every other term gets rendered into the target language: devanapiya becomes the Greek or Aramaic equivalent, laja (king) becomes the Greek or Aramaic equivalent. Every term, that is, except one. Piyadasi is not translated. It's phonetically transliterated — reproduced in Greek or Aramaic letters as a sound, not converted into a foreign-language equivalent of its meaning.1
Think about what that means. A translator faced with a title ("His Majesty," roughly) would translate it — titles carry meaning across languages, that's what titles are for. A translator faced with a personal name wouldn't translate it — you don't translate "Alexander" into "defender of men" when rendering a Greek document into Aramaic, you just write the sound of his name in the new script.
The Kandahar scribes treated Piyadasi exactly the way you'd treat a proper name: they kept the sound and dropped the meaning. That's the strongest available evidence that Piyadasi functioned, for Ashoka's own contemporaries, as his actual name — not an epithet, not a title, a name.
Further confirmation comes from Harry Falk's analysis of an inscription at Panguraria, long misunderstood, which records that "King Piyadasi," while still a prince (kumara), visited the site on a pleasure tour (viharayatra) — probably with his consort.
It reads almost like nostalgic graffiti, a young man's memory of an outing he'd later, as emperor, publicly condemn — the same kind of pleasure-hunt he would replace with dharma-pilgrimages once he took the throne. The name appears here attached to a private, pre-reformation memory, exactly the register a personal name occupies and a title doesn't.1
Olivelle offers a literary explanation for Ashoka's evident fondness for the compound title devanapiye piyadasi, and it's the kind of close reading that only a philologist notices: the middle two words alliterate. Devanapiye — piyadasi. The repeated p-sound links them. And the compound as a whole opens and closes on the same consonant too — the first and last syllables both begin with d.
Beyond the alliteration, Indian poetics classifies syllables as "heavy" or "light," and Olivelle notes that devanapiye runs four heavy syllables while piyadasi runs three or four light ones — a deliberate rhythmic contrast, heavy giving way to light, weight giving way to lift.
"Ashoka must have liked the poetic impact of his title," Olivelle concludes, and flags that this is the first of several places in the book where he'll point out Ashoka's rhetorical craftsmanship as a writer, not just a ruler.1 A king choosing his own self-designation for its sound, centuries before Sanskrit poetics formally theorized alliteration as a virtue, is doing something closer to composition than proclamation.
So where does the name everyone actually uses come from? All four occurrences cluster in the Minor Rock Edicts of the south (modern Karnataka) — twice in Minor Rock Edict I, twice in Minor Rock Edict II. Every other version of these same edicts, everywhere else, carries only the title "Beloved of Gods," with no personal name attached at all.1
The four outliers aren't just adding a name — they change the grammar around it in ways that mark them as later insertions. Two versions (at Maski and Gujarra) put the king's name in the genitive case — devanapiyasa asokasa, "of the Beloved of Gods, of Ashoka" — a construction that reads as an addition bolted onto an existing phrase, rather than the phrase's original design. Everywhere else, the standard formula is nominative: devanampiye heva aha, "The Beloved of Gods says thus."
Two other versions (Nittur and Udegolam) go further, writing laja ("king") before Ashoka's name — backwards from the normal word order, where the title follows the name, as in piyadasi laja. All of this creates the strong impression that "Ashoka" was appended after the fact, by scribes or local patrons somewhere down the transmission chain, wanting to attach a recognizable name to inscriptions the king himself had issued without one.1
This directly extends the vault's existing genealogical-silence and title-architecture material — see Ashoka's Deliberate Genealogical Silence and Devanampiya Piyadasi Title Architecture — but from a different angle. Those pages establish that Ashoka omitted his lineage and used the title-pattern strategically. This page adds a further layer: even his personal name, as far as the primary corpus goes, is mostly not there either. The self-erasure runs deeper than genealogy alone.
The sparseness of "Ashoka" in the primary record has fed at least one genuinely fringe position. The historian Christopher Beckwith argues Ashoka was an obscure, minor king inflated to legendary stature entirely by later Buddhist hagiographers — essentially, that "Ashoka" as history knows him barely existed.
Olivelle dismisses this as an outlier no mainstream Ashokan scholar accepts, and the counter-evidence is decisive: an inscription by the Western Kshatrapa ruler Rudradaman, carved around 150 CE — roughly three centuries after Ashoka's death — sits directly below Ashoka's own famous rock inscription at Girnar, Gujarat.
Rudradaman was plainly aware of the older inscription beneath which he placed his own, and he explicitly credits both Chandragupta and Ashoka with constructing and repairing the same reservoir. A king writing just three centuries later not only knew Ashoka's name but knew specific, verifiable facts about his construction projects. The theory that Ashoka is a later invention doesn't survive contact with a near-contemporary who clearly thought he was real, named, and responsible for identifiable public works.1
The tidiest resolution, given everything above: Ashoka had two names, however he acquired them. This isn't unique to him. His father Bindusara appears under that name in Indian sources but is called Amitrochates (Sanskrit Amitraghata, "killer of enemies") in Greek texts — likely a name used specifically for diplomatic purposes with the Hellenistic west.
Olivelle notes a small but suggestive pattern: both father's and son's alternate names begin with the same negative/privative Sanskrit particle, a- — Amitraghata and Ashoka ("without sorrow," the literal meaning of the name that won out). Chandragupta, by contrast, seems to have had only the one name across every tradition, Greek included.1
Which leaves the real puzzle unresolved: how and why did "Ashoka" — a name he barely used himself — eventually eclipse "Piyadasi" so completely in later memory that a book about him in 2023 has no choice but to call him Ashoka in the title, because "nobody would have known whom I was talking about" otherwise?1 Olivelle's best guess is that Buddhist tradition simply preferred the name, and that preference propagated outward into general historical memory — but he offers this as a guess, not a settled answer.
The victory of the minor, four-times-attested name over the king's own preferred self-designation is total, irreversible, and not fully explained by anything in the evidence.
The methodological payoff of this page is a transferable skill: how to treat a name itself as evidence, not just a label. Here's the protocol Olivelle's argument implicitly runs.
Count occurrences before assuming dominance. "Ashoka" feels like the obvious name because it's the name everyone uses today. The primary corpus says otherwise — four occurrences against a title used hundreds of times. Frequency in the primary record and frequency in later memory are two different measurements, and conflating them is the trap.
Use translation behavior as a name/title diagnostic. When a term travels across languages, watch whether translators translate it or transliterate it. Meaning-carrying titles get translated; personal names get carried over as sound. The Kandahar bilingual inscriptions turn this into a testable signal rather than a guess.
Read grammatical irregularity as a chronological marker. The genitive-case and reversed-word-order versions of "Ashoka" aren't just variants — they're tells. A phrase that breaks the standard formula's grammar is a strong candidate for a later insertion into a formula that was originally built without it.
Take a fringe rebuttal seriously enough to name the specific counter-evidence. Dismissing Beckwith's theory outright would be as undisciplined as accepting it uncritically. The Rudradaman inscription is a specific, checkable fact — a named third-century-CE king citing named Mauryan predecessors — and it's the fact that actually does the rebutting, not just scholarly consensus asserting itself.
The strongest evidence in this page is the translation-refusal argument — Piyadasi's phonetic preservation across Greek and Aramaic renderings while every surrounding term gets translated is close to a clean philological proof, not merely suggestive.
The Rudradaman counter-inscription similarly closes off the Beckwith fringe position about as decisively as ancient epigraphy allows — an independent, near-contemporary source naming Ashoka and citing verifiable facts about him is about as strong as ancient corroboration gets.
The grammatical analysis of the four "Ashoka" insertions (genitive case at Maski and Gujarra, reversed word order at Nittur and Udegolam) is internally consistent evidence from within the corpus itself, requiring no external corroboration to make its case.
The unresolved tension sits with the "Ashoka" name's dominance: Olivelle can explain, in granular grammatical detail, why the four occurrences of "Ashoka" look like later insertions, but he cannot explain why that minority name became the name history settled on.
The book states the puzzle and declines to resolve it — a genuine open question, not a rhetorical one. Olivelle's own guess (Buddhist tradition preferred it, and the preference propagated) is offered explicitly as a guess, which is itself methodologically consistent with the portrait-not-biography discipline documented elsewhere in this vault.
This material sits almost entirely outside Lahiri's coverage — her existing vault page on Devanampiya Piyadasi Title Architecture documents the title's usage patterns (who used it, where, humility-as-self-presentation) without engaging the grammatical tatpurusha/karmadharaya debate, the translation-refusal proof, or the four-insertion textual-critical argument for "Ashoka" itself.
The two pages are complementary rather than competing — Lahiri's asks what the title accomplished politically; Olivelle's asks what the two names actually were, etymologically and evidentially. Cross-link both directions rather than merge, since neither page's core argument depends on or contradicts the other's.
Creative Practice: Non-Fiction Writing Craft Hub — A king choosing his own public name for its alliteration and syllable-weight rhythm is doing the same craft work a poet does when selecting a pen name for its sound. Ashoka's evident fondness for devanapiye piyadasi's heavy-to-light rhythm anticipates, by Olivelle's own account, the kavya aesthetic tradition's later formal preoccupation with alliteration and sound-patterning by centuries.
Walk the mechanism on both sides. A poet choosing a pen name weighs the same variables Ashoka apparently weighed: how the syllables land in the mouth, where the stresses fall, whether the sounds repeat in a pleasing rather than clumsy way. A modern writer might reject a technically accurate byline in favor of one that simply sounds better read aloud — meaning takes a back seat to sound. Ashoka's title does exactly that trade: devanapiye piyadasi isn't the most information-dense way to say "beloved of the gods, of pleasant appearance" — it's the version that alliterates and shifts from heavy to light syllables, a rhythmic choice layered on top of the semantic one.
The insight neither domain produces alone: rhetorical self-branding — choosing how your own name sounds, not just what it means — is not a modern invention of pen names and stage names. It's at least 2,300 years old, and its earliest attested large-scale practitioner was doing it in service of political self-presentation carved into stone across an empire, not literary publication in a book. The craft tradition of sound-conscious naming has an imperial, epigraphic origin point most writers assume is purely literary.
Behavioral Mechanics: Manipulation and Influence Hub — The generic, transferable honorific "Beloved of Gods" — usable by any legitimate king, later adopted by Ashoka's own grandson — functions as a piece of reusable political branding infrastructure, distinct from the personal name underneath it.
Walk the mechanism concretely. Ashoka uses "Beloved of Gods" for former kings in Rock Edict VIII, treating it as a generic designation any legitimate ruler could claim — not something unique to him. His grandson Dasharatha then reuses the exact title in three separate inscriptions of his own, with no personal claim on Ashoka's specific achievements, simply inheriting the institutional prestige the title carries. That's structurally identical to how a modern corporate or political title works: "the Presidency" or "the Crown" carries authority and legitimacy that transfers to whoever currently occupies the role, independent of that person's individual record.
The insight neither domain produces alone: title-as-transferable-brand and name-as-private-identity is not a modern bureaucratic invention. Ashoka's own grandson borrowing his title while having no personal claim on "Piyadasi" — the actual personal name — shows the separation between institutional brand and individual identity was already operative and useful in the third century BCE, deliberately engineered into the very grammar of how the king referred to himself and his predecessors.
Sharpest Implication
If the name history actually settled on is the one Ashoka used least and seemingly didn't originate deliberately — a name added by later scribes and patrons rather than the king's own chosen self-designation — then the "Ashoka" the entire modern world knows, reveres, and puts on flags and currency is, at the level of the name itself, already a posthumous reconstruction layered onto the historical figure, not his own voice. Even readers deeply committed to inscriptional evidence over legend are, every time they say "Ashoka," using the name legend and later scribal addition won, not the name the king actually wanted.
Generative Questions