Chapter 1

Chapter 1 · Where the books came from

Draft 1 — 25 August 2026. Opening chapter of Part 1. Sanskrit quotations verified against holy-bhagavad-gita.org; replace with a public-domain scholarly edition (Telang, SBE vol. 8, 1882) before print. Tamil saints to be quoted from Kingsbury & Phillips, Hymns of the Tamil Śaivite Saints (1921, public domain) — not yet pulled. Written to the spine rules: history and plain words only, nothing tagged as anyone's religion yet, positive framing throughout, and short. Dates are approximate and must be checked against research/hinduism_development/CHRONOLOGY_STANDARD.md before print.

India has more sacred books than any other country on earth. Nobody planned it that way.

They were not written in one place, or in one century, or by people who had met each other. Some are older than writing itself in India. Some are younger than the Roman Empire. They were gathered into one library long afterwards, by people who thought the gathering was worth doing.

If you have ever opened two of them and found that they disagree, this chapter is the reason.

The oldest layer: songs before books

The oldest material is the Veda — not a book at first, but songs. Families of poets composed them and handed them down by voice, parent to child, for centuries before anyone wrote them down. Each family kept its own collection. There was no single edition because there was no single place to keep one.

Around three thousand years ago, when one kingdom in the north grew large enough to gather the families together, those scattered collections were arranged into an order.1 That is when a Veda in the sense we mean became a thing you could point at.

In those songs the god at the top is Indra — a storm-god, a fighter, the one you call on before a battle. Remember him. Who stands at the top will change twice, and each time the whole religion gets a different name.2

What came next: the forest, and the handbooks

What follows the songs is usually taught as a tidy ladder — handbooks, then forest books, then philosophy. It was not tidy, and the order most textbooks give is not the one this book follows.3

What seems to have come first is a turn to the forest. Some among the priestly community moved out of the settlements to practise, and out of that came the Āraṇyakas, the "forest books" — less concerned with performing a ceremony correctly than with what the ceremony was supposed to mean.4

Around and after that turn, a second kind of text was assembled: the Brāhmaṇas, handbooks of ceremony.5 They explain how each rite is performed, in what order, with what words, and what it achieves. If the Veda is the score, the Brāhmaṇas are the conductor's notes.

By this period the figure at the top is no longer the storm-god. It is Brahmā, a creator. The religion of this layer is what scholars call Brahmanism — named, as you would expect, for who is on top.

The third layer: the questions

Out of the forest books came the Upaniṣads,6 and they are a different kind of writing altogether. They ask questions. What is a person, underneath? What continues, and what does not? Is the ceremony the point, or is something behind the ceremony the point?

And here the oldest line stops. Veda, Brāhmaṇa, Āraṇyaka, Upaniṣad — this set has a name, śruti, "what was heard", and by long agreement nothing was ever added to it again.7 Everything that came later, however loved, belongs to a second category: smṛti, "what was remembered."

That distinction is not an insult to the later books. It is a date stamp. It tells you which side of a very old line a text stands on, and it is the single most useful thing to know when two books disagree.

How much fell outside the line

It is worth seeing how much. Once the line closes, Indian religious literature does not slow down — it multiplies, for another thousand years and more. A partial list, and every one of these is a body of literature rather than a single book:8

  • The six Vedāṅgas — the auxiliary disciplines a priest needed: phonetics, grammar, metre, etymology,

ritual procedure, astronomy.

  • The Kalpa-sūtras — manuals of practice, in three kinds: Śrauta for the great public rites, Gṛhya for

the household ones, and Dharma for conduct and law.

  • The Dharmaśāstras — the law codes that grew out of those, the Manusmṛti the best known of them.
  • The two epics — the Mahābhārata, which carries the Bhagavad-gītā inside it, and the Rāmāyaṇa.
  • The Purāṇas — eighteen major collections and eighteen minor ones, the storehouse of the gods, the

genealogies and the world-ages.

  • The six darśanas — six full philosophical systems, each with its own root-text and centuries of

commentary: Sāṃkhya, Yoga, Nyāya, Vaiśeṣika, Mīmāṃsā, Vedānta.

  • The Āgamas, Tantras and Pāñcarātra Saṃhitās — the revealed books of those who worshipped Śiva and those

who worshipped Viṣṇu.

  • The Tamil canons — the four thousand verses of the Āḻvārs, and the Śaiva Tirumuṟai.

Stand back and look at that. It is an enormous library, and almost none of it existed when the line closed.

Now the part that surprises people. In the reckoning of those who guarded the tradition, not one item on that list was scripture. Their canon was the Veda and the manuals explaining it, and it stopped there. The epics, the Purāṇas, the philosophy, the revealed books of Śiva and Viṣṇu — all of it stood outside.

So we should be careful about the word. It is easy to picture all of this as "Hindu scripture" sitting on one shelf, because that is how a bookshop arranges it now. For most of the time these works were being written, there was no such shelf. There were several traditions, each with its own books, living alongside one another.

What put them on one shelf happens later, and it is the subject of the next chapter.

What came after, and came from somewhere else

The later material is enormous: the great epics, the Purāṇas, the Āgamas, and the devotional literature. Much of it did not grow out of the old priestly world at all. It grew out of movements of ordinary devotion — people who loved Viṣṇu, or people who loved Śiva — and who produced their own scriptures, in their own languages, with their own saints.9

In the Tamil country these movements produced collections so treasured that people called them the Tamil Veda.10 That name is worth pausing on. It tells you two things at once: that the communities held these books as highly as the oldest ones, and that they knew perfectly well the books were not those.

These movements also did something the older system had not. Their saints came from every level of society, including the very bottom, and their central claim was that devotion counted for more than birth. That was a genuinely new proposal, and it spread because people found it true.

Were these part of "Hinduism" yet? Not as one thing. The formal scholarly system of the day, Mīmāṃsā, which existed precisely to settle what counted as authority, was built around the Veda.11 The devotional scriptures stood outside that frame — not condemned, simply not inside it. They were their own traditions, with their own books, alongside the older one rather than inside it.

What the new movements would not accept

It is worth being precise about what was new in them, because it was not only the god.

The old system ran on birth. What you were entitled to do — which ceremonies, which texts, which duties — followed from the family you were born into. The devotional movements did not accept that, and they did not merely say so. They put it in their scriptures, and they put it in their saints.

The Tamil Vaiṣṇava tradition counts twelve Āḻvārs among its founding poets. One of them, Tiruppāṇ Āḻvār, came from the Pāṇar — a community of drummers and singers held to be so impure that they were kept out of the temples. His verses are in the canon. He is one of the twelve.12

The Tamil Śaiva tradition counts sixty-three Nāyaṉmār. The eighteenth is Nandanār, a leather-worker.13 The tradition's own account is exact about his position: he supplied the great temple with the leather straps for its drums and the gut for its strings, so the music of the worship depended on his hands — and he was not allowed through the door.

Read that twice. The tradition did not bury this story. It canonised the man who had been shut out, and it kept the detail about the door. A tradition that wanted the memory gone would not have written it down.

Later, and further north, the same argument was made in the open. In twelfth-century Karnataka, Basava built a movement — the Vīraśaiva, whose descendants are the Liṅgāyats of today — on the plain claim that caste was not real, that ritual and priesthood were not necessary, and that a woman stood equal to a man.14 He wrote it in Kannada, in short poems called vacanas, so that anyone could hear them. It cost him the support of his king.

None of this is a modern reading imposed on old texts. It is what the traditions say about themselves.

The two verses

The clearest case of all sits inside a single book, and one of the best known in India.

The Bhagavad-gītā contains this:

cāturvarṇyaṃ mayā sṛṣṭaṃ guṇa-karma-vibhāgaśaḥ "The four orders were created by me, apportioned by quality and by work." — Gītā 4.1315

And, five chapters later, this:

māṃ hi pārtha vyapāśritya ye 'pi syuḥ pāpa-yonayaḥ / striyo vaiśyās tathā śūdrās te 'pi yānti parāṃ gatim "Those who take refuge in me, Pārtha — even those of low birth, women, vaiśyas, and śūdras — even they reach the highest goal." — Gītā 9.3216

One verse states that the four orders are divinely made. The other states that the people those orders place at the bottom arrive at the same destination as everyone else.

We are not going to tell you which verse wins. Both are in the book. Both have been read carefully by serious people for two thousand years, and those people have reached opposite conclusions.17

That is not a flaw we have discovered. It is the ordinary condition of any library assembled over a thousand years from many hands. And it is the whole reason this book exists, because if a tradition can hand you two answers, then at some point the question comes back to you.

One thing this book is not doing

Everything in this chapter comes from the texts themselves and from the traditions' own records of their own saints. Nothing here is anybody's opinion about anybody else.

It also settles something before it can be got wrong. The refusal to accept that birth decides a person's worth was not the invention of any one tradition. It is in the Tamil canons. It is in Basava's poems. It is in a verse of the Gītā. It had been argued in India, from inside several traditions, for a very long time.

The gathering — and a book nobody wanted

By the sixth century these traditions stood side by side without a shared roof. What put them under one is the subject of the next chapter, and it turns on a single book that neither side would claim, and on a man who saw that it was good and gave it a home.

The word Hindu, incidentally, was never a religious word to begin with. It comes from Sindhu, the river — the Indus. It was what outsiders called the people who lived beyond it. A geographical label became the name of a faith, very late, and largely because other people needed a word for it.18

A king had already tried it

Here is the part that should make you sit up.

Long before any of the gathering described in the next chapter, someone had already done the same thing — not in a monastery, not in a commentary, but in law, across an empire, and carved into rock where anyone could check it.

He had every reason not to. He had won his wars. He commanded a territory larger than any Indian ruler before him, containing every one of the traditions in this chapter and several that have since vanished. He could have picked one and made everyone else follow it. That is what kings did.

He did something else instead, and he wrote down why.

He is the subject of this book, and we reach him in Chapter 3.

Notes

  1. 1
    The gathering of the regionally composed Family Books (Maṇḍalas 2–7) into a canonical order at the Kuru-Pañcāla centre: Michael Witzel, "Early Sanskritization: Origins and Development of the Kuru State," Electronic Journal of Vedic Studies 1, no. 4 (1995); Witzel, "The Development of the Vedic Canon and its Schools," in Inside the Texts, Beyond the Texts, ed. Witzel (Cambridge, MA: Harvard Oriental Series, 1997), 257–345. Compression note: "one kingdom grew large enough" stands in for a long institutional process — a dynastic consolidation producing a paramount overlord with an editorial centre. The short form is for readability and is not a claim that one king ordered one edition.
  2. 2
    The division of this long history into three religions — Vedism (the religion of the Saṃhitās), Brahmanism, and Hinduism — is standard in the field, not an invention of this book. What this book takes from the author's Mechanism of the Vedic Lineage (in preparation) is the key used to mark the divisions: the migration of the summit, i.e. which figure stands at the top. On that reading the periodisation is not arbitrary — it tracks a real and traceable change in the pantheon's structure. On the rise of the successive supreme figures see Jan Gonda, Prajāpati's Rise to Higher Rank (Leiden: Brill, 1986). Compression note: the stages overlap heavily, older deities persist alongside newer ones throughout, and "who is at the top" was never uniform across regions or schools in any period. The body gives the outline; the volume named above gives the stratigraphy.
  3. 3
    This book does not follow the conventional textbook sequence of Saṃhitā → Brāhmaṇa → Āraṇyaka → Upaniṣad. It follows the reconstruction in the author's Mechanism of the Vedic Lineage (in preparation), in which the Āraṇyaka watershed precedes the Brāhmaṇa-corpus consolidation — the forest turn being the earlier institutional development, with the Brāhmaṇa stratum subsequently elaborating what the Āraṇyaka emergence had begun; the Āraṇyaka corpus is used there as the boundary marker of the pre-Brāhmaṇa transitional period. Two cautions. First, "before" here concerns the institutional emergence of forest practice; the crystallisation of the Āraṇyakas as a recognised free-standing text-category belongs to the late-Brāhmaṇa period, so the two corpora overlap substantially and no clean succession is claimed. Second, this ordering is not the standard textbook presentation, and a reader who has met the conventional sequence elsewhere has not been misinformed — the disagreement is real and is argued at length in the volume named above.
  4. 4
    Aitareya, Taittirīya and Kauṣītaki Āraṇyakas. Compression note: "less concerned with performing a ceremony correctly than with what the ceremony was supposed to mean" summarises a real shift of emphasis; these remain ritual texts and are not philosophical treatises.
  5. 5
    Sixteen principal Brāhmaṇa texts across the Indo-Gangetic plain — Aitareya and Kauṣītaki (Ṛgveda); Pañcaviṃśa and Jaiminīya (Sāmaveda); Taittirīya, Maitrāyaṇī, Kāṭhaka, Kapiṣṭhala (Black Yajurveda); Śatapatha (White Yajurveda); Gopatha (Atharvaveda). School affiliations and relative dating in Witzel, "Development of the Vedic Canon," 257–345. On dates: this book gives no absolute figure for the corpus. Every published range is an estimate under active dispute, and the argument here needs only the order, which is not.
  6. 6
    The body gives no date for the Upaniṣads, deliberately. Where a date is needed elsewhere in this project, the standard adopted is the anchored cascadeBṛhadāraṇyaka 2.1 at approximately 475 BCE, running to the Maitrāyaṇīya at approximately 328 BCE — with Patrick Olivelle, The Early Upaniṣads: Annotated Text and Translation (New York: Oxford University Press, 1998), Introduction, cited as the secondary position for the conventional ranges. Readers should know what the anchored dating implies: on it, the early Upaniṣads are contemporary with or later than the Buddha rather than earlier, which bears on which way influence ran between the Upaniṣadic and the śramaṇa currents. That is a contested position, argued at length in the author's Mechanism of the Vedic Lineage (in preparation) and consonant with Johannes Bronkhorst, Greater Magadha: Studies in the Culture of Early India, Handbook of Oriental Studies 2.19 (Leiden: Brill, 2007), Part II. A school text has no business adjudicating that question in passing, so this one does not.
  7. 7
    On śruti/smṛti and canonical authority: Sheldon Pollock, "The Idea of Śāstra in Traditional India," in Ideology and Status of Sanskrit, ed. Jan E. M. Houben (Leiden: Brill, 1996). Compression note: "nothing was ever added again" states the normative rule. In practice the boundary was argued over, and particular schools have treated particular later texts as functionally revealed. The sentence describes the tradition's own rule, not universal practice.
  8. 8
    The list is a survey rather than a catalogue and each entry names a corpus, not a work. On the Vedāṅgas and the Kalpa-sūtra literature: Jan Gonda, ed., A History of Indian Literature (Wiesbaden: Harrassowitz, 1975–), the volumes on ritual sūtras and on the dharma literature. On the Dharmasūtras and Dharmaśāstra: Patrick Olivelle, trans., Dharmasūtras: The Law Codes of Āpastamba, Gautama, Baudhāyana, and Vasiṣṭha, Oxford World's Classics (New York: Oxford University Press, 2000); Olivelle, trans., Manu's Code of Law (New York: Oxford University Press, 2005). On the six darśanas and the criterion by which mutually irreconcilable systems are alike filed as orthodox — nominal acceptance of the Veda's authority rather than derivation of content from it — see the standard treatments of āstika/nāstika. On the Śaiva and Vaiṣṇava canons, Sanderson, "The Śaiva Age" (n. sectarian-canons). Compression note, and it matters here more than anywhere else in the chapter. These bodies of literature are not one after another in a queue: they overlap, they run for centuries each, and several are still growing while others are complete. The list is arranged to be legible, not to assert a sequence. What the list is evidence for is the one thing the body claims from it — the sheer quantity of religious literature produced in India outside the closed line of śruti.
  9. 9
    The Śaiva Āgamas and Tantras and the Vaiṣṇava Pāñcarātra Saṃhitās claim revelation directly from Śiva or from Viṣṇu rather than derivation from the Veda. On the scale of the Śaiva traditions in this period — the basis for naming the era the "Śaiva Age" — Alexis Sanderson, "The Śaiva Age," in Genesis and Development of Tantrism, ed. Shingo Einoo (Tokyo: Institute of Oriental Culture, University of Tokyo, 2009), 41–350.
  10. 10
    The Vaiṣṇava Nālāyira Divya Prabandham is traditionally styled Drāviḍa Veda, the Tamil Veda; the Śaiva Tirumuṟai holds a comparable standing. Methodological note: the body reads the epithet as evidence of two things at once — the standing claimed for these collections, and an accompanying awareness that they were not the Sanskrit Veda. That reading is the body's own, offered as the plainest sense of the title, and is not attributed to a cited authority.
  11. 11
    On Mīmāṃsā drawing its canon-line short of the later devotional and liberation literature — the school most jealous of Vedic authority admitting that layer latest and least: Johannes Bronkhorst, Greater Magadha, Part II; Bronkhorst, ed., Mīmāṃsā and Vedānta: Interaction and Continuity (Delhi: Motilal Banarsidass, 2007). Compression note: "not condemned, simply not inside it" fairly summarises the canon-line's effect. It is not a claim that Mīmāṃsā authors were indifferent to these traditions. The dating consequence of that refusal is taken up in Chapter 2.
  12. 12
    On Tiruppāṇ Āḻvār (eighth–ninth century) as very probably the earliest figure of an "untouchable" community counted a saint within the Hindu tradition: Eleanor Zelliot and Rohini Mokashi-Punekar, eds., Untouchable Saints: An Indian Phenomenon (New Delhi: Manohar, 2005). Compression note: the caste identification rests on hagiography rather than contemporary documentation, and the precise community (Pāṇar or Paṟaiyar) is discussed rather than settled. What is not in doubt is that the tradition itself transmits him as of excluded birth and counts him among the twelve — which is the only thing the body claims.
  13. 13
    Nandanār, also Tirunāḷaippōvār, eighteenth of the sixty-three Nāyaṉmār. The account of his occupation, of his supplying leather for the temple's instruments, and of his exclusion from the temple is given in Cēkkiḻār's Periya Purāṇam (twelfth century), the tradition's own hagiographic compendium. Public-domain English renderings of the Tamil Śaiva hymns: Francis Kingsbury and G. E. Phillips, Hymns of the Tamil Śaivite Saints, Heritage of India Series (Calcutta: Association Press, 1921). Compression note: the Periya Purāṇam is devotional literature composed centuries after its subjects. It is evidence for what the tradition preserved and honoured — which is exactly the use the body makes of it — and not a contemporary record of a life.
  14. 14
    Basava (traditionally born 1131 CE), Karnataka; the vacana corpus in Kannada. On the movement's rejection of caste, of Vedic ritual authority and of priestly mediation, and on the consequences for Basava at the court of Bijjala II, see the standard treatments of Vīraśaiva/Liṅgāyat origins. On translation: A. K. Ramanujan, Speaking of Śiva (Harmondsworth: Penguin, 1973), is the best-known English rendering and remains in copyright; a complete vacana quoted in this book must come from a public-domain rendering or a fresh translation, since a short poem quoted whole is a whole work. ⚠️ Outstanding: page-level citation for this note is not yet pinned.
  15. 15
    Bhagavadgītā 4.13: cāturvarṇyaṃ mayā sṛṣṭaṃ guṇa-karma-vibhāgaśaḥ. Canonical locus. Public-domain English: Kāshināth Trimbak Telang, trans., The Bhagavadgîtâ, Sacred Books of the East vol. 8 (Oxford: Clarendon Press, 1882).
  16. 16
    Bhagavadgītā 9.32: māṃ hi pārtha vyapāśritya ye 'pi syuḥ pāpa-yonayaḥ / striyo vaiśyās tathā śūdrās te 'pi yānti parāṃ gatim. Telang, SBE vol. 8. On the rendering: pāpa-yonayaḥ, given here as "of low birth", is literally "of sinful wombs", and that phrase is itself part of what is at issue — the verse opens the highest goal to these groups while describing their birth in those terms. The body quotes and does not resolve this; the tension is the point.
  17. 17
    The two verses have been read together in opposite ways for centuries. One line holds that 4.13 grounds the four orders in quality and action (guṇa-karma) rather than in birth, and so does not underwrite hereditary caste at all; another holds that the Gītā supplies a philosophical defence of the graded order rather than a departure from it. Compression note: the body reports the disagreement without adjudicating it. This is not evasion. The chapter's argument requires only that one text yields both readings to careful readers, and that is disputed by no one.
  18. 18
    On Hindu as originally a geographical designation from Old Persian Hindu (Sanskrit Sindhu), and on the later construction of "Hinduism" as a single category over cognate but distinct traditions: Heinrich von Stietencron, "Hinduism: On the Proper Use of a Deceptive Term," in Hinduism Reconsidered, ed. Günther-Dietz Sontheimer and Hermann Kulke (New Delhi: Manohar, 1989); Richard King, Orientalism and Religion: Post-Colonial Theory, India and "The Mystic East" (London: Routledge, 1999). Disagreement stated: David N. Lorenzen, "Who Invented Hinduism?," Comparative Studies in Society and History 41, no. 4 (1999): 630–59, argues against the strong constructionist position, holding that a Hindu self-identity existed well before the colonial period. The body claims only that the word began as geography and became religious late — which both sides of that debate accept.
Chapter 2

Chapter 2 · The gathering

Draft 1 — 25 August 2026. Third chapter written, second in order. Completes Part 1. Built on the author's own §9★ in research/mechanism_vedic_lineage/DRAFT.md, where the full apparatus and footnotes stand. Positive framing throughout; nothing tagged as anyone's religion.

We left the last chapter with a country full of scriptures that had never been introduced to one another.

The oldest line — Veda, Brāhmaṇa, Āraṇyaka, Upaniṣad — had closed. Around it stood the epics, the Purāṇas, and the revealed books of the devotional movements: the Āgamas and Tantras of those who loved Śiva, the Pāñcarātra Saṃhitās of those who loved Viṣṇu — scriptures held to come not from the Veda but directly from Śiva, or directly from Viṣṇu.

That last point is the one to hold. These were not branches of the old tree. They were separate trunks, with their own roots, and by the sixth and seventh centuries they carried so much of the country's devotion that historians call the period the Śaiva Age.

So: several traditions, several canons, no shared roof. How did they end up under one?

A book nobody wanted

Around 500 to 550 CE a work appeared under the name of Gauḍapāda — a commentary in verse on one of the shortest of the Upaniṣads, the Māṇḍūkya. Its four parts are quite possibly not the work of one hand.

What it did was unusual. It did not take one tradition and defend it. It took pieces from several and fitted them into a single picture.

And here is what happened to it: nobody would claim it.

The Buddhists of the day filed it as tīrthika — an outsider's work, not one of theirs. Later Vedāntins of other schools called it pracchanna-bauddha, "crypto-Buddhist" — too Buddhist to own. That charge was not made by Buddhists against Hindus. It was made inside the Hindu world, by Vedāntins about other Vedāntins; Bhāskara pressed it, and Rāmānuja fixed it in writing around 1100 CE.

A homeless book, disowned from both directions — and it rose anyway.

What was actually in it

The piece that holds the whole assembly together is a claim, and it is worth reading slowly:

"the same Ātman is called prāṇa by those who know prāṇa … called Indra and the other gods by those who know the gods; called the Vedas by those versed in the Vedas; called the goal of the yajñas by those expert in yajñas…" — Māṇḍūkya-kārikā II.20–23

Read what that sentence does. It goes down the list of everything people in India argued about — the life- breath, the gods, the Veda, the sacrifice, the philosophical principles — and says of each one: you are all naming the same thing. You are using different words.

That is the proposal. Not "my school is right." Not even "let us tolerate each other." Something much more ambitious: there is one truth underneath all of these, and each tradition has hold of it by a different handle.

The tradition says this about itself, and says it proudly

We do not have to characterise this from outside. Advaita Vedānta — the school that grew from this book — describes its own achievement in exactly these terms.

Here is V. Subrahmanya Iyer — a Hindu, a Vedāntin of that school — introducing a modern edition of the very text. This Vedānta, he writes, is

"a truth, 'free from all dispute' and 'not opposed to any school … equally good for all beings,'" one that "assigns to religious faith, theology, scholasticism, mysticism, art and science their respective places in the one grand edifice … even the views of his opponents he welcomes and accepts as parts."

Free from all dispute. Not opposed to any school. Equally good for all beings. Even opponents' views welcomed and accepted as parts.

That is a Hindu describing the crown of his own tradition, with pride, and it is a fair description.

Note in particular the third of those phrases — equally good for all beings. A Hindu wrote that, about a Hindu text. We will need it again, at the very end of this book.

Keep those four phrases in mind for about two pages.

The man who built the house

Around 700 CE, Śaṅkarācārya took up the homeless book and its method.

He did the opposite of what everyone else had done with it. Where the Buddhists had disowned it and rival Vedāntins had held their noses, he saw that it was good, and he gave it a home — taking Gauḍapāda into his own lineage as his teacher's teacher, and building the whole architecture of Advaita on the procedure the Kārikā had run on.

Then he used that architecture to do something no one before him had managed. He set out a framework in which the old śruti and the separate devotional traditions could stand as parts of one whole rather than as competitors. The Brahma-sūtra is explicit about the method: the divergent statements of the scriptures are to be read as pointing at one thing — tat tu samanvayāt, "but that, because of harmonious construing" (1.1.4).

He chose what would let people live together. Given a shelf of books that disagreed, he built a reading in which they could all be shelved together and each still count.

The man who said no

We can say almost exactly how late this was, because somebody objected — and we know when he was writing.

Kumārila Bhaṭṭa worked in the first half of the seventh century, a generation before Śaṅkara. He belonged to Mīmāṃsā, the school whose entire business was guarding the authority of the Veda, and he was the fiercest defender it ever produced.

By his lifetime the Purāṇas had been circulating for centuries. The Trimūrti was in them. The Śaiva and Vaiṣṇava canons stood beside them, carrying most of the country's devotion.

And Kumārila would not count any of it. His Veda was the old śruti and the ritual manuals that explain it, and nothing beyond. Not the Purāṇas. Not the avatāras. Not the liberation-metaphysics that the Vedānta lineage was building.1

Think about what that tells us. The strictest guardian of the tradition, in the century Śaṅkara was born, was still saying that the Trimūrti's scriptures were not the tradition's scriptures.

So the joining was not settled by the Purāṇas merely existing. Someone still had to make it hold.

Why this is where the age turns

This book marks the eras by one thing: who the canon puts at the top. Indra in the songs, Brahmā in the handbooks, the Trimūrti after that.

By that measure the Trimūrti was available from the Purāṇas onward — but Kumārila is dated proof that it was not yet accepted at the tradition's own centre. What changed that was the consolidation.

So this is where the third age begins, and where "Hinduism" — one thing with many rooms rather than several separate things — starts to be true.2

And the figure at the top changes for the last time. Not Indra of the old songs. Not Brahmā of the ritual handbooks. The Trimūrti: Brahmā, Viṣṇu and Śiva as three faces of one — the older god and the two great devotional traditions each given a place at the summit rather than a place in the queue.

What kind of act that was

Now put the two things side by side.

Before: separate traditions with separate revelations, some of them holding positions the old system did not accept — including, as we saw, positions about whether birth decides a person's worth.

After: one framework, wide enough to hold all of them, built on the claim that they are naming one truth by different names.

The move that made a single tradition possible was an act of accepting difference. Not the erasing of difference — the differences are all still there, and a Śaiva and a Vaiṣṇava and a Vedāntin still differ today — but the decision that difference need not mean division.

That is not our reading. It is Iyer's: not opposed to any school; even the views of his opponents welcomed and accepted as parts.

Where you have heard this before

A thousand years before Gauḍapāda went looking for the truth beneath the schools, a king had this cut into rock, and made his officers read it aloud in every district of an empire:

"Devānāṃpriya does not value either gifts or honours so highly as this, (viz.) that a promotion of the essentials of all sects should take place… Therefore concord alone is meritorious, (i.e.) that they should both hear and obey each other's morals. For this is the desire of Devānāṃpriya, (viz.) that all sects should be full of learning, and should be pure in doctrine." — Rock Edict 12

Set that beside the Advaitin's four phrases.

A promotion of the essentials of all sects — the truth beneath the schools. Concord alone is meritorious — free from all dispute. Hear and obey each other's morals — not opposed to any school. All sects full of learning and pure in doctrine — equally good for all beings.

Same move. A thousand years earlier. In law, across an empire, in public, in the local languages, on stone.

Nobody is claiming that Gauḍapāda read the edicts. By his century nobody in India could read them; the script had gone dark, and would stay dark for another twelve hundred years. This is not a case of one man copying another.

It is the more interesting case. It is a move that a civilisation arrives at more than once, because it answers a problem that will not go away: what do you do when the people you must live with do not believe what you believe?

India has answered that question twice at the highest level — once from a throne, once from a monastery — and both times the answer was the same. Look for what is shared, and let the labels be.

The claim this book will now make

Here is where the two chapters of Part 1 have been going.

If you honour Śaṅkara — and hundreds of millions do, as the man who gave the tradition its shape — then you already hold the principle. Not opposed to any school. Even opponents' views welcomed as parts. You did not get that from us. It is in the foreword to your own book, written by one of your own.

And the principle is older than the man who is honoured for it. It was state policy in this country in the third century BCE, and there is stone to prove it, standing where it has always stood.

So this book will spend the rest of its length on what that king actually said a person should do — and on why he thought each item was worth carving.

Not because it belongs to some tradition or other. Because on the evidence, it belongs to all of them.

That is what the next chapter begins.

Notes

  1. 1
    Kumārila Bhaṭṭa, first half of the seventh century CE — John Taber, A Hindu Critique of Buddhist Epistemology (London: RoutledgeCurzon, 2005), Introduction. That his canon stops short of the Purāṇic, avatāra and liberation material, and that of all the Brahmanical schools Mīmāṃsā took in the liberation-doctrine latest and least: Johannes Bronkhorst, Greater Magadha: Studies in the Culture of Early India, Handbook of Oriental Studies 2.19 (Leiden: Brill, 2007), Part II; Bronkhorst, ed., Mīmāṃsā and Vedānta: Interaction and Continuity (Delhi: Motilal Banarsidass, 2007). Śaṅkara, flourishing about 700 CE on the modern scholarly dating rather than the traditional 788–820, is a younger contemporary. Compression note: the body calls Kumārila "the strictest guardian", which is fair but broad — Mīmāṃsā is one school, and an unusually conservative one. His refusal shows that the material was still contestable at the orthodox centre in the seventh century; it does not show that no Brahmanical party anywhere had accepted it earlier, and the argument here needs only the first.
  2. 2
    This book places the Brahmanism-to-Hinduism transition at the consolidation rather than at the Gupta and early-Purāṇic centuries, where the standard histories place it. The conventional placement dates the appearance of the theistic religion of image and temple; this book applies the criterion set out in Chapter 1 — which figure the canon designates as supreme — and on that criterion the boundary falls where the new summit is accepted at the tradition's own centre, which Kumārila's refusal dates to after him. The fuller argument, with its apparatus, is in the author's Mechanism of the Vedic Lineage (in preparation), §9. On the conventional periodisation see Gavin Flood, An Introduction to Hinduism (Cambridge: Cambridge University Press, 1996), and Axel Michaels, Hinduism: Past and Present (Princeton: Princeton University Press, 2004). Two things this claim does not say. It does not say the Purāṇic religion was absent before — it plainly was not. And it does not say everyone agreed afterwards: the Śaiva and Vaiṣṇava traditions went on holding their own revealed canons alongside the consolidation, as they do to this day. A further caution the reader is owed: the word Hinduism itself carries a well-known modern-construct critique — its boundaries, and the name, are partly a much later formation — and it is used here only as the conventional label for the period. See Heinrich von Stietencron, "Hinduism: On the Proper Use of a Deceptive Term," in Hinduism Reconsidered, ed. Günther-Dietz Sontheimer and Hermann Kulke (New Delhi: Manohar, 1989); against the strong constructionist position, David N. Lorenzen, "Who Invented Hinduism?," Comparative Studies in Society and History 41, no. 4 (1999): 630–59.
Chapter 3

Chapter 2 · The king who wrote it down

Draft 1 — 25 August 2026. Second chapter of Part 1. All quotations from E. Hultzsch, Inscriptions of Aśoka (1925), public domain — see sources/hultzsch_translations.md. Hultzsch translates dhaṃma as "morality"; where a quotation says morality, the stone says dhaṃma. OCR text: check each quotation against the page scan before print.

Around 261 BCE, an army marched east1 from the Ganges plain into a country called Kalinga, on the coast of what is now Odisha. Kalinga had never been conquered. By the end of the campaign it had been.

We know the numbers because the man responsible published them.

"…one hundred thousand in number were those who were slain there, (and) many times as many those who died." — Rock Edict 13 (B)2

Then, in the same inscription:

"…slaughter, death, and deportation of people, this is considered very painful and deplorable by Devānāṃpriya." — Rock Edict 13 (E)3

Devānāṃpriya Priyadarśin, "Beloved-of-the-Gods, He-Who-Looks-Kindly", is the name he used for himself in public. We know him as Ashoka.4

Stop and look at what has just happened

Kings publish victories. That is what monuments are for. Every empire in the ancient world left behind stone with a ruler's triumphs cut into it — the enemies broken, the cities taken, the tribute delivered.

This one published the cost, in figures, and called it deplorable.

Nothing forced him to. He had won. He was, by that point, the most powerful man in the subcontinent, ruling territory from Afghanistan to the far south.5 No one could have compelled him to write anything, and no one was ever going to check.

He had it carved on rock faces and on polished stone pillars, in the ordinary spoken languages of each region rather than in the language of scholars, and he ordered his officers to read it aloud to people who could not read.6

He went to considerable trouble to make sure the worst thing he ever did could not be forgotten.

That is the first thing about him worth knowing, and you should hold on to it, because everything else in this book grows out of it.

The villain and the hero are the same person

It would be easy to write this chapter with a wicked king somewhere in it for contrast. There is no need. Ashoka is his own contrast.

The man who wrote Rock Edict 13 is the man who ordered the invasion. Both are one person, roughly eight years apart. You do not have to imagine what he was like before; he tells you.

This matters for how you read the rest. If the bad king were somebody else, you could put this book down thinking I am not like that man. Because it is the same man, the question is harder and better: which of the two is the real one — and how would you know, about yourself?

What changes a person

Something has to break before anyone rebuilds. For Ashoka the break is on record and needs no guessing: he saw what he had done, at a scale he could not put out of his mind, and he could not undo any of it. Nobody can raise the dead. That door is shut and stays shut.

But a broken thing is not the same as a direction. Knowing that what you did was wrong tells you nothing about what to do next, and there were many things a shaken king could have done. Conquer more, and drown it. Perform ceremonies until he felt clean. Give up the throne. Decide it was necessary and think no further.

He did none of those. He went and studied.

Why he chose that, out of everything available to him, we do not know from the stone. Later accounts — written in Sri Lanka some seven hundred years afterwards, so treat them as tradition and not as evidence — say a young renunciant crossed his path, and that this was what started him reading.7

Whether that account is true we cannot say. But the shape of it is worth noticing, because it is not unusual. People do not often change from pain alone. They change when, on top of the pain, they see with their own eyes that some other way of living is actually possible — not described, not promised, but standing in front of them.

You will meet exactly this shape once more in this book, in a very different century.

The road he did not take

There was, at that time, a well-worn way for a man in his position to respond. He could have walked out. India already had a long tradition of rulers' sons who left everything, and he would have been admired for it.

He stayed.

And he made a point of saying what he was — in a small inscription at Maski he calls himself an upāsaka, a layman who follows a teaching, not a renunciant who has left the world.8 He does not claim to be holy. He claims to be an ordinary man in an extraordinary job.

That decision is the reason this book is about him rather than about somebody wiser.

Because you are not going to leave the world either. Whoever is reading this has a family, a school, a country, and obligations they did not choose — which is precisely the position Ashoka was in, at scale. The question he had to answer is the one that is actually in front of you: what do you do about goodness while staying exactly where you are?

Then he made something of his own

He studied — and then he did not simply repeat what he had studied. He produced his own summary, in his own words, and put it where his whole empire could read it.

This is the thing we now call Ashoka's dhamma, and it is worth being exact about what it is and what it is not.

It is not a religious doctrine. Across everything he carved, in a reign of some forty years, there is no account of the world you are required to accept, no doctrine you must hold, no test of faith anywhere in it. He takes for granted that good conduct earns merit and that there is another world after this one — that much runs right through the edicts — but he never sets out what either is, and he never asks anyone to agree with him about it.9

It is a list of things to do. Here it is, in his own wording:

"(To practise) morality is meritorious; but what does morality include? (It includes) few sins, many virtuous deeds, compassion, liberality, truthfulness, (and) purity." — Pillar Edict 2 (B–C)10

Six items. And the version he ordered his officers to proclaim on tour, every five years, in every district:

"Meritorious is obedience to mother and father. Liberality to friends, acquaintances, and relatives, to Brahmanas and Śramanas is meritorious. Abstention from killing animals is meritorious. Moderation in expenditure (and) moderation in possessions are meritorious." — Rock Edict 3 (D)11

And a fuller statement, which adds the people most easily left out:

"…proper courtesy to slaves and servants, obedience to mother (and) father, liberality to friends, acquaintances, and relatives, to Brahmanas and Śramanas, (and) abstention from killing animals." — Rock Edict 11 (C)12

Read those three passages again and notice what is absent. Nobody is asked to change what they believe. A brahmin could do every item on that list without ceasing to be a brahmin. So could a Jain, an Ājīvika, a follower of a local cult, or a person with no views at all.

That is not an accident. He was running an empire containing all of them.

The part that goes furthest

The most striking thing he wrote is not about behaviour at all. It is about how you treat people who disagree with you — and he did not settle for telling people to be polite.

"(A) King Devānāṃpriya Priyadarśin is honouring all sects, both ascetics and householders; both with gifts and with honours of various kinds he is honouring them. (B) But Devānāṃpriya does not value either gifts or honours so highly as this, (viz.) that a promotion of the essentials of all sects should take place. (D) But its root is this, viz. guarding (one's) speech — (i.e.) that neither praising one's own sect nor blaming other sects should take place on improper occasions… (F) If one is acting thus, he is both promoting his own sect and benefiting other sects. (G) If one is acting otherwise than thus, he is both hurting his own sect and wronging other sects as well… (I) Therefore concord alone is meritorious, (i.e.) that they should both hear and obey each other's morals. (J) For this is the desire of Devānāṃpriya, (viz.) that all sects should be full of learning, and should be pure in doctrine." — Rock Edict 1213

Take clause (G) slowly, because it is not the argument people expect.

He does not say do not attack other people's beliefs, because it hurts them. He says attacking them hurts your own side. The man who runs down his neighbour's tradition in order to build up his own has just damaged the thing he was defending.

And clause (J) is further still. He is not asking for the different traditions to be tolerated, or endured, or left alone. He wants all of them to become learned and sound — including the ones he does not follow, and he spent state money on them.

Two thousand two hundred years later, this is still further than most people are willing to go.

He also built the machinery

He did not leave it as a sentiment. Thirteen years into his reign he created a new office of state, which had not existed before:

"In times past (officers) called Mahāmātras of morality did not exist before. But Mahāmātras of morality were appointed by me (when I had been) anointed thirteen years. These are occupied with all sects in establishing morality… They are occupied with servants and masters… They are occupied in supporting prisoners… (or) with those who are bewitched (i.e. incurably ill?), or with the aged." — Rock Edict 5 (H–L)14

Officers of dhamma, whose brief covered every sect, and whose duties reached prisoners, the incurably ill, and the old. That is a welfare service, in the third century BCE, written into the constitution of an empire.

And he made himself reachable:

"Reporters are posted everywhere, (with instructions) to report to me the affairs of the people at any time, while I am eating, in the harem, in the inner apartment, even at the cowpen, in the palanquin, and in the parks." — Rock Edict 6 (D)15

At any time. Including at dinner. The most powerful man in India made himself permanently interruptible by anyone with a problem, and put that in writing so that he could be held to it.

What the rest of this book is for

So we have a list. It came from a man who had done something unforgivable, who studied afterwards, and who then wrote down — in plain language, for everybody, in his own words — what he had concluded a person ought to do.

The obvious question is the one this book spends the rest of its length on.

Why those things and not others? What is it about honouring your parents, or guarding your speech, or holding back your hand from an animal, that he thought was worth carving into rock and maintaining an office of state to promote?

He does not explain the reasoning on the stone. He had no need to; he was issuing instructions, not teaching a class.

But we know what he had been studying. And in those texts, the reasoning is set out at length.

That is where we go next: one item at a time, the stone first, and then the machinery underneath it.


All edict quotations are from E. Hultzsch, Inscriptions of Aśoka, Corpus Inscriptionum Indicarum vol. 1, new ed. (Oxford: Clarendon Press, 1925), public domain. Edicts are cited by their standard designation and version, a canonical locus needing no page. Hultzsch renders the Prakrit dhaṃma throughout as "morality": where a quotation reads morality, the stone reads dhaṃma.

Notes

  1. 1
    The Kalinga campaign is conventionally placed in the ninth year after the consecration, about 261 BCE. Compression note: Ashoka's absolute dates rest on the synchronisms of Rock Edict 13, which names five contemporary Hellenistic kings — the anchor that makes Mauryan chronology firmer than almost anything else in early Indian history, but an anchor with a margin of a few years all the same. Standard treatments: Romila Thapar, Aśoka and the Decline of the Mauryas, 3rd ed. (New Delhi: Oxford University Press, 2012); Nayanjot Lahiri, Ashoka in Ancient India (Cambridge, MA: Harvard University Press, 2015).
  2. 2
    Rock Edict 13 (B), Girnar version. The Girnar stone is damaged here, which is why the quotation begins mid-sentence; versions at other sites preserve a fuller reading giving a hundred and fifty thousand deported as well as the hundred thousand slain. The body quotes only what the Girnar stone carries.
  3. 3
    Rock Edict 13 (E), Girnar. The same edict names the Hellenistic kings and sets out conquest by dhamma; the body uses the remorse passage alone and does not treat the rest.
  4. 4
    Devānāṃpriya Priyadarśin is the style used throughout the edicts. The personal name Aśoka appears in only a handful of inscriptions — the Maski minor rock edict among them — which is how the corpus was securely attributed to him; the Buddhist chronicles supply it in full.
  5. 5
    The extent is reconstructed from where the inscriptions stand — Kandahar in the north-west to Karnataka in the south — together with the border peoples the edicts name (Rock Edict 2 lists the Cholas, Pandyas, Satiyaputa, Keralaputa and Tāmraparṇī). Compression note: find-spots record where edicts were set up, which is evidence of reach and administration rather than a surveyed frontier. No Mauryan boundary line survives.
  6. 6
    The edicts are in local Prakrits, in Brāhmī and — in the north-west — Kharoṣṭhī, with Greek and Aramaic versions at Kandahar. Not Sanskrit. On proclamation by officers on circuit, Rock Edict 3 (C). Compression note: "read aloud to people who could not read" is an inference from the instruction to proclaim, not a separately attested practice. It is, though, the plainest reading of why a text is ordered to be announced on tour.
  7. 7
    The conversion narratives are in the Sri Lankan Pali chronicles — the Dīpavaṃsa, about the fourth century CE, and the Mahāvaṃsa, about the fifth — and in the Sanskrit Aśokāvadāna; see John S. Strong, The Legend of King Aśoka: A Study and Translation of the Aśokāvadāna (Princeton: Princeton University Press, 1983). This is why the body labels it tradition rather than evidence. These accounts postdate Ashoka by six or seven centuries, and the edicts name no monk in connection with the change: Rock Edict 13 gives the remorse at Kalinga directly, with nothing in between. The body neither accepts nor rejects the story, and a reader should notice that it does not.
  8. 8
    Minor Rock Edict 1, Maski version, where the king names himself upāsaka. His personal Buddhist affiliation is documented elsewhere as well: the Bhabra Edict recommends seven Buddhist texts by name, and the Lumbinī pillar records his pilgrimage to the Buddha's birthplace. The body's point is the distinction he maintained, not any doubt about his faith — what he promulgated as public policy is dhamma, and dhamma is not the doctrine he personally followed.
  9. 9
    ⚠️ A correction to an earlier draft of this chapter, kept visible because it matters. An earlier version said the edicts contained "no account of the afterlife you must accept". That was too strong. The other world is assumed openly and often — "endless merit is produced in the other (world) by that gift of morality" (Rock Edict 11 E); "(happiness) in this (world) and in the other (world) is difficult to secure" (Pillar Edict 1 C); "this (action conduces) to my (happiness) in this (world), that other (action) to my (happiness) in the other (world)" (Pillar Edict 3 H). The accurate claim is narrower and still holds: merit and a further world are taken for granted, yet nowhere defined, argued for, or made a condition of anything, and assent to any account of them is never required. A brahmin, a Jain, an Ājīvika and a follower of a local cult could each bring their own understanding of what "the other world" meant and still perform every item on the list.
  10. 10
    Pillar Edict 2 (B–C), Delhi-Topra version. In the Prakrit the six are approximately apāsinave (few faults), bahu-kayāne (many good deeds), dayā (compassion), dāne (giving), sace (truthfulness), socaye (purity). ⚠️ Outstanding: these Prakrit forms are to be checked against Hultzsch's transliteration pages before print.
  11. 11
    Rock Edict 3 (D), Girnar. The clause is given in the edict as the wording officers are to proclaim on the five-yearly circuit set out at (C).
  12. 12
    Rock Edict 11 (C), Girnar. Clause (E) of the same edict carries the merit-in-the-other-world statement discussed at n. no-doctrine.
  13. 13
    Rock Edict 12, Girnar. Clause (D) grounds the whole in guarding one's speechvacaguti — which is worth marking: the root of the toleration is a discipline of what one says, not a theory about religion. Compression note: the body reads clause (G) as the argument that attacking another sect damages one's own. That is what the clause says. Whether Ashoka intended it as prudential counsel to the sects or as a claim about how sects in fact fare is not settled by the text, and the body does not decide it.
  14. 14
    Rock Edict 5 (H–L), Girnar. The office is the dhaṃma-mahāmāta, instituted in the thirteenth year after consecration. Compression note: the passage is damaged in places, and some of Hultzsch's renderings of the categories of person the officers assisted — the bracketed "incurably ill" among them — are reconstructions he marks as uncertain. The body follows him and flags nothing further, since the general scope is not in doubt.
  15. 15
    Rock Edict 6 (D), Girnar. Compression note: what the edict establishes is a standing instruction to his reporters — a claim about the system he set up and about how he wished it known, not independent evidence that he was in fact always available. The body's point is the commitment he made in public, and that is verifiable.