Whose Voice Is the Veda?

I. One Question, Set Apart

Ask a traditional Saṃskṛta scholar whether the Veda has an author, and the answer you get will depend entirely on which school you have asked. The debate that runs through Nyāya, Mīmāṃsā and Vedānta on this point is often read together with a second, quite different debate — whether śabda, the phonemes that make up spoken language, is a permanent kind of thing or a fresh event that arises and perishes with each utterance. That second question, the nityatva/ anityatva question, is the subject of this essay's companion piece. This essay isolates the first: does the Veda have an author, a puruṣa who composed it, in the way Kālidāsa composed the Raghuvaṃśa? This is the pauruṣeyatva/ apauruṣeyatva question, and three schools that never doubt the Veda's authority reach three incompatible verdicts on it.

It is tempting to assume that a school's answer here must track its answer to the eternity question — that an eternal Veda is automatically authorless, and a produced Veda automatically someone's composition. Nearly every school resists exactly that assumption at some point. Nyāya holds the Veda to be produced and authored, and treats both facts as compatible with — indeed as the ground of — its authority. Mīmāṃsā holds the Veda to be authorless and needs its companion doctrine of eternity to secure that claim. Advaita Vedānta, in the most technically interesting move in the whole debate, holds the Veda to originate anew with each cosmic cycle and yet insists it is not authored at all. Watching how each school defends its verdict on authorship — and where that verdict does or does not lean on a claim about sound's own nature — is the real content of this essay.

II. Why This Is Not a Question About Age

A modern reader's first instinct is to hear “does the Veda have an author?” as a historical question — how old is this text, really, and has its origin simply been forgotten? — and to file the entire debate under pious exaggeration. That instinct misreads what is at stake. Nobody in this literature is arguing about how many centuries the Veda has existed. The question is whether a puruṣa stands behind its ānupūrvī — its fixed sequence of syllables, words and sentences — at all, at any point, ever.

The reason this matters is epistemic, not antiquarian. A speaker, however wise, can be mistaken (bhrama), careless (pramāda), or deceptive (vipralipsā). If a text has a composer, these three possibilities automatically live inside it, and the text's reliability becomes only as good as the person standing behind it. Remove the person — genuinely remove him, not merely forget his name — and that entire class of defect has nowhere to enter from. Apauruṣeyatva, on the Mīmāṃsaka reading, is thus an argument about the structural impossibility of error, not a claim about the manuscript's date. Everything downstream in this essay follows from that reframing: once the real prize is seen to be reliability rather than antiquity, it becomes clear why three schools that never doubt Vedic authority can still fight this hard about who, if anyone, stands behind it.

III. Nyāya–Vaiśeṣika: Authority Through a Trustworthy Author

Nyāya–Vaiśeṣika's route into this question runs through its physics of sound. In its analysis of substances and qualities, śabda is classified as a produced, perishable quality (guṇa) of ākāśa — a conclusion argued for in detail in this essay's companion piece. That conclusion sets up rather than settles the question asked here: given that Nyāya holds the Veda to be a produced text, built out of produced sounds, who produced it?

It would be a mistake to read Nyāya's anityatva thesis as scepticism about the Veda's status. Nyāya's own account of Vedic authority is, if anything, more direct than the Mīmāṃsaka's. The Vedāntaparibhāṣā states the Naiyāyika position compactly:

वेदानां नित्यसर्वज्ञपरमेश्वरप्रणीतत्वेन प्रामाण्यम् इति नैयायिकाः

The Veda is authoritative because it was composed by an eternal, omniscient Parameśvara — Naiyāyikas hold. Authorship is not a liability on this account; it is the very ground of authority. A statement is trustworthy when its speaker is āpta — competent, truthful, and directly acquainted with what he reports — and no speaker could be more āpta than an omniscient one. Testimony, for Nyāya, is only as good as the witness, and here the witness cannot be improved on.

The Veda's authority is therefore, for Nyāya, derivative — it depends on an independent proof that Īśvara exists and is competent — but once that dependency is granted, the account is simple and requires no special metaphysics of sound at all. This is exactly the point where the Mīmāṃsaka presses hardest, as the next section shows.

IV. Mīmāṃsā: No Author, Because None Is Needed

The Mīmāṃsaka's opposing thesis is stated in the Vedāntaparibhāṣā in the same breath as Nyāya's:

वेदानां नित्यत्वेन निरस्तसमस्तपुंदूषणतया प्रामाण्यम् इत्यध्वरमीमांसकाः

The Veda is authoritative because, being eternal, every defect traceable to a person has already been excluded from it — so hold the Ādhvarīṇa (ritualist) Mīmāṃsakas. Where Nyāya makes the speaker maximally trustworthy, Mīmāṃsā removes the speaker altogether, on the reasoning that no speaker at all is more reliable than even the best speaker, because no speaker means no possibility of error entering the chain in the first place. That this rests on a claim that the Veda's phonemes are kūṭastha-nitya — eternal in the strongest sense, static and never produced or destroyed — is the metaphysical half of the argument, and it is examined in full in this essay's companion piece. What concerns this essay is the other half: why Mīmāṃsā thinks the Veda shows every sign of being authored, and how it answers each sign in turn.

Three marks of a made text — and why they fail

Before defending its own position, the Mīmāṃsāmañjarī lays out, fully and fairly, the case that the Veda looks exactly like something composed — a case shared, interestingly, by Nyāya and by non-Vedic schools alike (vedo'nityaḥ pauruṣeyaśca, “the Veda is non-eternal and authored”). Among the marks offered, three bear directly on authorship rather than on the nature of sound, and each is dismantled in turn.

1. The grammatical mark: śākhā names as taddhita derivatives.

This is the sharpest of the marks. The Vedic recensions (śākhās) are named Kāṭhaka, Kālāpa, Kauthuma, Paippalāda, Taittirīya — formations from personal names. Pāṇini's rules for such derivatives (taddhita suffixes) apply in the senses tena proktam (“spoken by him”) and kṛte granthe (“a text made [by him]”). If the very grammar used to name the Veda's branches encodes authorship, the Veda is kārya (a product), and every product needs a maker.

The Mīmāṃsaka reply distinguishes a maker from a teacher. A samākhyā (a name of this kind) can attach to whoever is distinctively associated with transmitting a branch, not only to whoever invented it. Vaiśampāyana is remembered as one who knew all the branches; Kaṭha taught only this one. Where many scholars command many śākhās, the one who devotes himself to a single branch acquires an atiśaya — an exclusive association — through unbroken teaching succession, without ever having composed a syllable of it. The name records who carried the branch forward, not who first spoke it.

2. The mark of content: mortal names inside the text.

Passages such as babaraḥ prāvāhaṇir akāmayat and kuruvinda audālakir akāmayat name individuals who are born and who die — Audālaki is, by derivation, “descendant of Uddālaka.” A sentence naming him cannot have existed before he was born; a text containing such names is dated by them.

The reply works by re-deriving the words. Prāvāhaṇaḥ is read from pra + vah + i — “that which drives forward” — and babara as an anukṛtiśabda, an imitative sound; both then denote Vāyu, the wind — an eternal referent, not a dead man — and the appearance of a patronymic simply dissolves under closer etymology.

3. The mark of absurdity: trees and serpents in sacrificial session.

Vanaspatayaḥ satram āsata; sarpāḥ satram āsata — “the trees sat in sacrificial session; the serpents sat in sacrificial session”. Attributing a multi-day ritual session to plants and snakes reads, on its face, like the speech of a madman or a small child — exactly what one expects from a fallible human author, not from a beginningless, authorless corpus.

The Mīmāṃsaka classifies such lines as arthavāda — commendatory or explanatory material attached to an injunction, meant to be stutipara (in service of praise), not read literally. The rhetorical logic is a fortiori: if even insentient trees are described as sitting in session, how much more should learned brāhmaṇas do so. The extravagance is the point of the praise, not evidence of a fallible speaker; nothing in the underlying ritual injunction is disturbed by it.

A fourth mark, drawn from ordinary speech about sound itself (utpanno gakāraḥ, “the sound g has arisen”), belongs to the nityatva question rather than the authorship question and is examined in Part 2.

The formal inference — and why Mīmāṃsā thinks it fails

These marks are gathered into a single formal inference (anumāna):

विमतं वेदवाक्यं पौरुषेयम्, वाक्यत्वात्, कालिदासादिवाक्यवत्

The Vedic sentence under dispute is authored, because it is a sentence — like the sentences of Kālidāsa. Two lines of response follow.

First, a counter-inference of equal formal standing (satpratipakṣa):

वेदः प्रागभावप्रतियोगितावच्छेदकधर्माभाववान् अस्मर्यमाणकर्तृकत्वात् आत्मादिवत्

The Veda lacks the very property that would make it the counterpositive of a prior absence — because no author of it is remembered, just as with the self (ātman) and other beginningless things.

Second, an attack on the reasoning itself: vākyatva (“being a sentence”) is sopādhika — it only works as evidence of authorship when accompanied by an unstated rider, janyajñānamūlakatva (“being rooted in a produced cognition”). Ordinary sentences are authored not simply because they are sentences, but because they arise from a cognition someone first acquired and then chose to express. Where that condition is absent — as the Mīmāṃsaka claims it is for the Veda — the inference has no grip; a reason that works only through a hidden, unproven condition is aprayojaka, incapable of establishing anything.

Why the author could not simply have been forgotten

Nyāya's obvious rejoinder — that no author is remembered simply because memory faded, as with the digger of an old well — is met with unusual force. A forgotten well-digger is still known, in general, to have existed; the disregard that erased his particular name is itself visible in how carelessly the well is treated. The Veda's situation is the reverse: universal reverence, unbroken transmission, lifelong study — and still no author at all. High regard combined with total absence of memory, the Mīmāṃsaka argues, is not what forgetting looks like.

The argument sharpens further. The forgetting would have to be attributed to the very ṛṣis who fixed the pada- and krama-recitations syllable by syllable and who are themselves credited with omniscience (sarvajña) — to say that these men were too indifferent to preserve the name of the Veda's author is, in the source text's own word, atisāhasa, recklessness. Nor could the memory have simply lapsed from disuse: since the efficacy of Vedic ritual is pramāṇāntarāgocara — reachable by no other means of knowledge at all — anyone performing that ritual would be relying wholly on confidence in the speaker's word. Under those conditions, the one thing that could not be allowed to lapse is precisely the speaker's identity.

Two supporting inferences close off any temporal starting point:

पूर्वकालो न वेदशून्यः, कालत्वात्, वर्तमानकालवत्

प्राक्कालिकं वेदाध्ययनं गुर्वध्ययनपूर्वकम्, अध्ययनत्वात्, अद्यतनाध्ययनवत्

Earlier time was not empty of the Veda, being time, like the present; and study of the Veda in any earlier period was itself preceded by a teacher's study, being study, like study today. Each act of learning presupposes an earlier act of learning, and the regress has no first term at which composition could be located.

A further, almost sociological argument is offered: where an author is genuinely remembered — the Mānava Dharmaśāstra, the Bhārata, the Buddhist texts — tradition converges without dispute. Where the Veda is concerned, proposed authors multiply endlessly and disagree: Īśvara, Hiraṇyagarbha, Prajāpati, or some plurality of beings. Persistent disagreement of that kind, the Mīmāṃsaka argues, is not what a real, if half-forgotten, memory looks like; it is what an inference dressed up as a memory looks like.

Even the apparently plain śruti passage trayo vedā ajāyanta — “the three Vedas were born,” from Agni, Vāyu and Āditya — is reread as jñāpita rather than janita: Agni made the Ṛgveda known, did not generate it. This reading is backed by other śruti passages describing speech as beginningless:

अनादिनिधना नित्या वागुत्सृष्टा स्वयम्भुवा। स्वयम्भूरेष भगवान् वेदो गीतः स्वयम्भुवा॥ शिवाद्या ऋषिपर्यन्ताः स्मर्तारोऽस्य न कारकाः॥

From Śiva down to the ṛṣis, all are only rememberers (smartāraḥ) of the Veda, none its makers (kārakāḥ).

The final vikalpa: who, exactly, could the author be?

The Mīmāṃsaka closes with a dilemma addressed to any candidate author. A human being? Impossible, since knowledge of dharma and its unseen fruit (apūrva) arises only from the Veda itself — no human faculty could have furnished the content in advance, and what a man never conceived, he cannot have composed. A yogin, then, using some extraordinary faculty? Then we must ask by which instrument he apprehended dharma: not an external sense, since dharma is not the kind of object the senses register; not the internal organ (manas) either, since manas cannot generate cognition of anything beyond the self and its own qualities. Īśvara, then, as Nyāya proposes? Here the objection is structural rather than empirical: Īśvara is himself established on the strength of the Veda (through scriptural testimony to his existence and qualities), while the Veda is said to be established by Īśvara — a mutual dependency (anyonyāśraya) that cannot be lifted from either side. A smaller point rides along with it: several Vedic passages describe worship and meditation directed at Īśvara, and an account of one's own devotional practice sits oddly inside a text one has supposedly composed oneself.

With every candidate closed off, the Mīmāṃsaka's conclusion is not that the question of authorship has gone unanswered. It is that the absence of an answer is itself the answer: apauruṣeyatva is not a gap in the record, but what the record, examined honestly, actually shows.

V. Advaita Vedānta: Originating Without Being Authored

Śaṅkarācāryacārya's treatment of this question, in the Devatādhikaraṇa of the Brahma-sūtra-bhāṣya at Brahma Sūtra 1.3.28–30, and its later systematization in Dharmarāja Adhvarīndra's Vedānta Paribhāṣā, produces the most technically surprising position in the whole debate: the Veda originates anew at the start of every cosmic cycle — the argument for which belongs to this essay's companion piece — and yet it remains apauruṣeya, authorless. This essay picks up at the point that combination requires defending: does origination not simply mean authorship?

The Vedānta Paribhāṣā meets the objection with a technical redefinition rather than a denial of origination. Two candidate senses of pauruṣeyatva are tried and rejected first. Being uttered by a person will not do, since on the Vedāntin's own view the Veda is transmitted through a succession of teachers who utter it in study and recitation, which would make it pauruṣeya by that test too. Having origination dependent on a person will not do either, since the Vedāntin already grants that the Veda originates from Īśvara — the Naiyāyika's inference to pauruṣeyatva on those grounds would then prove only what is already accepted, and collapse into siddhasādhana, proving the already-proven. What survives both tests is a narrower, technical definition:

पौरुषेयत्वं सजातीयोच्चारणानपेक्षोच्चारणविषयत्वम्

Pauruṣeyatva is being the object of an utterance that does not depend on a prior utterance of the same kind. On this definition, the Veda escapes the charge. At the start of each creation, Parameśvara composes a Veda whose sequence is identical to that of the Veda established in the cycle before it (पूर्वसर्गसिद्धवेदानुपूर्वीसमानानुपूर्वीकं वेदं विरचितवान्) — not a different one, invented independently. His utterance is therefore never independent of a prior, homogeneous utterance; it reproduces rather than originates. The Mahābhārata and comparable texts, by contrast, have no such prior utterance standing behind them — their authors compose freely, without modelling their words on any preceding text of the identical kind — and are pauruṣeya in the strict, technical sense. Āgama accordingly divides cleanly into the pauruṣeya and the apauruṣeya, and the Veda falls on the apauruṣeya side without any need to claim that it is eternal in the strong, static sense at all.

The mechanics of that recurrence — how the same Veda can arise afresh each cycle and still count as a repetition rather than a new composition — rest on a distinction between two kinds of eternity that this essay's companion piece unpacks in full. What matters here is the result: Advaita is the one school in this debate where nityatva and apauruṣeyatva visibly come apart. An authorless text can nonetheless originate, because what makes a text authored, on this account, is not that someone spoke it, but that the speaking was not itself a re-speaking of something prior.

VI. Three Schools, Compared

Laid side by side, the three positions differ sharply on who, if anyone, stands behind the Veda — and on what, in consequence, grounds its authority.

Nyāya shows that pauruṣeyatva and reliability can go together, provided the author is flawless — at the cost of making validity derivative on an independent proof of Īśvara, which is precisely where the Mīmāṃsaka presses a charge of circularity (Īśvara proved by the Veda; the Veda validated by Īśvara).

Mīmāṃsā shows that apauruṣeyatva can be established on structural grounds alone, at the cost of a standing burden: explaining away every apparent mark of composition, one by one, as the argument above shows it doing. Advaita's account is the most surprising of the three: an authorless text that nonetheless originates, because authorship, on its technical definition, was never about origination in the first place.

VII. What Remains Open

Three schools, one shared conviction that the Veda can be trusted, and three incompatible accounts of who, if anyone, is speaking. Nyāya answers by making the speaker maximally trustworthy: an omniscient author cannot be wrong, so his testimony transfers its reliability whole. Mīmāṃsā answers by removing the speaker: with no person behind the words, no personal defect has anywhere to attach, and the text's reliability needs no external guarantor at all. Advaita answers by removing the speaker's originality while leaving the speaker himself firmly in place: Īśvara does speak the Veda, cycle after cycle, but he never once speaks it for the first time.

But a second question has been deliberately set aside throughout this essay: whether the sound that composes the Veda — the phonemes themselves — is eternal or a fresh event with each utterance. Nyāya's anityatva thesis, Mīmāṃsā's kūṭastha-nitya varṇas, and Advaita's three-level distinction between token, type and structure were each invoked here only to the extent authorship required them. Each deserves the same close reading given to authorship in this essay — and each is taken up in turn in Part 2: Is Sound Eternal? Śabda Nityatva Across the Darśanas.

VIII. A Short Glossary, for Readers New to the Terms

Terms specific to the eternity-of-sound question — nitya/anitya, kūṭastha-nitya, pravāha-nitya, guṇa, sphoṭa, varṇa, and others — are glossed in full in Part 2.

IX. Sources and Editorial Notes

Primary sources drawn on

Thangaswami Sarma, R. (1996). Mīmāṃsāmañjarī (prathamaṃ kusumam). Indian Council of Philosophical Research. (Distributed by Munshiram Manoharlal Publishers)

Used for: §6 (pūrvamīmāṃsādarśane vedasyāpauruṣeyatvaṃ nityatvaṃ ca), pp. 22–26 — the pūrvapakṣa (marks of authorship) at pp. 22–23, the Mīmāṃsaka reply from p. 23 to p. 26.

Dharmarājādhvarīndra. (n.d.). Vedāntaparibhāṣā [with ‘Prakāśa’ Hindi commentary] (G. S. Musalgaonkar, Comm.; S. S. Musalgaonkar, Ed.) (The Vidyabhawan Saṃskṛta Granthamala, No. 100). Chowkhamba Vidyabhawan.

Used for: the Āgama chapter, for the Advaita summary statements and the redefinition of pauruṣeyatva.

Śaṅkarācāryacārya. (1938). The Brahmasūtra Śaṅkarācāryacārya Bhāṣya with the commentaries Bhāmatī, Kalpataru and Parimala (Maha-M. A. Sāstrī, Ed.; re-ed. Bhārgava Sāstrī, 2nd ed.). Nirnaya Sagar Press.

Used for: Brahma Sūtra 1.3.28–30 (the Devatādhikaraṇa) — for the background to Advaita's position on origination without authorship.

Viśvanātha Nyāyapañcānana Bhaṭṭācārya. (1932). Kārikāvalī, saṃvalita with the [author's own] Nyāyamuktāvalī [commonly cited as Nyāya-siddhānta-muktāvalī], together with the Dinakarī and Rāmarudrī glosses (Ātmārāma Śarmā, Rev.; 2nd ed.). Nirnaya Sagar Press.

Used for: the Śabda-khaṇḍa, for the Nyāya pūrvapakṣa/siddhānta referenced in brief; treated in full in Part 2.