Ṛg Veda 1.164.46: One Truth, Many Names
"They call it Indra, Mitra, Varuṇa, Agni; and it is the divine, well-winged Garutmān. The existent is one; the wise speak of it in many ways — they call it Agni, Yama, Mātariśvan."
This is the strongest pūrvapakṣa in Part III, and it is stated here as its own proponents state it, not as a caricature.
The verse is the clearest statement of unity of referent in the entire corpus and it long predates any of the schools now quarrelling over it. Ekaṁ sat — the one existent — is stated without qualification and without personal attributes. The plurality of names is explicitly a matter of how the wise speak, not of how things are: a statement about language, not about a person. Any reading that inserts a specific named deity as the referent has done exactly what the verse warns against — taken one name as privileged. Reading Viṣṇu into ekaṁ sat is the most obvious form of the very error the verse identifies.
The honest response begins by conceding what the pūrvapakṣa gets right.
1. Conceded. The verse does assert unity of referent behind plurality of name. Anyone who reads it as asserting a plurality of independent supreme beings is contradicting it, and no reading offered here does so.
2. What the verse does not say. It does not say the one existent is impersonal, formless, or without attributes. It says the wise call it by many names — and then it supplies a name of its own: suparṇo garutmān, the divine well-winged one. That is a description of a form. A verse whose point was that the referent has no form has chosen a very strange half-line to include.
3. The word sat does work. Ekaṁ sat is “one existent” — a present participle of √as, a term for that which is, not a term of negation. The Chāndogya develops exactly this word: sad eva saumya idam agra āsīt ekam evādvitīyam, where the same one existent is immediately said to have looked and willed — tad aikṣata bahu syām. Willing is not a property of an attributeless absolute. Read in the corpus rather than alone, the ekaṁ sat line describes something that deliberates.
4. Samanvaya (§6.1) decides between (b) and (c). Both surviving readings account for this verse. They diverge on the rest of the corpus. Reading (b) must treat every passage assigning attributes, form, will, and personality to the supreme as provisional or as concession to weak intellects — a very large body of text, including the Gītā entire, demoted by a rule the corpus never states. Reading (c) requires no such demotion. Under samanvaya, the reading requiring fewer passages to be strained wins. That is the whole argument, and it is an argument about accounting, not about assertion.
5. The corpus names its own referent. The accounting argument does not have to end in a tie broken by preference, because the corpus supplies the name in the first person, by direct statement — rank one of six on the evidence hierarchy of §8.3:
"By all the Vedas, I alone am to be known. Indeed, I am the compiler of Vedānta, and I am the knower of the Vedas."
Sarvaiḥ refuses any partition of the corpus; eva refuses any second ultimate object of knowledge. ṚV 1.164.46 states that the one existent is spoken of under many names; BG 15.15 states whose names they are. Neither statement requires the other to be bent, and both are Grade A.
6. The lineage supplies the mechanism. Madhva's Aitareya Upaniṣad Bhāṣya extends the principle past articulate names to the whole acoustic field — samudra-megha-vṛkṣapatana-bherītāḍanādi sarvaghoṣā api tasyaiva nāmāni, all sounds whatsoever are names of Him alone (§7.2, Grade B, pending collation). The Garuda Purāṇa 3.24.44–45 narrates the same structure as an act of delegation, and the Bharadvāja Saṁhitā 2.1.79 names the mode in which a devatā name reaches Puruṣottama, vyāmiśra, the intermixed. Both are Grade C (§7.3) and are cited here as illustration only. The adjudication below does not rest on them.
Reading (c). One existent, personally qualified, denoted by many aspectual names. Grade A on the śruti-and-sūtra chain: ṚV 1.164.46 with VS 1.1.4 and 1.4.28 for the method, and BG 15.15 for the referent by direct statement. Reading (a) is excluded by the verse's own ekam. Reading (b) is not excluded by this verse and is excluded, if at all, only by the wider corpus.
The names are not synonyms but aspectual: each denotes Him under a distinct quality — “Indra” as splendour, “Agni” as the conveyor of oblation. Information is preserved; only exclusivity is denied. This is the conclusion stated at width 2 of §7.6.
What this does not license. Reading (c) does not support the popular gloss that all worship is equivalent and all destinations identical. That gloss contradicts a passage that could not be more explicit: Bhagavad-gītā 9.25 states that worshippers of devas reach devas, of ancestors reach ancestors, of spirits reach spirits, and His own reach Him. Unity of ultimate referent and identity of proximate destination are different claims, and the corpus asserts the first while denying the second.
We record plainly that (b) is the reading of an opposing school; that its defenders have answered the accounting argument at great length; and that this case file demonstrates where the hermeneutic divergence is located. The Vaiṣṇava conclusion upholds the real distinction between the Lord and the individual soul. A summary of the opposing pūrvapakṣa treatment, published as a pūrvapakṣa node, is register entry D-08.
Separately: popular misreadings treating this verse as polytheistic syncretism or as voidist non-differentiation are corrected wherever they appear on this surface — but correcting a popular misreading is not the same as answering the school, and the two must not be conflated in review.