Four Verbs in the Past Tense: The Rigveda's Only First-Person Report of Immortality

The sentence that should not be there

Almost everything the Rigveda says about deathlessness is said in the optative, the mood of wishing. Make me immortal. May we cross over. Let us live a hundred autumns. The poets ask. That is what the collection largely is: a body of requests, addressed upward, by people who knew perfectly well they were going to die.

Then, in the eighth book, a poet stops asking.

ápāma sómam amṛ́tā abhūma áganma jyótir ávidāma devā́n

“We have drunk the soma; we have become immortal; we have gone to the light; we have found the gods.”

Rigveda 8.48, verse 3. Jamison and Brereton (Oxford, 2014), vol. 2, p. 1129.

Four verbs, four clauses, no conjunctions, no petition, no addressee until the second half of the stanza. And every one of the four verbs is in a past tense. Not may we become. We have become. The claim is not that immortality is coming. The claim is that it has already happened, that it happened a few minutes ago, and that the speaker is currently inside it.

There is nothing else quite like this in the Rigveda. Jamison and Brereton, whose 2014 translation is the standard scholarly English version, call it “the famous exultant boast.”[1] They are not being dismissive; they are being accurate about the register. Something has been drunk, and the person who drank it is reporting.

What follows is a close reading of that hymn, RV 8.48, and of the argument it has generated. The argument is unusually good, because the verse is unusually exposed: it is short, it is famous, it is grammatically strange, and it refuses to explain itself. Fourteenth-century Sanskrit commentators worried about it. So did Victorian translators, a Dutch Indologist writing in 1994, a Cambridge philologist demolishing a mushroom theory in 1971, and a banker turned mycologist who thought he had solved the whole thing. They reached incompatible conclusions from the same eleven Sanskrit words. My argument is that the disagreement is the interesting part, and that the hymn as a whole points somewhere none of the famous readings quite go.

A soma hymn in the wrong book

Nearly all the Rigveda’s soma poetry sits in the ninth maṇḍala, 114 hymns to Soma Pavamāna, “self-purifying Soma,” almost all of them concerned with what happens to the plant during the ritual: the crushing, the straining, the flowing. I have written elsewhere about why an entire book was built around a liquid passing through wool. Book 9 is process poetry.

RV 8.48 is not in book 9, and this matters. Jamison and Brereton note that it “is one of the very few dedicated to Soma outside of the IXth Maṇḍala, and as such does not concern ‘self-purifying’ Soma.”[1] The hymn barely mentions how soma is made. There is a single passing reference to it as “pressed” in verse 7. Everything else is about the drink after it is drunk, and specifically about what it does inside a body.

The metre is worth a second of arithmetic, because it tells you the shape of the thing on the page. A triṣṭubh stanza is four lines of eleven syllables:

$$4 \times 11 = 44 \text{ syllables per stanza}$$

The single jagatī verse, verse 5, runs to four lines of twelve, giving 48. That one stanza is metrically wider than its neighbours, and it happens to be the stanza where the poet describes soma binding his joints together. Whether the metrical swell is deliberate is not something anyone can prove. It is the kind of coincidence Vedic poets were entirely capable of engineering; I have made the case elsewhere that these were craftsmen who thought of a hymn as a built object, and a change of gauge in one course of the wall is a carpenter’s move.

The grammar is the argument

Here is the crux, and it is a real one rather than a manufactured puzzle.

The four verbs in ápāma sómam amṛ́tā abhūma, áganma jyótir ávidāma devā́n are aorists: ápāma from pā, “drink”; abhūma from bhū, “become”; áganma from gam, “go”; ávidāma from vid, “find.” English translators render them with the present perfect (we have drunk, we have become) because the Vedic aorist characteristically reports an action just completed and still in force. That is the grammar. It is why the verse reads like a live dispatch rather than a story about something that happened once.

But a past tense making a present claim is exactly the sort of thing that invites reinterpretation, and the Indian commentarial tradition got there first. Sāyaṇa, writing his Rigveda commentary in fourteenth-century Vijayanagara, cites the Taittirīya Saṃhitā for the principle that a statement of past fact can be used to express a wish: āśaṃsādyotanāya bhūtārthanirdeśaḥ.[2] On that reading, the poet is not reporting. He is wishing, in a rhetorical past tense, the way English speakers say “that’s settled then” about something not yet settled.

This single grammatical decision splits the entire translation history in half.

The modern consensus is indicative, and I think it is right. Sāyaṇa’s rule is genuine but it is a licence, not an obligation, and nothing in the hymn’s surrounding verses reads like petition at that moment. The petitions come later, in verses 4 and 10, and they ask for something quite different, which is the point I want to build toward.

Still, notice what Sāyaṇa’s discomfort tells us. A learned reader in the fourteenth century, fully inside the tradition, found the plain sense of this verse difficult enough to require a grammatical manoeuvre. He was not reading it as unremarkable devotional language. He was reading it as a claim strong enough to need managing.

Aside. There is a second, quieter piece of evidence for how difficult this verse is. Jamison’s detailed published commentary on Book 8 contains notes on verses 8.48.1, 2, 4, 5, 6, 9, 10 and 11. There is no note on verse 3.[3] The most famous line in the hymn draws no philological comment from the most careful living commentator on the text. Sometimes the loudest thing in a scholarly apparatus is a gap.

The word that means not-dead

Amṛta (Sanskrit: अमृत) is a negative compound: a-, “not,” plus mṛta, “dead.” The Rigveda’s word for eternal life literally means not-dead, and its shape is inherited. The adjective reconstructs to Proto-Indo-European *n̥-mr̥-tó-, giving Vedic amṛ́ta-, Avestan aməša- (as in the Zoroastrian Aməša Spənta, the “Bounteous Immortals”), and Greek ámbrotos. That equation is uncontroversial.

What is not established, and what popular writing routinely overstates, is the next step. Greek ambrosía, the noun naming the food of the gods, is a Greek-internal derivation from ámbrotos. It does not follow that Proto-Indo-European had an institution called “the undying stuff.” Joel Brereton puts the moderate position carefully: soma “likely inherits Indo-European traditions of the drink of deathlessness,” with the hedge doing real work.[4] Paul Thieme’s celebrated 1952 proposal that Greek néktar means “death-overcoming” is elegant and widely repeated, but Beekes’s standard modern etymological dictionary of Greek lists néktar as “of uncertain origin” and canvasses Pre-Greek and Semitic alternatives instead.[5] The family resemblance across Indo-European drink-of-immortality traditions is real at the level of the adjective and thin at the level of the substance.

Two philologists set the terms for how amṛta should be read in the Rigveda specifically. Thieme argued the word means non-dying rather than divine eternity, and Jan Gonda glossed it as “life” in the sense of “continuance of life, vitality, attainment of old age, being secure against a premature death.”[6] Henk Bodewitz, whose 1994 study of life after death in the Rigveda is the single most useful thing written on the subject, endorses the line bluntly: “stray references to ‘immortality’ in the oldest books are nowadays rightly interpreted as ‘non-dying,’ ‘deathlessness,’ i.e. remaining alive on earth.”[7]

A crude but reproducible count supports the same picture. Running a word-initial search for amṛt- over the complete saṃhitā returns roughly 210 tokens, spread across all ten books but concentrated in the tenth, the latest and most speculative. The concrete genitive amṛtasya, “of the deathless (thing),” outnumbers the abstract amṛtatvam, “deathlessness,” by about two to one.

Methods note. These are surface counts over the van Nooten and Holland electronic text, not lemma counts from a published concordance, and sandhi hides instances: pavamānāmṛte at 9.113.7 is invisible to a word-initial search. Treat 210 as a floor, and cite Grassmann’s 1873 Wörterbuch zum Rig-Veda if you need a figure that will survive a referee.

The shape of that distribution is the argument in miniature. Deathlessness in the Rigveda is mostly a substance or a property that something else has, and only later and more rarely an abstract state a person might enter. On the wider vocabulary of amṛta and the Rigveda’s remarkable preference for asking for a long life rather than an eternal one, see the companion piece on the hundred-autumn prayers.

So did they mean heaven?

No, and the reasons are worth laying out, because this is where the scholarship gets sharp.

The obvious modern reading of “we have gone to the light, we have found the gods” is a vision of paradise: the drinker’s soul rises, sees the celestial light, meets the deities, and reports back. That reading has a distinguished defender in Thomas Oberlies, who treats the hymn as the record of a Somarausch, a soma intoxication, in which the drink “gives a preview of life in heaven by producing visions or hallucinations.”[8] Light and the sun, on his account, are the central religious aims of the hymn, and the drinker gets a foretaste of them.

Against that stands the flat denial of Hyla Converse, who examined the hymn and concluded: “There is absolutely nothing in the hymn that refers to a life in heaven with the gods or fathers.”[9] On her reading, “light” is ordinary Rigvedic shorthand for prosperity and well-being, and “finding the gods” is standard sacrificial language for having successfully met the gods at the rite, which is what a sacrifice is for. The gods come to the ritual ground. You find them there. No ascent required.

Bodewitz, weighing both, lands somewhere deliberately unstable. He concedes that Converse is right about the rest of the hymn, that “the further information of this hymn does not concern life in heaven and that a realm of the blessed deceased is not mentioned,” and then adds: “However, a vision of immortality or life in heaven may have been expressed in this verse.”[7] That is a scholar refusing to close a question he cannot close. I find it the most honest sentence in the literature on RV 8.48.

graph TD
    A["RV 8.48.3: we have become immortal"] --> B["Literal: heaven attained"]
    A --> C["Ecstatic: vision of heaven"]
    A --> D["Ritual: gods met at the rite"]
    A --> E["Rhetorical: past tense as wish"]
    B --> F["Bodewitz: possible, unproven"]
    C --> G["Oberlies 1998"]
    D --> H["Converse 1971"]
    E --> I["Sayana, Wilson 1866"]
    C --> J["Brough 1971: liturgical style"]
    D --> J

The fourth position on that map deserves its own paragraph, because it is the most deflationary and the hardest to dismiss. John Brough, reviewing R. Gordon Wasson’s theory that soma was the fly-agaric mushroom, took RV 8.48.3 as his test case for whether the Rigveda records drug experience at all.[10] His answer was no. The poet speaks in exalted terms, Brough observed, but exalted language is normal in liturgical utterance, and we can hardly suppose that every soaring line in the Rigveda was composed under intoxication. Gods are invoked for similar blessings all over the collection by poets who were plainly sober. “The point need not be laboured,” he wrote, and moved on.

Aside. A widely repeated claim needs correcting here. Wasson’s Soma: Divine Mushroom of Immortality (1968) is often described as resting on RV 8.48.3. It does not. The title phrase comes from the Chinese ling chih, not from the Rigveda, and Wasson’s own citation of this hymn is to verse 6, not verse 3, in support of a soma-and-fire association. His urine argument rests on RV 9.74. The verse became famous through the debate about Wasson rather than through Wasson himself. On the botanical question and where it now stands, see the dedicated piece on the soma problem.

Four readings, then. The honest summary is that the verse underdetermines all of them.

What the rest of the hymn is actually about

Here is the part that changed my reading of the verse, and it comes from simply continuing past it.

If verse 3 were the report of a heavenly ascent, you would expect the hymn to develop that. It does not. What the hymn develops, with a persistence that becomes almost comic once you notice it, is the body. Not the soul, not the world of the fathers, not the ancestors. Joints. Feet. Limping.

“These glorious (drops), when drunk, seek wide space. As cows [= leather straps] do a chariot, it knots (me) together in my joints. / Let the drops guard me from my foot slipping, and let them keep me away from lameness.”

(RV 8.48.5, after Jamison and Brereton)

That is the widest metrical stanza in the hymn, and it is about not falling over. The image is precise and unromantic: as wet leather thongs shrink and bind a chariot’s frame tight at every joint, the soma binds the drinker’s joints. Then verse 11:

“These famines and diseases have gone off. Those allied to darkness have shied away; they have become afraid. / Soma has mounted us to his full extent. We have gone to where they lengthen lifetime.”

(RV 8.48.11, after Jamison and Brereton)

Note where the drinker says he has gone. Not to heaven. To where they lengthen lifetime. And in verses 4 and 10, twice, the actual petition surfaces, and it is the same petition both times: pra ṇa āyur jīvase soma tārīḥ, “lengthen our lifetime, for us to live, Soma.” Jamison and Brereton track this as the hymn’s governing wish, appearing in verses 4, 7, 10 and 11, and draw the conclusion I think is correct: “the emphasis throughout the hymn on lengthening our lifetimes (on earth) may provide a counterpoint to the ritually shaped afterlife inhabited by our ancestors.”[1]

So the structure of RV 8.48 is this. A man drinks. He announces that he has become deathless. And then, for twelve more stanzas, he asks not to limp, not to starve, not to fall ill, and to live longer. The most exalted claim in the Rigveda is embedded in a hymn about staying healthy.

I do not think that is bathos. I think it is the actual content of the claim. Amṛta means not-dead. The man is, at this moment, extremely not-dead: his joints are tight, his diseases have fled, his enemies cannot touch him, he is full of light and in the company of the gods at his own fire. He is as alive as a person gets. That state has a name in his language, and the name is deathlessness, and he uses it in the past tense because it has already arrived.

The other pole

To see how unusual 8.48 is, put it next to the one hymn in the Rigveda that genuinely does describe a paradise. RV 9.113, attributed to Kaśyapa Mārīca, ends with five stanzas of pure otherworldly wishing, each closing on the refrain indrāyendo pari srava, “flow, Indu, flow for Indra.”

“Make me immortal in that realm where they move even as they list, / In the third sphere of inmost heaven where lucid worlds are full of light. Flow, Indu, flow for Indra’s sake.

Make me immortal in that realm where happiness and transports, where / Joys and felicities combine, and longing wishes are fulfilled. Flow, Indu, flow for Indra’s sake.”

(RV 9.113.9 and 9.113.11, Griffith 1896)

That is a different religion from RV 8.48, or at least a different moment in one. There is a place, it is elsewhere, it is up, it is permanent, and the speaker is not there. The Sanskrit gives a nice structural detail that most readers miss: verse 7 says tasmin māṃ dhehi, “place me in that undying world,” while verses 8 through 11 switch to tatra mām amṛtaṃ kṛdhi, “there make me immortal.” The petition escalates from relocation to transformation partway through.

[!NOTE] The line ānandāś ca modāś ca mudaḥ pramuda āsate in RV 9.113.11 packs ānanda, mud and pramud into a single half-line. It is probably the densest statement of beatitude in the Rigveda, and it is the earliest plural, paradisiacal use of ānanda, a word with an enormous future ahead of it in Indian philosophy.

Both hymns are addressed to soma. They want opposite things from it. That is not a contradiction to be resolved; it is evidence that the Rigveda is a collection assembled across centuries, and that the wish for another world arrives in it late and unevenly. The tenth book, latest of the ten, carries the heaviest concentration of amṛta vocabulary. The apparatus of a real afterlife, with judgment and a road and a destination, gets built afterwards, in the funeral hymns and then in the Brāhmaṇas and Upaniṣads.

Reading it now

We are, at the moment, living through an unusually loud conversation about extending human life: biological age clocks, senolytics, cryonics, the persistent fantasy of uploading. Almost all of it optimizes a single variable, duration. More years. The implicit model is that death is a defect and life is the interval before it.

RV 8.48 optimizes something else, and I think this is why it survives being read three thousand years later. Its poet also wants more years; he says so four times. But when he reaches for the word deathless, he does not apply it to a future state of very long duration. He applies it to right now, to a body that has stopped hurting, in a place where the gods have shown up, at an hour when nothing hostile can reach him. He is not describing the absence of death at the end of a long line. He is describing the total absence of death from a present moment. Those are different quantities. One is measured in years and the other is not measured at all.

Three cautions before anyone builds a philosophy on this. First, we do not know what he drank, and the modern soma ritual uses an acknowledged substitute. Second, the poets were professionals paid to say elevated things, and Brough’s warning about liturgical register is not easily dismissed. Third, the Rigveda nowhere systematizes any of this; there is no Vedic doctrine of immortality, only Vedic poems that use the word.

But the sentence is there, in the aorist, in the first person plural. Someone in the Punjab in roughly the twelfth century BCE said we have become immortal and meant something by it, and the most probable something is not a promise about the future. It is a report about an evening.

Read the hymn straight through, all fifteen verses, without stopping at verse 3. The famous line comes early, and everyone quotes it alone, which is exactly why it gets misread. Its meaning is downstream, in the stanza about the leather straps and the joints and the foot that must not slip. For the wider question of how the Rigveda handles what it cannot answer, the hymn to the god named Who is the natural next stop; for the long afterlife of these ideas, see how a Rigvedic riddle grew into Advaita Vedānta.