The Gods Who Came to Collect: Debt in the Rig Veda

A Sinner, or a Man Who Owes Money

In the seventh verse of RV 1.169, the Maruts arrive. They are the storm troop, they come with a noise the poet can hear before he can see them, and they fall on a hostile mortal with a violence the verse compares to something from ordinary life. Ralph Griffith, translating in 1896, rendered the comparison this way:

“Heard is the roar of the advancing Maruts, terrific, glittering, and swiftly moving, who with their rush o’erthrow as ‘twere a sinner the mortal who would fight with those who love him.”

(RV 1.169.7, Griffith 1896) [1]

The Sanskrit word behind “a sinner” is ṛṇāvan. It does not mean sinner. It means a man who owes money.

Ṛṇa (Sanskrit: ऋण) is debt. Not metaphorical debt, not a stain on the soul, but the thing you have when you have taken twenty cows and not yet given back twenty cows. Ṛṇāvan is the possessive adjective: debt-having, in arrears, a debtor. The same word turns up in the Gambler’s Lament, where Griffith translates it correctly, because in a poem about a man ruined by dice there is no theology to protect: “in constant fear, in debt, and seeking riches, he goes by night unto the home of others” [1]. Same word, same translator, a lifetime of Sanskrit behind him. In the hymn about dice it is debt. In the hymn about gods it is sin.

That inconsistency is not a scandal, and Griffith is not the villain of this piece. It is a symptom of something more interesting: by the time Europeans began translating the Rig Veda, Indian religious literature had spent two and a half millennia building an enormous conceptual structure on top of the word ṛṇa, a structure in which every human being is born owing three debts that can never quite be settled. Read the Rig Veda through that structure and ṛṇa is obviously spiritual. Read the Rig Veda without it and ṛṇa is obviously money.

The word occurs in twenty verses. Laid out together, and read without the later apparatus, they are not a theology of debt at all. They are something rarer in the Rigvedic corpus: a glimpse of the ordinary economic fear of people who had no coins, no writing, no contracts, and no way to prove what anybody owed anybody, and who were frightened of creditors anyway.

Twenty Verses and a Homograph Problem

Counting ṛṇa in the Rig Veda is harder than it sounds. Vedic Sanskrit has a verb ṛṇóti, “he sets in motion, he opens,” from the root ṛ, which has nothing to do with debt and generates dozens of forms beginning with the same two syllables: Agni ṛṇvati the doors of heaven; Uṣas ṛṇavo the gates of the sky. A naive search for ṛṇ returns more than a thousand hits, almost all of them noise from unrelated roots (kṛṇoti, śṛṇoti, pṛṇāti, vṛṇakti) that happen to contain the sequence.

Filtered down, here is the whole inventory.

Methods note. These counts come from a string search over the romanised Saṃhitā text, checked verse by verse against the Devanagari and against Griffith’s translation, and they are counts of words, not of ideas. One further passage, RV 6.12.5, contains a form ṛṇo that Griffith took as debt (“swift as a guilty thief”); the analysis is contested and I have left it out of the twenty. Sandhi can also swallow a word into its neighbour, which is how ṛṇāvānam at 1.169.7 hides inside ūmair-ṛṇāvānaṃ, so treat twenty as a floor rather than a certainty.

Two things jump out of the table. The first is how small the vocabulary is: one noun, two or three derived epithets, and a verb of collecting. The second is who the epithets belong to. Five of the strongest powers in the pantheon are described, in passing and without any sense that this is odd, as agents who come after arrears.

What the Word Is Made Of

The dictionaries derive ṛṇa- from the root ṛ / ar, “to move, to set going,” with the ordinary Sanskrit past participle suffix -na (Monier-Williams, s.v.) [2]. A debt would then be, etymologically, “a thing set in motion”: something that has gone out from one hand and has not yet come back. I find that appealing and I would not press it, because participial etymologies of abstract nouns are easy to construct after the fact and hard to falsify.

What is harder to dismiss is the family resemblance to ṛta, cosmic order, truth, the way things fit. Ṛta is also standardly taken as a participle of the same root, formed with -ta instead of -na. If both derivations hold, then the Vedic words for “the order of the world” and “what you owe the man next door” are the same verb wearing two different suffixes. That is suggestive rather than probative, and Vedic scholarship has long been wary of etymological arguments that arrive at conclusions this tidy. Still, the two words do keep company. At RV 2.27.4 the Ādityas are ṛtāvānaḥ, possessed of ṛta, in the same breath in which they are cayamānā ṛṇāni, exacting debts. Order and accounts sit in one line of verse. Readers following the argument of Varuṇa and the bind of cosmic guilt will recognise the territory.

The derived words are precise in a way that tells you the underlying practice was real. Ṛṇayā is “debt-going,” one who goes after debts. Ṛṇa-cit uses the root ci, which means both “to notice” and “to requite,” so the epithet is close to “he who keeps the count.” Ṛṇaṃ-caya is “debt-gathering.” Ṛṇa-cyut is “debt-shaking,” said of a son, a man who shakes a debt off. Languages do not develop four distinct agent-nouns for an activity nobody performs.

The clinching evidence that this vocabulary is financial rather than moral is RV 8.47.17, where a bad dream is disposed of by an accounting procedure:

yathā kalāṃ yathā śaphaṃ yathā ṛṇaṃ saṃnayāmasi

“As we gather up a sixteenth, as we gather up an eighth, as we gather up a debt, so we transfer the whole bad dream onto Āptya.”

RV 8.47.17. After Griffith (1896), with the fractions given their standard values [1].

Kalā is a sixteenth part, śapha is a “hoof” and conventionally an eighth. The poet is not reaching for a spiritual analogy. He is reaching for the most familiar arithmetic he has, the summing of fractional payments into a single settled amount, and using it to move a nightmare off his own books. The full machinery of that hymn is the subject of the Rig Veda’s bad dream problem; what matters here is that the poets thought in part-payments, and that the counting apparatus behind them is the same one discussed in the Rig Veda’s arithmetic of 3,339 gods.

The Fear of Another Man’s Debt

Now the two verses that made me want to write this piece.

In RV 2.28, one of the great Varuṇa hymns, the poet works through a sequence of requests: loosen the bond, do not cut the thread of my song, do not strike me with the weapon that wounds the wrongdoer, protect me from the wolf and the thief who come at night. In the middle of that sequence, verse 9:

para ṛṇā sāvīr adha matkṛtāni māhaṃ rājann anyakṛtena bhojam

“Drive away the debts I have made myself; and let me not, O king, be made to pay for what another has made.”

RV 2.28.9, my rendering from the Sanskrit. Griffith (1896) has “Move far from me what sins I have committed: let me not suffer, King, for guilt of others” [1].

Read Griffith’s version and you have a familiar religious sentiment, the plea of a guilty conscience, at home in any hymnal. Read the Sanskrit and you have something much more specific and much stranger: a man asking a god to clear his account, and then, urgently, asking not to be held liable for somebody else’s obligation. Anyakṛta, “made by another.” That is not a theological worry. That is the worry of a man in a small kin-based society where liability ran along family lines and a creditor who could not find his debtor might reasonably come looking for the debtor’s brother.

We know it is not an isolated formula, because the same fear appears in a hymn to a different god, in a different family’s collection. RV 4.3.13 asks Agni not to go to the feast of a treacherous neighbour or an unworthy relative, and then:

mā bhrātur agne anṛjor ṛṇaṃ ver

“Do not, Agni, pursue the debt of a crooked brother.”

RV 4.3.13, my rendering. Compare Griffith (1896): “Punish us not for a false brother’s trespass” [1].

Two hymns, two gods, one anxiety. In each case the god is being asked to behave like a fair creditor: collect from the man who borrowed, and leave his relatives alone.

What makes this worth pausing over is what Indian law eventually did with exactly this question. The dharmaśāstra literature, a thousand years and more downstream, settles it in the opposite direction. The son inherits the father’s debt along with the father’s property, and discharging it becomes a religious duty rather than merely a legal one. The Rigvedic poet is praying against the rule that his descendants would codify. Whether that represents a real historical shift in liability or merely the difference between a hymn and a lawbook, I cannot say; but the hymn is evidence that the question was live.

Aside. There is a strong temptation, reading 2.28.9, to hear a doctrine of inherited sin, a Vedic ancestor of original guilt. Resist it for one verse longer. The poet says anyakṛta, “done by another,” not pūrvakṛta, “done before,” and the hymn’s other requests are all about present dangers: wolves, thieves, nightmares, weapons. The most economical reading is the literal one. Whether the Rigvedic poets also used ṛṇa figuratively, for moral obligation, is a genuinely open question; my point is only that the literal sense is primary and the figurative one has to be argued for, not assumed.

Five Gods With a Ledger

Here is the pattern the concordance exposes. The epithet ṛṇayā, debt-avenger, is not a minor or localised usage. It is spread across four maṇḍalas and attached to some of the most powerful figures in the pantheon.

Notice what this is not. It is not a doctrine that sin is a debt to the gods. In every one of these verses the god is the collector and somebody else, usually an enemy, is the debtor, and the relation is an image of force rather than a theory of morality. When the Rigvedic poets want to say that a god is formidable, one of the pictures they reach for is a man walking up your path to ask about the cattle you borrowed last spring.

Griffith’s translations of this cluster are worth putting side by side, because they show a nineteenth-century sensibility making a consistent choice.

The rule is almost mechanical. Where the surrounding verse is about enemies or property, Griffith writes debt. Where it is about gods and human standing before them, he writes sin or guilt. He was working within a long tradition: Sāyaṇa, the fourteenth-century commentator whose readings shaped every early European translation, glosses ṛṇa as pāpa, sin, in several of these places. The Sanskrit commentators were themselves reading the Rig Veda through the theology that the Brāhmaṇas had built, which is exactly the problem.

One more example shows how far the drift can go. At RV 2.24.13 Brahmaṇaspati is ṛṇam ādadiḥ. Griffith translates “remitting at his will the debt,” which makes the god a forgiver of loans. But ādadi- is built on ā-dā, to take or seize, not on dā, to give. The god is taking the debt at will, not waiving it. A single preverb turns a merciful deity into a collections agent, and the mercy is the translator’s.

Aside. I am not accusing Griffith of piety in bad faith. Anyone who translates four hundred pages of Vedic Sanskrit will make choices under the pressure of context, and “sin” was the available English word for a category that Victorian religious vocabulary had. The point is methodological rather than moral: when a translation is consistent in its inconsistencies, the pattern is usually telling you about the translator’s world rather than the text’s. Stephanie Jamison and Joel Brereton’s 2014 translation, working from a different set of assumptions, generally keeps the accounting language in place [3].

Settled Accounts

If the gods are creditors, the Rigvedic ideal turns out to be a cleared balance. This is the argument’s least obvious move and its most interesting one.

na nūnaṃ brahmaṇāṃ ṛṇaṃ prāśūnām asti sunvatām | na somo apratā pape

“There is now no debt outstanding from the formulators, the vigorous ones, who press soma. Nor has the soma been drunk without return.”

RV 8.32.16, my rendering; compare Griffith’s “No debt is due by Brahmans now, by active men who press the juice” [1]. The epithet prāśū is unclear: Griffith takes it as “swift, active,” others as “having eaten.”

Apratā means “without a counter-gift,” without requital. The verse is a statement that the books balance: the priests have consumed, the god has been given, nothing is outstanding on either side. The whole Rigvedic economy of sacrifice, the one traced in detail through the deified priestly fee Dakṣiṇā, runs on this principle. You give, the god gives back, and the aim is a closed transaction rather than an open-ended obligation.

The same instinct shows up in an unexpected place. At the end of the hymn to Night, RV 10.127.7, the poet asks the darkness to be settled up:

upa mā pepiśat tamaḥ kṛṣṇaṃ vyaktam asthita | uṣa ṛṇeva yātaya

“The dark has come close and settled on me, black, distinct. O Dawn, clear it away as one clears debts.”

RV 10.127.7, my rendering. Griffith has “O Morning, cancel it like debts” [1].

Night is a liability. Dawn is the settlement. The metaphor only works in a culture where debts routinely get cleared, which tells us something about how normal the transaction was.

And the negative case is just as revealing. The Rig Veda’s stock villains include the Paṇis, the figures who hoard cattle in a cave and will not release them, and whose name is traditionally connected with the verb paṇ, to bargain or haggle. In RV 1.124.10 and 4.51.3 the Dawn is asked to wake the generous and let the Paṇis sleep on. At RV 6.47.21 the enemy is described with the participle vasnayantaḥ, “haggling over the price,” from vasna, price. To the Rigvedic poets, the man who drives a hard bargain is not a shrewd businessman. He is on the same moral shelf as the demon.

The Only Word for Usurer, and Nobody Knows What It Means

If people were lending, somebody was lending at a profit. The Rig Veda has exactly one word that might name him, and it appears once.

indro viśvān bekanāṭān ahardṛśa uta kratvā paṇīn abhi

“Indra overcomes all the bekanāṭas who see only the day, and by his will the Paṇis too.”

RV 8.66.10, my rendering; Griffith has “Indra excels all usurers who see the day” [1].

Bekanāṭa is a hapax legomenon, a word that occurs once in the entire corpus. Sāyaṇa glossed it as kusīdin, moneylender, and Macdonell and Keith concluded in the Vedic Index that “usurer” is its natural sense, largely on the strength of the company it keeps: it stands in parallel with paṇi in the same verse [4]. That is an inference from context, and it is all we have. The word has no convincing Indo-European etymology, and it looks, to several scholars who have examined the Rig Veda’s stock of unanalysable vocabulary, like a loan from a non-Indo-Aryan language of the region, one of the 383 items F. B. J. Kuiper judged non-Indo-Aryan, some four per cent of the Rig Veda’s vocabulary [15]; the general problem is the subject of the Rig Veda’s foreign words.

Put that beside the rest of the evidence and a curious picture assembles:

• No word for interest. Kusīda, the standard later Sanskrit term for an interest-bearing loan, does not occur in the Rig Veda at all.
• No coinage. The niṣka of RV 2.33.10 and 5.19.3 is a gold ornament worn at the neck, a store of value rather than a medium of exchange. Struck coin arrives in India the better part of a millennium later.
• No writing. Nothing to record a loan on, and nothing to sign.
• But four agent-nouns for collecting debts, a stock simile of the creditor falling on the debtor, and a verse begging not to be charged for a brother’s arrears.

This is a credit economy with no instruments whatsoever. Everything ran on memory, witness and reputation, in a society small enough that everyone knew who had borrowed what. The Rig Veda’s own institutions for this are visible elsewhere: RV 2.24.13, the verse about Brahmaṇaspati taking the debt, also calls him sabheya, “fit for the assembly,” which puts the debt language one step away from the deliberative bodies discussed in the sabhā and samiti. A debt with no paper is a debt adjudicated in public, in front of people who remember.

A King Called Debt-Collector

The strongest single piece of evidence that ṛṇa was ordinary vocabulary is a man’s name.

RV 5.30 ends with a dānastuti, a praise of gifts, in which the poet thanks his patrons for an enormous payment. The patron is a king of the Ruśama people, and his name is Ṛṇaṃcaya, “Debt-Collector.”

auchat sā rātrī paritakmyā yāṃ ṛṇaṃcaye rājani ruśamānām

“That night, the one at the turning point, dawned for Ṛṇaṃcaya, king of the Ruśamas.”

RV 5.30.14, my rendering. Griffith has “Night, well-nigh ended, at Rnancaya’s coming, King of the Rusamas, was changed to morning” [1].

Two verses earlier the poet records the fee: four thousand cattle [1]. A king named Debt-Collector, who gives away four thousand cows. The name is not ironic and not, as far as anything in the hymn suggests, remarkable to anyone. It is formed exactly like other Rigvedic royal names of the aggressive type, and it sits in the corpus’s onomastic record alongside all the other accidental social data surveyed in the Rig Veda’s census of personal names. Cultures name their children after things they admire. Somebody looked at a man who got his money back and thought: that is what I want my son to be.

The counterpart is RV 6.61.1, where the river goddess Sarasvatī gives the childless Vadhryaśva a son described as rabhasaṃ ṛṇacyutam, “impetuous, a shaker-off of debts.” This is a birth-gift, and the quality being promised is that the boy will clear the family’s obligations. More than a thousand years before the lawbooks make it a rule, the hope is already there in a single compound.

The Debt You Are Born With

So where does the famous doctrine come from, the one that says every person is born owing three debts?

Not from the Rig Veda. The twenty verses above are its complete testimony on ṛṇa, and none of them says anything of the kind. Every debt in the Rig Veda belongs to somebody in particular, was incurred at some particular time, and can in principle be discharged. The Rigvedic ambition is to get clear.

The doctrine appears a few centuries later, in the ritual prose. The classic formulation is in the Taittirīya Saṃhitā:

jāyamāno vai brāhmaṇas tribhir ṛṇavā jāyate

“Being born, a brahmin is born with three debts.”

Taittirīya Saṃhitā 6.3.10.5; the doctrine is developed further at Śatapatha Brāhmaṇa 1.7.2.1-6, which counts four [5].

The three are a debt of studentship to the seers, of sacrifice to the gods, and of offspring to the ancestors. The Śatapatha adds hospitality to fellow men. Patrick Olivelle’s history of the āśrama system treats this material under the heading “The Theology of Debts” and shows how it became the structural argument for the whole householder ideal: you cannot renounce the world until the debts are paid, and the debts are paid by living a full social life [6].

Notice the inversion. In the Rig Veda, debt is a contingent misfortune that a careful man avoids and a god may help him clear. In the Brāhmaṇas, debt is the condition of being born at all. It is no longer something you have; it is something you are. Charles Malamoud’s essay on this transformation, the best thing written on it, traces how the ritualists turned an economic relation into the grammar of the human situation [7]. Once that move is made, the whole later architecture follows: ṛṇa as a structural obligation, and eventually the karmic bookkeeping in which every act creates a balance to be settled, a development traced from its Rigvedic roots in karman before karma.

graph TD
    A["Rigveda: rna = a specific debt"] --> B["Epithets: gods as collectors"]
    A --> C["Fear: liability for another's debt"]
    A --> D["Ideal: cleared accounts, nothing owed"]
    B --> E["Brahmanas: born with three debts"]
    C --> E
    E --> F["Asrama theology: householder pays"]
    E --> G["Dharmasastra: law of loans"]
    G --> H["Manu: interest at 1/80 per month"]
    F --> I["Renunciation only after settlement"]
    D --> J["Karma: acts as a balance"]
    E --> J

The legal track runs in parallel and is startlingly concrete. By the time of the Mānava Dharmaśāstra, the rate is fixed: the moneylender may take monthly “the eightieth part of a hundred,” which is one and a quarter percent a month, fifteen percent a year, a rule the text itself attributes to Vasiṣṭha [8]. Interest is capped: it may not exceed the principal, so a debt may at most double [8]. The arithmetic is worth doing.

$$\frac{100}{15} \approx 6.7 \text{ years to reach the cap at simple interest}, \qquad \frac{\ln 2}{\ln(1.0125)} \approx 55.8 \text{ months if compounded}$$

Roughly four and a half to seven years, depending on how the rate was actually applied, and then the lender’s claim stops growing. That is a debt-forgiveness rule, written by people who had thought hard about what unbounded compounding does to a farming household. Between Ṛṇaṃcaya’s four thousand cattle and Manu’s eightieth part lies the entire development of an economy: coinage, cities, written law, and professional lenders.

Aside. David Graeber’s Debt: The First 5,000 Years brought this Vedic material to a large readership outside Indology, deploying the three-debts doctrine against “primordial debt” theories of money, which hold that we owe an unpayable debt to society and that taxation is its repayment [9]. One caution for readers who meet the Vedic evidence there first. The three debts belong to the ritual prose, not to the hymns, and between the two lies several centuries and a complete change of genre. Whatever the Brāhmaṇas are doing, the Rig Veda’s debts are not primordial, not cosmic and not unpayable. They are cattle.

What the Ledger Was For

Set the twenty verses side by side and what you have is not a doctrine but a habit of mind.

The Rigvedic poets were people for whom obligation was the most available image of seriousness. When they wanted to say that a god was not to be trifled with, they said he collects debts. When they wanted to say that darkness would certainly end, they said it gets cleared like a debt. When they wanted to dispose of a nightmare, they did the arithmetic of part-payments. When they were frightened, one of the things they were frightened of was being billed for a relative’s arrears. None of this is theology. It is the mental furniture of people who lived close enough together that what you owed was known, and far enough from any institution that nothing enforced it except reputation, kinship, and the conviction that the powers holding the world together are also, in some sense, in the collections business.

Three observations to take away. First, the word ṛṇa did not become spiritual by growing; it became spiritual by being generalised, and the generalising happened in the Brāhmaṇas, not the hymns. Second, the Rigvedic attitude to debt is the opposite of the later one: get clear, stay clear, and do not pay for anyone else. Third, and this is the part that lasts, the English language does exactly the same thing with the same word. We speak of paying a debt to society, of owing it to ourselves, of a debt of gratitude that can never be repaid. We moralised a bookkeeping term too, and we did it so thoroughly that we no longer hear the ledger underneath. Griffith’s “sinner” was not a mistake so much as an act of translation between two cultures that had each made the same metaphor and forgotten it.

A suggestion for reading. Open RV 2.28 and go through it twice, once with Griffith’s vocabulary of sin and guilt and once substituting “debt” wherever ṛṇa stands. The hymn changes character completely. In the first reading it is a penitential psalm. In the second it is a man in trouble asking a powerful figure to sort out his accounts and to leave his brother’s mess out of it, and then, in the very next verse, to keep the wolves off. It is much less exalted and far more human, and I am fairly confident it is what the poet meant.

For the wider setting, the companion pieces on the gambler ruined by dice, the assembly where such matters were argued, and the Rig Veda’s own unbelievers, who priced Indra at ten cows, fill in the economic world this vocabulary belonged to.