Man Overboard: The Rig Veda's Shipwreck and the Sea That Should Not Be There

Nothing to Hold On To

The poet needs three words to describe the place where the man is drowning, and all three are negatives. Anārambhaṇe: with nothing to grab. Anāsthāne: with nowhere to stand. Agrabhaṇe: with no hold, no purchase, no handle on anything. Only after emptying the scene of every solid thing does he name it: samudre, in the sea.

The man is Bhujyu, son of Tugra. In the standard telling he is abandoned in deep water, out of sight of land, and the Aśvins, the divine twins, come for him. What they come in changes every time the story is told. Sometimes it is a flying ship that swims through the air and never gets wet. Sometimes it is four boats. Sometimes birds. Once, memorably, it is a vessel with a hundred oars.

Here is the difficulty. The people who composed these hymns lived in the Punjab and the upper Indus plain, a landscape of rivers, cattle, horses and dust. The nearest salt water was several hundred kilometres away, across country that nobody in the text seems to have visited. They left no sailing directions, named no port, and had no word for tide or for salt. And yet a word normally translated “sea” turns up in one hundred and fifty-nine verses, and a poet who has in all likelihood never seen a coastline reaches for a shipwreck when he wants an image for being past all help.

That gap is the subject here. Not “did the Vedic people sail”, which is the version of the question that has been fought over for a century and a half for reasons that have little to do with the Rig Veda, but something more interesting: what is actually in the text, what a careful count of it looks like, and what a very large body of water was doing in the imagination of a landlocked river culture.

The Most Retold Rescue in the Collection

Fourteen hymns spread across seven of the ten maṇḍalas return to this man in the water. No other Aśvin rescue is repeated so often, and none of the great Indra myths gets retold with so little agreement about the physical facts. Read the versions side by side and the effect is like listening to a story that has passed through a dozen tellers, each of whom kept the emotional core and improvised the machinery.

Two things fall out of this table. The first is that the vehicle is unstable and the predicament is not. Ships, birds, chariots, a raft, a tree: the poets are plainly not working from a fixed account of a real voyage. They are working from a fixed situation, a man in deep water with no bottom under him, and reaching for whatever conveyance the metre and the moment allow. In an oral tradition this is exactly what you would expect of a story that is old, universally known, and no longer attached to anything anyone can check.

The second is RV 7.68.7, the only version that supplies a motive:

And that Bhujyu, O Aśvins, whom his comrades, evil in their ways, abandoned in the middle of the sea: he was carried out, the devoted one who was yours.

(RV 7.68.7, my rendering, following the Sanskrit closely)

Madhye samudre, in the middle of the sea, put there by his own sakhāyaḥ, his companions, described as durevāsaḥ, of evil ways. Every other telling is about miraculous transport. This one is about betrayal. Whatever the historical kernel, someone decided the interesting part was that a man’s own crew left him to die.

And then there is the riddle, which is the strangest verse in the cycle:

What tree was that which stood fixed in surrounding sea to which the son of Tugra supplicating clung? Like twigs, of which some winged creature may take hold, ye, Aśvins, bore him off safely to your renown.

(RV 1.182.7, trans. Griffith 1896)

A tree standing in the middle of open water, which a drowning man holds on to. The poet asks the question and does not answer it. Sāyaṇa, commenting two millennia later, has to reach for allegory. It reads less like a memory of the sea than like a dream about one.

Aside. The Bhujyu story has a respectable Indo-European pedigree. The Greek Dioskouroi rescue sailors, and twin saviours who arrive when a vessel is foundering turn up across the family. This blog has treated the Aśvins and their Indo-European parallels at length. What matters for the present question is the direction of inference: an inherited rescue-at-sea motif is evidence about the ancestors of the language, not about the geography of the poets who kept it alive.

How Much Sea Is Actually in There?

Since the argument is usually conducted with numbers, it is worth generating some honestly. I counted over the romanised Saṃhitā text of all ten maṇḍalas, looking for any token containing the string samudr.

The result is 159 verses out of 10,524. That is higher than the figure of 133 often quoted, because it includes compounds and derived forms such as samudriya, “of the sea”, and samudra-vāsas, “sea-clothed”, which the lower count leaves out. Neither number is wrong. They answer slightly different questions.

Methods note. Counts like these are approximate by nature. The transmitted text runs words together under sandhi, so samudrāduta and sisicur utsamudriṇam are single tokens on the printed line even though each contains more than one word. String matching catches those but also risks the occasional false positive, and it cannot by itself distinguish the sea from the soma vat. Treat every count in this article as a good estimate rather than a census, and treat any argument that turns on the difference between 133 and 159 as an argument not worth having.

More revealing than the total is the distribution. Normalise by the number of verses in each maṇḍala and one book stands out sharply.

Maṇḍala 9 carries roughly twice the density of any other book. Maṇḍala 9 is also the one book with almost no narrative, no battles and no geography: a thousand-odd verses addressed to Soma Pavamāna, soma in the act of being pressed, filtered and poured. Its “sea” is a wooden tub.

These somas, past the fleece of wool, like heavenly buckets raining from the cloud, have run down to the pitchers as rivers run down to the sea.

(RV 9.88.6, my rendering)

That verse is doing two jobs at once, and both matter. Ritually, the vat into which the filtered juice pours is a samudra. Descriptively, the poet reaches for rivers running down to a sea as the obvious image for many streams converging into one large body. He would not use the comparison if his audience found it exotic. Neither would he use it if he expected them to picture a particular coastline.

So the frequency data cuts both ways, which is the honest summary. The single largest concentration of “sea” in the Rig Veda is metaphorical plumbing. What remains once you set Maṇḍala 9 aside is still a word used steadily, and without any air of novelty, across every family book. Anyone who quotes the raw count as proof of maritime experience has not looked at where the count comes from. Anyone who dismisses the word as pure metaphor has not explained why rivers keep flowing into it.

What a Rig Vedic Ship Is Made Of

Here the evidence gets thin in a way that is genuinely informative. Run the same kind of search across the vocabulary a seafaring culture would need, and this is the yield.

No mast. No sail. No rigging, anchor, harbour, hull or keel. No tide, no swell, no salt, no landfall. The Rig Veda knows the far shore, pāra, because rivers have far shores, and it knows the ford, tīrtha, because you cross rivers at fords. The one piece of nautical equipment it names is the oar, and it names it four times.

Look at how those oars are used and the picture sharpens. RV 1.140.12 asks Agni for a boat “with its own oars and with feet”, a vessel that walks, which is not a description of anything that floats. RV 2.18.1 describes a chariot as daśāritra, “ten-oared”. The poets are not distinguishing carefully between things that carry you across land and things that carry you across water. They have one vocabulary of conveyance and they apply it wherever a crossing is happening.

The most technically specific boat in the whole collection is not on any sea:

Well-guarding Earth, faultless Heaven, sheltering Aditi of good guidance, the divine ship, well-oared, blameless, that does not leak: may we climb aboard, for well-being.

(RV 10.63.10, my rendering)

Asravantīm: not leaking, not letting the water in. Whoever wrote that line had been in a boat that did. It is a small, unglamorous, entirely convincing detail, and it points at exactly the vessel Konrad Klaus argued for in the standard monograph on Vedic watercraft: a modest rowed or punted river craft for crossing the many rivers of the Punjab, not a ship for open water.[6]

The Argument, and Why It Is Not Really About Boats

The dispute over samudra is older than most of the disciplines that now conduct it. Heinrich Zimmer, in Altindisches Leben (1879), concluded that the Vedic Indians did not know the ocean and that samudra meant the broadened lower course of the Indus after it had gathered the Punjab rivers.[5] Arthur Berriedale Keith took a similar line. On the other side, Radhakumud Mookerji opened his History of Indian Shipping (1912) with the hundred-oared ship of RV 1.116.5 and built an ancient maritime tradition on top of it.[7]

The positions have hardened since, and it is worth saying plainly why. If the Rig Veda describes a maritime, urban, seafaring society, it becomes easier to identify its authors with the Indus civilisation and to argue that Indo-Aryan speakers were indigenous to the subcontinent. If it describes an inland cattle-herding society that had heard of a big water somewhere downstream, that identification fails. The verses became counters in an argument about origins, and both sides have quoted them selectively ever since.

Modern translators have quietly settled the matter of the word without settling the matter of the world. Stephanie Jamison and Joel Brereton render samudra as “sea” throughout their 2014 translation, because that is what the word means; whether any given poet had stood next to one is a separate question their translation does not pretend to answer.[1]

graph TD
  A["A verse says samudra"] --> B["Ocean"]
  A --> C["Lower Indus or confluence"]
  A --> D["Soma vat"]
  A --> E["Waters of the sky"]
  B --> F["Poets saw a coast"]
  C --> G["Poets saw big rivers"]
  D --> H["Ritual image only"]
  E --> H
  F --> I["Needs independent detail"]
  G --> I
  H --> I
  I --> J["That detail is missing"]

What can actually be concluded is narrower than either camp would like, and it comes to three things. The poets had a word for a very large water and used it without explanation, so their audience knew what it meant. They put it at the downstream end of rivers: RV 7.49 calls the waters samudrārthāḥ, “having the sea as their goal”, and RV 7.95.2 has the Sarasvatī going “from the mountains to the sea”. And they never once describe anything you could only know by being on one. That third observation is mine rather than a consensus finding, but it is easy to test: read every samudra verse and look for a single sensory fact about salt water. There isn’t one.

Aside. The Rig Vedic sea is also, repeatedly, not on the earth at all. RV 9.33.6 asks Soma to pour out “four seas” full of wealth. RV 4.58.5 describes streams of ghee flowing “from the ocean of the heart”. RV 10.136.5 says the long-haired ascetic dwells “in both seas, the eastern and the western”. These are not coastal reports. They are what a culture does with a word for immensity once it has one.

The Merchant Who Appears Twice

If the Rig Vedic people were running sea trade, the economy in the hymns should show it. It does not. The word vaṇij, merchant, occurs exactly twice in 10,524 verses.

Once at RV 1.112.11, where the Aśvins pour out honey for “Dīrghaśravas the merchant, son of Uśij”, a man in some kind of distress whose story we do not otherwise have. Once at RV 5.45.6, in a dense and difficult verse about a vaṇig vaṅkur, a “roving” or “winding” trader, who obtained water. Two occurrences, both oblique, neither describing a transaction.

Against that, the Rig Veda has an entire genre devoted to the movement of wealth: the dānastuti, the praise of the gift, in which a poet itemises what a patron handed over. Cattle, horses, gold, chariots, servants. Nothing is bought. Wealth moves by raiding, by herding and above all by gift and counter-gift between poet and patron. This is not a commercial society writing about itself.

The strongest single verse on the other side is a simile, and it deserves a fair hearing:

To him the guidance-following songs of praise flow full, as those who seek gain go in company to the flood.

(RV 1.56.2, trans. Griffith 1896)

The Sanskrit is samudraṃ na saṃcaraṇe saniṣyavaḥ, “like gain-seekers travelling together to the sea”. A poet reached for people who band into a convoy and head for a large water in hope of profit, and expected his listeners to recognise the picture at once. That is a real datum, and more suggestive than any number of raw samudra counts. It is also a single half-line, and what it establishes is that the image was available, not that the audience had made the voyage.

Meanwhile, on the Other Side of a Gap

There is no doubt that seafaring existed in the subcontinent. It may simply not have existed at the same time as these hymns.

Mesopotamian records from roughly 2350 BCE onward name Meluhha, almost certainly the Indus region, among the lands whose ships tied up at Akkadian quays. Carnelian, lapis, timber and ivory moved up the Gulf through Dilmun and Magan. There was a Meluhhan settlement in Mesopotamia and at least one official whose job title was interpreter of the Meluhhan language. Shereen Ratnagar’s work on that trade remains the standard treatment.[9]

Indus boats are attested visually as well as commercially: a steatite seal and a terracotta tablet from Mohenjo-daro, a graffito on a potsherd, a terracotta boat model from Lothal. The seal excavated by George Dales in 1964 shows a high-prowed craft with a central cabin and, apparently, birds carried aboard for finding land, a practice known from several ancient traditions.

Lothal itself is contested. S. R. Rao, who excavated it, identified a large brick-lined basin as a tidal dockyard, which made Lothal a Harappan port town.[8] Lawrence Leshnik argued in American Anthropologist in 1968 that the basin was a reservoir for drinking water and irrigation, and that its inlet was too shallow for sea-going hulls.[10] The debate has never fully closed, which is itself instructive about how much interpretive weight a single structure can carry.

Now put the two datasets next to each other. Harappan maritime contact peaks in the middle centuries of the third millennium BCE and is largely finished by about 1900 BCE. The Rig Veda, on the linguistic and archaeological evidence, belongs somewhere in the range of roughly 1400 to 1000 BCE, and its geography is the Punjab and the upper Indo-Gangetic plain, several hundred kilometres inland. Between the last Meluhhan ship and the first Rig Vedic hymn there is a gap of centuries and a distance of provinces.

The gap does not make transmission impossible. Words outlive their referents all the time, and a term for “great water” could easily have travelled inland with people who had never applied it to the thing it originally named. But it does mean that pointing at Lothal to explain RV 1.116.5 is not an argument. It is two facts placed near each other in the hope that they will bond.

The Ship Becomes an Idea

Follow the boat vocabulary away from the Bhujyu story and something more interesting happens than a maritime debate. The ship stops being transport and turns into the central image for getting through.

The ritual itself is a vessel. RV 10.44 ends with a hard verse about the people who could not get aboard: those “who could not ascend the ship of sacrifice sink down in desolation”. In RV 10.101, a hymn otherwise concerned with yoking ploughs and sinking wells, the poet turns to his colleagues and says, in exactly the same practical register: build a ship fitted with oars for crossing, prepare the implements, let the sacrifice go forward. The hymn and the boat are the same kind of object, something you construct carefully so that it will carry weight.

When the poets ask for help, they ask to be ferried. RV 2.39.4 asks the Aśvins to “bear us across like two boats”. RV 1.97.7 asks to be carried past enemies “as in a ship”. RV 10.63.10, the well-oared vessel that does not leak, is a prayer to board the divine ship and be taken over safely.

The counter-image to Bhujyu is the loveliest water passage in the collection, and there is nothing frightening in it at all:

When Varuṇa and I embark together and urge our boat into the midst of ocean, We, when we ride o’er ridges of the waters, will swing within that swing and there be happy.

(RV 7.88.3, trans. Griffith 1896)

Vasiṣṭha and his god take a boat out onto open water and rock on the swell, and the poet compares it to a child’s swing. The next verse says Varuṇa “placed Vasiṣṭha in the boat” and made him a seer. Whatever the geography, the psychology is exact: the same element that drowns Bhujyu carries Vasiṣṭha, and the difference lies entirely in whose company you are keeping. This is the mind that would later produce the whole Indian vocabulary of crossing: the tīrtha that is both a river ford and a place of pilgrimage, the raft you abandon once you reach the far bank, the ocean of existence that somebody ferries you over.

What the Sea Was For

Set the counting aside and the pattern is consistent. The Rig Veda’s boats are small, rowed and prone to leaking. Its named ship hardware amounts to an oar. Its merchants can be counted on one hand with fingers to spare. Its geography, wherever it becomes specific, is rivers.

And yet the word for a great water runs through all ten books, sits at the downstream end of every river the poets know, holds the pressed soma, fills the sky, appears in pairs at the eastern and western ends of the world, and opens under a man who has been left behind by his friends. The sea in this text is not a place the poets went. It is the category they kept for whatever was too large to be crossed on foot.

Which is why the hundred-oared ship is the wrong verse to argue about. The line that stays with you is the one before it, the triple negative: no hold, no standing, nothing to grasp. Three thousand years later that is still an unimprovable description of being out of your depth, and it was produced by people who, on the best available evidence, had never seen the ocean and did not need to.

Open RV 1.182 and read the last three verses aloud. Four boats in the dark, a tree standing in open water, and a question the poet asks and then refuses to answer. That is not a report from a coastline. It is something better: the record of an old story that a landlocked people would not let go of, because they knew perfectly well what it was about.