The Sky Was Made of Stone

A Treasure Hidden Inside Endless Rock

In the first book of the Rigveda, a poet describes Indra finding something valuable. The place he finds it is the interesting part.

avindad divo nihitaṃ guhā nidhiṃ ver na garbhaṃ parivītam aśmany anante antar aśmani

“He found the treasure of heaven, set down in hiding, wrapped like the embryo of a bird in the stone, inside the endless stone.”

RV 1.130.3, rendering mine from the Saṃhitā text.

Read it slowly. The treasure is divo, of heaven, deposited up there. And it is sealed inside áśman, stone, which the poet then specifies twice over, as though the word alone were not emphatic enough: aśmany anante antar aśmani, in endless stone, within the stone.

A modern reader reaches for metaphor. The poets mean clouds, surely, and a cloud is called a rock the way a beloved is called a rose. But when you collect every passage in which this particular word appears, the metaphor stops behaving like decoration. The sun turns out to be a stone, set in mid-heaven. Thunder turns out to be a stone that rolls and resounds. Indra’s weapon turns out to be a piece of rock thrown down from above, in one verse a piece of metal rock. Fire lives inside stone, which anyone with a flint already knew, so lightning needs no further explanation.

And then there is the fact that clinches the interest of the whole problem. The Rigveda’s closest linguistic relative, the Avesta of Zoroastrian Iran, uses the cognate of this same word, asman-, as its ordinary noun for the sky. Not a poetic flourish. The word for sky.

This article follows áśman through about thirty Rigvedic verses, out into its Indo-European cousins, and then forward to the point where the Indian and Iranian branches of the same word go in opposite directions, one keeping the rock and losing the sky, the other keeping the sky and losing the rock. What it will not do is claim that the Vedic poets wrote down a cosmology. They did not. They did something more useful for us: they talked as if.

Thirty Verses and a Word That Will Not Settle

Áśman is not a common word. By my count it occurs about thirty times in the Saṃhitā, which puts it far behind the Rigveda’s workhorse terms for rock. What makes it worth the attention is the distribution of those thirty passages. They do not cluster around a single image. They fan out across five quite different jobs, and the same word does all five without apparent strain.

The categories overlap, which is the point rather than a defect of the table. When Indra hurls “the stone of heaven”, the missile and the sky are the same substance. When Bṛhaspati hauls cattle out of the rock, the prison and the cloud are the same substance.

Three compounds are worth pulling out of the list, because compounds show what a speaker takes for granted. The Maruts, the storm troop, are aśma-didyavaḥ, “having stones for missiles” (RV 5.54.3). Indra and Soma strike with weapons called aśma-hanman, “stone-slaying”, and the verse says explicitly that these come divás pári, from across heaven (RV 7.104.5). The captive cows in two passages are aśma-vrajāḥ, “having a pen of stone” (RV 4.1.12, 10.139.6). Nobody builds a cattle pen out of rock in the Punjab. The pen is in the sky.

Methods note. The counts here come from string searches over the romanised Saṃhitā text, sorting forms of áśman from the unrelated and far commoner raśmi (“rein, ray”), whose inflections look similar in transliteration. I make it thirty-one tokens in thirty verses. Treat the figure as accurate to within a verse or two rather than as a concordance entry, and treat the five-way sorting as my reading of what each passage is doing, not as a lexicographer’s ruling. The Vedic Index and Mayrhofer’s etymological dictionary are the places to check me.[10][6]

One more detail before leaving the word list, and it is the kind of thing that tells you a translator has hit a wall. At RV 7.88.2, Vasiṣṭha wants to see a light that is svar yad áśmann, “the svar that is in the áśman”. Griffith, in 1896, rendered the phrase as “the light in heaven”.[2] He did not translate áśman as “stone” there, because the sentence would not let him. He translated it as “heaven”.

The Cousin Word That Means “Sky”

Hold the Vedic evidence still for a moment and look sideways, at Iranian.

Avestan asman- means “stone”. It also means “sky”, and in the Zoroastrian tradition that second sense becomes the standard one. Middle Persian has āsmān, New Persian āsmān, and from Persian the word was borrowed into Hindi and Urdu, where āsmān is the everyday word a speaker in Delhi or Lahore uses for the sky today. Somewhere along that line the stone dropped out of view and only the sky was left.

Hans Reichelt put the problem on the table in 1913 in a paper with a title that leaves nothing to guess at: Der steinerne Himmel, “The Stone Sky”.[3] His argument was that the shared sense was not an accident of one branch but an inheritance: that Indo-European speakers conceived the firmament as a vault of rock, and that the word for stone therefore came to mean the thing the stone vault was. The reconstruction usually written today is *h₂éḱmon-, from a root meaning something like “sharp, pointed”, with the derived sense “stone, stone tool”.

$$*h_2e\hat{k}mon\text{-} \;\longrightarrow\; \begin{cases} \text{Indic: stone, sky only by implication} \ \text{Iranian: sky, stone in the older layer} \end{cases}$$

The cognates outside Indo-Iranian are where the case gets genuinely entertaining.

Old Norse hamarr is the one that should make a modern reader sit up, because it is the ancestor of English hammer, and because it means “crag” as readily as it means “hammer”. The thunder god of the Germanic north swings a weapon whose name means, underneath, a rock. The thunder god of the Rigveda hurls áśmānam, a rock. That is not two traditions borrowing from each other at a late date; that is two traditions keeping the same very old piece of furniture.

Greek preserves the stone without the sky, in ákmōn, an anvil. But Greek epic keeps the metal firmament in other words, calling heaven brazen, and in places iron, a set of passages Martin West collects next to the Indo-Iranian evidence in his survey of inherited poetic material.[5] Hesiod even uses the anvil to measure the cosmos: a bronze ákmōn dropped from heaven would fall for nine days and nights before reaching earth (Theogony 722-725). The sky is high, and the unit of measurement is a lump of worked stone.

Two readings of all this are live, and honesty requires both. The strong reading, Reichelt’s, is that the inherited word itself meant “stone and therefore stone heaven”, and that Vedic divó áśman is a fossil of it. The cautious reading is that the inherited word meant “stone” alone, that Iranian innovated the sense “sky” on its own, and that the Vedic “stone of heaven” is nothing more exotic than a thunderbolt, since thunderbolts do come from up there. The cautious reading is hard to refute. It is also hard to square with a sun described as a stone sitting in the middle of the sky.

What Follows If Heaven Is Rock

Here is where the exercise earns its keep. Assume, for the length of this section, that the Vedic poets inherited a sky of rock and never thought to argue about it, the way nobody in an English poem argues that the sky is blue. Then look at what the Rigveda says about weather and see how much of it stops needing explanation.

The sun is a stone set in the middle of heaven

A hymn to the All-Gods describes something crossing the sky:

ukṣā samudro aruṣaḥ suparṇaḥ pūrvasya yonim pitur ā viveśa | madhye divo nihitaḥ pṛśnir aśmā vi cakrame rajasas pāty antau

“Bull, sea, ruddy, fine-winged, he has entered the womb of the primal father. Set down in the middle of heaven, the dappled stone, he has strode out; he guards the two ends of the middle realm.”

RV 5.47.3, rendering mine from the Saṃhitā text.

Pṛśnir áśmā, the dappled or speckled stone, madhye divó nihitaḥ, set down in mid-heaven. The subject, on the usual reading, is the sun, described in the piled-up appositional style these poets favour: bull, sea, bird, stone. No part of that sequence is stranger than the others to the poet. The sun is a bright rock fixed in a rock ceiling.

Thunder is a stone rolling and sounding

Twice, both times in the fifth book, the Rigveda uses a fixed phrase: áśmānaṃ cit svaryám, “even the svaryá stone”. Once it is rolling, once it is being shaken loose.

áśmānaṃ cit svaryam parvataṃ girim pra cyāvayanti yāmabhiḥ

“With their journeyings they shake loose even the resounding stone, the mountain, the hill.”

RV 5.56.4, of the Maruts; rendering mine.

Svaryá is ambiguous in a way that happens to be convenient. It can be taken from svar, “to sound”, giving “resonant, resounding”, or from svár, “sun, light, heaven”, giving “heavenly”. Griffith took the second option at RV 5.30.8 and translated “the rolling stone that is in heaven”.[2] Either way, a stone is up there, it moves, and storm gods are what move it. If you already believe the ceiling is rock, you do not need a theory of thunder. You need only point upward.

Fire lives inside stone, so lightning needs no theory either

The Rigveda is entirely clear that fire comes out of rock, because the poets struck it out of rock themselves. Indra “generated fire between the two stones” (RV 2.12.3), in the great hymn whose refrain names him after each feat. Agni is born “from the stone” in the opening verse of the second book (RV 2.1.1), along with being born from the waters, the plants and the wood. A fire drilled by arm-power “passes by the stones, burning the grass” (RV 3.29.6). See the companion piece on Agni for the full range of the fire god’s birthplaces.

Now combine the two facts the poets hold at once: fire is latent in stone, and the sky is stone. Lightning is then not a mystery at all. It is what happens when something large strikes the ceiling.

I want to mark this clearly, because it is the one move in this article that the text does not make for you. No Rigvedic verse says “lightning is fire struck from the sky-rock”. The inference is mine, built from verses that are each separately explicit. Treat it as a hypothesis that explains an unusual amount, not as a translation.

Everything good is locked inside rock and has to be broken out

This is the largest group in the table above, and it is the spine of Rigvedic myth. Light, cattle, water, dawn and treasure are all inside stone, and the central divine act is splitting the stone open.

graph TD
    A[Sky of stone] --> B[Sun: dappled stone in mid-heaven]
    A --> C[Thunder: the resounding stone rolls]
    A --> D[Fire latent in stone]
    A --> E[Goods sealed inside the rock]
    A --> F[Weapon: a fragment hurled down]
    D --> G[Lightning, by inference]
    E --> H[Cattle and dawn: Vala myth]
    E --> I[Waters: Vrtra myth]
    E --> J["Treasure: RV 1.130.3"]
    F --> K["Divo asmanam, 3 verses"]

Indra “split even the stone with his force and found the pen of the ruddy cows” (RV 5.30.4). The cows are “held fast within the rock” until soma makes him exuberant enough to free them (RV 6.43.3). Bṛhaspati, with his singing crew, bursts “bonds made of stone” (RV 10.67.3) and “lifted the cattle out of the rock, and split the earth’s skin as if with water” (RV 10.68.4). Hanns-Peter Schmidt’s monograph on Bṛhaspati and Indra remains the detailed study of how these two rock-breakers divide the same myth between them.[17] For the wider habit of the Rigvedic gods hiding valuables and the poets’ fascination with concealment, see what the Rigveda hides.

What I find persuasive is not any single verse but the absence of a competing image. The Rigveda never has Indra tearing open a bag, a skin, a curtain or a door of cloth to release the rain. The barrier is always hard. It is rock, or a rock-walled fort, or a stone pen, or a stone fetter. A tradition that imagined the sky as fabric would have reached for a different verb than “split”.

The Metal Stone of Heaven

One verse in the collection goes further than the rest, and it is worth sitting with.

tvam āyasaṃ prati vartayo gor divo aśmānam upanītam ṛbhvā

“Thou hurledst forth from heaven the iron missile, brought by the Skilful, from the sling of leather.”

RV 1.121.9, tr. Griffith (1896); the literal phrase is āyasam divó áśmānam, “the metal stone of heaven”.[2]

The syntax of the rest of the verse is genuinely difficult and translators differ. The noun phrase is not difficult. Áyas is the Rigveda’s word for metal, and Wilhelm Rau’s study of Vedic metal vocabulary concluded that in this period it means copper or bronze, with iron arriving later under the qualified names “dark metal” and “black metal”.[11] So the poet is describing a metallic stone that comes from the sky. Our companion article on Rigvedic metallurgy sets out what the text knows about smelting, casting and gold.

Set that phrase beside how the rest of the Bronze Age talked about metal from above. Egyptian scribes, from roughly the thirteenth century BCE, called iron bjꜣ n pt, “metal of the sky”. Mesopotamian terminology for iron also carries the sign for heaven. And the reason was not mysticism but supply: before smelting, the only iron anyone could work was iron that had fallen. Albert Jambon’s systematic chemical survey of Bronze Age iron artefacts found the meteoritic signature in every case he could test, with terrestrial smelted iron appearing only later.[12] The most famous object in that class is the dagger in Tutankhamun’s tomb, whose nickel and cobalt proportions match an iron meteorite.[13]

So an ancient poet saying “metal stone of heaven” is using a phrase that, across a wide stretch of the Bronze Age world, had a concrete referent.

Aside. Three reasons to keep the brakes on here. First, the Rigveda nowhere describes a fall: no fireball, no smoking hole in a field, no recovered object. Second, divó áśman is Indra’s weapon, and a god’s weapon can come from heaven without anyone having picked one up. Third, “metallic” may be doing poetic work, since the Rigveda likes metal as a byword for hard and durable. What can be said without overreach is narrower and still interesting: the poets had a phrase for a metal stone that belongs to the sky, and in the world they lived in, metal stones from the sky were the only iron going.

There is an irony in the timeline that I cannot resist pointing out. European natural philosophy spent the eighteenth century treating reports of falling stones as peasant credulity. Ernst Chladni argued in 1794 that the things were extraterrestrial and was not thanked for it. The argument closed only after a shower of more than three thousand stones fell on L’Aigle in Normandy in April 1803 and Jean-Baptiste Biot collected enough witnesses to make denial absurd.[14] The claim that stones come from the sky was, for a while, more respectable in a Bronze Age hymn than in the Académie.

Three Cautions

An argument this tidy should be tested against the evidence that resists it, and there are three pieces worth naming.

The Rigveda never states the thesis. Its ordinary words for the sky are dyú-/div- (cognate with Greek Zeús, Latin Iuppiter), nāka, “vault”, and rocana, “shining space”. None of them means stone. The sky in the Rigveda is propped up, spread out, measured, filled with light and occasionally declared to need no support at all. For that whole architecture, and for the poets’ surprisingly empirical confidence that the thing will not fall, see the sky does not fall. A cosmology reconstructed from vocabulary alone is always a reconstruction.

The rock idiom mostly runs on other words. Áśman is the rarest member of a crowded family, and if you only read the commonest members you would never notice it.

Ádri does most of the cloud-splitting work, and grāvan is the named, praised, almost personified stone of the ritual, which gets its own hymns and its own noise; the pressing-stone workshop article follows that one. The sky sense attaches to the word that happens to have the Iranian cousin. A sceptic is entitled to call that suspicious, and to suspect the comparative method of finding what it went looking for.

Ordinary masonry gets the same word. Indra threw down “a hundred forts made of stone” for Divodāsa (RV 4.30.20), and a river in the tenth book is simply named Aśmanvatī, “the Stony One”, with the poet telling his companions to hold on to each other and cross (RV 10.53.8). A well has a stone mouth (RV 2.24.4) and elsewhere a stone wheel for the rope to run over (RV 10.101.7), which is machinery rather than cosmology; the wells article works through that equipment. Those hundred stone forts also raise a separate problem, since nothing in the archaeology of the Punjab of that period offers a hundred stone fortresses. Wilhelm Rau’s study of púr concluded that the Vedic word denotes ramparts and enclosures rather than cities of cut stone.[18] Poetic hyperbole and ordinary building stone are both live options, and neither needs a sky of rock.

The Two Careers of One Word

What happened next is the part that makes the whole business feel less like philology and more like a story.

In India, the stone sky quietly disappeared. It was not refuted; nobody wrote a treatise against it. The later Vedic and Upanishadic traditions developed a different answer to the question of what the sky is, and the answer was ākāśa, space, the element that is nothing but extension. Search the Rigveda for that word and you get nothing: ākāśa is absent from the entire Saṃhitā, and arrives only in the later texts, where it becomes one of the five elements and then the favourite analogy of philosophers who want something that pervades without resisting. Classical Sanskrit kept aśman in its plain sense. Stone. Rock. The thing you trip on.

In Iran the opposite happened. The stone sky hardened into doctrine. In Zoroastrian cosmology āsmān is the first of the material creations, and the Pahlavi texts describe its substance as hard and bright, gleaming like stone or metal; A. Tafażżolī’s survey of the term in the Encyclopaedia Iranica lays out that material sky, personified as a divinity in shining armour.[4] Mary Boyce’s history of the religion sets the same picture in its wider frame: a created sky conceived as a shell around the world, shaped so that the hostile spirit, once inside, is trapped in it.[15] The sky is not a roof in that system. It is a container with walls.

graph LR
    A["PIE *h2ekmon-: stone"] --> B[Indo-Iranian: stone, sky-vault]
    B --> C[Vedic asman: stone]
    B --> D[Avestan asman: stone and sky]
    C --> E[Classical Sanskrit: stone only]
    D --> F[Middle Persian asman: sky]
    F --> G[New Persian asman: sky]
    G --> H[Hindi and Urdu asman: sky]
    E --> I[Sky becomes akasa: space]

And then the loop closes in a way no one planned. The Persian word came into India with Persian, and āsmān is now a perfectly ordinary Hindi and Urdu noun. Film songs are full of it. A speaker in Lucknow who says āsmān is using, without knowing it, an Indo-Iranian word for rock that the Sanskrit side of the family never let mean “sky”, restored to India after a detour of three thousand years through a language that kept the heavens and forgot the stone. For the broader pattern of words travelling into and out of this vocabulary, see the piece on the Rigveda’s foreign words.

Reading With a Pebble in Your Hand

The Rigveda is not a physics text and gains nothing from being read as one. But it is a record of people paying sustained attention to a sky they had no instruments to examine, and what the áśman verses show is that their attention produced a model rather than a set of unrelated images. A hard ceiling. Things of value sealed inside it. Fire latent in the material, and therefore available to anyone who strikes it hard enough. Gods whose main work is breaking the container open so that what is inside can come down, which is what rain and dawn and daylight all are in this worldview: goods extracted from storage.

That is why Indra’s career looks so repetitive to new readers and so coherent to old ones. He is not a generic warrior with an arbitrary list of enemies. He is the force that gets things out of rock, and nearly every myth attached to him is a variation on that one operation, which is also why the Rigveda can use the same verb for splitting a cloud, a fort, a mountain and a cattle pen. Our reappraisal of Indra takes up what that does to a reader’s sense of the god; the Varuṇa article covers the very different deity who keeps the whole arrangement lawful.

If you want to feel the idea rather than follow it, read RV 1.130 and RV 4.16 one after the other with a flint pebble or any hard stone in your hand. Notice how many of the poet’s verbs are things you do to rock: split, shatter, open, prise, break out, shake loose. Then go outside during a thunderstorm, ignore everything you know about electrical discharge, and listen to the sound as if something enormous were rolling overhead on a stone floor. The Vedic poets had no better theory available. Their word for it is the one hundreds of millions of people on the subcontinent still use, every day, for the sky.