The Mountain Is Afraid Because the Gods Are Playing

Something Is Coming Down the Valley

A poet of the Atri family is watching a storm come in over broken country, and the line he reaches for is not about rain or fear or destruction. It is about recreation.

parvataś cin mahi vṛddho bibhāya divaś cit sānu rejata svane vaḥ | yat krīḷatha maruta ṛṣṭimanta āpa iva sadhryañco dhavadhve

“Even the mountain, grown great, has taken fright. Even the ridge of heaven shook at the sound of you. When you play, Maruts, with your spears in your hands, you run all together like waters.”

Rigveda RV 5.60.3, my rendering from the Saṃhitā text. Compare Griffith (1896): “when, armed with lances, ye are sporting, Maruts.”

The verb in the middle of that is krīḷatha, “you play”, from the root that classical Sanskrit writes krīḍ and that the Rigveda, following its own sound habits, writes with an ḷ. It is the ordinary word. It is the word a later author will use for children with a ball, for lovers, for a king amusing himself in a garden, for Kṛṣṇa in the forest at night. Here it is what the thunderstorm is doing while a mountain shakes.

Run that root through the whole collection and the results are stranger than the single verse suggests. Twenty-nine verses out of roughly ten and a half thousand contain it. Thirteen of those twenty-nine describe a pressed drink moving through a sheep’s-wool filter. Seven describe the Maruts. Three describe fire or its rays. Two describe objects on the ritual ground that have no business being alive at all. One describes arrows in a battle. Exactly one describes human beings, and in that one the humans are playing at a sacrifice.

Nobody in the Rig Veda plays a game. No child in the text is shown playing, though a child at play turns up three times as a simile for something else. There is no Rigvedic word for a toy and, as far as I can find, no occurrence of the abstract noun krīḍā, “a game, sport, amusement”, which later Sanskrit cannot do without. Play in this text is not a thing you have. It is a way certain powers move, and the question that makes the whole pattern interesting is which powers are allowed to move that way.

A Verb Without a Noun

Start with the shape of the word, because it is unusual. Vedic nava, “new”, sits on a reconstruction so secure that you can line up Latin, Greek, Hittite and English behind it. Krīḍ has nothing like that behind it. The etymological handbooks take it back to Indo-Aryan and stop; there is no agreed Indo-European cognate set and no Iranian partner that the comparative grammars reach for. [1] For a concept this basic, that absence is worth registering. Either the word is an Indo-Aryan innovation, or it came from somewhere the comparative method cannot follow.

Aside. The Rigveda writes the root krīḷ, not krīḍ, for a purely phonetic reason: in the transmitted Śākala text a ḍ between vowels is pronounced and written as a retroflex ḷ, so krīḍati surfaces as krīḷati, just as īḍe and īḷe, “I invoke”, are the same word. Dictionaries lemmatise under krīḍ; the text in front of you will say krīḷ. If you search a romanised Saṃhitā for the dictionary form you will find nothing at all, which is one of the more frustrating ways to conclude that a word is absent from the Rig Veda.

The forms that actually occur are verbs and adjectives: krīḷanti, “they play”; krīḷan, “playing”; krīḷu-, krīḷi-, krīḷa-, “playful”. The vocabulary describes behaviour, never an institution. That is already a small finding. A culture that has a noun for “game” has fenced off a region of life where ordinary consequences are suspended. The Rig Veda has not done that fencing. When something plays in this text, it is playing in the same world where everything else happens, with the same teeth.

Table 1. Distribution of the root across the collection. Verse totals are from the romanised Śākala text used for this article; editions differ by a few dozen verses depending on how the Vālakhilya hymns and some long stanzas are counted.

Three of the six family books, the oldest core of the collection, never use the word once. The three books traditionally assigned to the Gṛtsamadas, the Viśvāmitras and the Bharadvājas carry 1,815 verses between them, and not one of those verses calls anything playful. Then maṇḍala 9 uses it thirteen times in 1,108 verses. Expressed as a density ratio against the rest of the text:

$$\frac{13/1108}{16/9420} \approx 6.9$$

Soma hymns are nearly seven times as likely to call something playful as anything else in the Rig Veda. That is not a fact about soma’s theology. It is a fact about what the word is for.

Methods note. Every count here comes from a stem search for krīḷ over the romanised Śākala Saṃhitā, with each of the twenty-nine hits inspected by hand against Griffith’s translation and the deity ascriptions. The method finds what it finds: forms buried inside compounds, or spellings the romanisation handles differently, will have escaped, so read these as floors rather than census figures. For numbers that will survive a referee, Grassmann’s 1873 dictionary lists every occurrence of every stem, and Lubotsky’s concordance does the same for every word form. [2] [3]

The Storm Troop at Play

Seven verses, one subject, and the portrait is consistent enough to be a definition. The Maruts are a band of young male storm gods who travel together, arrive at speed, carry spears, wear ornaments, make an enormous noise and do not take orders. Their hymns describe them doing damage and being admired for it in the same breath. A Kaṇva poet opens a hymn by naming the quality first:

krīḷaṃ vaḥ śardho mārutam anarvāṇaṃ ratheśubham

“Sing out, Kaṇvas, to your playing Marut troop, unassailable, splendid on its chariot.”

Rigveda RV 1.37.1, my rendering. Griffith (1896) has “Sporting, resplendent on their car.”

Krīḷam śardhaḥ, “the playing troop”. It is the first adjective in the hymn, which is where a Vedic poet puts the thing he wants heard. Elsewhere they are krīḷayaḥ dhunayaḥ, “playful, roaring”, in the same line that reports the earth trembling under them (RV 1.87.3). The war-band reading of the Maruts, their resemblance to the young men’s raiding societies reconstructed for other Indo-European traditions, is laid out in The Storm Troop; what the play vocabulary adds is the mood. These are not soldiers on a mission. They are the storm as a gang of boys, and the poet watching them is not sure whether to be frightened or delighted, so he is both.

The most revealing of the seven puts the word inside the ritual enclosure:

nityaṃ na sūnuṃ madhu bibhrata upa krīḷanti krīḷā vidatheṣu ghṛṣvayaḥ

“Carrying the honey as though it were their own son, they play, playful ones, eager, at the ritual gatherings.”

Rigveda RV 1.166.2, my rendering from the Saṃhitā text; Griffith (1896): “they sport in sportive wise gay at their gatherings.”

Gods bringing an offering as carefully as a man carries his only son, and simultaneously horsing around. Both things are true of the same beings at the same moment, and the poet does not experience this as a contradiction needing resolution.

Aside. Vidatha, the word translated here as “ritual gatherings”, is one of the genuinely contested items in the Rigvedic vocabulary. Proposals have ranged from a distribution of goods, to a clan assembly, to a ritual session with a verbal contest at its heart. Nothing in this article depends on which is right, but it matters for the picture: whichever institution a vidatha was, the Maruts are described as playing inside it. The assembly words are mapped in The Assembly That Lost Its Name.

What Play Looks Like When You Have to Describe It

A Vedic poet almost never leaves an abstraction standing on its own. He attaches it to something visible. So when the text says that soma plays, or that the Maruts play, it usually says at once what the playing resembles, and the vehicles it chooses are remarkably few.

Table 2. Every simile attached to the play verb. The field is narrow: young animals, young humans, and moving water.

Three classes of image, and no fourth. Play in this text looks like a calf, a foal, a child, or running water. Those are the four things in the Rigvedic world that move hard without going anywhere in particular, and the poets clearly recognised them as a single class. Tatyana Elizarenkova’s study of Vedic similes makes the general point that the comparison is doing cognitive work rather than decoration: the vehicle tells you which features of the subject the poet wants foregrounded. [4] Here the feature is surplus. Something has more motion in it than its task requires.

Notice what this means for the children. They appear at play three times, and all three times they are the simile, not the subject: a storm god or a drink is what is being described. No actual Vedic child is ever shown playing. We know from the similes that the poets had watched calves butting at an udder and children tumbling around a settlement, with enough attention to find the comparison ready to hand. The text simply never thought those scenes worth recording for their own sake. That is a useful reminder of what kind of document this is. The Rig Veda is not a description of Vedic life. It is a liturgy that occasionally has to reach into Vedic life for an image, and the reaching is where the daily world shows up.

Fire Plays, and Something Is Eaten

Three verses give the word to Agni, and the darkest one in the whole set is among them. A poet of maṇḍala 10 opens with an accusation phrased as a question:

kiṃ deveṣu tyaja enaś cakarthāgne pṛchāmi nu tvām avidvān | akrīḷan krīḷan harir attave ‘dan vi parvaśaś cakarta gām ivāsiḥ

“What abandonment, what offence have you committed among the gods? I ask you, Agni, not knowing. Not playing, playing, the tawny one, toothless, eating, has cut it apart joint by joint, as a knife cuts an ox.”

Rigveda RV 10.79.6, my rendering. Griffith (1896): “Playing, not playing, he gold-hued and toothless, hath cut his food up as the knife a victim.”

Akrīḷan krīḷan: not playing, playing. The hymn is a riddle poem about fire, and this is its sharpest turn. Flame moves the way a playing animal moves, loose and purposeless and quick, and what it is actually doing while it moves is butchering. The poet frames the whole image with a question about guilt, as though the fire’s lightness were itself the thing that needs explaining to the gods.

Hold that against the sunnier use in the fifth maṇḍala, where a poet asks a flame to come to him krīḷan, in a playing mood, and immediately adds that these same flames are “sharp as arrows, well whetted” (RV 5.19.5). The Rigvedic poets do not use play as a softening word. When they call something playful they are often telling you to watch it.

The Drink That Plays

Maṇḍala 9 is a book of 114 hymns addressed to a liquid in motion: soma pressed between stones, strained through a sheep’s-fleece filter, running into wooden vats. Thirteen of its verses say that the liquid plays. Read a few in sequence and you can watch the metaphor doing something precise.

somaḥ punāno avyaye vāre śiśur na krīḷan pavamāno akṣāḥ | sahasradhāraḥ śatavāja induḥ

“Soma, being purified in the sheep’s wool, flowing clear, has run like a playing child: the drop of a thousand streams, of a hundred prizes.”

Rigveda RV 9.110.10, my rendering; Griffith (1896) has “like a playful infant.”

Elsewhere he “plays in the wood beyond the fleece” (RV 9.6.5), “plays on the ox-hide with the stones” (RV 9.66.29), runs “like a playing horse” and sloughs his skin “like a snake” in the same stanza (RV 9.86.44). The verb is attached, every time, to the moment of free flow: after the pressure of the stones, before the drink is poured out and drunk. What plays is a substance between constraints. The physical setting of all this, the stones and the fleece and the vats, is reconstructed in The Soma Workshop, and the theology of the ninth book in The Pressed God.

The point is not that soma is playful by nature. It is that the Vedic poets had one word for motion that is fast, loose, repetitive and not aimed, and liquid running through wool is the most available example of it in their working lives. They were standing over it every morning.

Nobody Plays Except the Weather

Now the negative result, which is where the pattern earns its keep.

Varuṇa does not play. Mitra does not play. The Ādityas, the stern gods of oath and contract and observed behaviour, never once appear with the verb. Neither does Uṣas, across the twenty or so hymns addressed to her. Indra, who receives roughly a quarter of the collection, is never said to play; the single verse in his hymns that uses the word applies it to his sūnṛtāḥ, his acts of generosity, which “play like waters running down a slope” (RV 8.13.8). His gifts play. He does not.

And yet Indra dances. The vocative nṛto, “O Dancer”, is thrown at him in the middle of his most violent exploits, in the verse that recounts the shattering of ninety forts (RV 1.130.7) and again where he is praised for releasing the waters (RV 2.22.4). Uṣas dances too, in one of the most quoted similes in the text: adhi peśāṃsi vapate nṛtūr iva, “she puts on her embroidered things like a dancer” (RV 1.92.4). And in a creation hymn the gods, standing in the primordial water, kick up a thick dust nṛtyatām iva, “as if from dancers” (RV 10.72.6).

Table 3. The two vocabularies of non-productive motion, and who gets which.

One trap in that table is worth flagging, since it has caught people with searchable texts. The Maruts are repeatedly called nṛtama, which looks like a superlative of “dance” and is not: it belongs to nṛ, “man”, and means “most manly”. The two stems are homographs in romanisation and unrelated in origin.

The two words are not interchangeable, and the split is not random. Dancing in the Rig Veda is display: it is done in front of someone, it is ornamental, it belongs with jewellery and embroidered cloth and the self-adornment vocabulary the Maruts also attract. Playing is the opposite. It has no audience built into it. It is what a thing does when its energy exceeds its errand.

Which gives the distribution its sense. The powers that play are the powers of motion, intensity and transformation: wind, fire, flowing drink. The powers that never play are the powers of order: the gods who hold contracts, watch behaviour, count offences, keep the sun on its track. Varuṇa cannot be playful for the same reason a judge cannot be whimsical and still be a judge. The Vedic understanding of what makes something sovereign, examined from another direction in Varuṇa and the Cosmic Order, turns out to include a constraint on how it is allowed to move.

graph LR
    R["Root krid-, 29 verses"] --> S["Soma in the filter: 13"]
    R --> M["Maruts, the storm troop: 7"]
    R --> F["Fire, flames, rays: 3"]
    R --> O["Objects: dice and press-stones: 2"]
    R --> W["Missiles in battle: 1"]
    R --> G["Indra's bounties: 1"]
    R --> A["Apsarases, in a simile: 1"]
    R --> H["Human worshippers: 1"]

The One Time People Play

It happens in an Agni hymn of the fourth maṇḍala, the Vāmadeva family book, and it is easy to miss because the standard English translation does not keep it.

iha tvā bhūry ā cared upa tman doṣāvastar dīdivāṃsam anu dyūn | krīḷantas tvā sumanasaḥ sapemābhi dyumnā tasthivāṃso janānām

“Here let it come to you abundantly, of itself, at dusk and dawn, to you shining day after day. Playing, in good spirits, let us serve you, standing above the splendours of the peoples.”

Rigveda RV 4.4.9, my rendering. Griffith (1896) renders the key phrase “content and joyous”, which drops the verb entirely.

Krīḷantas tvā sapema: playing, let us serve you. This is the only verse in the Rig Veda where human beings are the grammatical subject of the play verb, and the activity they are playing at is the fire ritual. Not a festival, not a dance, not a children’s game. The daily sacrifice.

Set beside it the two verses where ritual equipment plays. The pressing stones, in their own hymn, raise their voice over the pressed juice “like playful ones butting their mother” (RV 10.94.14), which turns a row of heavy rocks into a litter of calves at an udder. And the dice, which are the subject of the next section. On the ritual ground, the things play and the people mostly do not; the one time the people do, they are described as playing at the work itself.

Frits Staal’s much-argued claim that Vedic ritual is “meaningless”, by which he meant that it is a self-contained, rule-governed activity whose correctness is internal rather than referential, reads differently next to this verse. [5] Staal reached for dance as his analogy. The Rig Veda had already reached for play, which is a better fit: an activity with exact rules, total absorption, no external product, and consequences that are nonetheless real to the participants. F. B. J. Kuiper’s reconstruction of the Vedic verbal contest, and Johan Huizinga’s argument that the brahmodya riddle duels are one of the clearest cases anywhere of a culture’s most serious business being conducted as a game, both push in the same direction. [6] [7] The Vedic sacrifice is not solemn in the Victorian sense. It is competitive, exacting, and occasionally very funny, as anyone who has read the frog hymn knows: see The Frog Hymn.

The Dice Are the Ones Playing

The gambler’s hymn is the Rig Veda’s great piece of first-person wreckage: a man who has lost his wife, his house and his standing to a handful of nuts from the vibhīdaka tree, talking about it. Its eighth verse does something to the grammar that is easy to read past.

tripañcāśaḥ krīḷati vrāta eṣāṃ deva iva savitā satyadharmā | ugrasya cin manyave nā namante rājā cid ebhyo nama it kṛṇoti

“Their troop of three and fifty plays, like the god Savitṛ whose ordinances are true. They do not bow to the anger even of the fierce; even a king makes obeisance to them.”

Rigveda RV 10.34.8, my rendering; Griffith (1896): “Merrily sports their troop, the three-and-fifty.”

The dice play. The man does not. He is the board. And the word for their formation is vrāta, the same word used for a band of men moving together, so the dice are a troop in the field, with the gods’ own authority behind them and a king bowing as they pass.

Heinrich Lüders devoted a monograph to Indian dicing in 1907, and Harry Falk argued in 1986 that the game sits structurally close to the Vedic sacrifice itself rather than at its disreputable edge. [8] [9] Whatever one makes of that argument, the Rigvedic poet’s grammar supports the instinct behind it: dice are not equipment in this hymn, they are agents, and the verse compares their play to the action of a god who keeps fixed laws. The hymn is read in full in The Gambler’s Hymn.

So the pattern at the human end is consistent. People are not players in this text. They are what gets played with, and the one exception is a priest in a good mood at the morning fire.

How Play Got to the Middle of Everything

Jump forward. By the time Indian philosophy is writing systematically about why a complete being would bother to make a world, the answer is a single word, and it is not krīḍā.

The Chāndogya Upaniṣad describes the self that has risen out of the body into the highest light, and what it does there is jakṣat krīḍan ramamāṇaḥ, “eating (or laughing), playing, taking his pleasure”, with women or chariots or kin, remembering nothing of the body it came from (Chāndogya 8.12.3). [10] The verb has survived, and it has moved: it now describes the condition of release itself. Liberation looks like playing.

Then the Brahmasūtra takes the last step. An objector points out that a perfect being can have no motive, since motive implies lack, and therefore cannot be the cause of the world. Bādarāyaṇa’s answer is four words long:

lokavat tu līlākaivalyam, “but it is mere play, as in the world” (Brahmasūtra 2.1.33). Śaṅkara’s commentary glosses it with the image of a king who wants for nothing and goes out to amuse himself anyway. [11] The world has no purpose because purpose would be a defect in its maker. What is left when you subtract purpose from action is līlā.

Table 4. The migration of a word from the periphery of the Vedic world to the centre of Indian theology.

graph LR
    A["Rigveda: krid-<br/>storm, flame, drink"] --> B["Chandogya 8.12.3<br/>the freed self plays"]
    B --> C["Brahmasutra 2.1.33<br/>creation is lila"]
    C --> D["Bhakti: the god's<br/>whole life is lila"]
    A --> E["Classical Sanskrit:<br/>krida, 'a game'"]

Līlā itself appears nowhere in the Rig Veda; its own etymology is disputed, and Norvin Hein’s history of the term traces its rise as a theological category well after the Vedic period. [12] [13] What travelled was not the word but the idea underneath it: that there is a kind of action which is intense, exact and utterly unaccountable, and that this is a mark of power rather than a lapse from it. The Vedic poets found that kind of action in the weather. Later theologians found it in the ground of being.

What the Word Was Actually For

Two things are worth holding onto from twenty-nine verses.

The first is that the Rigvedic poets had a precise ear for a category modern languages handle clumsily. English “play” covers a child with blocks, a violinist, a striker on a wing and a gambler at a table, and the common thread is hard to state. Vedic krīḷ- is narrower and sharper: motion with more energy in it than the task needs, performed by something that cannot be held responsible. Wind qualifies. Fire qualifies. A liquid running loose through wool qualifies. A judge does not. That is a serious piece of observation, and it is sitting in a text usually described as an archive of ritual formulae.

The second is the shape of the exclusion. The Rig Veda is often read as a world of sacrifice, debt, obligation and exact reciprocity, and it is one: the poets count their gifts, calculate their returns, and worry about offences in the way described in Debt in the Rig Veda. Play is what that world has instead of a holiday. It does not get its own enclosure, its own season, or its own noun. It is allowed only to things that cannot be billed: to the storm, to the fire, to the drink, to the dice, and once, for a single line, to a man at his morning fire who is in a good enough mood to enjoy the work.

Read RV 5.60 next time a squall line comes over and the first dust lifts off the ground ahead of it. Then read the eighth verse of RV 10.34 and notice who is playing and who is being played. The distance between those two verses is the whole argument, and it was already there three thousand years before anybody wrote the word līlā down.