A Hymn to Gruel: What the Rigveda Actually Ate
Sweet Food, Honeyed Food
Near the end of the first book of the Rigveda, after hundreds of hymns to fire, storm, dawn and the drink of the gods, a poet stops and addresses a different kind of power. Not a deity with a chariot and a myth. Not an ancestor. He addresses lunch.
Now let me praise Food, the great sustainer, the strength by whose power Trita tore Vṛtra apart at the joints.
Sweet food, honeyed food, we have chosen you. Be our helper.
(RV 1.187.1 and 1.187.2, translated from the Sanskrit)
The hymn is RV 1.187, eleven verses long, conventionally called the Pitu Sūkta after its addressee, pitu (पितु), “nourishment, food”. It has never been famous. It is not chanted at weddings, it produces no theology, and translators tend to hurry through its harder words. What it does is stranger than fame: it treats eating as a cosmic force worth flattering, in the same register the poets use for Indra, and it names its subject with unembarrassed specificity. By verse ten the poet is talking about gruel.
That single hymn is a door into something the Rigveda is never advertised for. The text has a food vocabulary, and it is small, oddly weighted, and unexpectedly informative. Parched barley and clarified butter turn up constantly. Meat appears in three verses out of roughly ten and a half thousand. Rice, wheat and salt do not appear at all, in any form, anywhere. None of this is decoration. The words on the Rigvedic table work as a chronological instrument, a technology inventory, and, in one late hymn, the closest thing the collection has to a social ethic.
Methods note. The counts in this article come from my own scan of the romanised Saṃhitā text of all ten maṇḍalas, tokenised on word boundaries and matched by stem. Vedic sandhi fuses words together, so a phrase like karambhiṇam apūpavantam can surface as a single token, and any count of this kind is approximate at the edges. Two traps are worth naming. The grain word yava looks identical in its opening syllables to yavasa, “pasture, fodder”, which is commoner and means something else entirely; and most of the apparent hits for dadhi, “curds”, are in fact forms of the verb dhā, “to place”. I have separated both by hand. Where I give a number, read it as “roughly this many verses”, not as a concordance entry.
The Hymn Nobody Quotes
The Pitu Sūkta is built like a praise hymn to a god, which is exactly what makes it peculiar. It opens with a mythological credential, crediting Food with the strength that let Trita dismember Vṛtra. It moves through invocation (upa naḥ pitav ā cara, “come to us, Food”), through attributes (Food is śiva, kindly; mayobhu, refreshing; sakhā suśeva, a delightful friend), and it closes with the poet claiming to have sweetened his addressee by singing at it.
The Rigveda does this occasionally. It has hymns to Wrath, to Faith and to the priest’s fee, a small set of abstractions raised into addressable powers, which I have written about separately in Hymns to Nouns. Food belongs in that company, but it behaves differently from the others. Manyu and Śraddhā stay abstract. Pitu does not. Halfway through, the hymn stops praising the idea of nourishment and starts listing what is in the bowl.
| Verse | Term | What it is |
|---|---|---|
| 1.187.1 | pitu | Nourishment, food, the addressee of the hymn |
| 1.187.4 | rasa | Juice, sap, the extractable essence of a plant |
| 1.187.5 | svādman | Sweetness, the taste of the juices |
| 1.187.8 | apām oṣadhīnām | “Of the waters and the plants”, the source of the food |
| 1.187.9 | gavāśir, yavāśir | Soma cut with milk, soma cut with barley |
| 1.187.10 | karambha | Gruel, groats stirred into curds or milk |
| 1.187.11 | sadhamāda | A feast eaten together, shared with the gods |
Two of those entries do real work. Gavāśir and yavāśir are technical terms for how the pressed soma was mixed: with milk, or with grain. The poet drops them into a hymn about food because to him they are food and not liturgy, and the machinery of the pressing shed, which I have described in Green Mouths and Singing Stones, sits on the same shelf as porridge.
The last word matters more. Sadhamāda means a feast at which people eat together, and the hymn’s final line offers Food to the gods and to us in the same breath, using that word twice. Hold on to it. Nine hundred hymns later the Rigveda comes back to the idea with an edge.
Aside. Pitu has nothing to do with pitṛ, “father”, despite a resemblance that trips up every first-year Sanskrit student. Pitu is nourishment, cognate with Avestan pitu, and it survives into the later ritual vocabulary. The hymn also carries a refrain addressed to something called Vātāpi, whose identity is genuinely unsettled: a name for the food itself, a spirit of swelling, or something the tradition had already lost by the time of the commentators. Griffith and Geldner differ. This is one of the places where the Rigveda is older than its own interpreters.
Barley, and the Grain That Is Not There
The Rigvedic poets knew one cereal well enough to name it. Yava (यव), barley, appears in about nineteen verses, always as something sown, cut, eaten or given: cattle grazing a barley field, a sickle going through the ripe stalks, a god who arrives the way ripe barley arrives.
He holds the household steady, a delight, like ripe barley, a conqueror of peoples.
(RV 1.66.2, of Agni, translated from the Sanskrit)
Around that stem sits a small processing vocabulary. Dhānā is parched or roasted grain, eleven verses, five of them inside a single hymn. Saktu, meal or flour, occurs exactly once in the entire collection. Karambha, a gruel of coarse groats, gets six verses. Apūpa, cake, gets two. Puroḷāśa, the pressed sacrificial cake, gets eighteen. Odana, which in later Sanskrit means rice boiled in milk, occurs twice, and where it is qualified at all it is kṣīra-pāka, “cooked in milk”, with no grain named.
The hymn where all of this arrives at once is RV 3.52, addressed to Indra, which reads less like a prayer than a breakfast order.
With parched grain, with gruel, with cake, with recitation: Indra, take your pleasure in us at daybreak.
(RV 3.52.1, translated from the Sanskrit)
For you and Pūṣan we have made the gruel; for you of the tawny horses, the parched grains. Eat the cake with the Marut troop, drink the soma, dragon-slayer, knowing hero.
(RV 3.52.7, translated from the Sanskrit)
Now the absences. Search the whole Saṃhitā for vrīhi, rice, and nothing comes back. Not one occurrence, not inside a compound, not in a simile. The same for godhūma, wheat, for lavaṇa, salt, and for ikṣu, sugarcane. These are not rare words in later Sanskrit; they are ordinary ones. Vrīhi is standard in the Atharvaveda and the Brāhmaṇas, where barley and rice appear as the paired staples. Between the Rigveda and the texts that follow it, the pantry changes.
| Term | Meaning | Rigveda | Later Vedic |
|---|---|---|---|
| yava | Barley | ~19 verses | Standard, paired with rice |
| dhānā | Parched grain | 11 verses | Standard offering |
| saktu | Meal, flour | 1 verse | Common |
| karambha | Gruel of groats | 6 verses | Common |
| puroḷāśa | Sacrificial cake | 18 verses | Central to śrauta ritual |
| vrīhi | Rice | Absent | Atharvaveda onward |
| godhūma | Wheat | Absent | Later Vedic prose |
| lavaṇa | Salt | Absent | Later Vedic prose |
| ikṣu | Sugarcane | Absent | Atharvaveda onward |
What follows from that? Michael Witzel’s linguistic work on Indian domestic plants argues that vrīhi is not inherited Indo-Aryan at all but a word picked up in South Asia, with relatives in the Nuristani languages, so its arrival marks a contact event rather than an accident of poetic taste. The archaeobotany points the same way from the other side. Barley, wheat, lentil and pea form the winter crop package of the northwest, well attested in Indus-period settlements, while the domestication of rice in South Asia is a separate and largely Gangetic process worked out slowly across the second millennium BCE. A text whose grain is barley, whose cattle graze yavasa, and whose rivers are the Punjab’s is describing the northwest in the season when barley grows.
Aside. Three cautions before anyone builds a chronology on a missing word. First, the Rigveda is liturgy and not a household inventory; it names what gets offered, and rice may simply not have been offerable. Second, arguments from silence in a corpus of ten thousand verses are strong but not decisive, and the silence has to be total to count, which is why zero occurrences of vrīhi matter more than the rarity of apūpa. Third, some scholars have read yava in the Rigveda as “grain in general” rather than barley specifically, which would loosen the inference; the mainstream reading keeps it as barley, on the strength of the Indo-Iranian cognates and of later usage. The absence of rice is evidence. It is not a date. For what dates actually rest on, see Dating the Rig Veda.
Everything the Cow Could Be Turned Into
If the grain vocabulary is thin, the dairy vocabulary is not. This is where the Rigvedic kitchen shows its real sophistication: not in variety of ingredients but in depth of processing applied to one of them.
Milk is payas. Curdled milk is dadhi, which the poets use to cut soma. Churning is manthana. The product of churning, clarified over heat, is ghṛta (घृत) or sarpis, and ghṛta alone turns up in around a hundred and fifty-five verses, which is more attention than most gods receive. Gharma is the hot milk draught and also the vessel it is heated in, roughly thirty verses. The Rigveda has a technical language for taking a liquid that spoils within a day and converting it into a fat that keeps for a year.
| Term | Sense | Notes |
|---|---|---|
| payas | Milk | Also “sap, vital fluid”; used of rain and of soma |
| kṣīra | Milk | Rare in the Rigveda, common later |
| dadhi | Curds, sour milk | Mixed into soma as dadhy-āśir |
| manthana | Churning | The same verb manth is used for the fire-drill |
| sarpis | Melted butter | Often in sarpir-āsuti, “having ghee for a draught” |
| ghṛta | Clarified butter | ~155 verses; poured on the fire, and a metaphor for nearly everything |
| gharma | Hot milk, and its cauldron | ~30 verses; ancestor of the later Pravargya rite |
That last column is not incidental. Once a substance is made in a kitchen, the poets start thinking with it. The Rigveda has an entire hymn, RV 4.58, in which ghee stops being a cooking fat and becomes the medium of a riddle.
A honeyed wave rose up out of the sea; along with the stalk it reached immortality. That which is the secret name of ghee is the tongue of the gods, the navel of the deathless.
(RV 4.58.1, translated from the Sanskrit)
Read that next to the fact that ghṛta is what you skim off boiled butter, and the move these poets keep making comes into focus. Their metaphysics is built out of their groceries. Honey, madhu, works the same way: it shows up in something like two hundred and twenty verses, far more often than any solid food, because it has become the word for anything good, sweet or well made. The madhu verses of RV 1.90 run the whole world through that register.
Honey the winds pour out for the man who keeps the order; honey the rivers run. Let the plants be honey for us. Honey by night and at dawn, honey the earthly realm, and honey let Heaven our father be.
(RV 1.90.6 and 1.90.7, translated from the Sanskrit)
Nothing in those lines is about bees.
graph LR
A[Barley yava] --> B[Parched grain dhana]
A --> C[Meal saktu]
C --> D[Gruel karambha]
D --> E[Cake apupa]
E --> F[Offering cake purolasa]
G[Milk payas] --> H[Curds dadhi]
G --> I[Butter clarified to ghrta]
H --> D
I --> F
F --> J[Fire Agni]
The Cooked Inside the Raw
Here is a formula that runs through the Rigveda and that almost nobody outside Vedic studies has heard of. Nine times, spread across eight different maṇḍalas, a poet praises a god for having put the cooked inside the raw.
In the raw ones you have set the cooked, the milk inside, gleaming white, in the black cows and in the red.
(RV 1.62.9, of Indra, translated from the Sanskrit)
A great light is laid inside the belly. Raw, the cow walks about carrying the cooked.
(RV 3.30.14, translated from the Sanskrit)
The Sanskrit is pakvam āmāsu, and it recurs at RV 1.62.9, 1.180.3, 2.40.2, RV 3.30.14, 4.3.9, 6.17.6, 6.44.24, 6.72.4 and 8.89.7, with a tenth variant at 8.32.25. It is credited to Indra, to Indra and Soma together, to Soma and Pūṣan, and to the Aśvins. Formulaic repetition on that scale is how the Rigveda flags something it considers a marvel.
So what is the marvel? The word pakva covers both “cooked” and “ripe”: the verb pac applies equally to a pot over a fire and to fruit on a branch. The cow, then, is āma, raw, unheated, alive; and inside her is milk, which is pakva, finished, ready, done. Nobody lit anything. The poets look at a cow and see a kitchen that runs without flame, and they cannot get over it. That is also why the phrase attaches to gods who open things up. The one who split the Vala cave to let the cattle out is the one who put the finished thing inside the unfinished one.
Indra and Soma, you two set the cooked inside the raw, in the bellies of the cows; you held it fast, untied, in those speckled shining ones.
(RV 6.72.4, translated from the Sanskrit)
I find this the most revealing thing in the Rigveda’s food language, because it shows that cooking was not a domestic category in this world. It was a cosmological one. Ripening, digesting, lactating, heating and baking are one process under one verb, and the fire that performs it in a hearth is the same power that performs it in a cow, in a fruit and in the sun. When the later ritualists build an entire rite around a milk cauldron, and when the early Upaniṣads start describing the body as a cooking vessel, they are extending a metaphor the Rigveda had already frozen into a formula.
Fifteen Oxen, Twenty
Now the question everyone actually wants answered, and the honest version of the answer.
The word māṃsa, “flesh, meat”, occurs in three verses of the Rigveda. Two of them, RV 1.162.12 and 1.162.13, sit inside the horse sacrifice hymn, where the flesh is being tested with a fork and boiled in a pot called an ukhā, and where the poet is anxious that not a scrap be wasted. I have read those hymns closely in The God Inside the Slaughtered Horse. The third is a Ṛbhu verse. That is the entire lexical footprint of meat in the collection.
Lexical footprint is not dietary footprint, though, and the evidence arrives sideways. Animals are offered, and what is offered to the gods gets cooked and eaten by people. RV 10.91.14 lists the fire’s menu without comment: horses, bulls, oxen, barren cows, rams. And in the strange, boastful Vṛṣākapi dialogue, Indra says something no translator has managed to soften.
They cook for me fifteen oxen, twenty at a time. And I eat, and I get fat. They fill both my flanks.
(RV 10.86.14, translated from the Sanskrit)
A verse later he is drinking mantha, the stirred barley-and-milk mixture, and calling it sweet to his heart. Elsewhere Viṣṇu is credited with fetching “a hundred buffaloes, and a mash cooked in milk”, after which Indra shoots a boar (RV 8.77.10).
| Animal | Term | Context |
|---|---|---|
| Ox, bull | ukṣan, ṛṣabha | Cooked for Indra, RV 10.86.14; offered, RV 10.91.14 |
| Horse | aśva, vājin | The Aśvamedha hymns, RV 1.162 and 1.163 |
| Barren cow | vaśā | Offered to Agni, RV 10.91.14 |
| Ram | meṣa | Offered, RV 10.91.14; Indra’s animal |
| Buffalo | mahiṣa | RV 8.77.10, a hundred of them |
| Boar | varāha | RV 8.77.10, shot by Indra |
| Goat | aja | Attached to Pūṣan; a ritual animal |
Where does that leave the argument? Three things can be said with confidence and a fourth cannot. It is clear that cattle were slaughtered in Rigvedic ritual, because the vocabulary of sacrifice assumes it. It is clear that the sacrificial animal was eaten, because the hymns describe the carving, the pot and the portions. It is clear that the cow is at the same time the most protected form of wealth in the text, which is the subject of The Cow as Currency, and that slaughter and reverence coexisted without the poets registering a contradiction. What cannot be settled from the Rigveda is how often ordinary people ate meat outside ritual occasions, because the text has almost no interest in ordinary meals. D. N. Jha’s survey of the later evidence makes the ritual case at length and drew a fierce reaction on publication; Stephanie Jamison’s work on the ritual household shows how carefully the killing was framed and euphemised. The heat of the modern argument is real. The Rigvedic evidence, taken on its own, is thin, specific and mostly liturgical.
A Sieve, a Pot, and Almost Nothing Else
The equipment is sparser than the menu. Ukhā, the cooking pot, appears twice in the whole collection. The pressing stones for soma get a hymn of their own, but the quern for grain gets none. There is no word for a stove, no oven, no ladle outside the ritual set.
And yet the finest kitchen sentence in the Rigveda sits in a hymn about none of this. RV 10.71 is the great hymn on the origin of language, on how the wise made speech and how friendship is recognised in it. Its second verse reaches for an image, and what it reaches for is somebody sifting flour.
Where the wise fashioned speech with their minds, sifting it as barley-meal is sifted through a sieve, there companions know their companionship; their good fortune is laid down within their speech.
(RV 10.71.2, translated from the Sanskrit)
Saktum iva titaünā punantaḥ. The word saktu, meal, occurs nowhere else in the Rigveda, and it survives here only because a poet needed a metaphor for the invention of language and picked the finest sorting operation he knew. That is the pattern across this entire vocabulary. The Rigvedic kitchen is barely described and constantly borrowed from. Its poets thought about speech through a sieve, about the cosmos through a cooking fire, and about immortality through skimmed butter. For what they built out of the metaphor of speech, see The Goddess Who Is Her Own Hymn.
One more example, this one comic. Pūṣan, the god of roads, carries the nickname karambhād, “gruel-eater”, apparently because he was toothless and got the soft food. A poet in the sixth maṇḍala takes offence on his behalf.
Whoever points at him and calls Pūṣan “porridge-eater” should not point at that god so.
(RV 6.56.1, translated from the Sanskrit)
The joke was old enough by then to need defending. More on the god who ate the mush in Pūṣan, Lord of the Paths.
The Man Who Eats Alone
Seven verses in the Rigveda use the word kṣudh, hunger. One of them opens a hymn in the tenth maṇḍala that has no real parallel anywhere else in the collection, and that changes what all of this adds up to.
RV 10.117 is about giving food away. Not offering it to gods: handing it to people. It is late, linguistically and thematically, and it argues, which the Rigveda seldom does.
The gods have not made hunger our only death: deaths of many kinds come even to the man who has eaten his fill. The wealth of the generous man does not run out, while the one who gives nothing finds nobody to comfort him.
(RV 10.117.1, translated from the Sanskrit)
The hymn works through cases. The man who has food and turns away the thin, begging stranger. The man who feeds him and finds a friend for later. Riches, verse five says, roll from one person to another like the wheels of a chariot, an image that assumes wealth is in motion and that hoarding is a category error. Then comes the line a reader does not forget.
The fool wins food for nothing. I tell the truth: it is his death. He nourishes neither patron nor friend. The man who eats alone is guilty alone.
(RV 10.117.6, translated from the Sanskrit)
Kevalāgho bhavati kevalādī. The Sanskrit is a pun that English cannot quite carry: kevala-ādī, eating alone, and kevala-agha, guilty alone, built on the same word for solitary. Eating by yourself and sinning by yourself are given the same shape.
Set that against the closing verse of the Pitu Sūkta, nine books earlier, where food is called sadhamāda, the shared feast, twice in one line: for the gods, and for us. The Rigveda’s food language starts in the pantry and ends in an ethic, and the ethic is not about what is eaten. It is about whether anyone else is at the table. The gift economy this hymn belongs to, with its praise-poems for open-handed patrons, is the subject of The Price of Praise.
What the Table Was For
There is a temptation, reading a text this old, to want the food to be exotic. It is not. Barley, milk, curds, clarified butter, honey, cakes baked on coals, gruel, and meat on the days when an animal was killed for a god. Any pastoralist across the Eurasian steppe belt in the second millennium BCE would have recognised most of that meal, and a Punjabi farmer in a bad winter three thousand years later would have recognised a fair amount of it too.
What is unusual is not the menu but the attention. A tradition that produced the Nāsadīya hymn, and asked whether even the highest overseer knows how creation happened, also produced a hymn that says sweet food, honeyed food, we have chosen you, and meant the porridge. The same poets who could not stop noticing that a living cow contains finished milk are the ones who concluded that ripening, digesting and cooking are a single act. Nothing in that world was too ordinary to be thought about at full strength, and the food vocabulary is where the habit is most unguarded, because nobody was performing profundity while talking about groats.
The absences do their own work. No rice, no wheat, no salt, no sugarcane, in a text of ten thousand verses that names dozens of rivers and hundreds of people. That is not an oversight. It is a map of where and when these hymns were made, drawn in negative, and it agrees with what archaeobotanists have independently reconstructed about the winter crops of the northwest. Two unrelated methods, philology and seed identification, converging on one answer is about as close to proof as this field gets.
Read RV 1.187 before a meal sometime, in Jamison and Brereton if you can get hold of it, and then read RV 10.117 straight afterwards. Eleven verses praising food as a power, and nine verses saying that keeping it to yourself is a way of dying. Between them sits nearly everything the Rigveda has to say about being a body that needs feeding, and the distance between the two hymns is about as long a moral argument as the Bronze Age has left us.
References
Aufrecht, Theodor. Die Hymnen des Rigveda. 2nd ed. Bonn: Adolph Marcus, 1877.
Bates, Jennifer, Cameron A. Petrie, and Ravindra N. Singh. “Approaching Rice Domestication in South Asia: New Evidence from Indus Settlements in Northern India.” Journal of Archaeological Science 78 (2017): 193-201. ScienceDirect.
Doniger O’Flaherty, Wendy. The Rig Veda: An Anthology. Penguin Classics, 1981.
Fuller, Dorian Q. “Finding Plant Domestication in the Indian Subcontinent.” Current Anthropology 52, no. S4 (2011): S347-S362. University of Chicago Press.
Geldner, Karl Friedrich. Der Rig-Veda aus dem Sanskrit ins Deutsche übersetzt. Harvard Oriental Series 33-36. Cambridge, MA: Harvard University Press, 1951.
Griffith, Ralph T. H. The Hymns of the Rigveda. Benares: E. J. Lazarus, 1896. archive.org.
Jamison, Stephanie W. The Ravenous Hyenas and the Wounded Sun: Myth and Ritual in Ancient India. Ithaca: Cornell University Press, 1991. Cornell University Press.
Jamison, Stephanie W. Sacrificed Wife / Sacrificer’s Wife: Women, Ritual, and Hospitality in Ancient India. New York: Oxford University Press, 1996.
Jamison, Stephanie W., and Joel P. Brereton. The Rigveda: The Earliest Religious Poetry of India. 3 vols. New York: Oxford University Press, 2014.
Jha, Dwijendra Narayan. The Myth of the Holy Cow. London: Verso, 2002.
Keith, Arthur Berriedale. The Religion and Philosophy of the Veda and Upanishads. Harvard Oriental Series 31-32. Cambridge, MA: Harvard University Press, 1925.
Kuiper, F. B. J. Aryans in the Rigveda. Amsterdam: Rodopi, 1991.
Macdonell, Arthur A. Vedic Mythology. Strassburg: Karl J. Trübner, 1897. archive.org.
Macdonell, Arthur A., and Arthur B. Keith. Vedic Index of Names and Subjects. 2 vols. London: John Murray, 1912. archive.org.
Oldenberg, Hermann. Prolegomena zur Metrik und Textgeschichte der Ṛgveda. Berlin: Wilhelm Hertz, 1888.
Om Prakash. Food and Drinks in Ancient India: From Earliest Times to c. 1200 A.D. Delhi: Munshi Ram Manohar Lal, 1961. archive.org.
Renou, Louis. Études védiques et pāṇinéennes. 17 vols. Paris: Institut de Civilisation Indienne, 1955-1969.
Witzel, Michael. “Early Indian History: Linguistic and Textual Parameters.” In The Indo-Aryans of Ancient South Asia, edited by George Erdosy, 85-125. Berlin: de Gruyter, 1995.
Witzel, Michael. “The Linguistic History of Some Indian Domestic Plants.” Journal of Biosciences 34, no. 6 (2009): 829-833. Springer.
Comments (0)
No comments yet. Sign in to start the discussion.