Danu

She gets one verse.

Rigveda 1.32 is a hymn about Indra, fifteen stanzas long, attributed to a poet of the Angiras clan named Hiraṇyastūpa. Indra fills himself with soma, takes up the bolt his smith made for him, and kills the serpent Vritra, who has been holding the world’s waters shut. The rivers break loose and run. It is the central story of the older Vedic religion, and it is told with a lot of relish.

In the ninth stanza the poet looks down at the battlefield and notices someone else lying there. Wilson, following the medieval commentator Sāyaṇa, renders it:

The mother of Vṛtra was bending over her son, when Indra struck her nether part with his shaft; so the mother was above and the son underneath, and Dānu slept (with her son), like a cow with its calf.

Griffith’s 1889 version keeps the same picture and drops the bracket: the mother was above, the son was under and like a cow beside her calf lay Danu.

That is the whole of her, in the text where she is oldest.

Appearance

There isn’t one. No hymn describes her face, her form or her dress, and there is no Vedic image of her to describe. What the verse gives is a position: above her son, then beside him, in the posture of a cow lying with a calf. In this poetry a cow with her calf is the ordinary domestic image of care, which is a strange thing to reach for at the end of a killing, and the poet reaches for it anyway.

The word

Her name is a common noun. In Vedic Sanskrit dānu means fluid, drop, dew, rain. The water word itself, used as a name.

Sāyaṇa, writing his commentary centuries later, already offered two ways to take it: from a root do, to cut or destroy, or simply as the name of the wife of Kaśyapa who bore the Danavas. The ambiguity is not something modern scholars introduced. The oldest surviving reader of the verse was already unsure.

In Avestan, the closest relative of Vedic on the language tree, the same word means river.

The rivers

Follow the word outward and it turns up across the map. Philologists reconstruct a Proto-Indo-European root, usually written *déh₂nu-, for running water, and read it in the Don and the Donets, in the Dnieper and the Dniester, in the Daugava, and in the Latin Dānuvius that became the Danube.

How the Danube got its name is not settled. It may be Celtic, or it may be an early Scythian loanword that Celtic picked up, and specialists hold both positions.

A poet in the Punjab, three thousand years ago, named the mother of the serpent who withheld the waters with the word for water. Speakers moving west along the great rivers used that same word for the rivers themselves. Neither the goddess nor her story went with it.

Her children

The Danavas are hers, a line of asuras, and Vritra is the one everybody remembers. He is the drought in serpent form: he lies on the mountain, the waters stand still behind him, and the ninety-nine strongholds hold until Indra breaks them. People who needed the monsoon sang this.

Calling the asuras devils imports a later quarrel. The hard opposition between deva and asura sharpens over centuries, and in the older layer the two words name two kinds of power rather than two moral sides. Danu is the mother of one of those kinds.

The family tree usually quoted for her, in which she is a daughter of Prajāpati Dakṣa and Asiknī and a consort of the sage Kaśyapa, comes from the Puranas and the epics. Those texts organised a crowd of older, looser figures into one genealogy. It is a real tradition and a much later one, and none of it is in the Rigveda.

The Irish question

Anyone who has met the name before has met it in Ireland, as Danu the mother of the Tuatha Dé Danann, and the temptation to run the two together is strong.

The nominative Danu does not appear in any medieval Irish text. The form that survives is the genitive Danann, in the phrase Tuatha Dé Danann, and Danu is a reconstruction that modern scholars made by working backwards from it. Anu, genitive Anann, may be the same name in another shape.

The Vedic connection was popular with nineteenth-century scholars and has been contested ever since. Eric Hamp argued in 2002 that the water etymologies are wrong and that the Celtic name comes instead from the root behind Latin bonus, older duenos, giving a Proto-Celtic form meaning something like aristocrat. On that reading the Irish name has no water in it at all. John T. Koch, working on the Welsh Dôn, derives her from the Indo-European word for earth rather than the one for rivers.

The river names really do share a root. Whether the Irish mother-goddess belongs to that family is a separate question, and it has not been answered.

After the Rigveda

The later handbooks did not leave her that bare. In the Puranas and the epics she has a father, a mother, a husband and a long list of named Danava sons, all fitted into the genealogy of Kaśyapa’s wives that those texts use to account for where every class of being came from. Kinsley’s survey of the Hindu goddesses still treats her as a background figure even there.

Sources

Bibliography. The same list is held in the article’s frontmatter for the citation tools that read it programmatically.

• Rigveda 1.32, attributed to Hiraṇyastūpa Āṅgirasa (hymns dated c. 1400-1000 BCE)
• H. H. Wilson, Rig-Veda Sanhitá, English translation (1866), following Sāyaṇa’s commentary
• Ralph T. H. Griffith, The Hymns of the Rigveda (1889)
• David Kinsley, Hindu Goddesses: Visions of the Divine Feminine in the Hindu Religious Tradition (University of California Press, 1987)
• David Leeming and Jake Page, Goddess: Myths of the Female Divine (Oxford University Press, 1994)
• Eric P. Hamp, 2002 proposal deriving Celtic *Danu from Proto-Celtic *Duonū rather than from the water root (article title unverified)
• John T. Koch, on the derivation of Welsh Dôn from Proto-Indo-European *dʰéǵʰōm