Elephants that attend her, cavalry that encircles her

Sampatkarī and Aśvārūḍhā are named goddesses in the Lalitopākhyāna, not personifications invented for the verse, and both have an independent life in Śrīvidyā practice, with their own mantras and their own place in the ritual invocation of Lalitā's retinue. Their function in the narrative is military and specific: one commands the elephant division, the other the cavalry. The retellings give names to their mounts as well, though such details vary between versions and are the first thing to shift in transmission. What does not vary is the structure — the goddess does not fight alone, and the powers that fight for her are individuated, named, and given commands.

The classical Indian army had four arms, and the word for it, caturaṅga, means simply four-limbed: elephants, chariots, cavalry, and foot. The word travelled westward with the game modelled on that army and became, by a long chain of borrowings, the English word chess. Naming the elephants first is conventional and reflects their actual standing, since an elephant corps was the most expensive and the most prestigious component of a kingdom's force, the thing a ruler was assessed by. That standing is why the goddess who leads them is called the maker of prosperity. The association between elephants and wealth is not decorative in Indian thought: Śrī herself is shown flanked by elephants pouring water over her, and the elephant is the standing emblem of royal abundance. Sampad, the word inside her name, covers prosperity, attainment and abundance in the broad sense, and pairing it with the elephant division makes the point that the wealth of a kingdom and the force of a kingdom were, in practice, the same asset.

The vocabulary of the first name repays attention. Sindhura is one of the many Sanskrit words for elephant, and vraja is a herd or pen — the same word that gives Vraja, the cowherd country of Kṛṣṇa's childhood, a place named after its cattle enclosures. Sevitā, waited upon, is from the root sev, to attend or serve, which also gives sevā, the ordinary word for service in a temple or to a teacher. So the elephants are not merely present; they are described in the vocabulary of liturgical attendance.

The second name is built out of the same materials and does something different with them. Adhiṣṭhita, presided over, is formed from the root sthā with the prefix adhi, over — the same root that supplied the words for abiding and standing in the verses about her dwelling, now shifted from location to command. Koṭi is a real number, ten million, the largest of the older Indian numerals in common use, and doubling it into koṭikoṭi is an idiom for a quantity past counting rather than a vague gesture at largeness; the tradition prefers hyperbole that has been arithmetically specified. And āvṛtā, surrounded, comes from the root vṛ with the prefix ā, to cover or enclose — which is precisely the root behind āvaraṇa, the technical term for the nine enclosures of the Śrīcakra. The word chosen to say that cavalry rings her is the word used for the concentric rings of the diagram. Whether the hymn intends the echo or simply reaches for the natural word, the effect for a practitioner is the same: the army in formation and the yantra on the altar have the same shape, a centre with rings around it, and the enclosing is done in both cases by her own powers.

One textual matter. The nāmāvalī transmitted with this text writes the elephant word as sindura rather than sindhura, dropping the aspiration. Sindūra is a genuine Sanskrit word — it means vermilion, the red lead used as a cosmetic and in worship — but it is not what the compound requires, and the reading is most likely a slip of transmission rather than a variant with meaning behind it. It is the sort of thing worth knowing about before comparing two printed texts and concluding that one of them is wrong about something important.

These two names are the best-matched pair in this part of the hymn, and the efficient way to learn them is together. Both open with a goddess's name, continue with a participle meaning mounted, then give the animal and the multitude, and close with a participle describing what that multitude does. Hold that frame in mind and the sixteen syllables stop being a wall.

The joints of the first name are sampatkarī, samārūḍha, sindhura, vraja, sevitā; of the second, aśvārūḍhā, adhiṣṭhita, aśva, koṭikoṭibhiḥ, āvṛtā. The second involves three sandhis, all worth seeing rather than memorising. Long ā plus short a gives long ā, so the join in aśvārūḍhādhiṣṭhita is silent; two short a's likewise give one long ā, giving adhiṣṭhitāśva; and a visarga after i, before a voiced sound, becomes r, hence koṭikoṭibhir āvṛtā.

That last point carries a grammatical fact: this name is not one word. The compound ends at koṭikoṭi, in the instrumental plural, and āvṛtā is a separate participle governing it. Knowing where the boundary falls makes the name easier to hold.

The consonants are the hardest in this stretch of the stotra. Ḍh, the voiced aspirated retroflex, appears in both names: tongue curled back to the roof of the mouth, voiced, with breath released after it. Ṣṭh in adhiṣṭhita is retroflex sibilant into aspirated retroflex stop, both with the tongue back, and is the most difficult conjunct here. Śv appears twice, palatal sibilant into v. Ndh in sindhura is dental n into aspirated dh. Vr in vraja is v into a full r, and vṛ in āvṛtā is v into the vocalic ṛ, itself a syllable.

Sampatkarī conceals a discrepancy between how Devanāgarī writes the name and how it is said. The script groups it as sa, mpa, tka, rī, but the syllables are sam, pat, ka, rī: the m closes the first syllable and the t the second. Written conjuncts often straddle a syllable boundary in this way, and a learner who treats the written units as spoken ones will produce something unrecognisable.

Koṭikoṭi is worth noticing as sound and not only as arithmetic. The o is long and the i of ṭi is short, so the four syllables come out long, short, long, short over a repeated retroflex ṭ. In a name about cavalry it falls into a gallop, and there is no reason to resist that.

For archanā both names take the ordinary dative: sampatkarīsamārūḍhasindhuravrajasevitāyai namaḥ, aśvārūḍhādhiṣṭhitāśvakoṭikoṭibhirāvṛtāyai namaḥ. If a printed nāmāvalī gives sindura, that is the transmission slip described above rather than a choice to be made.

What the verse asks is symmetry. Two names of identical length and near-identical construction, delivered at the same weight, produce the effect the poet built — two divisions parading past in matched order, and rushing the second because the first was hard destroys it. There is one thing to hold while chanting that second name: the word saying the horses surround her is the word used for the rings of the diagram, so the name tells you what those enclosures are made of.