Section II · Gati-citra
Around 1300 CE, the philosopher-poet Veṅkaṭanātha — known to the tradition as Vedānta Deśika — composed the Pādukāsahasram: a single overnight composition of 1,008 verses in praise of the sandals (pāduka) of Raṅganātha at Śrīraṅgam. In its thirtieth chapter (the Citra-paddhati) sit two verses of thirty-two syllables each, written in anuṣṭubh metre, which together form the most spectacular known instance of the knight's tour in poetry: the second verse can be derived from the first by performing a knight's tour over a 4×8 chessboard.
Five hundred years later, Leonhard Euler would publish the first western analysis of the same problem. The Sanskrit example is not merely earlier; it is a fundamentally different artefact — a Hamiltonian path that is a poem, and a poem that is a Hamiltonian path. Below is an interactive reconstruction. Press Step to walk the knight, and watch the second verse assemble syllable by syllable.
स्थिरागसां सदाराध्या विहताकरताम्भुजा ।
स्थलनासंनिवेद्यासि गलगर्दभरांगणा ॥ — Pādukāsahasram, ch. 30 (Chaturaṅga-turaṅga-bandha · verse 1 of the pair). Source verse, written in serial order on the half-board.
The knight begins at the top-left square, syllable 1. Press Step to advance one move. The path traced is a real Rudraṭa-style closed half-board tour; the syllables shown approximate the placement Vedānta Deśika used.
Three lenses on the same artefact
It helps to look at the chaturaṅga-turaṅga-bandha through three lenses, because the tradition treats them as one and modern readers separate them by habit.
The 32 squares of a 4×8 board form a graph; an edge joins two squares iff a knight can leap between them. A knight's tour is a Hamiltonian path through this graph. Vedānta Deśika's tour is one of millions of valid solutions — but its specific permutation is forced, because the second verse must be Sanskrit.
Each verse has 32 syllables, four pādas of eight syllables each — the classical anuṣṭubh metre. The first verse, written in normal reading order, fills the half-board row by row. The second verse, equally metrical and equally meaningful, is the readout of the same syllables re-ordered by the knight's path.
The whole composition praises the sandals of the deity at Śrīraṅgam. That this praise is delivered as a chess-puzzle is not coincidental: in Vedānta Deśika's theology, the world is itself a play of constraint and grace. The artefact says, in shape, what the verse says in word.
Why this matters for computing
The chaturaṅga-turaṅga-bandha is a useful object for software people because it sharpens an intuition we tend to lose. A modern program is generally written for one purpose; we minimise constraints. A bandha is written so that the constraint is the point: the value of the artefact is proportional to how tightly several specifications agree on the same sequence of bits — here, syllables. The cost of finding such a sequence is enormous, but the resulting object is dense: it carries multiple readings in the same data.
Read this way, the project's broader claim — that classical Indian poetry anticipates a "new computing paradigm" — looks less ornamental and more practical. It is a tradition that took multi-objective compression seriously, eight centuries before we built the machines for which we now relearn it.
