Yantra: Form as a Tool

Here's the distinction that gives it away: sound is temporal, form is spatial. A mantra exists only while you're making it — it's a stream, it vanishes the instant you stop, you have to keep generating it. A yantra sits there. It persists. And because it persists in space, it can do the one thing sound structurally cannot: show you a whole structure at once. You can't hear a totality — sound only ever gives you one moment, then the next. You can see a totality. [inference]My reasoning from the sources — mine, not the tradition's.

That's the yantra's job. The bīja gives you the seed — the deity compressed to a single sound, a point. The yantra gives you the full scope — the same deity unfolded into her complete structure, held spatially so the whole field is visible together. Sound gives the seed; form gives the whole. [inference]My reasoning from the sources — mine, not the tradition's.

The elements are named and described in the scholarship. The bindu, the point at the centre, often read as the principle from which all form radiates. Downward-pointing triangles, which authors like Bhāskararāya read as Śakti, against upward-pointing ones read as Śiva or fire. In yantras of Kālī specifically, five triangles at the centre. And around the whole thing an outer square called the bhūpura, the earth-citadel. [attested]A named text, edition, and verse — linked. If it isn't linked, it isn't attested.

That's not a picture of Kālī. I read it as a map of her scope: source at the centre, the descent into form, the thresholds you cross. [inference]My reasoning from the sources — mine, not the tradition's.

The draft I wrote before this one said Tārā's yantra differs in its geometry, and that the difference tracks her different function — the crossing, the guide, against Kālī's dissolution. I went looking for a source describing Tārā's yantra geometry, or contrasting it with Kālī's, and I could not find one. It may well be true. I can't show you that it is, so it comes out of the argument and stays here as a question. [gap]I looked and could not find it. What I searched and where is named.

So the general claim — that the geometry encodes the specific scope of the specific power — is my reading and not a finding. Bühnemann's own introduction warns against exactly this kind of generalisation across traditions, and she is right to. What she gives me is the named structures. What I do with them is mine. [inference]My reasoning from the sources — mine, not the tradition's.

This is what my father has always felt in his practice, and I think he's right: the bīja names her; the yantra shows that this named power is not an isolated transaction but opens onto the whole — the total field, the one Ma. The yantra is the seed drawn as part of the whole house, so you never forget it's one house. [devotional]Personal conviction, not claimed as proof.

So, in the frame of the body-lab: the body is where you work, the mantra is the operation compressed to a seed, and the yantra is that same operation unfolded into its full structure — the persistent, spatial form that holds the totality the seed can only point at. Three distinct jobs, no overlap. [inference]My reasoning from the sources — mine, not the tradition's.

If the yantra really is the deity's structure made visible, then it should be logically linked to the traditional image of the deity — the icon — because the icon is also a representation of that same deity. Two encodings of one structure: one geometric, one figurative. And that link should be consistent — the elements of the yantra should map to the elements of the icon by a repeatable rule, across many deities, not just by a pretty coincidence in one case.

I've looked at Kālī and the correspondence is striking … the central source, the descent, the threshold-crossing show up in both her yantra and her icon. But one case proves nothing — you can always find a rhyme in a single example. The real test is whether the mapping holds consistently across many deities. I haven't run that test rigorously. So I'm naming it as an open question, not asserting it as true. [open]An attribution I cannot yet source. Named so it can be closed, not left implied.