Devangari conjunct formation: which ja + nya + ca conjunct is correct?

James Puckett

Below are two outcomes for forming the Devanagari conjunct ja +halant + nya + halant + ca. Which is correct? The two on the left use features code compiled by Glyphs, the two on the right are Murty Hindi and Adobe Devanagari.

Kent Lew

I’m no expert, but I definitely think you need to use the half-jña form, as I believe the jña has nearly independent status, like kśa.

So, one of the forms on the right. Not sure why the Murty doesn’t align. John would have to say what the rationale was.

Kent Lew

And I would guess this is an *extremely* rare conjunct. But I suppose someone must have an attested example?

James Puckett

Thanks Kent. That’s what I thought, but I’ll do some more digging to be sure.

You are right about it being rare. I’m not sure this would ever appear in text, but since there are different approaches t

John Hudson

James, this is beyond rare: it doesn't occur in any of the sources from which we've compiled our Sanskrit conjunct set. This is why, as is evident in your screenshot, the conjunct is being formed with a half form of JNy-, none-to-elegantly connecting with the full Ca.

I can confirm, though, that if this or similar conjunct were to occur, then it absolutely should form with the JNya shape, as this, with KSsa, is one of the two pseudo-letters whose shaping always takes precedence. This is the whole point of the <akhn> feature: to form the full forms of these pseudo-letters as an initial shaping step, whereafter they behave like letters, e.g. taking a half form.

The Glyphs shaping is simply wrong, because the J- shouldn't be separable from the -Ny- at any stage after the <akhn> feature.

Georg Seifert

The problem is that the fonts doesn't contain a j_ny glyph. And I added that to the feature code generator.

karambir S Rohilla

Hi
Last two option is wright conjuncts.

karambir singh rohilla
fontdesigner@gmail.com
font designer
India