Imagine you have a cake, Swedish style birthday cake, two layers of sponge cake, since cut up in the middle, and strawberry jam between the two, strawberry jam and whipped cream on top, and a few whole strawberries on top of that as decoration. Plus candles.

Ooops, it wasn't a Swedish birthday, it was a US American wedding. Take off candles and strawberries. Scrape off all cream and jam. Now add rhum, add some buttercream with chocolate and hazelnut flavour, make another storey or two, add glazing, add a doll of a wedding couple on top. Done.

But the sponge cake is still sponge cake. You didn't turn it into pancake or puff pastry.

Or imagine you have a house that is being constantly repainted, from one year to ten years later, the painting may have changed from a mix of blue, green and purple colour fields to a mix involving mainly red, yellow and orange.

And a pair of trousers can be remade (as I'm known to do) adding more and more stuff to replace the holes and widening the knees that way and taking stuff from lower legs to make insertions of Renaissance type or mendings at the butt.

That doesn't start turning the trousers into a house or the house into trousers.

If I bring up that language can't have developed from ape like communications (shrieks, grunts etc) I have reason to fear someone will bring up the absolute facepalm in this situation : stating that linguists deal in language development.

Now, there are other studies in linguistics (as linguistics proper) than historic linguistics, but historic linguistics does indeed deal with one thing that is often nicknamed "language development" - like the "development" from Latin to French. But this has no bearing on developing human language from non-human communications any more than changing the glazing could make your sponge cake into a puff pastry.

Between Latin and French, language of Plautus and language of Molière, we can suppose there were c. 20 or 30 intermediaries. But each one would be perfectly viable, just as the house remains habitable while you repaint it. It is a bit hard to know for each intermediary - like the one between 450 and 550 and by peasants in Gaul - which changes were already made and which weren't. But it's not hard to imagine how the language remained viable through the changes. I'll give you one, from Latin six case system to Old French two case system, in the masculine type declinsion, i. e. II declinsion:

bonus		bonos		bonos		(later bons)
boni		boni/bone		-
bono		bono		bono		(later bon)
bonum		bono		bono
bone		bone		-
bono		bono		bono

You start out with nominative, genitive, dative, accusative, vocative and ablative-instrumental. You get nominative still there, replacing also vocative, and you get accusative taking the place of dative, ablative and even genitive.

In some declinsions, distinguishing between nominative and accusative becomes hard, like first declinsion, nominative bona, accusative bonam ... bonam becomes bona, and later bona becomes bone / bonne. Now, Latin not only had nominative and accusative to distinguish the doer and the done-to, they also had (optionally) word order. In some cases, like neuters or plurals of third declinsion, the option became obligatory, and the development of "bonne" just added to the situations where this was so.

All other types of change similarily left the language useful as human communication along the way too. It's like changing the glazing.

But there is a fact which is true of all human language, and of no bestial communication, including that of apes. A message is subdivided into morphemes. In Greenlandic, there might be one lexical morpheme and a lot of endings on that, in Chinese all morphemes have nearly the status of lexical items - or separate words (not quite true, some words need certain positions to be used as grammatical morphemes, where real lexical ones would be used in other positions) and in French or Latin or English or German you have a situation in between. But the message is subdivided into morphemes. And a morpheme is subdivided into phonemes, that, unlike morphemes, don't mean anything on their own. Changing from message = morpheme = phoneme to this is like trying to turn sponge cake into puff pastry or trousers into a house. It simply won't work. There are no intermediate situations imaginable that this would work with.

Hans Georg Lundahl
Paris
Sts Zachary and Elisabeth
father and mother of the
precursor
5.XI.2021

Sancti Zachariae, Sacerdotis et Prophetae, qui pater exstitit beati Joannis Baptistae, Praecursoris Domini. Item sanctae Elisabeth, ejusdem sanctissimi Praecursoris matris.