Hello everyone! I thought I'd make a little post about the history of a few personal pronouns & personal determiners in the Emaic languages. First I'll talk about Proto-Emaic, and then move on to three different modern Emaic languages belonging to the three different primary branches of the Emaic family tree.

You might wonder what "personal determiners" are. Just like any other determiner, they provide more information about the noun phrase it occurs with. In particular, the Emaic 1st and 2nd person personal determiners specify that the noun phrase refers to the speaker(s) or listeners(s). In English, a similar effect can be achieved by apposition, e.g. "You, my son, will accomplish something great one day". Sometimes "we" and "you" are also used like determiners, e.g. "We conlangers share a great hobby". In the Emaic languages it's very common to use a determiner together with a title or kinship term instead of a simple pronoun. In some languages like Atłaq the determiners are actually a lot more common than the pronouns.

In this post I'll only be talking about the 1st and 2nd person. This is because the 3rd person forms are identical to distal demonstratives both in Proto-Emaic and in most descendants, and the demonstratives are still a work in progress. I'm also restricting myself to talking about the basic forms of the pronouns (i.e. the absolutive/nominative depending on the language). I don't want this post to be too long, so let's get into it.

Proto-Emaic

There are several things to notice here. First, the 1PL pronoun and determiner are the only ones that are identical. The other pronouns look as if they're derived from the corresponding determiners by means of suffixes *-ta and *-i. In that case, the 2SG pronoun *uhi has an epenthetic *h, which is expected as Proto-Emaic disallows vowel hiatus. The change from *β to *b in *reʰbi is an example of the DZ-alternation, where many word-final voiced continuants turn into lenis stops after the addition of a vowel-initial suffix.

It's not known what *-ta and *-i meant, if they even existed. It's equally likely that the determiners are originally reduced forms of the pronouns. This would be especially likely if they cliticized to the following noun, which they do in many of the modern languages.

Atłaq

All four determiners are the expected reflexes of the Proto-Emaic forms. The pronouns, less so. All of them end in the clitic =š, which is extremely common in Atłaq and has a range of different functions. You can read more about it here (it's only slightly outdated).

The 2PL pronoun ziv-š is by analogy with the determiner, as the expected reflex of *reʰbi would be ^xzib [jɪp]. Similarly, na-š is an alternative variant of natła-š, also by analogy with the determiner. It's used mostly by younger speakers in casual speech.

Zikkou

In Zikkou the pronouns and determiners have all become formally identical with each other. The 1SG and 2PL pronouns were replaced with their corresponding determiners, while the 2SG determiner was replaced with the corresponing pronoun. A 2SG determiner o (< *u) can still be found in some set phrases, however.

Bǫgal

In contrast with Zikkou and Atłaq, none of the pronouns or determiners have been replaced. In fact, all forms are the expected reflexes of the Proto-Emaic counterparts, with the exception that rùuf should've gotten a high tone instead of the actual low. This might either be by analogy with rìi or an irregular tone assimilation from an earlier LH sequence.

Complications with the reconstructions

One issue with the Proto-Emaic form *reʰbi is that it's only attested in a single primary branch, and those forms could go back to either *reʰbi or *reʰpi. The only reason *reʰbi is chosen is because of the DZ-alternation which presumably takes place between the *b and the *β in *reʰβ. On the other hand, if *reʰβ is really a reduced variant of the pronoun then *reʰpi seems basically as plausible as *reʰbi. There have also been a few proposed alternations between fortis stops and voiced continuants in Proto-Emaic, such as between *kuʰtu "four-legged animal" and *kuʰl "four". Maybe we shouldn't be so quick to discard *reʰpi as the original form after all.

As we've seen there's been a lot of analogy between the pronouns and determiners over the years. We can therefore ask how we know the Proto-Emaic 1PL pronoun really was *hamɡāʰl and not, for instance, something like *hamɡāʰli or *hamɡāʰlda (running with the idea that the other pronouns had *-ta/-i). Maybe it was replaced by the determiner independently in the different branches? After all, without the branch Bǫgal belongs to we'd reconstruct the 2PL pronoun as *reʰβ, the same as the determiner. Well, that's certainly a possibility. But we have to work with the evidence we actually have, and that evidence tells us that it was *hamɡāʰl.

The end

thanks for reading! I hope this was understandable, as I sometimes get lost in the weeds and forget to adequately explain the bigger picture. Anyway, happy conlanging!