Wiktionary talk:About Proto-Bantu

Noun prefixes

@Metaknowledge I think we can treat these the same as we do with PIE. We can list the full word if it's reconstructable and clear, and list just the root otherwise. The issue with listing just the root is that a language can have multiple derivations from a root, and we'll want to separate them. For example, I doubt that Zulu isintu has a Proto-Bantu ancestor, but umuntu very clearly does. Furthermore, we can be sure that Proto-Bantu had at least one noun from a particular root, since bare roots didn't exist. If we follow the usual principle of "reconstruct all you can", then that speaks for including the prefixes. Likewise, we'd include tone marks if known, omit them if not. —CodeCat 21:50, 13 November 2016 (UTC)

Thanks for making the page. About roots: isn't in somewhat untrue to say that in a language which has switched all class 6 words over to class 2, that its class 2 word for "water" comes from the class 6 PB word? Its root comes from the PB root, and its NC prefixes come from the PB prefixes, but it doesn't assign the same prefixes to the root that PB did. (Oh, and about orthography, you'll have to find any outstanding links to PB forms and correct them as needed.) —Μετάknowledge^{discuss/deeds} 21:55, 13 November 2016 (UTC)

That's a wholesale grammatical shift, similar to replacing inherited plural -en with -s in English. It can also be compared to the loss of a gender, where, as in most Romance languages for example, the neuter gender might be shifted to the masculine gender. In Romance, it can be argued to be the result of loss of a phonetic distinction in the nouns, but not in the Eastern Baltic languages where the merger was accompanied by a change in morphology as well. In any case, we can still list them as descendants but make a note of any non-predictable morphological changes. This is what we do for some PIE entries as well (but not all, e.g. Proto-Indo-European *mḗms does not naturally evolve into Proto-Germanic *mimzą, there's a shift to a thematic inflection). —CodeCat 22:17, 13 November 2016 (UTC)

Okay, that convinces me. We could also use {{bnt-noun}} and a template for linking to BLR3 for referencing the reconstructions. —Μετάknowledge^{discuss/deeds} 22:35, 13 November 2016 (UTC)

I created a barebones {{R:bnt:BLR}}. Ideally the template would link to the proper entry, but I don't know how to do that. —CodeCat 14:10, 15 November 2016 (UTC)

Each page has a unique number that can be used to link to it; you could make that number the first argument. For example, -kú- is 2089, so you can link directly thus. —Μετάknowledge^{discuss/deeds} 18:53, 15 November 2016 (UTC)

Ok, done. —CodeCat 19:06, 15 November 2016 (UTC)

I've created entries including the prefixes, for those classes where I felt confident enough about what the prefixes were in Proto-Bantu, following the Wikipedia article. But some are less clear and so we're still missing those: 5, 8, 9, 10, 11. @Metaknowledge Would you be able to provide these? —CodeCat 20:11, 18 November 2016 (UTC)

5 is *ì-, 8 is *bì-, 9 and 10 are *ǹ-, and 11 is *dʊ̀-. Are you sure you don't want me to send you some reading? —Μετάknowledge^{discuss/deeds} 20:22, 18 November 2016 (UTC)

Where does Zulu's class 5 li- come from then, I wonder? —CodeCat 20:33, 18 November 2016 (UTC)

Also, do class 9 and 10 really have a toned consonant? —CodeCat 20:41, 18 November 2016 (UTC)

I haven't studied Zone S, but I think that li- comes straight from *ì-. Nasals can behave as vowels; syllabic nasals are an extremely common feature of Bantu languages. —Μετάknowledge^{discuss/deeds} 05:35, 19 November 2016 (UTC)

Sorry, I was mistaken — Zulu li- comes from the PB PPx. In class 5, the PPx widely replaced the NPx, apparently at least partly due to its odd shape (V- when all the other prefixes are CV-). —Μετάknowledge^{discuss/deeds} 04:46, 20 November 2016 (UTC)

I'd like to know more about the other concords as well. I don't think the sources you gave me really elaborated on their exact forms, though I may have missed it.

There's also the question of the augment, which is widespread in Nguni, but oddly only partially in Swazi. Rwanda-Rundi seems to have it as well, despite being very far away. Did it exist in PB already in some form? —CodeCat 13:05, 20 November 2016 (UTC)

That's all in the sources! You have a lot of reading to do. :) Check out the 'Historical linguistics' section in Nurse & Phillippson (I think Schadeberg wrote it) for those. Most of that section is based on the BGR, which is Meussen 1967, so you might as well go straight there. However, Meussen has some guesses and opinions that are no longer in favour, and though his conceptualisation of augments arising from PPx being used as pronouns seems sound, his idea that such a usage was standard in PB is questionable. —Μετάknowledge^{discuss/deeds} 19:28, 20 November 2016 (UTC)