Akana and the comparative method

[Nortaneous]

EDIT: looks like the voiceless nasals are reconstructed as *hN sequences. There’s also a series of reconstructed onsets *bʔ *dʔ *gʔ, which look like they could be the prenasalised series — especially since the grammar mentions a connection between the prenasalised and ejective consonants. In that case, Zju’s reconstruction does correctly find all the consonant series… but without their correct realisations, just as I predicted.

The idea was *mp > pʼ based on Sotho *mp *mb > pʰ pʼ.

[WeepingElf]

Is there any reason why this thread is here rather than in the Akana section?

I thought it would be best suited to here, since it’s really about studying the comparative method itself. The Akana aspect is incidental. But presumably it can be moved if it would work better there.

I understand.

[Man in Space]

1. Fantastic idea.
2. I think I do, actually, have it around somewhere. (EDIT: I saw you already did but misread which relay you referred to—I actually have Proto-T1 from 2014. I can share that, but it might also make a good reconstruction effort since no one ever really tackled it IIRC?)
3. I have further remarks but will save them for when I have the time to write them out.

[Man in Space]

OK. So, a few further thoughts:

1. Are we going to put a blanket declassification on Akana protolangs? For all or for only a few (q.v.)? I suppose the "statute of limitations", for lack of a better term, is expired on all but the most recent.
2. Do we want to maybe open a challenge to the forum at large to reconstruct the languages from the last relay or two? Might be something interesting for the board to do, kind of like a once-off event, and might generate some interesting data.

[bradrn]

1. Are we going to put a blanket declassification on Akana protolangs? For all or for only a few (q.v.)? I suppose the "statute of limitations", for lack of a better term, is expired on all but the most recent.

I sort of assumed they were all declassified already, hence my making this thread.

2. Do we want to maybe open a challenge to the forum at large to reconstruct the languages from the last relay or two? Might be something interesting for the board to do, kind of like a once-off event, and might generate some interesting data.

I thought they were reconstructed already?

[Man in Space]

1. Are we going to put a blanket declassification on Akana protolangs? For all or for only a few (q.v.)? I suppose the "statute of limitations", for lack of a better term, is expired on all but the most recent.

I sort of assumed they were all declassified already, hence my making this thread.

Ah. I was on T1 of the 2014 relay but IIRC there never was much of an attempt at it for one reason or another? Anyway it's still referred to as "T1" on the wiki. I have the original documents for the protolang (and the ones for Teyetáti if not Tked).

2. Do we want to maybe open a challenge to the forum at large to reconstruct the languages from the last relay or two? Might be something interesting for the board to do, kind of like a once-off event, and might generate some interesting data.

I thought they were reconstructed already?

I think some aren't as much as others (especially the last one which I really fell apart on after making the proto…my health and job difficulties frustrate me to no end). (And again, cf. Proto-T1 above.)

[bradrn]

1. Are we going to put a blanket declassification on Akana protolangs? For all or for only a few (q.v.)? I suppose the "statute of limitations", for lack of a better term, is expired on all but the most recent.

I sort of assumed they were all declassified already, hence my making this thread.

Ah. I was on T1 of the 2014 relay but IIRC there never was much of an attempt at it for one reason or another? Anyway it's still referred to as "T1" on the wiki. I have the original documents for the protolang (and the ones for Teyetáti if not Tked).

Ah, if some families never got reconstructed, it makes sense that the protolang was never revealed. In that case it might be interesting for people here to have another go at reconstructing them.

[kodé]

The idea is this: we now have ~15 years of reconstruction relays from the Akana project.

15 years?!

Of course I wouldn’t really know, but I got the impression it started around 2010.

I think it was 2006? I have some vivid memories of working on deriving a daughterlang from Gezoro during Thanksgiving 2007, and I think the relay had been going on for about a year at that point. So around 17 years or so, then !

[bradrn]

Thinking about this again… does anyone know if there’s any list of which protolangs have been reconstructed and which ones haven’t? It would help me keep track of things, at least.

[Corumayas]

This is a neat idea.

15 years?!

Of course I wouldn’t really know, but I got the impression it started around 2010.

I think it was 2006? I have some vivid memories of working on deriving a daughterlang from Gezoro during Thanksgiving 2007, and I think the relay had been going on for about a year at that point. So around 17 years or so, then !

The first reconstruction game actually happened in 2005; I got involved during the "cursed" derivation relay the following year (around Thanksgiving 2006, as it happens). So Akana will turn 19 this year...

Thinking about this again… does anyone know if there’s any list of which protolangs have been reconstructed and which ones haven’t? It would help me keep track of things, at least.

There's a list of reconstructed languages on the Akana wiki: https://web.archive.org/web/20210927012 ... _languages. But it's likely that more recent ones just didn't get listed there. (For example, there's also a wiki page for the reconstruction of Proto-Leic, though it doesn't seem to have gotten very far: https://web.archive.org/web/20240317161 ... Proto-Leic)

Although I haven't followed the more recent games as closely, I think the reconstruction of Proto-Dumic might be the most successful and complete to result from one of these games. The reconstructed grammar and lexicon were published on the wiki, as was the actual proto-language grammar—with a link to WeepingElf's original work-in-progress text file, which includes a lexicon—and a list of the sound changes actually applied in each daughter, which can be compared with the parallel list in the reconstructed grammar. All of that (as well as the daughter language descriptions) can be found here: https://web.archive.org/web/20210927000 ... _languages

It also happens that one of the forum pages that got saved on the Internet Archive is the Dumic reconstruction thread: https://web.archive.org/web/20160426183 ... e97a8e24bd
(And the google sheet that thread linked to is even still online, too: https://docs.google.com/spreadsheets/d/ ... li=1#gid=0)

[bradrn]

This is a neat idea.

Glad you’re interested!

(And welcome back to the board!)

There's a list of reconstructed languages on the Akana wiki: https://web.archive.org/web/20210927012 ... _languages. But it's likely that more recent ones just didn't get listed there. (For example, there's also a wiki page for the reconstruction of Proto-Leic, though it doesn't seem to have gotten very far: https://web.archive.org/web/20240317161 ... Proto-Leic)

Although I haven't followed the more recent games as closely, I think the reconstruction of Proto-Dumic might be the most successful and complete to result from one of these games. The reconstructed grammar and lexicon were published on the wiki, as was the actual proto-language grammar—with a link to WeepingElf's original work-in-progress text file, which includes a lexicon—and a list of the sound changes actually applied in each daughter, which can be compared with the parallel list in the reconstructed grammar. All of that (as well as the daughter language descriptions) can be found here: https://web.archive.org/web/20210927000 ... _languages

It also happens that one of the forum pages that got saved on the Internet Archive is the Dumic reconstruction thread: https://web.archive.org/web/20160426183 ... e97a8e24bd
(And the google sheet that thread linked to is even still online, too: https://docs.google.com/spreadsheets/d/ ... li=1#gid=0)

This is all really helpful, thanks! Dumic sounds particularly promising. I’ll have to look through these when I get time.

[bradrn]

OK… looking through Dumic, this is extremely impressive. Practically everything is reconstructed correctly. I can find only very minor deviations, along the lines of:

• The reconstruction lists */b d ɡ/ for /w ð ɣ/, though they do suggest that ‘/b d g/ were probably realized more like fricatives [β ð ɣ]’
• Numeral ‘one’ is *kada for actual kara
• The reconstruction merges mood and evidentiality into one verb slot
• ‘hair’ is inexplicably reconstructed as *piNti rather than ðamki

So, this does confirm that the comparative method can get close to 100% accuracy, in ideal conditions. But on the other hand, it’s not such an interesting case for me… I’m interested in finding out when and in what ways it fails.

[Ketsuban]

• ‘hair’ is inexplicably reconstructed as *piNti rather than ðamki

It looks like the "hair" etyma for Jouki Stəy (đtəy) and Wokatasuto (tidoki) confused them; the relationship to *ðamki is far from obvious to me. (The latter in particular seems like it may descend from a compound; the reconstruction ends up listing -doki as an "unknown formant", and punts on Jouki Stəy entirely.)