> In addition to being a genitive, no was in OJ, and still is in NJ, an adnominal form of the copula. The genitive function of no seems to derive in pre-OJ from the function as adnominal copula, but at the OJ stage both functions of no were firmly established and fully independent. (A History of the Japanese Language, Frellesvig 2010, p.126)

Oh, you know how sometimes AであるB can be replaced with AのB?

Various linguists trace that back to an old copula with forms like に and の and にて ＞ で

Well, we've been sort of discussing it on and off over a period of months :-)

There's a couple older copulas, the n- one and the t- one

You'd recognize the difference from にある＞なる and とある＞たる, and from にして versus として

Martin call the n- one "objective" and the t- one "subjective"

It seems like に was originally a verb form, which explains why the conjunctive て was able to attach to it (as you know, て normally only attaches to verbs!)

So it's possible there was an older copula verb in pre-Japanese that it is a form of. I linked to a paper which suggested it was *ぬ "locative BE"

Probably not

I put a star next to it to indicate that it's unattested / reconstructed :-)

It may not have existed

But it doesn't seem to have a semantic relationship to いぬ・しぬ・ぬ, does it?

Hmm. True!

I like being able to make this all make sense in my head, fit into a bigger picture, although I'm afraid not all the conjecture I've internalized may be right :-)

No, a pre-Japanese copula which had already vanished (at least in that form) by Old Japanese

Most people agree on what that て is, but it seems odd to suggest that て attaches to a particle there and almost nowhere else

Especially since つ, from which て is thought to be derived, never attached to に

So the reconstruction of *ぬ is an attempt to address that problem

Explaining how て could attach to に

The idea is that it was reanalyzed from a locative verb form to a copula form

And that this took place before Old Japanese was first written down

Yes, this relates locative に and essive/copular に as originally the same thing

Vovin analyzes no and ni as forms of a defective pre-Japonic copula n-

Frellesvig says that the genitive no seems to have come from the adnominal copula no

No doesn't look (morphologically) like a verb form

To doesn't look like a verb form either. (Where did copular to come from?)

vsarpj.orinst.ox.ac.uk/files/frellesvig.EALS13.presentation.pdf

Hmm. In modern Japanese, の can also often be paraphrased with にある, with a locative meaning: 大阪の家　⇔　大阪にある家

I need a way to absorb books into my brain faster :-)

@ssb I think this Frellesvig 2010 book would be a good place to start :-)

Yes, -kyer- "modal past" < *-ki-ar-

He mentions it in his book :-)

Page 48

He gives two glosses, so perhaps he doesn't consider it clear which ki it is (I've seen previous discussion of this, but I've forgotten it already)

Oh, sorry, I don't have that open anymore

Incidentally, the case he builds for ki/si as a suppletive adjectival copula fit pretty well with Nishiyama's synchronic analysis where i is a suppletive form replacing ku ar-u most of the time

That makes sense, it gives a source for the voicing

Ah, here's the basic idea

Adjectives derive most of their forms from the analytic 〜く＋ある

So you get e.g. past 高かった　＝　高く＋あった taka-k'atta

When you insert a particle, you can see it clearly: 高くもあった・高くさえなかった

With just a few frozen forms

The main exception in modern Japanese is that we don't say 高くある, we say 高い

But there are a couple situations where 高くある is possible, too

As before 〜べき

The analysis I was giving is mostly a synchronic one, but it fits decently with the history

It's just a little more complicated in a historical perspective

い comes from き of course

The /k/ was lost

I'm not sure I follow--isn't loss of /k/ common in western dialects? As in ありがたく　＞　ありがとう

Yeah

Okay, I suppose I understood you backwards

Nishiyama's synchronic analysis likens く to で, so we have く＋ある and で＋ある

Which can fuse into い and だ, if I recall correctly how he phrases it

Markdown is disabled on multi-line messages

Probably because of programmers pasting code :-)

Stack Exchange chat is silly.

I have no explanation

Yeah, the 未然形 is viewed by many linguists as an artifact of analyzing Japanese at the mora level and not below

@EiríkrÚtlendi Identical in what sense? -(a)n- and -(i)n- seem to be different in form and meaning

Yeah, I see you're going with the traditional grammar analysis there

Although I have no problem with analyzing -a- as a stem, either

Frellesvig seems to suggest that originally it was part of a negative suffix *-an and the a at the beginning was reanalyzed as the mizenkei (in pre-Old Japanese?)

I'm going to have to read this more carefully later

This is a pretty interesting book, by the way :-)

Probably if I read through in order instead of flipping through I'll be less confused

Rest well!

@EiríkrÚtlendi Yes, Frellesvig thinks that perfective ぬ is related to this pre-OJ copula n-