Avarin (and other Tolkienian languages)

[WeepingElf]

This has come up in the altlangs thread where it does not really belong (Middle-earth is not an alternative history), so I shall open a new thread for it.

The Ardalambion page on Avarin lists six names of Avari tribes which are all cognates of Q. Quendi: Kindi, Cuind, Hwenti, Windan, Kinn-lai, Penni.

Kindi could be a "satem" language where the labiovelars were delabialized as in IE satem languages, is that the "Baltic-style" language?

Cuind could be the "Gaelic-style" language.

Hwenti seems to have undergone a stop shift analogous to Grimm's Law.

Penni, as Helge (the author of Ardalambion) says, seems close to Telerin, and may be the language of the Wood-elves of Lord of the Rings if these are actually Avari (and not Nandor). I mean the common people, not the royal family who were Sindar.

I have nothing to say about Windan and Kinn-lai.

[Creyeditor]

Did Windan undergo a change from *qu to /w/?

[Frislander]

I think Kinn-lai must be branched with Kindi because they both share the same delabialisation, with Kinn-Lai additionally seeing lenition of the *d (like that seen in Penni) and the accretion of the *-lai element.

[Creyeditor]

I think Kinn-lai must be branched with Kindi because they both share the same delabialisation, with Kinn-Lai additionally seeing lenition of the *d (like that seen in Penni) and the accretion of the *-lai element.

Could it also be that: *nd > nnl in Kinn-lai?

[sangi39]

I think Kinn-lai must be branched with Kindi because they both share the same delabialisation, with Kinn-Lai additionally seeing lenition of the *d (like that seen in Penni) and the accretion of the *-lai element.

Could it also be that: *nd > nnl in Kinn-lai?

Personally, I'd doubt it. I'd assume Tolkien is marking some sort of morpheme boundary with the hyphen, since I can't imagine him using a hyphen within a root morpheme (seems a tad... not his style)

[Creyeditor]

I think Kinn-lai must be branched with Kindi because they both share the same delabialisation, with Kinn-Lai additionally seeing lenition of the *d (like that seen in Penni) and the accretion of the *-lai element.

Could it also be that: *nd > nnl in Kinn-lai?

Personally, I'd doubt it. I'd assume Tolkien is marking some sort of morpheme boundary with the hyphen, since I can't imagine him using a hyphen within a root morpheme (seems a tad... not his style)

True I thought it would maybe mark a syllable boundary, but I see your point.
So, we would have to say that *nd > nn and the final vowel is lost?

[Salmoneus]

Middle-Earth pretty explicitly is an alt-history, actually.


Crey: yes. There are three ways this may have happened: (1) /kw/ > /w/ (plain loss of the initial segment); (2) /kw/ > /xw/ > /hw/ > /w/; or (3) /kw/ > /gw/ > /w/. The second option seems more plausible, and allows the Windan and Hwinti branches to be more closely related. [though nb 'windan' and 'hwinti' are names of nations, not of languages, strictly speaking...]

However, the problem here is that it's not as simple as just Grimm's Law. In Grimm, it's the "hardest" stops that end up as fricatives, presumably via an aspirated stage. But in Quendian, there's already an aspirated series. So what happened to them? Three models spring to mind:

2A:
/kh/ > /kh/
/k/ > /x/
Here, the unaspirated voiceless stop just fricativises for no apparent reason while the aspirate remains intact. Huh?

2B:
/kh/ > /kh/ > /x/
/k/ > /kh/ > /x/
Here, the distinction is lost through merger first, before a Grimmesque shift.

2C:
/kh/ > /x/ > /G/
/k/ > /kh/ > /x/
Here, BOTH series end up as fricatives - at least initially (either or both of these could have a restriction like 'initial' or 'stressed onset' or whatever). This would allow not only for a family with Hwinti, but also with Eldarin - because the first stage of that chain shift, /kh/ > /x/, also occurs in Sindarin and Quenya. [although, was that cannonically an Eldarin shift, or just parallel?]

2C was the option I went with in my Windan language originally.

HOWEVER! There is a fourth option here, for real heretics: what if Tolkien was wrong? What if what he thought were voiceless aspirates were in fact voiced aspirates (/breathy/etc)? That makes the equivalent of 2A suddenly seem plausible (the voiced aspirates are never 'in the way' to begin with).

[Salmoneus]

I think Kinn-lai must be branched with Kindi because they both share the same delabialisation, with Kinn-Lai additionally seeing lenition of the *d (like that seen in Penni) and the accretion of the *-lai element.

The -lai element is definitely a distinct word originally, meaning "people".

Regarding nd > nn, it's worth pointing out that Quendian *nd (etc) was either a prenasalised stop phoneme, or at least a cluster that in some ways acted like a single phoneme. For instance, *nd- is a common initial cluster, and it did indeed end up as a plain nasal at least some of the time in Quenya. [eg 'Melkor' starts out as 'Mbelekoro']

[Creyeditor]

So more like *ⁿd > nː?

[WeepingElf]

I think Kinn-lai must be branched with Kindi because they both share the same delabialisation, with Kinn-Lai additionally seeing lenition of the *d (like that seen in Penni) and the accretion of the *-lai element.

The -lai element is definitely a distinct word originally, meaning "people".

Yes - cognate to Q. lië, it seems.

[Lambuzhao]

I think Kinn-lai must be branched with Kindi because they both share the same delabialisation, with Kinn-Lai additionally seeing lenition of the *d (like that seen in Penni) and the accretion of the *-lai element.

Don't know if this was mentioned anywhere else, but my Proto-Italic LEDs tripped off at the sight of Kinn-Lai/Kindi :: Penni.

Similar to the k/p Celtic langs, Italic has that distinction:
E.g.

quis 'who'
:osc: pis

quattuor 'four'
:osc: petora


quinque 'five' (Cf. borrowed patronyms Pompeius, Pomponius ≅ Quintus 'fifth' )
:umb: pumpe
:osc: pompe

---

Now, another similarity lies in the analogy of /nd/ :: /nn/ -

operandam
:osc: upsannam = 'faciendam' 'about to be made'


Therefore, in an alt-hist kind of way, the Kindi (or more likely the Cuind) and the Penni could be seen as cousins/cognate, similar to the Latins, Oscans & Umbrians.

[Salmoneus]

All six names are intended to be cognate, yes.