Avarin Language Development

If you're new to these arts, this is the place to ask "stupid" questions and get directions!
Post Reply
Hugh_Capet
rupestrian
rupestrian
Posts: 10
Joined: 03 Dec 2022 09:05

Avarin Language Development

Post by Hugh_Capet »

This is my first time endeavoring to develop a constructed language, and I was wonder what, if anything, I might be able to do to aid the process. As I'm looking specifically to develop a language for Windan tribe of Avari elves. My main problem is I'm not sure what language, if any, Tolkien might have used as inspiration for it. The only foundation I might have is based upon an earlier conception where Avarin (called Lemberin) is split into West, North, and East branches. The branches are said to be inspired by the Goidelic, Baltic, and Finnic languages, respectively.

However, I lack sufficient knowledge of how to proceed with the language, as I don't think it would be as simple as haphazardly applying real-life linguistic elements and hoping they make sense. Any advice or ideas offered would be most appreciated.
Salmoneus
MVP
MVP
Posts: 3031
Joined: 19 Sep 2011 19:37

Re: Avarin Language Development

Post by Salmoneus »

Windan! My favourite!

No, but seriously, I've had a few half-hearted goes at deriving Windan - though never beyond a few pages of notes. As you've discovered, it's hard to really hang a whole language on just one attested word and have it feel 'right'. Particularly because the Elvish languages may look natural at first glance, but aren't really, and I was trying to squeeze together a bit of that unnaturality with a naturally-derived daughter language and found it hard to produce something that was simultanously 'like a realy language' and also 'recognisably Tolkienian'. Of course, if you abandon that last part, it's a lot easier!


I'm not aware of there being an obvious influence for Windan - with those six names, Tolkien was clearly just doodling around with ideas for how the same ancestor word could end up in different families. In the case of Hwenti, he's just applied the Germanic Consonant Shift, so if you were deriving Hwenti it would seem reasonable to think of it as vaguely Germanic (though obviously not modern Germanic, where that -i would have been lost eventually - in the 'English' descendent they would now be called the Whints, with the 'h' silent for most people...). [NOTE TO SELF: NO, DON'T START TRYING TO DERIVE WHINTISH...]

But with Windan, it's not obvious that there's an obvious inspiration, at least to me. Partly because it's not clear to me what soun changes have produced this form, which is more puzzling than it first looks. The good news is that that gives you a lot of freedom in making your own language! The bad news is, you don't have much to guide you.

There are three big questions here:
Why is it w- and not kw-?
The ancestral form is of course kwende. So where did the k- go? (particularly because this was probably originally *kw, with the rounding seen as an integral part of the consonant, not a cluster.

There are endless possible answers, but to me, four seem most obvious:
A: the k- has simply been lost (after the secondary articulation had already broken into a cluster). This seems weird, losing such a strong consonant in such a strong position. It's not something that strikes me as a 'European' sound change. but it's certainly possible; I'm sure it must have occurred somewhere in the world. Maybe find where?
[one intermediary route would be to strengthen /w/ to a fricative, lose the /k/ because that's a weird clust, and then weaken the fricative again; but that's kind of roudnabout, and still doesn't make any real language spring to mind]
More problematically, this localised change doesn't give a big enough rule to sink your teeth into, as it were, as it doesn't say anything about how the rest of the language would look.

B: the kw- has weakened to gw-, which has in turn weakened to w-. This seems more reasonable, although it's still odd to have initial lenition like this, particularly when that /d/ is still standing there in the middle of the word. Perhaps /d/ > /D/ and /t/ > /d/ everywhere, but blocked next to a consonant? (that *nd may actually be a prenasalised stop and thus automatically exempt from the lenition of plain *d anyway). This would obviously tend to give a very 'soft-sounding' language, and raises the question of where you could re-generated voiceless stops from (though since they're elves, maybe they ignore normal human universals and just don't have them!?). Well, I guess they could just be from sC clusters?

C: the kw- has weakened to hw-, as in Germanic, and then in turn to w-, as in (most) modern English. This would make Windan a close cousin of Hwenti. Now, clearly Hwenti has -t- where Windan has -d-, so why?
C1: They both underwent the GCS, but Windan softened *t back to *d adjacent to a nasal.
C2: It was a chain shift (t>th followed by d>t), and Windan broke away before the second change happened. Again, this leaves very few if any voiceless stops.
C3: It was a chain shift, and Windan DID strengthen *t, but broke away before Hwenti then analogised this to prenasalised *nd.

D: the *kw- shifted to *p- (making Windan instead a close cousin of Penni), and *p- was then lenited to a bilabial fricative, which then >w (maybe only before unrounded vowels, and >h elsewhere?). Lenition of p- can easily happen even without a more general lenition of stops. Witness: Celtic, which Tolkien would obviously have known about. Except that in Celtic it happens before (and 'causes') kw>p in Welsh. If it happens before that, it raises the question of what happened to original *p... did *p and *kw just merge? Or did *p do something else (maybe following the same route but further, now disappearing entirely)


When I've played with Windan before, I've gone with C, either C1 or C3. I'm not sure that's what I'd do now. Right now B is more appealing to me. But C does give you more of a 'hook' for real-world inspiration, because it does at least resemble Proto-Germanic, as well as creating an Avarin subfamily of Windan and Hwenti as siblings.

Why is it -i- and not -e-
Why did the vowel in kwend- change?

A: *e always raises to *i (or only when stressed?) So what happened to *i?

B: *e always raises to *i before a nasal. Not unreasonable.

C: *e raises to *i after *w. I'm not sure why, but not impossible.

D: *e raises to *i when umlauted by the *-i of the following syllable (if there was one, see below).

When I've done it, because I'd already decided that Windan was going to be pseudo-Germanic (or at least Germanic-influenced), I naturally went immediately for the umlaut option here. So in my version the singular form was Wend (which I liked the sound of, though it is already a real-world ethnonym). But the other options are also fine.


What's going on with the plural?

Clearly, the form Windan uses the plural element -m. But what happened to the original plural, -í?

A: the original plural has been there all along, it's just been reduced to schwa, written as (or in turn strengthening to) *a before nasals. Final -i: > final -i > final -e > final -@ > -a before a nasal. The general reduction has caused -m (now -n) to be added as reinforcement to prevent ambiguity. Alternatively, final -i is lost after heavy syllables, as in most Germanic languages, producing a syllabic final -n, which was made more pronounceable by epenthesis. Both routes are totally in line with what Germanic actually did (eventually), and they also provide a reason for umlaut of the root vowel, as noted above.

B: the original plural has been lost entirely, and *kwendem has simply > *windan. This is very straightforward, but does raise the question of why -en > -in but -em> -an. This could be because of the rounding of the nasal (perhaps -em>-om>-am), or it could be because of stress position, or it could be because of the initial *w. In any case, your answer here has to feed into your answer to the preceding question.

Obviously, following my Germanic theme, I went with A. Actually, I went with a variant, in which there are now two plural forms, a regular individual plural and a collective plural (including for the people as a whole). This gives the forms Wend, Winde, Windan (an elf, two elves, the elves as a group).


Is this Germanic route what Tolkien intended? I don't know, but the fact that all three obvious questions can be solved in Germanic ways makes me wonder whether it may have been. On the other hand, as I subsequently found out Proto-Elvish as a whole doesn't look that Germanic, so you can't just directly apply all the Germanic changes. For instance, since the final vowel often is an important derivational morpheme, too much reduction (without a huge amount of umlaut!) may give too much ambiguity. Or maybe it does and you just need new derivational processes. I also got too ambitious, and tried playing around with the weird metathesis/whatever of Proto-Elvish, combining it with Germanic's reduction of unstressed vowels to produce something approaching biconsonantalism (i.e. a sort of Germanic/Semitic), but got frustrated and wandered off before I got very far...



Anyway, I know I didn't really answer your question, but maybe
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Avarin Language Development

Post by Creyeditor »

I would definitely be interested in Baltic Elvish.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
Hugh_Capet
rupestrian
rupestrian
Posts: 10
Joined: 03 Dec 2022 09:05

Re: Avarin Language Development

Post by Hugh_Capet »

Salmoneus wrote: 03 Dec 2022 14:27 Anyway, I know I didn't really answer your question, but maybe
I'd say that you did, in a round-about way. If nothing else, you've given me inspiration off of which to base the language itself.
Salmoneus wrote: 03 Dec 2022 14:27 C: the kw- has weakened to hw-, as in Germanic, and then in turn to w-, as in (most) modern English. This would make Windan a close cousin of Hwenti. Now, clearly Hwenti has -t- where Windan has -d-, so why?
C1: They both underwent the GCS, but Windan softened *t back to *d adjacent to a nasal.
C2: It was a chain shift (t>th followed by d>t), and Windan broke away before the second change happened. Again, this leaves very few if any voiceless stops.
C3: It was a chain shift, and Windan DID strengthen *t, but broke away before Hwenti then analogised this to prenasalised *nd.
Of the options, C1 seems the most plausible to me. This does beg the question of what other stops might soften in like-manner.
Salmoneus wrote: 03 Dec 2022 14:27D: the *kw- shifted to *p- (making Windan instead a close cousin of Penni), and *p- was then lenited to a bilabial fricative, which then >w (maybe only before unrounded vowels, and >h elsewhere?). Lenition of p- can easily happen even without a more general lenition of stops. Witness: Celtic, which Tolkien would obviously have known about. Except that in Celtic it happens before (and 'causes') kw>p in Welsh. If it happens before that, it raises the question of what happened to original *p... did *p and *kw just merge? Or did *p do something else (maybe following the same route but further, now disappearing entirely)
I'm somewhat partial to this also, as it does line-up with the fictional history I've envisaged for the Windan. As certain non-canon sources endeavor to have them be the Avari said to inhabit the Taur-im-Duinath of Beleriand. Specifically, they would travel through Rhûn and commune with the Nandor of Greenwood before eventually settling in the forest by way of the southern Blue Mountains (as it seems implausible for them to enter through Ossiriand).
Salmoneus wrote: 03 Dec 2022 14:27Why is it -i- and not -e-
Why did the vowel in kwend- change?

A: *e always raises to *i (or only when stressed?) So what happened to *i?

B: *e always raises to *i before a nasal. Not unreasonable.

C: *e raises to *i after *w. I'm not sure why, but not impossible.

D: *e raises to *i when umlauted by the *-i of the following syllable (if there was one, see below).

When I've done it, because I'd already decided that Windan was going to be pseudo-Germanic (or at least Germanic-influenced), I naturally went immediately for the umlaut option here. So in my version the singular form was Wend (which I liked the sound of, though it is already a real-world ethnonym). But the other options are also fine.
Of the options you presented, B seems the simplest, but it does clash with the supposed Germanic influence of Windan. So, D might be a better choice.
Salmoneus wrote: 03 Dec 2022 14:27What's going on with the plural?

Clearly, the form Windan uses the plural element -m. But what happened to the original plural, -í?

A: the original plural has been there all along, it's just been reduced to schwa, written as (or in turn strengthening to) *a before nasals. Final -i: > final -i > final -e > final -@ > -a before a nasal. The general reduction has caused -m (now -n) to be added as reinforcement to prevent ambiguity. Alternatively, final -i is lost after heavy syllables, as in most Germanic languages, producing a syllabic final -n, which was made more pronounceable by epenthesis. Both routes are totally in line with what Germanic actually did (eventually), and they also provide a reason for umlaut of the root vowel, as noted above.
Then it seems utilizing umlaut would be best. Though, why would the basis for the plural be kwendim and not kwendem (as posited in B)? Wouldn't it be possible that the final -e- of kwende could be strengthened into -a-, or would that create linguistic problems I'm just not seeing?
Salmoneus wrote: 03 Dec 2022 14:27Obviously, following my Germanic theme, I went with A. Actually, I went with a variant, in which there are now two plural forms, a regular individual plural and a collective plural (including for the people as a whole). This gives the forms Wend, Winde, Windan (an elf, two elves, the elves as a group).
This does make me wonder what might have become of the primitive collective-plural suffix -stā and whether it might prove relevant.
Salmoneus wrote: 03 Dec 2022 14:27Is this Germanic route what Tolkien intended? I don't know, but the fact that all three obvious questions can be solved in Germanic ways makes me wonder whether it may have been. On the other hand, as I subsequently found out Proto-Elvish as a whole doesn't look that Germanic, so you can't just directly apply all the Germanic changes. For instance, since the final vowel often is an important derivational morpheme, too much reduction (without a huge amount of umlaut!) may give too much ambiguity. Or maybe it does and you just need new derivational processes. I also got too ambitious, and tried playing around with the weird metathesis/whatever of Proto-Elvish, combining it with Germanic's reduction of unstressed vowels to produce something approaching biconsonantalism (i.e. a sort of Germanic/Semitic), but got frustrated and wandered off before I got very far...
I'm not sure how I'd address the problem either, as I don't know much concerning PE's metathesis (except that it's apparently more prominent in Quenya, though I might be wrong about that). Did you apply gender, by chance, or did you endeavor to keep the language genderless (due to it being absent in PE and its derived languages)?
Last edited by Hugh_Capet on 04 Dec 2022 04:40, edited 1 time in total.
Hugh_Capet
rupestrian
rupestrian
Posts: 10
Joined: 03 Dec 2022 09:05

Re: Avarin Language Development

Post by Hugh_Capet »

Creyeditor wrote: 03 Dec 2022 21:43 I would definitely be interested in Baltic Elvish.
As would I. I'm not sure which of the known tribes would fit the bill, but I've seen attempts to create a distinctly "northern" Avarin dialect. Not one that would be Baltic in nature, but still different from what we're familiar with.
Post Reply