Classical Saarvian Scratchpad

[Rounin Ryuuji]

The phonology isn't terribly interesting, but that's ok. The main divergences from the IPA are /b d g/ which are [β ð ɣ], except when immediately before a voiceless stop, in which case they devoice to [ɸ θ x]. The [ɸ] produced by this merges with original /f/ [ɸ].

I do find myself rather unsure of how to pronounce it, notably I find myself assuming ä to be [ɛ], though I suppose it could be [æ]; also, is j intended to be [j], [ʒ], [d͡ʒ], [ʑ], [ɟ͡ʑ], [ʐ], [ɖ͡ʐ], or something else? it seems to pattern with both /s/ and /i/, depending on the context, so I'm guessing it must descend from multiple sources, and possibly be some sort of consonant in variantion between [j~(d)ʒ~(ɟ)ʑ] (unless, of course, the third-person inanimate accusative and directional(?) are suppletive, or the j- is a typographical error).

[KathTheDragon]

Have a complete sentence: se-peŋŋ se-kanjo se-mijom-käll-sä lowo ro-soo-nobuad "The fox ate the fish intended for the child"

And the gloss:

se-

ᴅᴇꜰ=

peŋŋ

fox{ᴀʙs}

se-

ᴅᴇꜰ=

kanjo

ʀᴇʟ.ᴀɴ-ᴀᴄᴄ

se-

ᴅᴇꜰ=

mijom

child-ᴅɪʀ

-käll

=ɢᴇɴ

-sä

=3

lowo

fish-ᴀᴄᴄ

ro-

ᴘꜰᴠ=

soo-

3>3=

nobuad

eat-ᴘsᴛ

I do find myself rather unsure of how to pronounce it

Alright, I can give a more complete exposition (though specific allophone distributions and phonetic details are subject to change). p t k are [pʰ tʰ kʰ] word-initially and after voiceless consonants, [p t k] before voiceless consonants, and [b̥ d̥ g̊] (weakly-voiced) in most other contexts. b d g are [ɸ θ x] before a voiceless consonant and [β ð ɣ] otherwise. f s h l j w m n ŋ are always [ɸ s h l j w m n ŋ]. I will not commit to a specific rhotic for r other than it's a coronal sonorant. ä a e o i u are [æ ɑ e o i u] in stressed syllables (the first syllable of non-clitics). In unstressed syllables only e o i u are written, and stand for [ə o ɨ u], though the back vowels are only weakly rounded. Doubly-written vowels are just the long counterparts of the short vowels, except that ee oo are [ɛː ɔː]. iä ie ua uo are [i̯æˑ i̯eˑ u̯ɑˑ u̯oˑ].

[KathTheDragon]

unless, of course, the third-person inanimate accusative and directional(?) are suppletive, or the j- is a typographical error

Neither of these. /j/ is deleted word-initially before a high vowel or one of the front diphthongs. ie- is thus *jie-.

[bradrn]

ä a e o i u are [æ e i ɑ o u] in stressed syllables (the first syllable of non-clitics).

Are you sure this shouldn’t be [æ ɑ e o i u]?

[KathTheDragon]

You're absolutely right. I changed the order of the graphemes and my brain ignored that when I wrote the transcription.

[quinterbeck]

Have a complete sentence: se-peŋŋ se-kanjo se-mijom-käll-sä lowo ro-soo-nobuad "The fox ate the fish intended for the child"

And the gloss:

se-

ᴅᴇꜰ=

peŋŋ

fox{ᴀʙs}

se-

ᴅᴇꜰ=

kanjo

ʀᴇʟ.ᴀɴ-ᴀᴄᴄ

se-

ᴅᴇꜰ=

mijom

child-ᴅɪʀ

-käll

=ɢᴇɴ

-sä

=3

lowo

fish-ᴀᴄᴄ

ro-

ᴘꜰᴠ=

soo-

3>3=

nobuad

eat-ᴘsᴛ

Very nice, what's -sä doing in se-mijom-käll-sä? It looks like the clause is relativised from "The fish (is?) intended for the child" - what would that look like by itself? I'm noticing there isn't a verb in the relative clause.

[KathTheDragon]

sä is one of the agreement clitics for predicates, it's there to mark the adverb as functioning predicatively. "The fish is intended for the child" looks like se-luow se-mijom-käll-sä with "fish" moved into the nominative.

[Moose-tache]

The order of the relative clause is melting my brain a little, but in a good way.
I like this language, and I'm watching closely, even if I don't have much to contribute yet.

[KathTheDragon]

As far as I know, preposed relative clauses are actually pretty normal in head-final languages like this. Nonetheless, I'll probably do a Syntax and have preposed relative clauses only where they are "short". When they're "long" they'd be placed after, and have their head clefted out of the main clause.

[KathTheDragon]

Another clitic, tis, which marks motion/direction away with the lative and directive cases, and position off in the locative.

wiekäälii

wall-ʟᴏᴄ

-tis

=AB

"off the wall"

wiekääles

wall-ʟᴀᴛ

-tis

=AB

"(moving) away from the wall"

wiegäälom

wall-ᴅɪʀ

-tis

=AB

"(in the direction) away from the wall"

[Rounin Ryuuji]

Do you have complicated, messy naturalistic etymologies for any of this? I love it when an inflection or a lemma has a story to tell if we only know where to look.

[Travis B.]

I really like this language, even though I don't have much to contribute myself.

[KathTheDragon]

Do you have complicated, messy naturalistic etymologies for any of this? I love it when an inflection or a lemma has a story to tell if we only know where to look.

Etymologies yes, complicated and messy, not really. I haven't gone back very far at all, just deep enough to get the mutations I wanted.

[Rounin Ryuuji]

Oh, well. Maybe post a few (and also give the language a name)?

[KathTheDragon]

I have half a mind to at some point derive some sisterlangs to this one and call it a reconstruction challenge.

I hope this explains why I'll keep the etymologies to myself. As for a name, I'm working on it! I'm trying to think of a fitting etymology for it.

[quinterbeck]

Are the loc/lat/dir cases inherently adverbial, or only adverbs when the clitics are affixed? Can the cases be used to modify nouns without being relativised?

[KathTheDragon]

Inherently adverbial, and they can't modify nouns directly. I'd already decided that when nouns are modified, all their clitics migrate to the first modifier,
se-ŋälbom-sää "into the room", se-sarfom-sää ŋälbom "into the red room". Using a word that's already hosting clitics as a modifier would conflict with this bit of syntax, so I don't want to allow it.

[Travis B.]

I was reading about Samic consonant gradation today, and it seemed quite neat - basically most medial consonants in Proto-Samic had two forms, short and geminate, but based on whether the next syllable was originally open or closed each of these had a weak and a strong form, resulting in four lengths. Note that these collapsed into three lengths in most Samic varieties aside from some dialects of Ume Sami. Also note that in present-day Sami varieties these original conditioning environments have been lost in many cases due to loss of final consonants.

[KathTheDragon]

Oh, that's interesting! I didn't actually know any specifics about what triggered gradation in any Uralic language when I started this, so what I ended up doing probably isn't like any of them.

[KathTheDragon]

I've added a series of possessive pronoun clitics: kääj- "my, our", nääj- "your", sääj- "his/hers, theirs", iej- "its". They're all indifferent to number, and make their possessand definite (so they replace the definite clitic se-).

kääj-sarfo lowo nobuad-soo

kääj-

1.POSS=

sarfo

red-ACC

lowo

fish-ACC

nobuad

eat-PST

-soo

=3>3

"He ate my red fish"