Frislander's IE-lang scratchpad

A forum for all topics related to constructed languages
User avatar
Frislander
runic
runic
Posts: 3304
Joined: 14 May 2016 17:47
Location: The North

Re: Frislander's IE-lang scratchpad

Post by Frislander » 24 Jan 2019 16:52

OK, so let's discuss phonotactics and distribution.

Unlike its immediate neighbours, this language does allow for word-initial consonant clusters. These are pretty much entirely restricted to Indo-European vocabulary, since Ugric as a family lacks initial clusters entirely. Here is a list of all such clusters that are in the vocabulary so far - this is subject to expansion as more IE vocabulary is derived.
  • pś: pŭk "bee"
  • pṣ: pṣul "brow, forehead"
  • px: pxet "leaf"
  • pł: płən "woman;s breast, teat"
  • kr: kriwe "neck"
  • xr: xrojət "clay"
  • śt: śtəxa "earth"
In native words, the roster of word-final clusters is similar if not identical to the roster of initial-clusters, e.g. xapx "tooth", pśəpṣ "beaver", ŏkś "snake", məśt "border region", əlp "aspen/white poplar", łəpt "seven". However, the roster of final clusters has been greatly expanded by the addition of Ugric loanwords, most notably nasal+ obstruent clusters which had been lost from the Indo-European vocabulary, e.g. uńś "buttocks", näŋk "larch". The set of intervocalic clusters on the other hand is essentially free, leading to alternations in certain nouns between absolutive unpossessed singulars containing an epenthetic schwa and other forms lacking the same: łŏpən ~ łŏpne "dream(s)", jĭpəɣ ~ jĭpɣixa "owl ~ with an owl", ənəm ~ ənmət "breath ~ your breath", xŏməs ~ xŏmsijən "(on a) marsh island". However, it is important to note that not all post-initial schwas behave this way; see for instance xrojət ~ xrojətəm "(my) clay".

There are some distributional restrictions on certain sounds. Notably l does not occur word-initially, instead only ł. Similarly, since ŋ does not occur word-initially in neighbouring languages it does not occur word-initially either. Similar restrictions hold for the vowels - the reduced vowels are prohibited from appearing in absolute word-final position, though otherwise vowels are fairly unrestricted in their distribution.

User avatar
Frislander
runic
runic
Posts: 3304
Joined: 14 May 2016 17:47
Location: The North

Re: Frislander's IE-lang scratchpad

Post by Frislander » 25 Jan 2019 18:33

Actually, about that velar nasal. If Avestan can get a velar nasal from intervocalic *s, then I can get one from coda *r, caus I've kind of been on the fence with what to do with it. I have previously been losing it with compensatory lengthening, but because of the re-alignment of the vowel length contrast this is a bit iffy with me. I think this would only apply to original coda *r, not new coda *r from lost final vowels, e.g. wir "young man", instead mainly forms like pəsəŋ "father", xŏŋ "door", but it'll still be fairly uncommon.

User avatar
Frislander
runic
runic
Posts: 3304
Joined: 14 May 2016 17:47
Location: The North

Re: Frislander's IE-lang scratchpad

Post by Frislander » 05 Feb 2019 16:29

OK, some stuff on nominal inflection.

Case inflection. Of the original PIE cases the nominative/accusative have fully merged as zero-marked. The dative survives as a suffix -i which deletes a vowel that immediately precedes it, and results from the merging of o-stem and eH2-stem datives. The genitive/survives as an ablative -(ə)l where the schwa is epenthesised after consonants, and is derived from genitive *-osyo. The last inherited case ending is the instrumental -(ə)pś, derived from consonant stem *-bhi.

Then added to this are some new case endings from old prepositions added to different case endings. The old accusative is continued in the ablative -as and perlative -aṣ derived from *-om-de and *-om-trHx respectively. The dative forms the basis for the comitative -ixa, inessive -ija and superessive -ijən, from *-ey-kom, *-ey-en and *-ey-h2enh2e respectively, and again in all these cases the vowel of the case ending replaces a stem-final vowel.

To sum up then we have the following endings for case.

Code: Select all

DAT -i
INS -(ə)pś
COM -ixa
INE -ija
ALL -as
ABL -(ə)l
PER -aṣ
SUP -ijən
Plural forms are somewhat less regular. Most nouns regularly take -e in the plural, which extends to -el- before other case endings besides the instrumental and ablative. However some nouns instead form their plural stem with an unalternating -əj or -əw, which sometimes descends from an old i- or u-stem, like ŏkś(əj) "snake", jək(əw) "horse", but not always, e.g. sur(ə)t~surtəj "newborn reindeer calf" (from Ugric). The equivalent forms of the dual are the more regular -et/-it/-ut respectively.

There are also some bisyllabic stems which syncopate the schwa when a suffix follows. Most of these are Ugric loans, e.g. tar(ə)ɣ "crane", xŏm(ə)s "marsh island", sŏr(ə)t "pike", though there are some IE-stems which do the same, e.g. łŏp(ə)n "dream", łew(ə)ɣ "sun". Similarly some roots insert a consonant before other suffixes, e.g. jŏk(n) "liver", łek(l) "pine" (from Ugric). All these take the regular -e/-et plural/dual endings.

Finally there are the possessive suffixes, which come after all the other suffixes. These show a three-way contrast in both person and number. The paradigm is as follows.

Code: Select all

  SING  DUAL PLUR
1 -(ə)m -mət -(ə)ne
2 -(ə)t -tət -we~-o
3 -(ə)l -lət (ə)le
The singular and 1st and 3rd plural forms insert the schwa after a consonant, while the 2nd plural -we monophthongises to -o in the same environment.

So some examples.

Pexŭn təmijal kəmen
Perkunos house-INE-3s.POSS sleep-1s.PRES
I will sleep in the Thunder God's house

kŏnem łəɣ juletəpś mŏxəlxəsĭn
woman-1s.POSS salt broth-DUAL-INS mix-DUR-3s.PRES
My wife is mixing salt into our stews

Post Reply