Zompist Bboard Again

Could be a number of reasons. Just off the top of my head…
- Morphological alternations
- Partial voicing/devoicing
- It’s actually slack voice or some other phonation
- To highlight particulars of the phonetic environment

Man in Space wrote: ↑Mon Nov 23, 2020 3:50 pm Could be a number of reasons. Just off the top of my head…
- Morphological alternations
- Partial voicing/devoicing
- It’s actually slack voice or some other phonation
- To highlight particulars of the phonetic environment

I used to write that way to indicate voiceless /b d dʒ g v ð z ʒ/ in English, partially because they alternate with voiced [b d dʒ g v ð z ʒ] and partially because they may not be fully devoiced or may be shorter than the "fully voiceless" counterparts (e.g. for many English-speakers it seems they distinguish final /s z/ which are not intervocalic on the basis of both vowel length allophony and consonant length, with /s/ being longer than /z/, rather than voicing).

Does Pali descend from Sanskrit? In particular, is it legitimate to say of most Pali words that they are inherited from Sanskrit? This is chiefly a question about what the term 'Sanskrit' includes and excludes in plain English.

Is there any language that has something a bit like grammatical gender, but instead of various objects and categories of objects and people having fixed gender, gender is assigned based on where in a short text a person or object appears? That is, the first person or object in a sentence or short text would always have Gender No. 1, the second person or object in a sentence or short text would always have Gender No. 2, the third person or object in a sentence or short text would always have Gender No. 3, and so on? I'm asking because I think that this might fulfill some of the functions that grammatical gender seems to fulfill in terms of reducing ambiguity, but might fulfill these functions even better than "standard" grammatical gender.

Jonlang wrote: ↑Mon Nov 23, 2020 2:43 am A good while ago now, I saw something either on this forum, or one of the many conlangy Facebook groups I follow, about the possibility that Welsh once had /s/ > /h/ as one of its initial consonant mutation, but it was 'abandoned' which led to some words becoming solidified with the initial /h/ (like 'Hafren' (the Welsh form of Sabrina and the name for the River Severn)) and some remaining with unlenited /s/ instead. Has anyone seen or heard of this anywhere?

Could it have been this website (not sure about reliability)? It links it to the initial mutations (and to the /s/ > /h/ mutation in Goidelic) but doesn't actually say it went through a phase of being one (though I'm not sure why saith should be an exception based on what it's saying).

Raphael wrote: ↑Tue Nov 24, 2020 2:22 am...first person or object...

Given that this is a linguistics forum, it took an embarrassingly long time to parse what this meant. This would probably not be called gender. It's clearly a type of anaphora. Plenty of languages use multiple layers of anaphora ("So I says to your man, that one's got to go, but him, he's alright!"). Generally these are more related to salience or topicality than strict order, but the two often overlap.

Raphael wrote: ↑Tue Nov 24, 2020 2:22 am Is there any language that has something a bit like grammatical gender, but instead of various objects and categories of objects and people having fixed gender, gender is assigned based on where in a short text a person or object appears? That is, the first person or object in a sentence or short text would always have Gender No. 1, the second person or object in a sentence or short text would always have Gender No. 2, the third person or object in a sentence or short text would always have Gender No. 3, and so on? I'm asking because I think that this might fulfill some of the functions that grammatical gender seems to fulfill in terms of reducing ambiguity, but might fulfill these functions even better than "standard" grammatical gender.

Proximate-obviate can work that way. (Though there are complications. The most salient constituent: animate, named, etc. generally gets the proximate forms anyway.)

I seem to recall a sign language (ASL?) doing something similar.

Thank you!

Does anyone know how a possessed case usually works, in terms of syntax? Wikipedia doesn't have much info about the languages that have a possessed case,* and googling "possessed case" mostly turns up stuff about possessive case (or demonic possession) instead. In particular, I'm wondering if the possessed noun is still treated as the head of the NP even though it's also the one that the morphology is applied to--and if so, where the accusative case for example gets applied if the NP is the sentence's object.

* it does have more info about Afro-Asiatic languages' construct state, but I'm looking for something about languages that have only a possessed case and not a genitive as well.

I wouldnt call it a noun case. Is this for a conlang? Most likely you'll want to put case endings on this noun, so unless your language allows case stacking I'd say possession is not a case like those others. That might explain why it's hard to search for.

It's probably better to think of possession as a category analogous to number and gender. You may also want to distinguish person, e.g. "my fish, your fish, their fish" rather than just a binary distinction of free vs. possessed. This would make it a dimension all its own. Though I dont think this is required.... free vs possessed seems plausible so long as your pronouns are conveniently nearby.

Richard W wrote: ↑Mon Nov 23, 2020 10:43 pm Does Pali descend from Sanskrit? In particular, is it legitimate to say of most Pali words that they are inherited from Sanskrit? This is chiefly a question about what the term 'Sanskrit' includes and excludes in plain English.

From most of my research: no. Pali and most other modern Indo-Aryan languages (outside of Daric (and perhaps Kashmiri and related language~dialects (and even more tenuously Romani, etc.)))* descend from a closely related language. Even Sanskrit itself is not monolithic as can been seen from "vedic" versus other forms distant in both time and place. There are a number of morphological and phonological differences within Sanskrit and between Sanskrit and especially middle indo-aryan. Leaving aside the morphological area which is more murky and statistical, I'd note some phonological differences. As background, I'd also note that modern Indo-Aryan languages in general have multiple tranches of borrowings from Sanskrit that have participated in various levels of subsequent phonological changes.

Key examples are /kS-/ clusters in Northeastern/Vedic Sanskrit versus other dialects. Source material for this cluster was ... immense including, notably, a neutralization in voicing in certain sources. This may not have occurred in more southeasterly dialects. Other large clusters such as /*(k)swe-/ in "six" and other forms were treated very differently. Northeastern forms trend towards /ʂa- ~ ʂo-/ which southeastern form trend towards /tʃʰa- ~ so-/. The latter form is continued in modern Indo-aryan languages (e.g., Hindi /tʃʰah/) which have not re-borrowed the Sanskrit form. The former seems to have been preserved in Gandhari, Dardic, and Romani. Notably, some of the NE dialects seem to show not just /ʂ-/ but /ʈʂ(ʰ)-/. This form in particular is interesting to compare to Nuristani forms / in e.g., ʂu-/ and extreme eastern-Iranian forms like Khotanese <kṣäta> /ʈʂ(ʰ)əta/.

Good article: https://www.academia.edu/8594335/Indo_Aryan_six_

Very dated but informative especially for lesser known Prakrit dialects: A grammar of the Prākrit languages by Richard Pischel (great example is a group of dialects changing the intials on second person singular pronouns /*tu-/ to /p-/).

In the last few years, a number of Dardic grammars have been published. Also more information about Gandhari has been released. E.g., https://stefanbaums.com/baums_grammar_outline.pdf

Also worth noting, Sanskrit itself is not monolithic itself--there appear to be dialectic differences between various books and schools. A phonological example is the treatment of /*l/ and /*r/. In more northwesterly dialects, both seem to collapse into /r/ and /l/ is extremely rare. In more southeasterly dialects, things are more complex but it appears in certain areas both collapsed into /l/ with /r/ reintroduced from other dialects. Between those extremes, it is thought that there may have been dialects that more closely preserved /*l/ and /*r/.

Great source in general on this area: https://www.researchgate.net/publicatio ... Consonants

*in part my opinion, but I think this holds up and other sources mention this.

offhand, does anyone know if the Spanish diminutive -ito can be attached to nouns with the agent suffix -or? e.g. actorito? how about -ador? I see a small number of hits for such words but I dont know of any such words that are considered part of the daily lexicon.

Same question goes for French, Spanish , Catalan, Portuguese, Italian, or any other Romance language anyone knows.
http://www.laspalabras.net/suffixe.php says that there are no such words in Spanish.

Pabappa wrote: ↑Wed Nov 25, 2020 11:29 pm offhand, does anyone know if the Spanish diminutive -ito can be attached to nouns with the agent suffix -or? e.g. actorito? how about -ador? I see a small number of hits for such words but I dont know of any such words that are considered part of the daily lexicon.

Yes it can, to both the -tor/-sor of Latin borrowings and also the native -ador. It is overall very productive.

LingEarth wrote: ↑Wed Nov 25, 2020 3:36 pmDoes anyone know how a possessed case usually works, in terms of syntax? Wikipedia doesn't have much info about the languages that have a possessed case,* and googling "possessed case" mostly turns up stuff about possessive case (or demonic possession) instead. In particular, I'm wondering if the possessed noun is still treated as the head of the NP even though it's also the one that the morphology is applied to--and if so, where the accusative case for example gets applied if the NP is the sentence's object.

* it does have more info about Afro-Asiatic languages' construct state, but I'm looking for something about languages that have only a possessed case and not a genitive as well.

Standard Arabic, Biblical Hebrew and Classical Nahuatl have a possessed (or "construct") form, and that this is independent of object (or subject, etc.) marking. It's on a different axis in these languages at least. In the two Semitic ones, it's the same as definiteness, and in Nahuatl it's just being possessed or not (construct vs. absolutive). Neither Hebrew nor Nahuatl have a genitive case.

In Standard Arabic, case exists on a different plane and is combined with the construct/possessed state: accusative dual possessed صحنَي sˤaħn-ai 'the two plates of', acc. dual definite الصحنَين asˤ-sˤaħn-aini 'the two plates', acc. dual indefinite صحنَين sˤaħn-aini 'two plates'. And so also nominative dual possessed صحنا sˤaħn-aa, nominative dual definite الصحنان asˤ-sˤaħn-aani and so on. In Hebrew, an NP containing possession that involves the construct state can be perfectly marked with the preposition אֵת ʔet (~ אֶת־ /ʔɛt/), which marks direct objects, if it's definite.

A possessed noun can perfectly remain the head of its NP. You can see this because if an NP containing possession this way is the subject of the verb, the verb can still get to agree with the possessed head noun in those three languages with subject marking.

הִנֵּה יַד־ יְהוָה הוֹיָה בְּמִקְנְךָ
/hinˈne jad jəhoˈwɔ hoˈjɔ bə-miqnə-ˈkɔ/
behold hand.CONST[F] lord[M] be.PTCP.SG.FEM in-livestock.CONST-2SG.MASC
'Behold, the hand of the LORD is in your livestock (in the field, in your horses, donkeys, camels...).' (Exodus 9:3)

Here, the participle agrees with a possessed noun, feminine 'hand'. As in English, quantifiers may be ignored even when they're found in constructions that resemble possession. Compare the following with "The majority of the people say~says this" (among people that don't treat "majority" as a collective that takes the plural).

וַתֵּצֶאןָ כָֽל־ הַנָּשִׁים אַחֲרֶיהָ
/wat-teˈsˤɛnɔ kɔl han-nɔˈʃim ʔaħaˈrɛ-hɔ/
and-go.out.PRF.3PL.FEM all.CONST[M] the-women after-3SG.FEM
'And all the women went out after her (in celebration).' (Exodus 15:20)

Note כֹּל kol 'the whole of, every instance of' is a masculine noun on its own...

Pabappa wrote: ↑Wed Nov 25, 2020 11:29 pm offhand, does anyone know if the Spanish diminutive -ito can be attached to nouns with the agent suffix -or? e.g. actorito? how about -ador?

I checked with my wife (native speaker, Peru), and she would say e.g. actorcito, amorcito, emperadorcito.

The Spanish diminutive is very productive, you can add it to just about anything.

zompist wrote: ↑Thu Nov 26, 2020 11:04 amand she would say e.g. actorcito, amorcito, emperadorcito.

Yes, me too. I should've really mentioned the insertion of this -c- as it's not obvious.

The Spanish diminutive is very productive, you can add it to just about anything.

Could you please ask her if she can apply it to -ía words like calcomanía 'sticker' and sabiduría 'wisdom'?

Kuchigakatai wrote: ↑Thu Nov 26, 2020 2:01 pm Could you please ask her if she can apply it to -ía words like calcomanía 'sticker' and sabiduría 'wisdom'?

Yes, though she hesitates between calcomaniita and calcomanita.

Pabappa wrote: ↑Wed Nov 25, 2020 11:29 pmoffhand, does anyone know if the Spanish diminutive -ito can be attached to nouns with the agent suffix -or? e.g. actorito? how about -ador? I see a small number of hits for such words but I dont know of any such words that are considered part of the daily lexicon.

Not **actorito but actorcito. We've also got cafecito (or even cafelito) instead of **cafeíto.

Butː

mēhar qoḏšô (Psalms 3:5)

from-hill holiness-his

from His holy hill

Richard W wrote: ↑Thu Nov 26, 2020 4:00 pmButː

mēhar qoḏšô (Psalms 3:5)

from-hill holiness-his

from His holy hill

Oh, that's pretty interesting. Why not 'from the hill of his holiness'? Is it because har qoḏš-ô is some sort of very lexified phrase? I guess I could eventually find this if I look for it in grammars, but do you happen to know?

My Hebrew is pretty bad, but you just reminded me that in Standard Arabic (which I'm more comfortable with), it is the case that very lexified possessed-possessor constructions, pragmatically often equivalent to English compounds, have a possessive clitic attached on the possessor modifier, e.g.:

فرشاة أسناني
furʃaatu ʔasnaan-ii
brush.CONST.NOM teeth.CONST.GEN-1SG
'my toothbrush'

...which is the exact same construction in mē-har qoḏš-ô. Normally, though, what you find is that the possessive construction is "broken up" by a preposition that expresses possession (such as لــ li- 'to (marks indirect objects), of (marks possessors)'), right after the possessive clitic.

مدرستي لتنس الطاولة
madrasat-ii li-tinis atˤ-tˤaawilat-i
school.CONST.NOM-1SG to-tennis.CONST.GEN the-table-DEF.GEN
'my table tennis school'
(literally: my school to/of the tennis of table)


(I took that last example from Reverso Context Arabic, and the actual example was: Turning my school of table tennis into a den of thieves. محولا مدرستي لتنس الطاولة إلى وكر من اللصوص. What a weird phrase. Was this taken from the subs of a table tennis anime or something???)

zompist wrote: ↑Thu Nov 26, 2020 2:50 pmYes, though she hesitates between calcomaniita and calcomanita.

I only have calcomaniita. In El Salvador we also have the more colloquial pronunciation calcomani[j]a ("calcomanilla", but etymologically this is not the diminutive -illa, but just an epenthetic [j]), and that can also become calcomani[j]ita.