felipesnark wrote:Shonkasika nouns decline differently according to animacy . For the nominative and accusative cases, animate nouns take the suffixes -s and -d while inanimate nouns take -k and -(nothing) respectively. The first and second person pronouns (animate) are different, taking nothing in the animate case.
I'm thinking of changing that so that the 1p and 2p pronouns take -dz in the nominative and -(nothing) in the accusative. Shonkasika is a pro-drop language due to the clear verb endings; thus, accusative forms will appear more often. So yay or nay for new (distinct) case forms for the 1p and 2p pronouns?
This satisfies the "universal" that,
"if one of the cases is unmarked, it includes as one of its meanings the subject of the intransitive",
only if your 'lang's Alignment of Case Marking of Pronouns is Ergative - absolutive .
felipesnark wrote:Shonkasika nouns decline differently according to animacy . For the nominative and accusative cases, animate nouns take the suffixes -s and -d while inanimate nouns take -k and -(nothing) respectively. The first and second person pronouns (animate) are different, taking nothing in the animate case.
I'm thinking of changing that so that the 1p and 2p pronouns take -dz in the nominative and -(nothing) in the accusative. Shonkasika is a pro-drop language due to the clear verb endings; thus, accusative forms will appear more often. So yay or nay for new (distinct) case forms for the 1p and 2p pronouns?
This satisfies the "universal" that,
"if one of the cases is unmarked, it includes as one of its meanings the subject of the intransitive",
only if your 'lang's Alignment of Case Marking of Pronouns is Ergative - absolutive .
Well, Shonkasika's alignment is nominative-accusative, so I don't know. But I already have a marked nominative for the inanimate nouns. Those nouns have -k nominative and no ending in the accusative.
Visit my website for my blogs and information on my conlangs: http://grwilliams.net/ It's a work in progress!
felipesnark wrote:Well, Shonkasika's alignment is nominative-accusative, so I don't know. But I already have a marked nominative for the inanimate nouns. Those nouns have -k nominative and no ending in the accusative.
Marked nominative with marked accusative is attested by several languages in WALS.info's database. I can't tell how many.
However, they included no languages that have marked nominative with unmarked accusative. There may be no such languages. According to WALS.info, languages that have marked nominative with unmarked accusative are very rare. Though languages that have marked absolutive with unmarked ergative are even rarer; they know of only one.
They've included 6 languages in their database that do that with their nouns; and 3 (all included among the aforesaid 6) that (also) do that with their pronouns.
Edit: I had misread the parts of the chapter texts for chapters 98 and 99 that had to do with marked nominatives.
Many languages have neither nominative/absolutive, nor accusative, nor ergative, marked on their nouns (98 languages), or on their pronouns (79 languages), or both (70 languages).
The overwhelming majority of such languages all rely on marking the verb instead. (69 out of 79 (~87%) that don't mark their pronouns do mark* their verbs; 76/98 (~78%) that don't mark their nouns do mark* their verbs. 50/70 (~71%) of the languages that mark neither their nouns nor their pronouns for these cases, do mark* their verbs.).
*(That is, mark their verbs in such a way that addressees can tell which participant the speaker means is the agent and which the patient.)
If you make your language marked nominative / unmarked accusative, you will be violating one or two or three "statistical universals".
Since all three are "statistical" rather than "absolute", that means there is some known exception to each one of them. WALS.info didn't include any such exceptions, as near as I can tell.
Also, there may be no language which is an exception to all three; nor even to any two of them.
So what?
Go for it!
If you can't make it work, or have trouble making it work, you'll gain insight into why it's so rare among natlangs.
If you can make it work, that'll probably be interesting too, especially to the rest of us.
I only wanted you to go into it "with eyes open";
that is, to be aware that it might be unnaturalistic and/or unrealistic, if you want your conspeaker community to consist of standard average modern humans.
However, there are other changes that would be "more naturalistic" and/or "more realistic".
You could make the markings on nouns and/or pronouns be "split-ergative";
or you could make the markings on nouns and/or pronouns be "neutral", and have the agreement-marking on the verb disambiguate for you instead.
Spoiler:
Agents just naturally tend to be animate; consequently addressees just tend to assume that animates must be agents, unless the speaker explicitly marks them as patients.
So you might assume nouns and pronouns for animates are unmarked nominative and marked accusative.
OTOH inanimates just naturally tend to be patients. If an inanimate is an agent, that's a remarkable fact, and the speaker should mark it for the addressee's benefit.
So you might assume nouns and pronouns for inanimates to be unmarked absolutive and marked ergative.
Or else, they are all neutral (no difference between nominative and accusative nor between ergative and absolutive).
Since inanimates are almost always 3rd-person (a 2nd-person inanimate is kinda weird, and a 1st-person inanimate is symptomatic of hallucination):
and also, 1st-persons are always animate, and 2nd-persons are sentient in any "felicitous" speech-act, and usually not only animate, but the same species as the speaker, if the speaker expects a reply or expects the 2nd-person to carry a message:
there's a strong correlation in at least one direction between 1st- and 2nd- person and animacy, and between 3rd-person and inanimacy.
That leads to a somewhat-less-strong correlation between 1st- and 2nd- -person agents and 3rd-person patients.
Your 1st- and 2nd- person pronouns, therefore, are likelier to be unmarked nominative / marked accusative.
Or maybe neutral (nominative/absolutive, accusative, and ergative are all equally unmarked).
3rd-person pronouns may be handled differently depending on the animacy of their referents.
Animate 3rd-person pronouns might be unmarked nominative / marked accusative, (or neutral);
Inanimate 3rd-person pronouns might be unmarked absolutive / marked ergative, (or neutral).
In several split-ergative languages, all animates are unmarked nominative / marked accusative and all inanimates are unmarked absolutive / marked ergative. [citation needed]
In several (5 in WALS.info) split-ergative languages, all pronouns are unmarked nominative / marked accusative and all nouns are unmarked absolutive / marked ergative.
I don't know that I've seen them mentioned, but I wouldn't be surprised to learn there are languages whose 1st- and 2nd- -person pronouns are all unmarked nominative / marked accusative, but whose 3rd-person pronouns are all unmarked absolutive / marked ergative.[citation needed]
Last edited by eldin raigmore on 25 Jul 2017 12:11, edited 6 times in total.
eldin raigmore wrote:However, they included no languages that have marked nominative with unmarked accusative. There may be no such languages.
Now that's just not true. The 6 marked nominative languages are exactly those languages. Languages with marked accusative and marked or unmarked nominative are all lumped into "Nominative - accusative (standard)". Please read the WALS again?
eldin raigmore wrote:However, they included no languages that have marked nominative with unmarked accusative. There may be no such languages.
Now that's just not true. The 6 marked nominative languages are exactly those languages. Languages with marked accusative and marked or unmarked nominative are all lumped into "Nominative - accusative (standard)". Please read the WALS again?
If that's true, my inanimate nouns (marked nominative with -k and unmarked accusative) are okay. I could also switch markings for the inanimate nouns, having nothing for nominative and -k for accusative. Right now, animate nominatives have -s and animate accusatives have -d. At the moment, 1p and 2p personal pronouns have nothing for nominative and -d for accusative. Shonkasika is apriori, but I'd prefer for pick naturalistic alternatives.
Visit my website for my blogs and information on my conlangs: http://grwilliams.net/ It's a work in progress!
felipesnark wrote:Are there systems that are split-ergative by animacy? Like animate nouns are nominative-accusative, and inanimate nouns are ergatative-absolutive?
I am pretty sure that some Australian languages do this.
Creyeditor "Thoughts are free." Produce, Analyze, Manipulate 12344 Ook & Omlűt & Nautli languages & Sperenjas Papuan languages, Morphophonology, Lexical Semantics
felipesnark wrote:If that's true, my inanimate nouns (marked nominative with -k and unmarked accusative) are okay. I could also switch markings for the inanimate nouns, having nothing for nominative and -k for accusative. Right now, animate nominatives have -s and animate accusatives have -d. At the moment, 1p and 2p personal pronouns have nothing for nominative and -d for accusative. Shonkasika is apriori, but I'd prefer for pick naturalistic alternatives.
It is true. I had misread the chapter.
in [url]http://wals.info/chapter/98[/url] Alignment of Case Marking of Full Noun Phrases, Bernard Comrie wrote:Note that the definition of the nominative–accusative system says nothing about how the distinction between S/A and P is marked. In Latvian, both nominative and accusative have overt markers. However, it is also possible for just the accusative to have an overt marker, as in Hungarian, where the word for ‘person’ is ember in the nominative, but ember-t in the accusative. Much less frequently cross-linguistically, it is the nominative that has an overt marker and the accusative that lacks one, as in Harar Oromo (Cushitic, Afroasiatic; Ethiopia) examples (3a–b).
....
Since the “marked nominative” type illustrated by Harar Oromo is a topic of current typological and theoretical interest, it has been given a separate encoding in the maps, contrasting with the standard type where either just the accusative or both nominative and accusative are marked.
Exactly the same paragraph is in http://wals.info/chapter/99 "Alignment of Case Marking of Pronouns". Unfortunately, it also has exactly the same examples; which is a shame, because the examples have only full noun-phrases in them; there aren't any examples of marked-nominative / unmarked-accusative pronouns in chapter 99, though three such languages are sampled in map 99A.
felipesnark wrote:If that's true, my inanimate nouns (marked nominative with -k and unmarked accusative) are okay. I could also switch markings for the inanimate nouns, having nothing for nominative and -k for accusative. Right now, animate nominatives have -s and animate accusatives have -d. At the moment, 1p and 2p personal pronouns have nothing for nominative and -d for accusative. Shonkasika is apriori, but I'd prefer for pick naturalistic alternatives.
It is true. I had misread the chapter.
in [url]http://wals.info/chapter/98[/url] Alignment of Case Marking of Full Noun Phrases, Bernard Comrie wrote:Note that the definition of the nominative–accusative system says nothing about how the distinction between S/A and P is marked. In Latvian, both nominative and accusative have overt markers. However, it is also possible for just the accusative to have an overt marker, as in Hungarian, where the word for ‘person’ is ember in the nominative, but ember-t in the accusative. Much less frequently cross-linguistically, it is the nominative that has an overt marker and the accusative that lacks one, as in Harar Oromo (Cushitic, Afroasiatic; Ethiopia) examples (3a–b).
....
Since the “marked nominative” type illustrated by Harar Oromo is a topic of current typological and theoretical interest, it has been given a separate encoding in the maps, contrasting with the standard type where either just the accusative or both nominative and accusative are marked.
Exactly the same paragraph is in http://wals.info/chapter/99 "Alignment of Case Marking of Pronouns". Unfortunately, it also has exactly the same examples; which is a shame, because the examples have only full noun-phrases in them; there aren't any examples of marked-nominative / unmarked-accusative pronouns in chapter 99, though three such languages are sampled in map 99A.
Thanks for the WALS info. I actually went and read over those chapters. While marked nominative is attested, I decided to go with Option B, with an unmarked nominative for inanimate nouns. I also decided to stick with the current system for 1p and 2p personal pronouns for now. Thanks for the conversation, guys.
Visit my website for my blogs and information on my conlangs: http://grwilliams.net/ It's a work in progress!
felipesnark wrote:If that's true, my inanimate nouns (marked nominative with -k and unmarked accusative) are okay. I could also switch markings for the inanimate nouns, having nothing for nominative and -k for accusative. Right now, animate nominatives have -s and animate accusatives have -d. At the moment, 1p and 2p personal pronouns have nothing for nominative and -d for accusative. Shonkasika is apriori, but I'd prefer for pick naturalistic alternatives.
It is true. I had misread the chapter.
in [url]http://wals.info/chapter/98[/url] Alignment of Case Marking of Full Noun Phrases, Bernard Comrie wrote:Note that the definition of the nominative–accusative system says nothing about how the distinction between S/A and P is marked. In Latvian, both nominative and accusative have overt markers. However, it is also possible for just the accusative to have an overt marker, as in Hungarian, where the word for ‘person’ is ember in the nominative, but ember-t in the accusative. Much less frequently cross-linguistically, it is the nominative that has an overt marker and the accusative that lacks one, as in Harar Oromo (Cushitic, Afroasiatic; Ethiopia) examples (3a–b).
....
Since the “marked nominative” type illustrated by Harar Oromo is a topic of current typological and theoretical interest, it has been given a separate encoding in the maps, contrasting with the standard type where either just the accusative or both nominative and accusative are marked.
Exactly the same paragraph is in http://wals.info/chapter/99 "Alignment of Case Marking of Pronouns". Unfortunately, it also has exactly the same examples; which is a shame, because the examples have only full noun-phrases in them; there aren't any examples of marked-nominative / unmarked-accusative pronouns in chapter 99, though three such languages are sampled in map 99A.
Thanks for the WALS info. I actually went and read over those chapters. While marked nominative is attested, I decided to go with Option B, with an unmarked nominative for inanimate nouns. I also decided to stick with the current system for 1p and 2p personal pronouns for now. Thanks for the conversation, guys.
felipesnark wrote:Are there systems that are split-ergative by animacy? Like animate nouns are nominative-accusative, and inanimate nouns are ergatative-absolutive?
This is basically the system in Hittite. "Common" nouns are nominative/accusative, and neuter nouns are ergative/absolutive. Though the "common" class includes inanimates as well, it's just that neuter includes only inanimates.
me wrote:Exactly the same paragraph is in http://wals.info/chapter/99 "Alignment of Case Marking of Pronouns". Unfortunately, it also has exactly the same examples; which is a shame, because the examples have only full noun-phrases in them; there aren't any examples of marked-nominative / unmarked-accusative pronouns in chapter 99, though three such languages are sampled in map 99A.
Presumably you can get examples for any language sampled for map 99A by going to http://wals.info/valuesets/99A-xxx where xxx is their "languoid" code for that language.
For instance, (as Dr. Prof. Dryer pointed out), http://wals.info/valuesets/99A-ana gets you to a list of four example sentences in Araona, a language with ergative-absolutive alignment of case-marking of pronouns.
Three of the four example sentences actually contain pronouns.
Sentence igt-1733 is intransitive with first-person-singular (1s-ABS) subject <ema>;
sentence igt-1734 is transitive with first-person-singular (1s-ERG) agent <yama> and second-person (ABS, I guess?) recipient <midya>;
sentence igt-1731 looks (I'm guessing here!) to be "double-subject intransitive" (or "reciprocal", maybe?) with the plural absolutive noun <tata cana> and the 1st-sing-ABS pronoun <ema> as subjects (though perhaps <ema> is the object of an adposition?);
sentence igt-1732 is ditransitive with three full noun-phrases; a plural ergative agent <mama cana ja>, a plural absolutive recipient <tata cana>, and an (ABS?) theme <jana>.
Omzinesý wrote:Should I add diphthong /ie/ to Vålkakil for Slavonic loan words with /e/ after a palatalized consonant? Vålkakil doesn't have diphthongs ATM.
I have a crazier idea, why not split neuter into 2 other genders? It could be mostly based purely on phonology or be something like food and drink split off from neuter.
Nachtuil wrote:I have a crazier idea, why not split neuter into 2 other genders? It could be mostly based purely on phonology or be something like food and drink split off from neuter.
The problem with neuter is that I have no way to derive it from masculine and feminine, since the case patterns are like thus:
So does the protolang have only masculine and feminine? I thought you were contemplating going from a 3 gender system to a 2. There must be a way. They say Proto Indo European added the feminine gender through collective plurals. Your language doesn't have classifiers does it? That may be a way.
![[<3]](./images/smilies/heartic.png "<3"){kind=small}
Papuan languages, Morphophonology, Lexical Semantics 
Śād Warḫallun (Vrkhazhian) [