(Conlangs) Q&A Thread - Quick questions go here [2010-2020]

[Chagen]

Ah, okay.

This should result in some interesting things happening when I apply sound changes--for instance, the proto plural form "Bhilubub" becoming "Bhiluvub" in the daughter languages.

[Lambuzhao]

In Rozwi, at a certain stage in its development, some verbs formed various principle parts by means of reduplication.

E.G.

Ētma – die (PRS)
Ētmamao – die (PST)

niddu - give (PRS)
niddudao gave (PST)

Vezt – live (PRS)
Veztiviza – alive (ADJZ)

At least with /Ētma/ and /niddu/, it seems to be the word-final reduplication, that is Final R → L copying as in natlang Sirionó.

with /vezt/ it seems to be word-initial elements redupled and tacked on to the end, sort of like Final L → R copying in natlang Chukchi.

Until your post, I didn't even know what to call it except for reduplication or repeating.
Gracias, Chagen~!

[Lambuzhao]

Totally other question -

We have a smiley/flag to represent conlang -- .
Do we have a smiley/flag to represent natlang (the concept in general, that is)?
I don't even have an idea what such a thing would look like (earth-globe with an "L" ?). Does ZBB or another Forum have such a symbol?

Is this an extraneous /over the top request?

Maybe i'm too into smileys/flags.

[eldin raigmore]

Yeah, a smiley that looked like a U.N. flag would work for that.

[Lambuzhao]

Yeah, a smiley that looked like a U.N. flag would work for that.

Excellent suggestion!
Have we got one?

[Iron]

Can an absolutive-ergative language feature a dative case?

[Omzinesý]

Can an absolutive-ergative language feature a dative case?

Yes, many do. I think Basque is one of them.

The other option is to encode the recipient (which would be in the dative) with the absolutive and have another case for the theme (what is given).

[Creyeditor]

Can an absolutive-ergative language feature a dative case?

Yes, for example this.

[Iron]

That helps a lot (both of you), actually. Thank you(s).

[Micamo]

Can an absolutive-ergative language feature a dative case?

Such a strategy appears to be the majority, actually.

[eldin raigmore]

Yeah, a smiley that looked like a U.N. flag would work for that.

Excellent suggestion! Have we got one?

Not yet. I thought you'd make one.
Somewhere on the CBB somebody posted the instructions for how to make a smiley, starting with an image of a flag (jpg or gif or what I don't remember).

[eldin raigmore]

Can an absolutive-ergative language feature a dative case?

Such a strategy appears to be the majority, actually.

But only barely -- 13 out of 25. (Which is 52%, but the sample size of languages in that table is fewer than 100 ergative languages and fewer than 100 indirect-object languages).
(And actually that chart talks about indirect objects; dative strategies, but not necessarily dative cases).

Anyway, to test whether that's significant, divide the 138 languages for which both features are recorded in that table, between ergative (25) and not-ergative (113), and between indirect-object (74) and other (64).

If the properties were independent of one another, one would expect:
(25 * 74)/138 = about 13.4 ergative-absolutive indirect-object languages (there are 13, quite close)
(25 * 64) /138 = about 11.6 ergative-absolutive other languages (there are 12, quite close)
(113 * 74)/138 = about 60.6 non-ergative indirect-object languages (there are 61, quite close)
(113 * 64)/138 = about 52.4 non-ergative other-strategy languages (there are 52, quite close)

Then run a chi-square test. I'm not going to, because between the nearly-negligible difference (0.4) and the very-small sample-size I'm pretty sure it is not significant even at a p = 0.5 level. That's a guess, but anyway, WALS.info's data doesn't support any correlation between ergativity and dativeness.

[Micamo]

Can an absolutive-ergative language feature a dative case?

Such a strategy appears to be the majority, actually.

But only barely -- 13 out of 25. (Which is 52%, but the sample size of languages in that table is fewer than 100 ergative languages and fewer than 100 indirect-object languages).
(And actually that chart talks about indirect objects; dative strategies, but not necessarily dative cases).

Anyway, to test whether that's significant, divide the 138 languages for which both features are recorded in that table, between ergative (25) and not-ergative (113), and between indirect-object (74) and other (64).

If the properties were independent of one another, one would expect:
(25 * 74)/138 = about 13.4 ergative-absolutive indirect-object languages (there are 13, quite close)
(25 * 64) /138 = about 11.6 ergative-absolutive other languages (there are 12, quite close)
(113 * 74)/138 = about 60.6 non-ergative indirect-object languages (there are 61, quite close)
(113 * 64)/138 = about 52.4 non-ergative other-strategy languages (there are 52, quite close)

Then run a chi-square test. I'm not going to, because between the nearly-negligible difference (0.4) and the very-small sample-size I'm pretty sure it is not significant even at a p = 0.5 level. That's a guess, but anyway, WALS.info's data doesn't support any correlation between ergativity and dativeness.

My apologies: When I said such strategies "were the majority" I meant "the strategy is a majority strategy in languages in general, and ergativity doesn't change that." I should have been clearer. Your point about a dative strategy not necessarily being a dative case is a good one, though; I had assumed the poster meant dativeness in general rather than a dative case in particular.

[eldin raigmore]

My apologies: When I said such strategies "were the majority" I meant "the strategy is a majority strategy in languages in general, and ergativity doesn't change that." I should have been clearer. Your point about a dative strategy not necessarily being a dative case is a good one, though; I had assumed the poster meant dativeness in general rather than a dative case in particular.

My apologies right back!
I didn't mean to imply that you should have been clearer.
I wasn't really even sure anyone else would want the elaboration.
I was only being clearer for my own self; I investigated and decided to post the result.

[Chagen]

In an Ergative-Absolutive language, which is more common: an un-marked Ergative or an un-marked Absolutive?

[sangi39]

In an Ergative-Absolutive language, which is more common: an un-marked Ergative or an un-marked Absolutive?

From what I remember hearing, it's usually the absolutive that's unmarked while the ergative is more usually marked (although some languages, like Basque IIRC, mark both, while others mark neither on the noun and use things like word order to mark them instead).

Just a quick search online turns up:

In languages with ergative–absolutive agreement systems, the absolutive form is usually the most unmarked form of a word

and:

If any case has zero realization, or a zero allomorph, it will be absolutive or nominative.

[Chagen]

Wierd.

I assumed that the Ergative is the one most usually un-marked, since in the term "Erg-Absol" it comes first. In "Nom-Acc", the Nominative comes first and it's usually unmarked, so I expected the Erg-Absol system to be named under the same logic.

[Micamo]

Wierd.

I assumed that the Ergative is the one most usually un-marked, since in the term "Erg-Absol" it comes first. In "Nom-Acc", the Nominative comes first and it's usually unmarked, so I expected the Erg-Absol system to be named under the same logic.

It comes first because, in transitive clauses, the ergative is the subject and the absolutive is the object. The name has nothing to do with markedness. The situation where the ergative is null-marked and the absolutive takes marking (the so-called "Marked Absolutive" paradigm) is likely unattested in natlangs.

[eldin raigmore]

@chagen:
what all of them just said, plus:

"Absolutive" comes from "absolute" which, in old-time morphology terms, meant "without inflection".

"Nominative" comes from the same root as "noun" and, in old-time morphology terms, meant "just the noun, ma'am. Just the noun". (But it doesn't mean that nowadays, at least not when discussing a "marked nominative" language.)

"Absolutive" and "nominative" are near-synonyms.
Both of them label the treatment (including the case if the language has cases) of the S sole participant of intransitive clauses, should it happen that in the overwhelming majority of intransitive clauses the only participant is always treated the same way.
If A the agents of transitive clauses are treated the same as S we call that treatment "nominative".
If P the patients of transitive clauses are treated the same as S we call that treatment "absolutive".
In split-S languages (including fluid-S languages and active/stative languages), there are two major types of intransitive clauses, and they treat their S differently.

"Ergative" means "the treatment (including the case, if the language has case) that's usually given to agents". It comes from a word that means "work", among other things. (from Greek ergatos "workman," from ergos "work"")

"Accusative" means "the treatment (including the case if the language hs case) that's usually given to patients". It comes from a Latin translation of a Greek term that was already unclear.

In tripartite languages (also split-transitive languages with "Austronesian-Philippine Alignment") the treatment of S could be called either Absolutive or Nominative. The treatment for all Agents in tripartite languages, or for nothing but Agents in split-transitive languages, could be called "ergative"; the treatment used for all Patients in tripartite languages, or for nothing but Patients in split-transitive languages, could be called "accusative".

[Chagen]

I recently added a volitional distinction to one of my conlangs, but I'm stuck on what to actually do with it. I mean, I know that it encodes the subject's volition in the action, obviously, but I'm stuck on how to make it interesting and not tacked-on. What exactly can a volitional distinction be used for?