My Avrelang

[tezcatlip0ca]

Also, in Soviet Russia, copula zeros you.

No, in Soviet Russia zeros copulate you.

Th[3:]ded.

[Bristel]

Also, in Soviet Russia, copula zeros you.

No, in Soviet Russia zeros copulate you.

Th[3:]ded.

f[ɔ˞]thed

[vec]

Hm?

Don't quite understand the last two posts.

Anyway, I got busy with school so this is low priority at the moment, but I will get back to it as soon as I have time on my hands.

[Bristel]

Hm?

Don't quite understand the last two posts.

Anyway, I got busy with school so this is low priority at the moment, but I will get back to it as soon as I have time on my hands.

Canepari wrote "thirded" with the vowel of his dialect of English.

I wrote "fourthed" with the vowel of my dialect of English.

I don't remember where Canepari is from, but I suspect from somewhere with a non-rhotic accent of English.

[tezcatlip0ca]

No, I use [M`=]. I said [3:] because of the number 3.

[tezcatlip0ca]

I've also used fou[4]thed.

[Nortaneous]

No, I use [M`=]. I said [3:] because of the number 3.

nitpick: [M] is [ɯ], which is a vowel, so you don't need the syllabicity marker. You probably meant [M\`=], which is [ɰ˞̩], although I don't see a reason not to just use [M`].

[tezcatlip0ca]

Yeah, I meant [M\`=].

[TomHChappell]

.... I got busy with school so this is low priority at the moment, but I will get back to it as soon as I have time on my hands.

Is that now, I hope?
If not now, when?
And btw how's school going, anyway?

[vec]

So, I'm going to try to figure out the possessive thingies.

They work opposite of our genitives:

My mother's dog is rendered as mother [dog-owned-by-her]

Right?

Anyone care to explain hebrew and arabic possessives for me which do this same stuff?

And then there are possessive suffixes, but I'm assuming those refer to the the possessor, not the possessee.

And thanks Tom, school's going great!

[TomHChappell]

So, I'm going to try to figure out the possessive thingies.
They work opposite of our genitives:
My mother's dog is rendered as mother [dog-owned-by-her]
Right?
Anyone care to explain hebrew and arabic possessives for me which do this same stuff?

It's called "construct state", if that helps.
I think I've seen a good explanation but that doesn't really mean I can explain it myself.
I remember it was discussed on the CONLANG@LISTSERV.BROWN.EDU mailing-list.
One person who was involved in that discussion is Carsten Becker aka Guitarplayer aka Dampantingaya Tatapyranga.
Perhaps you should ask him; he likely knows where to get a good explanation.
Grzegorz Jagodzinski also seems to have a good handle on it. I don't have a URL for him, though.
You probably also would benefit from asking the major contributors to the 3Cons thread in the L&L Museum on this ZBBoard.

And then there are possessive suffixes, but I'm assuming those refer to the the possessor, not the possessee.

In what language?

I'd say if the possessor comes before the possessum it seems natural for any construct-state to be shown by a prefix on the possessum; whereas if the possessor comes after the possessum it seems natural for the construct-state to be shown by a suffix on the possessum.

But the natlangs I know about that have a "construct state" are 3Cons (triconsonantal root) languages.
AFAICT IIRC the "construct state" is just the root with as few vowels as possible.

Though if you read Milewski's typology I think you'll see he thought there were several other natlangs that also had head-marking in "genitive constructions", that is, the head-noun, the one being modified by the other noun -- for instance, the head-noun would be the possessum in a possessive phrase -- is marked to show the relation, while the attribute-noun --- e.g. the possessor --- is left unmarked.

Look at WALS.info Feature/Chapter 24: Locus of Marking in Possessive Noun Phrases by Johanna Nichols and Balthasar Bickel and look at the head-marking languages (they've listed 78 of them); for instance, Acoma, Fijian, and Yoruba.

In Acoma, the possessor comes before the possessum, the possessor is unmarked, and the possessum is prefix-marked to agree with the person and number of the possessor.

In Fijian, the possessum comes before the possessor, the possessum is suffix-marked just to show that it is a possessum (it doesn't have to agree with anything about the possessor), and the possessor is unmarked.

In Yoruba, the possessum comes before the possessor, the possessum's final vowel is lengthened to mark it as a possessum (it doesn't agree with the possessor), and the possessor is unmarked.

So Yoruba's "construct state" is shown by ablaut or apophony on the final vowel of the possessum, not by an affix.

_____________________________________________________________________________________

You might also want to look at
http://wals.info/feature/combined?id1=24&id2=26
and/or
http://wals.info/feature/combined?id1=24&id2=86.

Perhaps also at http://wals.info/feature/combined?id1=26&id2=86

And thanks Tom, school's going great!

Good!

EDIT: as Cedh pointed out I should have said Dampantingaya.

[Cedh]

The construct state, when used possessively, is basically a grammatical marker that says "hey, this noun is possessed by something else", but it doesn't specify by whom it is possessed. There are many languages though which have true possessive affixes that differ by e.g. person, number, animacy of the possessor - this is quite similar to pronominal agreement on verbs, and in fact possessive affixes and verbal agreement morphemes frequently share the same etymology in languages that have both. Sometimes, possessive affixes make an additional distinction between alienable and inalienable possession; for instance, they might be used with only one type of possessive relationships, or there might be different affixes for different types of possession.

Using your example "my mother's dog", a language with a construct state might render that as 1SG.GEN mother dog.CONSTRUCT or 1SG mother.CONSTRUCT dog.CONSTRUCT, whereas a language with true possessive affixes (in this example, prefixes) might render that as 1SG-mother 3SG-dog. Unlike with a construct state, there would be two ways to simulate possessive affixes using only real English words: my-mother her-dog or the mother-of-mine the dog-of-hers.

One person who was involved in that discussion is Carsten Becker aka Guitarplayer aka Tatapyranga.

Dampantingaya. Tatapyranga is someone else.

[TomHChappell]

The construct state, when used possessively, is basically a grammatical marker that says "hey, this noun is possessed by something else", but it doesn't specify by whom it is possessed. There are many languages though which have true possessive affixes that differ by e.g. person, number, animacy of the possessor - this is quite similar to pronominal agreement on verbs, and in fact possessive affixes and verbal agreement morphemes frequently share the same etymology in languages that have both. Sometimes, possessive affixes make an additional distinction between alienable and inalienable possession; for instance, they might be used with only one type of possessive relationships, or there might be different affixes for different types of possession.

Using your example "my mother's dog", a language with a construct state might render that as 1SG.GEN mother dog.CONSTRUCT or 1SG mother.CONSTRUCT dog.CONSTRUCT, whereas a language with true possessive affixes (in this example, prefixes) might render that as 1SG-mother 3SG-dog. Unlike with a construct state, there would be two ways to simulate possessive affixes using only real English words: my-mother her-dog or the mother-of-mine the dog-of-hers.

Thanks.

One person who was involved in that discussion is Carsten Becker aka Guitarplayer aka Tatapyranga.

Dampantingaya. Tatapyranga is someone else.

Oops!
Thanks.
Fixed.

[vec]

Thanks for all of that. But it seems like I got confused reading the original article and overcomplicated the situation in my head.

In fact, Avrelang is dependent marking which is the same as English. So the possessor receives a suffix but the possessum can also be marked with a pronominal suffix; not both. This is a very similar situation to Turkish, if I'm not mistaken, rather than Arabic/Hebrew. The possessive suffix is going to have to be not a preposition and not a case suffix, though, to comply with other typological features, so probably a clitic like in English.

Adjectives are weird in that they can be used as both verbs and nouns without derivation in most cases, but they are also highly differentiated from genitives on the one hand and participles on the other. I'm thinking adjectives will have no number marking, even when used as "nouns", participles will indicate actors and patients. Adjectives as verbs will have limited TAM marking. Does that make sense?

[TomHChappell]

If it's the only case-marking and/or adposition you have, it won't break the averageness of your 'lang.

[vec]

If it's the only case-marking and/or adposition you have, it won't break the averageness of your 'lang.

But I thought "no case marking" won out? I feel like what you're referring to was discussed somewhere but I can't remember where. Why wouldn't it break the averageness, again?

[tezcatlip0ca]

Carsten Becker? Could Dampantingaya be a relative to Viktor77?

[TomHChappell]

If it's the only case-marking and/or adposition you have, it won't break the averageness of your 'lang.

But I thought "no case marking" won out? I feel like what you're referring to was discussed somewhere but I can't remember where. Why wouldn't it break the averageness, again?

Well, I could be wrong; but, read the chapter discussions on the features involved.

If it's the only adposition you have, it won't break the averageness for two reasons;
first, because the vast majority of languages have adpositions,
and second, because if the only adposition(s) is(are) used to mark genitive/possessive relationship, then Matthew Dryer (author of Feature 85) counted that language as a "no adpositions" language in Feature 85.

As for adpositions, http://wals.info/feature/description/48 suggests the vast majority of languages (209 out of 378) do have adpositions, but that those adpositions don't have person-marking.

http://wals.info/feature/description/85 also suggests the majority (1046 out of 1074) have adpositions; 520 have mostly postpositions, 467 have mostly prepositions, 7 have mostly inpositions, 52 have more than one type of adposition with none dominant, and only 28 have no adpositions.

A word is treated here as an adposition (preposition or postposition) if it combines with a noun phrase and indicates the grammatical or semantic relationship of that noun phrase to the verb in the clause. Some languages also employ adpositions to indicate a relationship of a noun phrase to a noun (especially in a genitive/possessive relationship); however, if the only candidates in a language for adpositions are in the genitive construction, they are not treated as adpositions here.

http://wals.info/feature/combined?id1=48&id2=85 lists only 13 languages that have "no adpositions" according to both features 48 and 85.

_____________________________________________________________________________________

As for cases,

The functional core of case morphology is the expression of the specific syntactic relations of clausal arguments.

which, to me, suggests that unless case marks at least the difference between subject and other, or between object and other, or between indirect object and other, then some people don't really count it as "case".

But it looks like Oliver A. Iggesen did count such languages has having "case", at least for purposes of WALS.info feature 49.

So I guess maybe if it was your only case-inflection, some people would say it's a derivation instead of an inflection, and therefore is not a "genuine case"; but WALS.info would say it is a "case", and so it would ruin your averageness.

BTW some authors, reputedly, don't count it as "case" unless at least one case is polysemous, having both a "syntactic" or "grammatical" use (such as nominative or absolutive or accusative or ergative or dative or dechticaetiative or genitive) and also a "semantic" use (such as locative or instrumental or ablative or allative or perlative).

_____________________________________________________________________________________

Adjectives are weird in that they can be used as both verbs and nouns without derivation in most cases, but they are also highly differentiated from genitives on the one hand and participles on the other. I'm thinking adjectives will have no number marking, even when used as "nouns", participles will indicate actors and patients. Adjectives as verbs will have limited TAM marking. Does that make sense?

I've forgotten, tell me again:
Why can adjectives be used as verbs with zero-derivation?
And why can they also be used as nouns with zero-derivation in the same language?
Is it really averageness, or is it something you've had to decide?

It makes sense for "adjectives used as nouns" (is that the same as "substantives"?) to get all the inflection the average common noun gets, at least in the positive degree.
Maybe you can't use the equative degree as a noun; maybe the comparative degree will also inflect for number; maybe the superlative degree can only be in the singular.
There may be other nominal accidents that "adjectives used as nouns" can inflect for, and other interactions between degree-of-comparison and those accidents.
http://wals.info/feature/combined?id1=60&id2=61 shows 37 natlangs in WALS.info's sample database for which both "adjectives, genitives, and participles are highly differentiated" and "adjectives can be used without nouns". Could you imitate some of those 37 natlangs?

A more substantive proposed correlation suggests that languages will allow adjectives without nouns to occur in bare form if and only if adjectives are the target of morphological agreement controlled by the noun. While this may hold as a statistical tendency, ....

I suggest, then, that "an adjective used as a noun" should be inflected for every nominal accident in which it would have had to agree with its head-noun, if it had had a head-noun.

It makes sense for "adjectives used as verbs" to get most of the inflection the average verb would get in the same situation. Aspect, modality/mode/mood, polarity, tense, voice, and maybe others also (e.g. evidentiality, mirativity, or pluractionality, or agreement with one or more participants in definiteness, gender, number, person, or pragmatic status).

Another possible correlation suggests that languages will allow adjectives without nouns to occur in bare form if and only if adjectives are themselves noun-like in their grammatical behaviour. To be in a position to test such a hypothesis, it is necessary to identify specific characteristics of noun-like adjectives, distinguishing them from other kinds of adjectives, such as verb-like ones.

So
(1) it suggests that languages that can use adjectives as nouns, whether or not they can do so without any marking, cannot also use the same adjectives as verbs (though perhaps they can use some adjectives as verbs and others as nouns); and
(2) it doesn't say anything about languages using adjectives as verbs without marking, nor anything about what fraction of languages can use adjectives as verbs.

Participles made from transitive roots definitely need to inflect for voice, so you can tell whether it means the modified noun was the agent or the modified noun was the patient. They probably also could need to inflect for either aspect or tense or both, especially if their root verb is monovalent, or maybe even if it's intransitive. I don't see why a participle would need to inflect for modality/mode/mood, but maybe I'm just unimaginitive.


_____________________________________________________________________________________

Carsten Becker? Could Dampantingaya be a relative to Viktor77?

Is Viktor a Becker? I didn't know.

Carsten lives in Deutschland and Viktor lives in Detroit, near Bucharest. So if they are related, it's probably as first-cousins at the closest.

[finlay]

I think it came up one time, and it's just an odd coincidence.

[tezcatlip0ca]

*bumpy bumpety bee*

[vec]

*bumpy bumpety bee*

Thanks. You might have to wait a while. So much to do.

[Jipí]

Carsten Becker? Could Dampantingaya be a relative to Viktor77?

Is Viktor a Becker? I didn't know.

Hi. No, I'm not related to Viktor as far as I know. Although Viktor is also a Becker IRL. However, Becker's only the 9th most common surname in Germany.

Regarding Dampantingaya ~ guitarplayer: I made that 2nd account some time ago because it was ~fashionable~. Also it seemed a good way to save those threads you wanted to go back to later at home after checking them out (instead of reading them whole in the lunch break at work) marked as "unread". I used it for a couple of months in more recent times because I borked my Guitarplayer account. Dampantingaya is what you get if you literally translate "guitar player" into my conlang (dap- 'to play (an instrument)' + manting 'lute-like instrument' + -maya 'masculine agent derivative suffix') and do some metathesis for euphony.

[cromulant]

Couple things I notice about Avrelang on CALS:

1. Initial velar nasal
2. Antepenultimate stress

These are both minority features cross-linguistically, so changing them will make Avrelang more average.

[vec]

Bumping in case it's getting pruned soon. I'm gonna get back on it soon, promise!

[finlay]

i think zompist turned off pruning for the main boards. (quickies and ephemera still have it) you can see this for yourself by the fact that the posts on the last page are now almost a year old.