Zompist Bboard Again

Richard W wrote: ↑Sat Jan 11, 2020 8:10 am

Knit Tie wrote: ↑Thu Jan 09, 2020 12:38 pm How can a common singular nominal suffix arise? Such as the Latin -us, for example?

Like this, I suspect:

"A common singular nominal suffix, how can it arise?"

At least, I've come across natives speaking English like that.

That even fits the PIE case, where *so was an animate nominative/accusative demonstrative stem.

Yes. Such constructions exist in colloquial German, too:

Mein Bruder, der restauriert einen alten Porsche.

I have little doubt that the PIE nom. sg.*-s arose from *so in just that kind of construction. No genitive involved here, let alone an ergative (pace Beekes).

In typological works, I have come across the comment that topics getting reinterpreted as subjects while retaining their distinct topical marking is totally a thing and has been observed in many intra-family language comparisons. Usually it ends up creating or modifying existing conjugations, like:

1. WeepingElf, he likes swimming.
2. WeepungEwf he-likes swimmung.
3. WeepnÖf elikes swimn.

...so the new 3S form of "to like" is now e-like-s. But I think it's safe to say a demonstrative or article can be reinterpreted as a nominative suffix in the same kind of process.


However, KnitTie didn't ask about singular nominative suffixes, but nominal suffixes (affixes that distinguish nouns from verbs/adjectives). Maybe he meant to say nominative considering he mentioned Latin (a language most often talked about in conlanglands for its cases), but just in case he didn't, nominal suffixes often come from derivational suffixes getting reinterpreted as part of the stem. It especially happens with diminutives (see French and Mandarin for the classic examples, also Dutch for a more intermediate stage).

Ser wrote: ↑Sat Jan 11, 2020 12:00 pm However, KnitTie didn't ask about singular nominative suffixes, but nominal suffixes (affixes that distinguish nouns from verbs/adjectives). Maybe he meant to say nominative considering he mentioned Latin (a language most often talked about in conlanglands for its cases), but just in case he didn't, nominal suffixes often come from derivational suffixes getting reinterpreted as part of the stem. It especially happens with diminutives (see French and Mandarin for the classic examples, also Dutch for a more intermediate stage).

But that doesn't work for stem suffixes like Latin -us, as in opus 'work', venus 'desire', also an adjective suffix in vetus 'old' and part of a vowel suffix in sumus 'we are'!

I recently started looking into this old conlang of mine: http://www.frathwiki.com/Proto-Kunnu-l%C5%ABjungo

I based it on Finnish so I could quickly create a protolang without having to think too much about grammar. I remembered it being a total relex of Finnish and the article being more or less a stub, but it turned out that it wasn't that bad. It has several things Finnish doesn't have, and I had written quite a lot about it.

But I can't quite make sense of all of this, so I wanted to ask if anyone here can. There are two things in particular that I can't understand now, years later.

One thing is that this is supposed to be a split ergative language. It has an ergative case and a nominative-absolutive case. What on Earth is a nominative-absolutive case? I must've gotten it from Wikipedia somewhere, but I can't find that anymore. I remember being told once on ZBB that in a normal ergative language the argument of an intransitive verb will typically be a patient, experiencer, undergoer or somesuch. Is this correct? And if the argument instead is usually an agent, then that would make it a split ergative language?

The other thing I don't understand is that this language has an active voice, a passive voice and an "unmarked voice". >_<

Simply put, the unmarked voice is used in transitive sentences and the active voice in intransitive[...]

In intransitive sentences, any voice except for the unmarked voice can be used.

So the active voice is the default voice used on intransitive verbs, and the unmarked is the default voice used on transitive verbs? What could the purpose of this be? It seems like the active voice just marks a verb as being intransitive. So the unmarked voice is not used on intransitives, but can the active voice be used on transitives? There doesn't seem to be any example sentences where that occurs. I don't know if that's because it is indeed disallowed, or because I just hadn't thought of making such examples.

Is anyone able to analyze this conlang better than me? If the way the voices work right now is dumb, is it possible to take this active-unmarked distinction and make something sensible out of it?

Richard W wrote: ↑Sat Jan 11, 2020 1:59 pm

Ser wrote: ↑Sat Jan 11, 2020 12:00 pm However, KnitTie didn't ask about singular nominative suffixes, but nominal suffixes (affixes that distinguish nouns from verbs/adjectives). Maybe he meant to say nominative considering he mentioned Latin (a language most often talked about in conlanglands for its cases), but just in case he didn't, nominal suffixes often come from derivational suffixes getting reinterpreted as part of the stem. It especially happens with diminutives (see French and Mandarin for the classic examples, also Dutch for a more intermediate stage).

But that doesn't work for stem suffixes like Latin -us, as in opus 'work', venus 'desire', also an adjective suffix in vetus 'old' and part of a vowel suffix in sumus 'we are'!

Nobody's ever said derivational affixes need to be unambiguous, especially once you place them in a syntactic context. Spanish -o marks its stereotypical masculine singular nouns and its 1S present-tense verbs, but nouns are very often preceded by an article, and verbs often end in another ending, etc., etc.

You can forge garden paths along the lines of Time flies like an arrow; fruit flies like a banana, where flies can be a noun or a verb, but those things are very rarely encountered, unless you're Linguoboy reading newspaper headlines.

Qwynegold wrote: ↑Sat Jan 11, 2020 2:28 pm One thing is that this is supposed to be a split ergative language. It has an ergative case and a nominative-absolutive case. What on Earth is a nominative-absolutive case? I must've gotten it from Wikipedia somewhere, but I can't find that anymore. I remember being told once on ZBB that in a normal ergative language the argument of an intransitive verb will typically be a patient, experiencer, undergoer or somesuch. Is this correct? And if the argument instead is usually an agent, then that would make it a split ergative language?

I have never heard of a nominative-absolutive case, but I can guess that it corresponds to the nominative in nominative-accusative contexts, and an absolutive in ergative-absolutive contexts. This means that your ‘ergative’ case corresponds to both the accusative case and the ergative case. This sounds pretty normal for a split ergative language, as both the nominative and the absolutive are usually unmarked, and both the accusative and the ergative are usually marked.

(You also seem a bit confused about what ergativity actually is. In a nutshell: intransitive verbs have one argument, the experiencer, usually denoted as S. Transitive verbs have two arguments, the agent and the patient, usually denoted A and P respectively. Nominative-accusative languages, like English, group S and A in the nominative case, and P is in a separate accusative case. Ergative languages group S and P in the absolutive case, and A is in a separate ergative case. So a nominative-accusative language has ‘I-NOM laugh’ and ‘I-NOM see you-ACC’, whereas an ergative-absolutive language has ‘I-ABS laugh’ and ‘I-ERG see you-ABS’. Split-ergative languages have nominative-accusative alignment in some contexts, and ergative-absolutive alignment in others.)

The other thing I don't understand is that this language has an active voice, a passive voice and an "unmarked voice". >_<

Simply put, the unmarked voice is used in transitive sentences and the active voice in intransitive[...]

In intransitive sentences, any voice except for the unmarked voice can be used.

So the active voice is the default voice used on intransitive verbs, and the unmarked is the default voice used on transitive verbs? What could the purpose of this be? It seems like the active voice just marks a verb as being intransitive. So the unmarked voice is not used on intransitives, but can the active voice be used on transitives? There doesn't seem to be any example sentences where that occurs. I don't know if that's because it is indeed disallowed, or because I just hadn't thought of making such examples.

Is anyone able to analyze this conlang better than me? If the way the voices work right now is dumb, is it possible to take this active-unmarked distinction and make something sensible out of it?

This doesn’t make sense. If anything, I would expect intransitive verbs to have unmarked voice, and transitive verbs to have active voice (since intransitive verbs often don’t have a passive). And generally, the active voice is unmarked.

Ser wrote: ↑Sat Jan 11, 2020 7:07 pm

Richard W wrote: ↑Sat Jan 11, 2020 1:59 pm

Ser wrote: ↑Sat Jan 11, 2020 12:00 pm However, KnitTie didn't ask about singular nominative suffixes, but nominal suffixes (affixes that distinguish nouns from verbs/adjectives). Maybe he meant to say nominative considering he mentioned Latin (a language most often talked about in conlanglands for its cases), but just in case he didn't, nominal suffixes often come from derivational suffixes getting reinterpreted as part of the stem. It especially happens with diminutives (see French and Mandarin for the classic examples, also Dutch for a more intermediate stage).

But that doesn't work for stem suffixes like Latin -us, as in opus 'work', venus 'desire', also an adjective suffix in vetus 'old' and part of a vowel suffix in sumus 'we are'!

Nobody's ever said derivational affixes need to be unambiguous, especially once you place them in a syntactic context.

You did, above, as part of speech indicators. I'll allow them to be unreliable - much in language is.

With regard to Latin -us, its most typical use is for nominative singular of masculine substantives, which embraces both nouns and adjectives. As a suffix in noun stems, it doesn't look like an ex-diminutive - they typically yield concrete nouns.

Perhaps KnitTie will tell us what he meant.

I have no idea what's going on here.

Are there any agglutinative languages where tense and aspect are expressed by completely different affixes without any fusion? Based on Dahl (1985), I would be surprised to find any language like this, but I’d like to know if that’s correct.

bradrn wrote: ↑Sat Jan 11, 2020 11:08 pm Are there any agglutinative languages where tense and aspect are expressed by completely different aspects without any fusion? Based on Dahl (1985), I would be surprised to find any language like this, but I’d like to know if that’s correct.

Do you mean different affixes? Because I'm sure there are some languages like that. It mostly comes down to definition, because almost any detailed treatment of aspect will have some connotations of tense and vice versa. In Choctaw, for example, there are tense suffixes that mostly ignore aspect, and aspect (to the extent that it is indicated morphologically) is dealt with through other means like stem gradation, adverbial suffixes, auxilliaries, etc. But if you wanted to, you could analyze the tense suffixes as having some aspectual implications, especially when it comes to the speaker's choice of what other morphology to include in the same word with those suffixes.

Moose-tache wrote: ↑Sat Jan 11, 2020 11:53 pm

bradrn wrote: ↑Sat Jan 11, 2020 11:08 pm Are there any agglutinative languages where tense and aspect are expressed by completely different aspects affixes without any fusion? Based on Dahl (1985), I would be surprised to find any language like this, but I’d like to know if that’s correct.

Do you mean different affixes?

Yes, I did indeed! Fixed now.

Because I'm sure there are some languages like that. It mostly comes down to definition, because almost any detailed treatment of aspect will have some connotations of tense and vice versa. In Choctaw, for example, there are tense suffixes that mostly ignore aspect, and aspect (to the extent that it is indicated morphologically) is dealt with through other means like stem gradation, adverbial suffixes, auxilliaries, etc. But if you wanted to, you could analyze the tense suffixes as having some aspectual implications, especially when it comes to the speaker's choice of what other morphology to include in the same word with those suffixes.

Thanks! That’s interesting, because the impression I got from reading about aspect was that almost all agglutinative languages fuse tense and aspect affixes to some degree.

Richard W wrote: ↑Sat Jan 11, 2020 8:10 am

Knit Tie wrote: ↑Thu Jan 09, 2020 12:38 pm How can a common singular nominal suffix arise? Such as the Latin -us, for example?

Like this, I suspect:

"A common singular nominal suffix, how can it arise?"

At least, I've come across natives speaking English like that.

That even fits the PIE case, where *so was an animate nominative/accusative demonstrative stem.

I don't get it, sadly.
Is the "it" supposed to be it, so to say?

In Richard's example, the indefinite article "a" becomes a nominative case marker.

Knit Tie wrote: ↑Sun Jan 12, 2020 1:56 am

Richard W wrote: ↑Sat Jan 11, 2020 8:10 am

Knit Tie wrote: ↑Thu Jan 09, 2020 12:38 pm How can a common singular nominal suffix arise? Such as the Latin -us, for example?

Like this, I suspect:

"A common singular nominal suffix, how can it arise?"

At least, I've come across natives speaking English like that.

That even fits the PIE case, where *so was an animate nominative/accusative demonstrative stem.

I don't get it, sadly.
Is the "it" supposed to be it, so to say?

Yes, the use of that pronoun, it is supposed to be an example of how a nominative suffix can arise. Now Ser, he thinks you may have been asking something different. Weeping Elf, he gave an example of the same construction in German.

bradrn wrote: ↑Sun Jan 12, 2020 12:02 am the impression I got from reading about aspect was that almost all agglutinative languages fuse tense and aspect affixes to some degree.

I think the larger issue here is that tense and aspect are largely the same thing. Essentially we're asking "which languages split their timey affixes into two arbitrary categories, one for timeyness with respect to change or completion, and one for timeyness regardless of change or completion?" You're not going to catch many fish with that bait. It's a bit like a Navajo asking "how many languages fuse the concepts of stick-throwing and ball-throwing into a single verb for stick-or-ball-throwing?" The question tells you more about the linguistic assumptions of the asker than any answer you could come up with.

Ser wrote: ↑Sun Jan 12, 2020 2:34 am In Richard's example, the indefinite article "a" becomes a nominative case marker.

I don't see that at all. What you might read into my example is that one should not expect a 3rd person pronoun to become a suffixed nominative marker in a VSO language. I have a vague collection that a definiteness marker might be reinterpreted as a subject marker, but I don't recall any examples.

Yeah, I definitely got that point wrong.

Moose-tache wrote: ↑Sun Jan 12, 2020 9:10 am

bradrn wrote: ↑Sun Jan 12, 2020 12:02 am the impression I got from reading about aspect was that almost all agglutinative languages fuse tense and aspect affixes to some degree.

I think the larger issue here is that tense and aspect are largely the same thing. Essentially we're asking "which languages split their timey affixes into two arbitrary categories, one for timeyness with respect to change or completion, and one for timeyness regardless of change or completion?" You're not going to catch many fish with that bait. It's a bit like a Navajo asking "how many languages fuse the concepts of stick-throwing and ball-throwing into a single verb for stick-or-ball-throwing?" The question tells you more about the linguistic assumptions of the asker than any answer you could come up with.

This is more of a philosophical discussion, but I’ve been reading a lot about tense and aspect recently, and would argue that tense and aspect are definitely distinct from each other. The difference is easy to describe: tense locates an action in time with reference to another fixed point in time, whereas aspect is about how the speaker wants to view the extent of an action in time. Importantly, for most non-punctual actions, tense is objective while aspect is subjective: even if it sounds weird, we can normally describe an action as being imperfective, perfective, habitual etc., whereas usually only one choice of past, present, future etc. is valid.

A more important argument is that there are many languages which do separate tense and aspect. Most commonly, there are languages which mark aspect but not tense (e.g. Standard Mandarin); if tense and aspect are simply arbitrary divisions, then there would be no reason for these languages to mark one arbitrary-division but not another arbitrary-division. Similarly, I believe there are languages which mark tense but not aspect; these are much rarer, but I believe Kabardian does this.

Of course, tense and aspect are both closely related, as they are both used to mark different parts of the general category of ‘time’. But I believe there is enough evidence to justify a separation of tense and aspect as distinct concepts.

bradrn wrote: ↑Sun Jan 12, 2020 5:16 pmOf course, tense and aspect are both closely related, as they are both used to mark different parts of the general category of ‘time’. But I believe there is enough evidence to justify a separation of tense and aspect as distinct concepts.

I believe so too, but it's also the case that languages are constantly turning tense systems into aspect and back into tense.

Your original question could be kind of answered by Germanic and Romance (I know they're not agglutinative but bear with me). At some point they expressed tense with inflections, and the perfect with auxiliaries. Except, quite a lot of the individual languages reinterpreted the perfect as a past tense; some of them then innovated a new perfect...

zompist wrote: ↑Sun Jan 12, 2020 5:25 pm

bradrn wrote: ↑Sun Jan 12, 2020 5:16 pmOf course, tense and aspect are both closely related, as they are both used to mark different parts of the general category of ‘time’. But I believe there is enough evidence to justify a separation of tense and aspect as distinct concepts.

I believe so too, but it's also the case that languages are constantly turning tense systems into aspect and back into tense.

Yes, but this is hardly restricted to tense and aspect. Languages are constantly turning just about everything into anything, as is evidenced by e.g. Knit Tie and Richard W’s discussion in this same thread.

Your original question could be kind of answered by Germanic and Romance (I know they're not agglutinative but bear with me). At some point they expressed tense with inflections, and the perfect with auxiliaries. Except, quite a lot of the individual languages reinterpreted the perfect as a past tense; some of them then innovated a new perfect...

But what about other aspects? I know very little about Romance or Germanic (except to the extent that I am a native English speaker), but don’t they both have a perfective/imperfective distinction in addition to a perfect? (Although I don’t know whether that distinction is also made with auxilliaries or if it is with inflections.)