Zompist Bboard Again

Also, reduplication of part of a word can become nonconcatenative when internal changes, infixing or deletion makes the repeated segments no longer identical.

Znex wrote: ↑Wed Jan 09, 2019 3:06 am

mae wrote: ↑Wed Jan 09, 2019 12:11 am The liaison evidence is only one issue. Another is that despite the common contention that it's "implausible" or excessively "complex" for people to memorize a multitude of essentially arbitrary added segments, the data from acquisition, etc. shows that people do memorize all those things anyway. The subtractive analysis seemed at least reasonable to me when I first looked at the data, so like you I was also surprised to find that no one really uses that analysis anymore.

I suppose the more basic solution is to presume that both forms still have an underlying final segment? So froid still has |frwad| [fʁwa], and froide has |fʁwadE| [fʁwad]. Final segment deletion is hence a synchronic change rather than diachronic; the feminine form still has an archiphoneme marking the gender.

But then aren't you positing a segment (/ə/) that exists only to be deleted?
Also, I'd like to point out that pretty much everybody seems to be suggesting that /frwad/ is the base form. The only difference seems to be whether it represents the underlying masculine or feminine (or both? it seemed like that's what mae was implying).
NINJA EDIT: didn't see mèþru confirm it when I was writing

Nonconcatenative morphology isn't exactly vastly common in natlangs, don't forget. And loads of people at least attempt triconsonantal root conlangs.

Curlyjimsam wrote: ↑Wed Jan 09, 2019 2:07 pm Nonconcatenative morphology isn't exactly vastly common in natlangs, don't forget. And loads of people at least attempt triconsonantal root conlangs.

Neither are initial consonant clusters...

missals wrote: ↑Tue Jan 08, 2019 9:32 pm
if you know that the feminine is epuz, you can apply the general rule observable in the other pairs - knock off the last consonant - and correctly predict the masculine form, epu. Moreover, it is simpler to state that "The masculine is generated via deletion of the final consonant from the feminine" than to posit that -z, -ʃ, -d, etc., are all realizations of a feminine suffix whose variant phonological forms are unpredictably distributed across different vocabulary terms. This is why it is possible to describe French (the language above, of course) as having a type of disfixation.

I forget the source, but I remember reading somewhere that less than half of French adjectives actually have phonetically different forms for different genders anyway (I think the data was for a "reference" kind of Parisian accent, where there is no difference in pronunciation between words ending in e.g. -ée and -é or -ie and -i, if that matters). More than a handful of adjectives end in a consonant for both masculine and feminine, such as jaune, rose, rouge, sage, malade, timide, vide, simple, calme, sale, triste, pauvre. You could maybe deal with some such "exceptions" to the supposedly regular deletion rule (of deleting the final consonant in the feminine to get the masculine) by postulating that it doesn't apply to certain consonants or in certain environments (e.g. the clusters in "triste", "calme", "pauvre" and "simple") but that weakens the argument that a synchronic deletion rule is a simple way to account for all of the data.

Well, anyways, I didn't mean to drop a bomb about French grammar, I just wanted to show what disfixation could look like if it did exist in a language. Just pretend the words I listed are from some unrelated language that coincidentally resembles French.

Subtractive morphology, nearly grammaticalised verbal marking for three arguments, nasal vowels... clearly French has a strong Muskogean substrate!

It seems that nobody addressed this:

Frislander wrote: ↑Tue Jan 08, 2019 4:06 amIn all seriousness though I do kind of think that; analytic languages appear to be the result of the imposition of one language on another group and the acquisition errors are passed on leading to reduced morphology, and this can be seen that morphologically analytic languages are basically restricted to two regions: West Africa and East Asia.

I have a couple issues with this:
- A language is not analytic as a yes/no property, rather, we speak of an analysis-synthesis continuum in which some languages are more analytic than others. Vietnamese is more analytic than English, which is more analytic than Spanish, which is (arguably) more analytic than Classical Arabic, which is more analytic than etc., etc.
- We know of languages that have demonstrably lost a lot of morphology without having had an exterior language imposed on, e.g. Florentine/Roman Italian.

Yeah, the "imposition of one language on another group" has a big problem: it describes nearly the entire world, on one side or another. The languages of China and Southeast Asia are analytic, and you could argue that it's somehow because Chinese, Burmese, Vietnamese, etc. are imperial languages (and thus reduced) whereas Shan, Hmong, and Zhuang are minority languages (and thus simplified due to acquisition errors), but then show me the language where this process didn't happen. Arabic, French, Turkish, Persian, and Russian are all imperial languages, and just about every African and South Asian language is the language of a colonized group.

For what it's worth, my own Tagorese is heavily analytic: morphology is largely limited to suppletion and some compounding. Another of my conlangs, Qôni, is nonconcatenative, but very "Semitic-like". I think non-concatenative morphology like triconsonantal roots is probably one of the hardest kinds of conlang to do "well" without ending up too much like Semitic. For example, Zompist's Old Skourene I feel is only partially successful int his.

Furthermore, there is evidence that some of the changes that made Chinese and Southeast Asian languages analytic were before the establishment of empires. Consider also that many of the those who are now minorities there at various times had their own empires and vice versa.

Curlyjimsam wrote: ↑Wed Jan 09, 2019 2:07 pm Nonconcatenative morphology isn't exactly vastly common in natlangs, don't forget. And loads of people at least attempt triconsonantal root conlangs.

Wrong, there is loads of it. If the language is tonal (which is very common), there is a good chance that they have nonconcatenative morphology. In fact, many language has something like triconsonantal root, but for tone. And they're not related. From Chalcatongo Mixtec, Dogon, Maasai, to even Standard Mandarin (although the latter one is not productive)

Akangka wrote: ↑Mon Jan 14, 2019 2:02 pm

Curlyjimsam wrote: ↑Wed Jan 09, 2019 2:07 pm Nonconcatenative morphology isn't exactly vastly common in natlangs, don't forget. And loads of people at least attempt triconsonantal root conlangs.

Wrong, there is loads of it. If the language is tonal (which is very common), there is a good chance that they have nonconcatenative morphology. In fact, many language has something like triconsonantal root, but for tone. And they're not related. From Chalcatongo Mixtec, Dogon, Maasai, to even Standard Mandarin (although the latter one is not productive)

I would be interested to see sources on this. WALS chapter 20 says only ~6% of languages have non-concatenative morphology for case and/or tense/aspect/mood. Obviously that's not a complete sample of all morphology but it seems likely to be fairly indicative. Of course, this suggests there are hundreds of languages with non-concatenative morphology - but against several thousand without it.

If (say) half of tonal languages had tonal morphology, then that would still only be about a fifth of all languages, so still not particularly common. I suspect tone is heavily underrepresented in conlangs, and that tonal morphology is maybe pretty common in conlangs which do have tone.

I'd probably want to distinguish "tonal morphology" from regular tone sandhi processes arising from the addition of concatenative morphemes. But that is possibly debatable and it may not always be easy to tell the difference.

There are also questions about more marginal processes, e.g. non-concatenative morphology that only marks a small number of (possibly rare) distinctions or which is lexically restricted / "irregular" in some way. In Europe non-concatenative morphology is generally like this; WALS doesn't count it. But even if a high proportion of languages had just a tiny bit of non-concatenative morphology, it would still be the case that non-concatenative morphology was overall pretty rare.

Likely partial reduplication should also count, and it's super common.

Curlyjimsam wrote: ↑Mon Jan 14, 2019 3:19 pm

Akangka wrote: ↑Mon Jan 14, 2019 2:02 pm

Curlyjimsam wrote: ↑Wed Jan 09, 2019 2:07 pm Nonconcatenative morphology isn't exactly vastly common in natlangs, don't forget. And loads of people at least attempt triconsonantal root conlangs.

Wrong, there is loads of it. If the language is tonal (which is very common), there is a good chance that they have nonconcatenative morphology. In fact, many language has something like triconsonantal root, but for tone. And they're not related. From Chalcatongo Mixtec, Dogon, Maasai, to even Standard Mandarin (although the latter one is not productive)

I would be interested to see sources on this. WALS chapter 20 says only ~6% of languages have non-concatenative morphology for case and/or tense/aspect/mood. Obviously that's not a complete sample of all morphology but it seems likely to be fairly indicative. Of course, this suggests there are hundreds of languages with non-concatenative morphology - but against several thousand without it.

If (say) half of tonal languages had tonal morphology, then that would still only be about a fifth of all languages, so still not particularly common. I suspect tone is heavily underrepresented in conlangs, and that tonal morphology is maybe pretty common in conlangs which do have tone.

I'd probably want to distinguish "tonal morphology" from regular tone sandhi processes arising from the addition of concatenative morphemes. But that is possibly debatable and it may not always be easy to tell the difference.

There are also questions about more marginal processes, e.g. non-concatenative morphology that only marks a small number of (possibly rare) distinctions or which is lexically restricted / "irregular" in some way. In Europe non-concatenative morphology is generally like this; WALS doesn't count it. But even if a high proportion of languages had just a tiny bit of non-concatenative morphology, it would still be the case that non-concatenative morphology was overall pretty rare.

Well, I'm actually thinking about including the marginal processes.

Alon wrote: ↑Mon Jan 07, 2019 7:36 pm Are there good examples of analytic and nonconcatenative conlangs? If I'm right and there are very few of them, why do you think that is?

How's this one? In writing and language, it's a mix of Classical Chinese, Modern Chinese, Oracle Bone Script Chinese, Bronze Script Chinese, and English. I make some references in it to "The Five Lost Classics" from around Confucius' time or earlier, discovered in 1973, which I've been studying recently.

Then I'm also including some art of a ford that I made. This is in a style of art I've been using to illustrate my translations of 1700s and 1800s Eastern Algonquian fairy tales into 1600s Massachusett, it's a bit based on Ancient Mayan iconography but also maybe M C Escher.

It would take a long time to explain the whole thing.

There's sentence-final particles that mark the sentences as Questions. That "attack" is Passive is marked by adding two EARTHS to it because the "Five Lost Classics" say that earth is organized under Yang in the Yin-Yang dichotomy. If I had to write it out, I think I'd make Passivization via a vowel change whereas in real Classical Chinese, it would be a separate particle, I think.

"Where" is two words, I think like in Modern Chinese. "We" is made of two clear morphemes, and "our" is made of three, like in Modern Chinese. But "fords of Jordan" is "Jordan s fords" in Classical Chinese and Modern Chinese and 1600s Massachusett, I went English for that one.

Names like Ephraim and Jordon, I didn't know or look up etymologies for, I just made some evocative Chinese characters.

( I do this recently for some conlangs, see how much of an ancient language I can remember and then make up the rest based again on what I can remember. )

This one is based on exciting, short quotes from around Judges 7 22 in the 1970s Picture Bible, a full graphic novel of the Protestant Bible. It is absolutely epic and I doubt there's anything like it. Maybe I subconsciously made a reference in it to the beginning of Michael Crichton's "Timeline" where the hero falls asleep to tapes on everyday Middle English on the plane in preparation for time-travel to the Middle Ages. A little worried, he still goes to sleep, though the last thing he hears is, "Where is my sword?" "Varr bee myne shvaard?" or something like that, right?

I'm seeing these attachments twice. Does anyone know how to stop that? I never had that problem before.

My langs mostly end up isolating even though I kind of always wanted to do agglutinating ones because a bunch of other preferences I have have kind of railroaded me into making them isolating.

- strong preference for head-initial, right-branching syntax and morphology
- a dislike of prefixes (gone now, since I fell in love with Swahili)
- dislike of a mix of head-modifier orders

My currently-focused-on* conlang Ngolu / Iliaqu is predominantly isolating except for a small closed class of highly inflecting words (person, number, definiteness/specificity, animacy, accessibility, proximity) and aside from that, the open word-class compounds extensively and has some derivational affixes (pre-, in- and suffixes)

*I do quite enjoy head final phrases in English

My conlang Vrkhazhian is a triconsonantal root language.

https://conworkshop.com/view_article.ph ... 4f538b754a

https://www.frathwiki.com/Vrkhazhian

Whimemsz wrote: ↑Tue Jan 08, 2019 12:38 amWhat I want to see is a good nonconcatenative conlang that doesn't look Semiticky.

Is there a natural non-concatenative language that doesn't look Semiticky. I was struck with how Semitic Tagalog looked - and that's despite the skeletons including vowels.

I'm not sure how helpful the concept of "nonconcatenative language" is in the first place. We can identify items of nonconcatenative morphology, but I don't think any real language is purely nonconcatenative... and a truly vast number of languages, surely the majority, have at least some nonconcatenative morphology (between ablaut, harmony, gradation, irregular sandhi, suppletion, etc...). So there isn't really a binary division to be made here.