Verbal morphology - relative time of grammatical evolution

[The Hanged Man]

I've read somewhere about "grammaticalization cycles" - changes leading from isolationg to agglutinating structures, from agglutinating to fusional ones, and from fusional back again to isolating. These cycles can be classified to groups, by time of duration of each of them:

"short-timed" - like Jespersen's Cycle
"medium-timed" - like verbal morphology (tenses, aspects, moods, etc.)
"long-timed" - transition of a whole language from isolating to its isolating descendant, through agglutinating and fusional daughter languages

My questions are:
1. Is evolution of verbal morphology always (or in most cases) as time consuming, as this is in case of Indo-european languages? Is it general truth, or just a property of indo europan langs?

2 . It seems that verbal morphology evolves more quickly in IE languages than nominal morphology. Am I right? And is this also a general tendency in other language families?

[vec]

I don't exactly understand the question. But the PIE verbal system was insane and got even insaner in Greek and Sanskrit. Other languages tried to make it slightly less insane. Eventually, all languages went with less insane.

[TaylorS]

This is a good question. I know that French and (IIRC) Bengali are already currently transitioning from isolating to agglutinative verbal morphology, and for French you have 2000 years from the definitely fusional Latin

I believe there has been almost a full cycle of Fusional to Isolating to Agglutinating and back to Fusional from Old Egyptian to modern Coptic, and so that is about 5000 years.

PIE as reconstructed was transitioning from Agglutinative to Fusional, which is why it was so morphologically insane.

[vec]

One of the problems with the question is the terms that are used in proposing it. They were invented in the nineteenth century at the dawn of modern linguistics and today, we definitely feel their age. The terms agglutinative, fusional (or inflected as it was originally called) and isolating, can really only be used as very general, vague, descriptions, not as over-all typology.

Let's take for example English.

English's deriviational processes are mostly agglutinative:

create
recreate
recreation
recreational
recreationalize
recreationalization
recreationalizational
recreationalizationalist
recreationalizationalistic
recreationalizationalisticism
recreationalizationalisticismic

father
fatherless
fatherlessness

and so forth. When words are Græco-Latinate in origin, they possibilities are endless, when words are Germanic, not as much, but the same principle applies on the whole.

English has a lot of fusional elements:

foot - feet
man - men
dance - dancing
dance - danced

English also has a lot of isolating aspects:

he was not to be
I haven't been able to be singing
out of the house of the mother

Which is English then?

You tell me.

Where do you get the idea that IE language's verbal systems have changed more or faster than other verbal systems around the world?

The idea that a language family could have some sort of global tendency is patently false. Sound change is entirely random and analytical change is pretty much random, too.

PIE o became u in a lot of cases in Latin. That's why we have the ending -us popping up all over the place from IE -os. Guess what that same ending is in most Romance languages? -o. The vowel o went through a cycle: o > u > o. That's not exactly "progress".

But I'm just talking out of my ass here. I still don't understand the question fully. I think you need to bring up some examples. Generalising about the world's largest attested language family is pretty impossible.

[zompist]

One \English has a lot of fusional elements:

dance - dancing
dance - danced

Your overall point is good, but the cited words are not an example of fusion. The -e in dance is just a spelling convention.

Better examples would be do/does or be/been.

And while I'm nitpicking, TaylorS, French didn't go through an isolating period, though like English it has plenty of isolating elements.

[vec]

I wasn't thinking of the -e, actually. But I do agree that -ing and -ed are not very good examples. I was thinking in terms of categorical weight: "present active participle" and "past passive participle", but thinking about it, it's not like English has a contrastive "past active participle" so the categorical weight of these morphemes is exaggerated.

[The Hanged Man]

Where do you get the idea that IE language's verbal systems have changed more or faster than other verbal systems around the world?

Did I have such an idea? No, I meant that IE languages' verbal system evolved more quickly that IE languages' noun morphology. Mentioning it as an example of "short-timed" processes should clarify that thing, I think... And I'm asking, if that is the case also in other language families: verbal forms evolving more quickly than noun forms. I also asked, whether ecolution of verbal morphology takes always medium amount of time, with respect to short and long timed changes, like Jespersen's cycle, and the full cycle form more analytical language to more fusional and more analytical one again.

Sound change is not an only factor to consider about evolution of verbal morphology. The other one is rate of appearance of new, analytical expressions, that could replace older, fusional ones. And I suspect, that new analytical expressions for tense and aspect of verb appear in language more often (also when older forms are still used), than new analytical expressions of noun forms.

[merijn]

I do think that some language families may develop new tenses quicker than others. No Germanic language has developed a new synthetic tense in 2000 years of evolution, but both Swahili (the -me- tense) and Zulu (the future tenses) have developed new synthetic tenses in the last couple of centuries. The reason is probably syntactic: in Bantu languages auxiliaries precede the main verb directly, which means that they can be reinterpreted as inflection, whereas in Germanic they are frequently separated from the main verb.

+++++

Comparing Zulu and Swahili (and this is consistent with other Bantu languages I have come across) the nominal morphology, mainly the noun classes, are still very similar, whereas the verbal morphology is much more divergent. What is different in the verbal morphology is not so much agreement (which is still quite similar), but the formation of tenses, though there are still quirks that show you they are related, as well as the general architecture of the verb forms that is very similar

[Bedelato]

I've read somewhere about "grammaticalization cycles" - changes leading from isolationg to agglutinating structures, from agglutinating to fusional ones, and from fusional back again to isolating. These cycles can be classified to groups, by time of duration of each of them:

"short-timed" - like Jespersen's Cycle
"medium-timed" - like verbal morphology (tenses, aspects, moods, etc.)
"long-timed" - transition of a whole language from isolating to its isolating descendant, through agglutinating and fusional daughter languages

My questions are:
1. Is evolution of verbal morphology always (or in most cases) as time consuming, as this is in case of Indo-european languages? Is it general truth, or just a property of indo europan langs?

2 . It seems that verbal morphology evolves more quickly in IE languages than nominal morphology. Am I right? And is this also a general tendency in other language families?

The thing about language change is that it's extremely chaotic—there's loads of external factors in play which makes it impossible to determine the final outcome based on internal information alone. Just like genetics That's why coming up with sound changes for conlang families is so tricky, and grammatical changes even more so. There are very few hard-and-fast rules; one can only speak in terms of "general tendencies".

There's no predicting how fast a change will take to complete; this applies to any process, not just verbal morphology. I don't know much about non-Indo-European languages (or even all IE languages), but my guess is that if there is a tendency at all, it's probably only a slight one. My point here is, it's hard to tell. There's too much chaos to confidently pick out any patterns.

[Cedh]

Sound change is not an only factor to consider about evolution of verbal morphology. The other one is rate of appearance of new, analytical expressions, that could replace older, fusional ones. And I suspect, that new analytical expressions for tense and aspect of verb appear in language more often (also when older forms are still used), than new analytical expressions of noun forms.

I think this is a crucial point. Sound change typically proceeds with no regard to syntax or semantics (except that clitics, due to their lack of independent stress, may undergo additional changes that do not happen to free morphemes), and syntactic change should also proceed in a way that doesn't really distinguish between different parts of speech. But of course any type of syntactic change depends on creating and switching to alternative constructions, and the possibility for such alternative constructions to arise is much higher for verbs because these inflect for a lot more categories cross-linguistically than nouns do, and in most languages the number of analytical expressions with verbs will also be higher than the number of analytical nominal constructions.