polysynthesis sans polypersonalism?

[Vardelm]

Oh gawd, I've started down the polysynthesis path in conlanging. It's like graduating from wine coolers to crack.

I've done a bit of reading so far, including Whimemsz's thread on polysynthesis from the old board, a few Wikipedia articles, etc. I read that most polysynthetic langs having polypersonalism, but not all. My question then: what languages are polysynthetic without having polypersonalism?

I'm considering making one of my conlangs polysynthetic. Mostly this means adding noun incorporation and some kind of adverbial markings on verbs, since verb already have a ton of inflections. However, personal agreement is something I want to stay away from on this lang, so the above question would help point me to languages that might be good examples.

[Ares Land]

Some examples here:

https://books.google.fr/books?id=0pA3Dw ... al&f=false

I'd suggest a very unexpected model... how about starting from some IE languages such as Sanskrit, Germanic langages, or Ancient Greek and slightly exaggerating various forms of compounding and inflection?

[Vardelm]

Some examples here:

https://books.google.fr/books?id=0pA3Dw ... al&f=false

Thanks! Will definitely take a look at the ones listed there.

I'd suggest a very unexpected model... how about starting from some IE languages such as Sanskrit, Germanic langages, or Ancient Greek and slightly exaggerating various forms of compounding and inflection?

Good idea! I hadn't thought to look at other places for compounding. In particular, if I can find one or more example languages where there is a catalog of compounds, what their elements are, & their meanings, it would be super helpful to steal creatively borrow from that.

I do think that it may take several different types of sources to construct the rough idea in my head.

[Kuchigakatai]

Oh gawd, I've started down the polysynthesis path in conlanging. It's like graduating from wine coolers to crack.

I've done a bit of reading so far, including Whimemsz's thread on polysynthesis from the old board, a few Wikipedia articles, etc. I read that most polysynthetic langs having polypersonalism, but not all. My question then: what languages are polysynthetic without having polypersonalism?

I'm considering making one of my conlangs polysynthetic. Mostly this means adding noun incorporation and some kind of adverbial markings on verbs, since verb already have a ton of inflections. However, personal agreement is something I want to stay away from on this lang, so the above question would help point me to languages that might be good examples.

As Whimemsz wrote in his thread, polysynthesis is more like a collection of properties such morphologically-heavy languages tend to have, but there aren't hard-and-fast rules as to how they're applied and to what degree.

It's a bit questionable whether you can have zero personal agreement and still call it polysynthetic since pretty much all languages that get called "polysynthetic" have that (and I'm including Ket, Coptic and the Gyalrongic Sino-Tibetan languages), but ehhhhh, things don't have to be a perfect fit either. Especially so if your verbal derivational morphology is pretty elaborate (say, if it's common to use multiple affixes to derive words' basic meanings, I'm not even talking about adverbial notions), if a lot of inflectional aspect/mood/modality gets marked, and if there's a lot of freedom about what can be incorporated.

[Vardelm]

As Whimemsz wrote in his thread, polysynthesis is more like a collection of properties such morphologically-heavy languages tend to have, but there aren't hard-and-fast rules as to how they're applied and to what degree.

It's a bit questionable whether you can have zero personal agreement and still call it polysynthetic since pretty much all languages that get called "polysynthetic" have that ...

Based on the link Ars Lande provided, plus some more looking this afternoon, it appears that Haida, Maidu, Arabana, Klamath, and Tlingit are examples where pronouns appear in free form as opposed to bound to the verb. I don't know to what extent each of those is "prototypical" polysynthetic. I think if I follow some examples there and (if needed) add in more noun incorporation, I'll be within the general ball park. I'm not fussy about matching an exact definition of polysynthetic, but it helps to know what the range of natural options are.

One place where I think I may ditch precedent is incorporating transitive agents. As I put it in my scratchpad thread, we'll see if that passes the ZBB sniff test once I post it.

[Pabappa]

I'm interested to see what you mean by incorporating transitive agents. Are they indefinite only? Do they replace the person markers or do they co-occur with person markers?

[Vardelm]

I'm interested to see what you mean by incorporating transitive agents. Are they indefinite only?

Yes. What I have in mind is basically the mirror image of incorporating patients. I'm thinking of it as basically a "passive" version of the usual incorporation. I think it might work well for this language because what I had worked up before made expressing participants a bit messy. I like what I've read so far on backgrounding & valence reduction. Once an agent has been introduced, it could be incorporated for further utterances. The language would still have the regular patient incorporation as well.

Do they replace the person markers or do they co-occur with person markers?

Nope, there are no person markers here. There would need to be either an agent or a patient/theme expressed separately from the verb, at least if the valence is 1 or higher. I do see having impersonal statements where an agent (or patient) could be incorporated, so you might have something like "there is tick-biting", implying that the speaker is having an issue with ticks.

[Richard W]

Once an agent has been introduced, it could be incorporated for further utterances.

That sounds a bit odd. Do patients work that way in any language? The nearest I can think of is languages like Thai that use classifiers to effectively provide a rich set of third person pronouns. Pragmatically, the introduction can associate the classifier with the participant, though the choice of classifiers for a given participant is very limited, so they aren't as flexible as 'the former' and 'the latter' in English. The classifiers reference a noun phrase rather than a participant.

[bradrn]

Once an agent has been introduced, it could be incorporated for further utterances.

That sounds a bit odd. Do patients work that way in any language? The nearest I can think of is languages like Thai that use classifiers to effectively provide a rich set of third person pronouns. Pragmatically, the introduction can associate the classifier with the participant, though the choice of classifiers for a given participant is very limited, so they aren't as flexible as 'the former' and 'the latter' in English. The classifiers reference a noun phrase rather than a participant.

My understanding is that this is pretty much exactly how noun incorporation works in quite a few of the languages which have it. I think Mithun’s four types of NI are relevant here:

• Type 1 incorporation is used only to make an activity more specific. In his polysynthesis post, Whimemsz gave the example of ‘I chopped a tree’ vs ‘I wood-chopped’. Another nice example from Comanche which I also think is Type 1: ‘to gamble’ is ‘to forcefully cloth-throw’.
• Type 2 incorporation is used to promote an oblique argument by incorporating the former patient; for example, ‘my(oblique) head hurts’ vs ‘I(subject) head-ache’, or ‘she hit my(oblique) knee’ vs ‘she knee-hit me(object)’.
• Type 3 incorporation is used at the discourse level; after a noun is introduced, it is backgrounded through incorporation. Whimemsz gave a Koryak narrative as example: ‘This is the first time that such a whale has come near us. It is good-whale. They whale-attacked.’ (Translating more literally than usual to give an impression of what this is like.) This is pretty much what Vardelm is describing.
• Type 4 incorporation is where a generic noun is incorporated into the verb to classify the participant (without reducing the valency of the verb). Whimemsz gives a Mohawk story (I’ll translate literally again): ‘My father fish-bought eight bullheads. My uncle then returned to fish-fix them. At home, he fish-cleaned and fish-fried them. And then as he finished fish-frying, he decided to fish-take them to his friend as a special treat.’ This is the same sort of classificatory use you mention with Thai.
• And also note that if a language has one type of incorporation, then it also has all the ‘lesser’ types — so something with, say, type 3 NI also has types 2 and 1.

(This is pretty much just a summary of Whimemsz’s post on the subject.)

By the way, one thing that helped me understand NI is that it always has some sort of semantic repercussion — specifically, the incorporated noun is backgrounded in some way. As far as I’m aware (which admittedly is not very), there are few (or quite probably zero) languages which have NI without some sort of semantic difference to the ‘equivalent’ non-incorporated sentence. This seems like quite an obvious concept to me now, but it caused me quite a bit of confusion when I first learnt about NI without realising this.

[Vardelm]

[*] Type 3 incorporation is used at the discourse level; after a noun is introduced, it is backgrounded through incorporation. Whimemsz gave a Koryak narrative as example: ‘This is the first time that such a whale has come near us. It is good-whale. They whale-attacked.’ (Translating more literally than usual to give an impression of what this is like.) This is pretty much what Vardelm is describing.

Thanks for the assist! I was getting ready to reread me some Mithun and make sure I hadn't misunderstood (a likely possibility) and then reply. This is even better since someone else understands that part of Mithun the same way I do.

By the way, one thing that helped me understand NI is that it always has some sort of semantic repercussion — specifically, the incorporated noun is backgrounded in some way. As far as I’m aware (which admittedly is not very), there are few (or quite probably zero) languages which have NI without some sort of semantic difference to the ‘equivalent’ non-incorporated sentence. This seems like quite an obvious concept to me now, but it caused me quite a bit of confusion when I first learnt about NI without realising this.

For me, understanding this actually helps clarify NI rather than confusing it. HOWEVER, that's within the context that I have just started learning about polysynthesis, NI, etc. and have the advantage of reading Whimemsz's thread, with its links to Mithun & other papers, plus Wikipedia articles which have generally improved a lot over the last 10 years.

[bradrn]

[*] Type 3 incorporation is used at the discourse level; after a noun is introduced, it is backgrounded through incorporation. Whimemsz gave a Koryak narrative as example: ‘This is the first time that such a whale has come near us. It is good-whale. They whale-attacked.’ (Translating more literally than usual to give an impression of what this is like.) This is pretty much what Vardelm is describing.

Thanks for the assist! I was getting ready to reread me some Mithun and make sure I hadn't misunderstood (a likely possibility) and then reply. This is even better since someone else understands that part of Mithun the same way I do.

Sorry to disappoint you, but I actually haven’t read Mithun myself — I was just summarising Whimemsz’s post.

By the way, one thing that helped me understand NI is that it always has some sort of semantic repercussion — specifically, the incorporated noun is backgrounded in some way. As far as I’m aware (which admittedly is not very), there are few (or quite probably zero) languages which have NI without some sort of semantic difference to the ‘equivalent’ non-incorporated sentence. This seems like quite an obvious concept to me now, but it caused me quite a bit of confusion when I first learnt about NI without realising this.

For me, understanding this actually helps clarify NI rather than confusing it.

That’s exactly what I meant: ‘it caused me quite a bit of confusion when [I first learnt about NI without realising this]’.

HOWEVER, that's within the context that I have just started learning about polysynthesis, NI, etc. and have the advantage of reading Whimemsz's thread, with its links to Mithun & other papers, plus Wikipedia articles which have generally improved a lot over the last 10 years.

That’s pretty much the context I was in as well when I realised that. (Except I haven’t read any of Mithun’s paper.)

[Richard W]

3. Type 3 incorporation is used at the discourse level; after a noun is introduced, it is backgrounded through incorporation. Whimemsz gave a Koryak narrative as example: ‘This is the first time that such a whale has come near us. It is good-whale. They whale-attacked.’ (Translating more literally than usual to give an impression of what this is like.) This is pretty much what Vardelm is describing.

That is odd. They're doing what pronouns generally enable one to avoid doing - massive repeats of a noun. What would happen in a stury about a duck-billed platypus? Would 'duck-billed platypus' be repeated massively, or would it be reduced to something snappier?

4. Type 4 incorporation is where a generic noun is incorporated into the verb to classify the participant (without reducing the valency of the verb). Whimemsz gives a Mohawk story (I’ll translate literally again): ‘My father fish-bought eight bullheads. My uncle then returned to fish-fix them. At home, he fish-cleaned and fish-fried them. And then as he finished fish-frying, he decided to fish-take them to his friend as a special treat.’ This is the same sort of classificatory use you mention with Thai.

For consistency, you should remove the word 'them' from your literal translation. It's work is already being done by 'fish'.

Note that the word for 'fish' has just two phonemes. It does have a sort of pronoun inclusion. It's like my joke about Thai having 3 cases and umpteen genders.

[bradrn]

3. Type 3 incorporation is used at the discourse level; after a noun is introduced, it is backgrounded through incorporation. Whimemsz gave a Koryak narrative as example: ‘This is the first time that such a whale has come near us. It is good-whale. They whale-attacked.’ (Translating more literally than usual to give an impression of what this is like.) This is pretty much what Vardelm is describing.

That is odd. They're doing what pronouns generally enable one to avoid doing - massive repeats of a noun. What would happen in a stury about a duck-billed platypus? Would 'duck-billed platypus' be repeated massively, or would it be reduced to something snappier?

Well, I don’t know terribly much about this subject, but I assume that the incorporated form would be a bit shorter.

4. Type 4 incorporation is where a generic noun is incorporated into the verb to classify the participant (without reducing the valency of the verb). Whimemsz gives a Mohawk story (I’ll translate literally again): ‘My father fish-bought eight bullheads. My uncle then returned to fish-fix them. At home, he fish-cleaned and fish-fried them. And then as he finished fish-frying, he decided to fish-take them to his friend as a special treat.’ This is the same sort of classificatory use you mention with Thai.

For consistency, you should remove the word 'them' from your literal translation. It's work is already being done by 'fish'.

No, that’s the whole point — in Type 4 incorporation, the NI doesn’t reduce valency, but rather acts as a ‘classifier’ of sorts.

Note that the word for 'fish' has just two phonemes. It does have a sort of pronoun inclusion. It's like my joke about Thai having 3 cases and umpteen genders.

I’m sure I’m completely missing your point here, but as far as I can tell, /fɪʃ/ has three phonemes and no pronoun.

[Pabappa]

Think of the incorporated object as taking the place of "it" in English. Yes, its shorter ... and i assume it's a many-to-one mapping, such that every species of fish will just use the morpheme for "fish" with the possible exception of one particular very important species ... e.g. if your culture gets 40% of its protein from sockeye salmon and 5% from all other fish species combined, then yes, the sockeye might have a verbal incorporation form unique to itself, but the other fish will just be "fish". so a duckbilled platypus will probably just use the morpheme for duck or bird, or there may be even just a generic morpheme for all animals otherwise uncategorizable.

All of these posts so far have addressed traditional object incorporation, but i still want to see example of subject incorporation when you develop the language to the point where you can show examples. i imagine this will be much moire difficult which is why i am so interested in it.

[Ares Land]

They say agent incorporation is unattested, but I'm sure it's completely feasible.
All you need is some French people and a few nuclear bombs.

Let me explain:

There are several ways you could say 'The dogs ate the leftovers.'

Les chiens ont mangé les restes.
Les restes, les chiens ils les ont mangés. (a bit weird)
Les restes, les chiens les ont mangés.

Nothing special here. You could use pretty much the same constructions in English.

But the thing is, the last one is pronounced something like /leˈʁɛst leʃjɛ̃lezɔ̃mɑ̃ˈʒe/
Almost as if les chiens les ont mangés. was a single word. I don't think it is, really, you can always add material: les chiens de chasse les ont mangés, les chiens de ma soeur les ont mangés and so on.

On the other hand, I could do it, but would it do it myself? Wouldn't I say something like C'est les chiens de ma soeur qui les ont mangés instead? I don't know really, I'd have to check a pretty large corpus of colloquial speech to be sure.

What I can say with some confidence, though, is that while des chiens les ont mangés is perfectly grammatical, spontaneously chances are I'd say something like y'a des chiens qui les ont mangés.

Note that this last remark is pretty close to that comment:

Once an agent has been introduced, it could be incorporated for further utterances.

Now, let's drop those atomic bombs. The apocalypse ensues, followed by ten years of nuclear winter. Literacy is practically nil in French, and anyway we're so busy fattening each other for foie gras and collecting old cork stoppers for currency that nobody reads anymore. In the post-apocalyptic landscape, where everyone's forgotten the non-head marking standard register with standard IE SVO constructions, well, it would be totally plausible for 'les chiens' to have turned into an incorporated morpheme. And you'd have agent incorporation, just as specified by Vardelm.

[Travis B.]

Maybe they could make a remake of Mad Max, but in postapocalyptic, polysynthetic, noun-incorporating French. Whether your present-day French speakers would need subtitles, though, is an interesting question.

[Pabappa]

I guess what I mean to say is that while agent incorporation certainly grammatically plausible, and while I like the idea a lot and hope to see more, I dont think its quite so simple to just use the same patterns we already use for object incorporation.

Object incorporation is wide open ... I can eat a banana, a fish, a pizza, ice cream, popsicles, etc .... but only one of those things can eat me and it is surely a million times more rare than the other way around. And you mention tick bites .... yes, thats perfectly reasonable, but ... what else do ticks do besides bite us, anyway? Probably the incorporated tick morpheme isnt going to appear in any other words in the language, and in that case, is it really agent incorporation or just a compound word?

Im not trying to discourage you, ... I *do* think this is a good idea and that you can make it work, but if you have incorporated morphemes that appear in only one word, I would analyze the language as a more traditional type.

[Richard W]

4. Type 4 incorporation is where a generic noun is incorporated into the verb to classify the participant (without reducing the valency of the verb). Whimemsz gives a Mohawk story (I’ll translate literally again): ‘My father fish-bought eight bullheads. My uncle then returned to fish-fix them. At home, he fish-cleaned and fish-fried them. And then as he finished fish-frying, he decided to fish-take them to his friend as a special treat.’ This is the same sort of classificatory use you mention with Thai.

For consistency, you should remove the word 'them' from your literal translation. It's work is already being done by 'fish'.

No, that’s the whole point — in Type 4 incorporation, the NI doesn’t reduce valency, but rather acts as a ‘classifier’ of sorts.

If you look at the inter-linear translation, there is no morpheme corresponding to 'them'. NI may not reduce valency, but it seems to supply a default argument.

Note that the word for 'fish' has just two phonemes. It does have a sort of pronoun inclusion. It's like my joke about Thai having 3 cases and umpteen genders.

I’m sure I’m completely missing your point here, but as far as I can tell, /fɪʃ/ has three phonemes and no pronoun.

Actually, I may have misread the 'underlining' scheme in the Type 4 example. Perhaps the morpheme is actually /tsya/ - four phonemes.

This system looks rather like redundant polypersonal marking. The verb is marked for having a fish argument, but in the absence of an explicit argument, that marking serves as the argument. Just as agreement markers can show gender - think of Semitic 3s markers y- and t- in the prefix conjugation - the 'fish' morpheme is there, and in the absence of an explicit argument, the marker serves as a pronoun. Now there are a lot of different possibilities in this system. The analogy with my Thai joke is that when using definite singular demonstratives in Thai, the classifier has to be included. A classifier is also needed with numbers in Thai, but their place can be taken by a unit of measurement. Classifiers in Thai thus act like a sort of gender marker - but there are a lot of 'genders', and there can be a freeish choice. For example, there are special classifiers for domestic elephants and for important people such as monks and government officials, but one can usually just use the ordinary classifiers for animals and people instead. (Elephants in royal service and monks are eligible for ranks of nobility. Civil servants now have civil service grades instead.)

[Vardelm]

They say agent incorporation is unattested, but I'm sure it's completely feasible.
All you need is some French people and a few nuclear bombs.

I found this immensely entertaining while working this afternoon. Thank you.

All of these posts so far have addressed traditional object incorporation, but i still want to see example of subject incorporation when you develop the language to the point where you can show examples. i imagine this will be much moire difficult which is why i am so interested in it.

It may well not be feasiable. I freely admit I don't really know WTF I'm doing with incorporation, so this might be like watching a conlang trainwreck. I'll post whatever I come up with though, so feel free to drop in, point, & laugh.

Probably the incorporated tick morpheme isnt going to appear in any other words in the language, and in that case, is it really agent incorporation or just a compound word?

Im not trying to discourage you, ... I *do* think this is a good idea and that you can make it work, but if you have incorporated morphemes that appear in only one word, I would analyze the language as a more traditional type.

This is an interesting point. Is noun incorporation actually converting arguments into morphemes? I assumed that it was essentially compounding that had various grammatical effects as a result (backgrounding, indefinite, etc.).

[bradrn]

For consistency, you should remove the word 'them' from your literal translation. It's work is already being done by 'fish'.

No, that’s the whole point — in Type 4 incorporation, the NI doesn’t reduce valency, but rather acts as a ‘classifier’ of sorts.

If you look at the inter-linear translation, there is no morpheme corresponding to 'them'. NI may not reduce valency, but it seems to supply a default argument.

Oh, sorry, I must have misread that.

This is an interesting point. Is noun incorporation actually converting arguments into morphemes? I assumed that it was essentially compounding that had various grammatical effects as a result (backgrounding, indefinite, etc.).

How are those two perspectives different? To me it seems fine to consider it as being effectively grammaticalised compounding with semantic and discourse effects.