(L&N) Q&A Thread - Quick questions go here [2010-2019]

[qwed117]

Well, "go wash" is "go and wash", or "go to wash", which is a common kind of thing you get with serial constructions. I'm not saying it is one for sure (it probably isn't since English is said not to have them) but what you said isn't a correct analysis.

This person says "go eat" is one, but they also say "I made him laugh" is one, so I would take it with a grain of salt: http://homes.chass.utoronto.ca/~cla-acl ... 07/Yin.pdf

However, it has been noticed that not any VV sequence or multi-verb structure can be regarded as an SVC. For lack of agreed-upon defining criteria, different linguists have given different classifications to these structurally similar constructions.

[HoskhMatriarch]

Does anyone have an English or German grammar of Chechen, or Tsakhur, or really any language with both productive umlaut and giant agglutinated strings of affixes? It occured to me I probably need to look at how umlaut works in languages not morphologically similar to Germanic instead of just shooting in the dark at it. And as I said, I mean from a morphological perspective, not phonological or diachronic, like I want to see what happens when you just pile affixes on for one thing, or what kinds of things you can mark that I haven't thought of.

[Keenir]

Does anyone have an English or German grammar of Chechen, or Tsakhur, or really any language with both productive umlaut and giant agglutinated strings of affixes? It occured to me I probably need to look at how umlaut works in languages not morphologically similar to Germanic instead of just shooting in the dark at it.

as we used to say in my family, I'm going to head you off at the pass there: its great that you want to look at more reference material, but I have this fear that you'll look at it, start work on your conlang, and say "I don't want to make a clone of Chechen/Tsakhur/other" and will head in another direction...so here's my advice: while we're looking for those grammars, you start on the conlang.

there's lots of options:
* save the umlaut stuff for later.
* make a guess how the umlaut works {see bottom of this post}
* save the affix strings for later.
* make a guess how the affix strings work.(and when you get the grammar, you can use both the guessed system and the IRL system - lots of natlangs have more than one strategy for things, after all)

you had some great thoughts in that other thread with the serial verbs; perhaps your new conlang's sentances - at their simplest - rely upon serial verbs...and that's what the affixes build upon?

And as I said, I mean from a morphological perspective, not phonological or diachronic, like I want to see what happens when you just pile affixes on for one thing, or what kinds of things you can mark that I haven't thought of.

pretty much anything can be marked.

and before you say nothing good can come of making wild guesses about things we haven't adequately studied, let me quote one of the conlang greats:

I created it because, at the time, I couldn't find any resources on the Hawai'ian language, which I'd fallen in love with. So, I created a language that sounded like Hawai'ian, vaguely. (Definitely Polynesian.)

[cedh]

Tsakhur, by Wolfgang Schulze (1997)

[Creyeditor]

Nimboran, by Anceaux, but I don't know where you can get a free pdf. It is really a giant agglutinating string of affixes lang, but Umlaut is a bit of a different, yet procuctive process there.

[HoskhMatriarch]

you had some great thoughts in that other thread with the serial verbs; perhaps your new conlang's sentances - at their simplest - rely upon serial verbs...and that's what the affixes build upon?

The serial verbs are pretty important to the language, but affixes are too. I said that there are lots of languages that are pretty synthetic that have serial verbs and mine is one of those, not a Chinese styled language. Verb agreement is obligatory, for one thing. And at this point verb agreement is done completely through suffixes, which tend to cause umlaut, although there are prefixes for other things. I guess I can just limit my sentences to neuter words and use zero-marked word forms for now.

[Ephraim]

They meant the "go wash" part, which actually seems pretty similar even if there's some technical reason it's not. Gonna/going to is obviously just an auxiliary verb.

It's still INF-go wash, just that the INF morpheme is cliticized onto the previous morpheme. The verbs are not morphological equivalents, and thus not a serial construction

The English go get-construction is actually really interesting. You could probably analyse "go get", "go wash" and "come visit" as two verbs of the same morphological form, or a finite form and an infinitive, but it's hard to test because the construction is apparently restricted to morphologically unmarked forms. In other words, you can say "he wants to come visit", "go wash up!" And "I go get coffee every day", but you can't say *"he goes gets coffee every day" or *"he goes get coffee every day". I don't know if these rules applies to all varieties of English.

http://individual.utoronto.ca/bjorkman/ ... LLT%5D.pdf

[HoskhMatriarch]

It looks like an infinitive in certain forms:

He goes to get coffee every day.
I came to visit.

Now I want to look at the go get construction in Middle English or Old English when infinitives had special marking...

[zyma]

I think I might understand them in terms of morphosyntax now, but cross-linguistically, what characterizes a serial verb construction semantically, if anything? I'm trying to read the book that was linked to, but I'm afraid I'll end up confusing myself, especially since I'm tired and it's getting late where I am.

If you're asking "why is it important" then, to my knowledge, you're opening a whole bag of worms that I will not touch.
If you're asking "what is it in more general terms" then, its two verbs in the same form slapped together with no connection whatsoever

I'm probably misunderstanding what you've said, but I don't think I was asking either of those things, at least not intentionally. I meant to ask if in most languages with serial verb constructions, they convey a similar variety of semantic connotations or not. Hopefully that makes sense; it's getting quite late again for me.

The something else would be an auxiliary verb. Anyways, the difference between that and a serial verb construction is that in a serial verb construction, one of the verbs isn't subordinate to the other, but rather they are two different clauses that are part of a single event, while in an auxiliary verb construction, the lexical verb is subordinate to the auxiliary verb in one clause. Some languages (generally moderately to highly synthetic languages) have serial verb constructions that are morphologically and/or phonologically one word, and that's harder to explain, but if you understand the ones with more than one word and how they're different from auxiliary verbs, then not as much. Here's a book about it: https://dl.dropboxusercontent.com/u/109 ... ctions.pdf

When you say "subordinate", do you mean "subordinate" as in "subordinate clause", or something else?

By subordinate, I mean that when you say "I will go", "go" is subordinate to "will". But when you say "I get up and go" they're not subordinate. Of course, in some languages you can express purpose with serial verb constructions, which is just confusing, but both of the verbs are still syntactically independent.

That is quite confusing. It seems like the meanings serial verb constructions convey are far from common to most or all of the language that use them.

I really hope that book can help.

[Trebor]

Is there a standard term in linguistics for the structure in (1b), an alternative to (1a)? I've just been looking into 'possessor raising', but the below phenomenon has not come up (so far).

(1)
a. As the economy faltered, the company's profits fell sharply.
b. As the economy faltered, the company saw its profits fall sharply.

How (if such is an option) do natlangs around the world decompose a subject genitive phrase as above?

Spoiler:

PS: In case anyone comes across it elsewhere, I've also raised this issue in a Facebook group.

[HoskhMatriarch]

@shimobaatar - There are lots of things serial verb constructions can mean. There are some that are nearly universal and some that are rarer. I forgot exactly which ones are the most common.


I've heard some dialects of Arabic are picking up /ʀ/ as a rhotic, and that in these dialects /ʀ/ contrasts with /ʁ ~ ɣ/. Is that true or was the person who said that misinformed? I can't really find much anything about Arabic dialects with the "guttural r" outside of what that one person said. Wikipedia says that some dialects have it, but that's Wikipedia. It is interesting how many languages outside of Europe have it though (even though probably a lot of those are due to European influence).

[Creyeditor]

I've heard some dialects of Arabic are picking up /ʀ/ as a rhotic, and that in these dialects /ʀ/ contrasts with /ʁ ~ ɣ/. Is that true or was the person who said that misinformed? I can't really find much anything about Arabic dialects with the "guttural r" outside of what that one person said. Wikipedia says that some dialects have it, but that's Wikipedia. It is interesting how many languages outside of Europe have it though (even though probably a lot of those are due to European influence).

That would probably be due to French influence, so you might want to look into former French colonies. I know, that doesn't help much

Is there a standard term in linguistics for the structure in (1b), an alternative to (1a)? I've just been looking into 'possessor raising', but the below phenomenon has not come up (so far).

(1)
a. As the economy faltered, the company's profits fell sharply.
b. As the economy faltered, the company saw its profits fall sharply.

How (if such is an option) do natlangs around the world decompose a subject genitive phrase as above?

Spoiler:

PS: In case anyone comes across it elsewhere, I've also raised this issue in a Facebook group.

Well, it seems that you just added a verb of perception in there, which means that the whole perspective changes and we get another tense and can use another pronoun. Could you elaborate on your examples? On thing you might want to look at are so called "Monsters". Some languages would use sentence (2b) with the meaning of English (2a), IIRC. Such pronouns are called monsters.

(2)
(2a) John saw his girlfriend leave.
(2b) John saw my girlfriend leave.

I don't know if that's what you meant, but it seems to come close.

[sangi39]

I've heard some dialects of Arabic are picking up /ʀ/ as a rhotic, and that in these dialects /ʀ/ contrasts with /ʁ ~ ɣ/. Is that true or was the person who said that misinformed? I can't really find much anything about Arabic dialects with the "guttural r" outside of what that one person said. Wikipedia says that some dialects have it, but that's Wikipedia. It is interesting how many languages outside of Europe have it though (even though probably a lot of those are due to European influence).

That would probably be due to French influence, so you might want to look into former French colonies. I know, that doesn't help much

It's also a feature of some Jewish Arabic dialects, but from what I can tell, in those dialects ghayn becomes unstable, merging into the now guttural ra' or into qaf as /q/ or even /g/. In some of these dialects, ra' is only backed as far as /ɣ/, but I haven't found anything yet to suggest is contrasts with ghayn as /ʁ/, although considering you can find such contrasts in Dutch, IIRC, I wouldn't find it implausible for that particular contrast to exist.

Hoskh seems to be right, though, in that there doesn't seem to be much easily accessible material out there on this particular feature. The main tendency I'm picking up on though is that it's a feature of Jewish dialects and Moroccan/Algerian dialects, sometimes, again, amongst Jewish Arabic speakers, but sometimes amongst Muslim speakers, so it could be French influence or Jewish. The other tendency is that the backing of ra' also appears to occur alongside some reworking of the uvulars, which, AIUI, is already a fairly common feature amongst basically all Arabic dialects.

[HoskhMatriarch]

Well, even if there's some language or dialect that contrasts /ʀ/ and /ʁ/, that's not as bad as the languages that have [ʕ] as a rhotic, or, even better/worse, [h] or [ɦ] or whatever like in some Portugese dialects (at least [ʕ] sounds vaguely rhotic even if it's weird). And if ra and ghayn merge sometimes, but only sometimes, that's basically the same situation as English t and d, which are still seperate phonemes, so it's still pretty remarkable.

[Creyeditor]

We did some phonetic measurements in our phonetics class and some German dialects seem to have either [ʕ] or even [q] (!!!) where other's have an uvular rhotic.

[qwed117]

I've heard some dialects of Arabic are picking up /ʀ/ as a rhotic, and that in these dialects /ʀ/ contrasts with /ʁ ~ ɣ/. Is that true or was the person who said that misinformed? I can't really find much anything about Arabic dialects with the "guttural r" outside of what that one person said. Wikipedia says that some dialects have it, but that's Wikipedia. It is interesting how many languages outside of Europe have it though (even though probably a lot of those are due to European influence).

That would probably be due to French influence, so you might want to look into former French colonies. I know, that doesn't help much

It's also a feature of some Jewish Arabic dialects, but from what I can tell, in those dialects ghayn becomes unstable, merging into the now guttural ra' or into qaf as /q/ or even /g/. In some of these dialects, ra' is only backed as far as /ɣ/, but I haven't found anything yet to suggest is contrasts with ghayn as /ʁ/, although considering you can find such contrasts in Dutch, IIRC, I wouldn't find it implausible for that particular contrast to exist.

Hoskh seems to be right, though, in that there doesn't seem to be much easily accessible material out there on this particular feature. The main tendency I'm picking up on though is that it's a feature of Jewish dialects and Moroccan/Algerian dialects, sometimes, again, amongst Jewish Arabic speakers, but sometimes amongst Muslim speakers, so it could be French influence or Jewish. The other tendency is that the backing of ra' also appears to occur alongside some reworking of the uvulars, which, AIUI, is already a fairly common feature amongst basically all Arabic dialects.

It might be a Berberic influence in the Moralgerian sphere. Most Berber languages seem to have R and r though.

[HoskhMatriarch]

I've noticed, you can whisper English words with fricatives and still be able to tell what they are in isolation (like sue vs. zoo). I also read somewhere that English fricatives not between vowels, like English stops, are only partially voiced, and sometimes they are even voiced for a minuscule amount of the duration of the fricative . A contrast between fully and partially voiced fricatives seems like it would be unstable and just turn into full voicing if that were really what the contrast is. So, since that doesn't appear to be what the contrast actually is in English, what is it?

[qwed117]

I've noticed, you can whisper English words with fricatives and still be able to tell what they are in isolation (like sue vs. zoo). I also read somewhere that English fricatives not between vowels, like English stops, are only partially voiced, and sometimes they are even voiced for a minuscule amount of the duration of the fricative . A contrast between fully and partially voiced fricatives seems like it would be unstable and just turn into full voicing if that were really what the contrast is. So, since that doesn't appear to be what the contrast actually is in English, what is it?

TBH, if I'm whispering, it becomes significantly closer to [sɯ] and [su]

[Sumelic]

I've noticed, you can whisper English words with fricatives and still be able to tell what they are in isolation (like sue vs. zoo). I also read somewhere that English fricatives not between vowels, like English stops, are only partially voiced, and sometimes they are even voiced for a minuscule amount of the duration of the fricative.

Well, as you may already know, for coda fricatives the duration of the preceding vowel is highly significant (it's shorter before phonemically voiceless consonants than it is before phonemically voiced consonants). I believe the duration of the fricative itself constitutes another secondary difference between English /s/ and /z/: apparently, voiceless fricatives are of longer duration ("Acoustic Characteristics of English Fricatives: I. Static Cues," by Allard Jongman, Ratree Wayland, and Serena Wong).

Also, see section 2.3.2 (page 57), "Other features of tense and lax fricatives," from "The phonetics of the fortis-lenis contrast," which seems to be part of "Laryngeal Contrast and Phonetic Voicing: A Laboratory Phonology Approach to English, Hungarian, and Dutch" by Wouter Jansen. In addition to the duration differences already mentioned, this paper mentions differences in tone (the parts of surrounding vowels that are adjacent to a fricative usually have lower tone if the fricative is voiced than if it is unvoiced).

According to Wikipedia, duration plays a similar role in the contrast between Navajo /s/ and /z/: the former is longer, and the latter is often devoiced to some degree.

Speaking of fricatives and voicing, one interesting fact that I learned only recently is that in French, phonemically voiceless fricatives apparently may be partially voiced after a vowel, even when the fricative is word-final or followed by a voiceless segment. The relevant literature is outlined in the following post (When is an “s” sound pronounced “z”?, French Language Stack Exchange).

A contrast between fully and partially voiced fricatives seems like it would be unstable and just turn into full voicing if that were really what the contrast is. So, since that doesn't appear to be what the contrast actually is in English, what is it?

Your reasoning here doesn't make sense to me. Partially voiced fricatives are one possible phonetic realization of English voiced fricative phonemes. There is no contrast in English between partially voiced and fully voiced fricatives, so I don't think anyone postulates that English voiced fricatives are phonemically marked as having the feature "partially voiced" as opposed to just plain "voiced." The surface realization of a phoneme doesn't have to correspond perfectly to its underlying featural classification (if you even believe such a thing exists). Surface realizations are affected by articulatory considerations.

Anyway, phonemes can contrast in multiple ways. Trying to unify all of these to just one thing that the contrast "actually is" seems to me to be a path to oversimplification, like the idea of "fortis" and "lenis" being explained as the result of greater and lesser "energy" or "muscular effort" in articulation (where it is not specified how to measure the degree of "energy," or the precise ways it is expressed). I mean, explanatory theories like this would be nice to have if they are actually true, but as that Wikipedia page outlines, there is little evidence that "fortis-ness" is actually a single thing, rather than a mixture of several phenomena that correlate in different ways in different languages. (Although even though it's vague and possibly ambiguous, the term "fortis" can be useful in language-specific discussions).

TBH, if I'm whispering, it becomes significantly closer to [sɯ] and [su]

That's really weird. I don't understand why there would be any connection between phonemic voicing and phonetic vowel rounding.

[Salmoneus]

a. As the economy faltered, the company's profits fell sharply.
b. As the economy faltered, the company saw its profits fall sharply.

How (if such is an option) do natlangs around the world decompose a subject genitive phrase as above?

Well, it seems that you just added a verb of perception in there, which means that the whole perspective changes and we get another tense and can use another pronoun.

That's not really what's happening, though. Because "saw" in this case doesn't relate to physical seeing, or even really to any perception at all. That's just the etymology, not really the meaning.

My immediate reaction would be that this might be parallel to so-called double-topic or double-subject constructions in more topic-oriented languages. These have many functions, but one of the most common is with possession. So instead of "the trees' leaves fell", the possessive is broken up into "the trees, their leaves fell", or even just "the trees, leaves fell" (with the possessive relationship not explicit). It's a way of pushing the topicality onto the trees, even though you're actually talking about the leaves.
And that seems to be exactly what's happening with "see". "When winter comes, those trees are going to see a lot of their leaves fall". It's keeping the topicality on the trees, even though you're talking about the leaves - and it does it by breaking up the possessive phrase. Obviously, it's much more restricted in use than in a topic-oriented language, but I think it's doing the same thing, just English needs this dummy verb.

Note also that where context is clear, we can even drop the possessive element altogether, just like in a topic language: "the company saw profits fall".



The other thing I'd say, though, is that "see" is actually a more interesting verb than that. Compare:
"Over the winter, we saw company profits fall" / "Over the winter, we saw a fall in company profits"
"Over the winter, investors saw company profits fall" / "Over the winter, investors saw a fall in company profits"
"Over the winter, the company saw profits fall" / "Over the winter, the company saw a fall in profits"
"Over the winter, company profits saw a fall"
"Over the winter, a fall was seen in company profits"
"The bad winter saw company profits fall" / "The bad winter saw a fall in company profits"

And it goes further!
"Over the winter, we saw investors see a fall in the value of their company bonds"
"Over the winter, the company saw investors see a fall in the value of its bonds"
"Over the winter, company bonds saw investors see a fall in their value"
"Over the winter, the value of company bonds saw a fall in value"
?"Over the winter, the value of company bonds saw investors see a fall" (maybe?)
etc

And:
"Over the summer, the company saw an increase in their sales of bananas"
"Over the summer, sales of bananas saw an increase"
"Over the summer, bananas saw an increase in sales"

"See" seems to be an extremely powerful dummy verb that allows a wide range of arguments, including even oblique, bystander arguments ("profits rose during this summer" > "this summer saw profits rise"; "cars frequently crash on this road" > "this road sees cars frequently crash") to be topicalised.