How common are rhotic approximants?

Discussion of natural languages, or language in general.
Post Reply
User avatar
Mike Yams
Sanci
Sanci
Posts: 32
Joined: Fri Aug 07, 2015 7:38 pm

How common are rhotic approximants?

Post by Mike Yams »

How common are rhotic approximants (as in most dialects of English)? Specifically, I mean ones which pattern as liquids? (I'm not quite sure if they ever don't actually.)

My main conlang (well, my main naturalistic conlang) has an alveolar approximant that in some environments is realized as an alveolar tap. I'm not sure how naturalistic or common this is.

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: How common are rhotic approximants?

Post by Zaarin »

Perfectly natural. /n r ɾ ɹ l/ have a tendency to merge in any direction. Wikipedia says Farsi has it as an allophone of /ɾ/ before certain consonants, as do certain dialects of Spanish, Portuguese, and Zapotac. I'm using the same rule (as an allophone of /r/) in one of my own conlangs at the moment.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

User avatar
Mike Yams
Sanci
Sanci
Posts: 32
Joined: Fri Aug 07, 2015 7:38 pm

Re: How common are rhotic approximants?

Post by Mike Yams »

Well that's reassuring, conlang-wise. Linguistics-wise that's somewhat intriguing though. You mentioned [n r ɾ ɹ l], but I'd imagine [d] and [t] to be common as well. In fact, I'm somewhat surprised that [n] is in there. I wonder how and why the nasality creeps in.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: How common are rhotic approximants?

Post by Nortaneous »

common in eurasia and australia, not so common elsewhere

depends on the environment. you'd prob have 4 syllable-initially and r\ syllable-finally
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: How common are rhotic approximants?

Post by Zaarin »

Mike Yams wrote:Well that's reassuring, conlang-wise. Linguistics-wise that's somewhat intriguing though. You mentioned [n r ɾ ɹ l], but I'd imagine [d] and [t] to be common as well. In fact, I'm somewhat surprised that [n] is in there. I wonder how and why the nasality creeps in.
Yes, [d] can easily become any of [n r ɾ ɹ l], though its change is often more conditioned (whereas [n r ɾ ɹ l] can often intermingle in unconditioned changes). For an example with [n], compare Hebrew ben with Aramaic bar. One of the distinguishing features of Seneca is that historical /r/ became /n/. And the Caananitic article ha- is reconstructed as han with the [n] assimilating to whatever consonant followed.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

User avatar
Mike Yams
Sanci
Sanci
Posts: 32
Joined: Fri Aug 07, 2015 7:38 pm

Re: How common are rhotic approximants?

Post by Mike Yams »

Zaarin wrote:Yes, [d] can easily become any of [n r ɾ ɹ l], though its change is often more conditioned (whereas [n r ɾ ɹ l] can often intermingle in unconditioned changes). For an example with [n], compare Hebrew ben with Aramaic bar. One of the distinguishing features of Seneca is that historical /r/ became /n/. And the Caananitic article ha- is reconstructed as han with the [n] assimilating to whatever consonant followed.
That's interesting. Isn't Proto-Iroquoian atypical for North America in having /ɹ/, or for that matter a rhotic in general?

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: How common are rhotic approximants?

Post by Zaarin »

Mike Yams wrote:
Zaarin wrote:Yes, [d] can easily become any of [n r ɾ ɹ l], though its change is often more conditioned (whereas [n r ɾ ɹ l] can often intermingle in unconditioned changes). For an example with [n], compare Hebrew ben with Aramaic bar. One of the distinguishing features of Seneca is that historical /r/ became /n/. And the Caananitic article ha- is reconstructed as han with the [n] assimilating to whatever consonant followed.
That's interesting. Isn't Proto-Iroquoian atypical for North America in having /ɹ/, or for that matter a rhotic in general?
Atypical, yes, but there are a few non-Iroquoian languages with rhotic consonants scattered across the continent, such as Tunica--which, I believe, is unique for North America in having both /r/ and /l/--Hopi, Cree (which also has ɹ), etc. On the opposite end of the spectrum, most of the languages of the Pacific Northwest have numerous lateral consonants; Tlingit famously has five lateral consonants but no lateral approximant. But it's certainly fair to say that rhotic sounds are rare in North America (discounting uvular fricatives, which are common in the West).
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

User avatar
Salmoneus
Sanno
Sanno
Posts: 3197
Joined: Thu Jan 15, 2004 5:00 pm
Location: One of the dark places of the world

Re: How common are rhotic approximants?

Post by Salmoneus »

Zaarin wrote:
Mike Yams wrote:Well that's reassuring, conlang-wise. Linguistics-wise that's somewhat intriguing though. You mentioned [n r ɾ ɹ l], but I'd imagine [d] and [t] to be common as well. In fact, I'm somewhat surprised that [n] is in there. I wonder how and why the nasality creeps in.
Yes, [d] can easily become any of [n r ɾ ɹ l], though its change is often more conditioned (whereas [n r ɾ ɹ l] can often intermingle in unconditioned changes). For an example with [n], compare Hebrew ben with Aramaic bar. One of the distinguishing features of Seneca is that historical /r/ became /n/. And the Caananitic article ha- is reconstructed as han with the [n] assimilating to whatever consonant followed.
It is sometimes confusing to give languages the same name as historical people. For a moment there I thought you were saying the guy had a speech impediment...
Blog: [url]http://vacuouswastrel.wordpress.com/[/url]

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!

User avatar
Mike Yams
Sanci
Sanci
Posts: 32
Joined: Fri Aug 07, 2015 7:38 pm

Re: How common are rhotic approximants?

Post by Mike Yams »

Nortaneous wrote:depends on the environment. you'd prob have 4 syllable-initially and r\ syllable-finally
What about intervocally? I had the idea of varying it in my conlang, depending on the vowels. For instance [iɾa], but [aɹi], or something like that.

(I'm new to the forum; so sorry if I'm bringing up my conlang too much in L&L.)

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: How common are rhotic approximants?

Post by Zaarin »

Salmoneus wrote:It is sometimes confusing to give languages the same name as historical people. For a moment there I thought you were saying the guy had a speech impediment...
Totally forgot about the Greek philosopher; should have clarified "the Seneca language."
Mike Yams wrote:
Nortaneous wrote:depends on the environment. you'd prob have 4 syllable-initially and r\ syllable-finally
What about intervocally? I had the idea of varying it in my conlang, depending on the vowels. For instance [iɾa], but [aɹi], or something like that.

(I'm new to the forum; so sorry if I'm bringing up my conlang too much in L&L.)
I'd expect /ɾ/ intervocaically, and /ɹ/ before some or all consonants. If you go with "some," it will probably be coronals.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

CatDoom
Avisaru
Avisaru
Posts: 739
Joined: Fri Sep 20, 2013 1:12 am

Re: How common are rhotic approximants?

Post by CatDoom »

Zaarin wrote:Atypical, yes, but there are a few non-Iroquoian languages with rhotic consonants scattered across the continent, such as Tunica--which, I believe, is unique for North America in having both /r/ and /l/--Hopi, Cree (which also has ɹ), etc.
Hopi and Cree contrast their rhotics with /l/ as well, though Hopi's is actually a retroflex fricative, which arguably means that /l/ is the only liquid in the language. The r-l contrast is fairly common in the native languages of California, and is found in Atsugewi, Chimariko, Costanoan, At least some of Takic (Uto-Aztecan), Wintuan, Wiyot (which has a rhotic approximant), Yuman (Kumeyaay apparently contrasts a rhotic approximant with a trill), and Yurok, which has retroflex approximants and rhotic vowels, and may be unique among the languages of the world in having rhotic vowel harmony.

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: How common are rhotic approximants?

Post by Zaarin »

CatDoom wrote:
Zaarin wrote:Atypical, yes, but there are a few non-Iroquoian languages with rhotic consonants scattered across the continent, such as Tunica--which, I believe, is unique for North America in having both /r/ and /l/--Hopi, Cree (which also has ɹ), etc.
Hopi and Cree contrast their rhotics with /l/ as well, though Hopi's is actually a retroflex fricative, which arguably means that /l/ is the only liquid in the language. The r-l contrast is fairly common in the native languages of California, and is found in Atsugewi, Chimariko, Costanoan, At least some of Takic (Uto-Aztecan), Wintuan, Wiyot (which has a rhotic approximant), Yuman (Kumeyaay apparently contrasts a rhotic approximant with a trill), and Yurok, which has retroflex approximants and rhotic vowels, and may be unique among the languages of the world in having rhotic vowel harmony.
I stand corrected; I'm not as familiar with the languages of California as I am the languages of the Northwest and Northeast. (Though Wikipedia says some varieties of Hopi have an alveolar flap while others have a retroflex fricative.)
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

CatDoom
Avisaru
Avisaru
Posts: 739
Joined: Fri Sep 20, 2013 1:12 am

Re: How common are rhotic approximants?

Post by CatDoom »

No worries; you almost certainly know more about the languages in those regions than I do. We all have our particular areas of interest. :)

California languages are really diverse; I wrote a post a while back about areal features in northern and southern California (though the linguistic areas naturally don't correspond to the borders of the state), if you're interested.

User avatar
Mike Yams
Sanci
Sanci
Posts: 32
Joined: Fri Aug 07, 2015 7:38 pm

Re: How common are rhotic approximants?

Post by Mike Yams »

I'll have to take a look at that too...

User avatar
Mike Yams
Sanci
Sanci
Posts: 32
Joined: Fri Aug 07, 2015 7:38 pm

Re: How common are rhotic approximants?

Post by Mike Yams »

Zaarin wrote:
Salmoneus wrote:It is sometimes confusing to give languages the same name as historical people. For a moment there I thought you were saying the guy had a speech impediment...
Totally forgot about the Greek philosopher; should have clarified "the Seneca language."
Mike Yams wrote:
Nortaneous wrote:depends on the environment. you'd prob have 4 syllable-initially and r\ syllable-finally
What about intervocally? I had the idea of varying it in my conlang, depending on the vowels. For instance [iɾa], but [aɹi], or something like that.

(I'm new to the forum; so sorry if I'm bringing up my conlang too much in L&L.)
I'd expect /ɾ/ intervocaically, and /ɹ/ before some or all consonants. If you go with "some," it will probably be coronals.
I think I'll go with [ɾ] word-initially and [ɹ] word-finally and in clusters, with [ɾ] and [ɹ] being in free variation intervocalically, but with [ɾ] being more common.

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: How common are rhotic approximants?

Post by Zaarin »

CatDoom wrote:No worries; you almost certainly know more about the languages in those regions than I do. We all have our particular areas of interest. :)

California languages are really diverse; I wrote a post a while back about areal features in northern and southern California (though the linguistic areas naturally don't correspond to the borders of the state), if you're interested.
Just an amateur enthusiast myself, but I find Tlingit and Haida in particular quite fascinating, so I've done some reading on them (mostly in a cultural context rather than a strictly linguistic one). :) Thanks for the link--it's interesting how opposite Californian languages are to their northern neighbors: presence of labials, absence of labiovelars labio-uvulars, few laterals and more rhotics...Am I correct in my understanding that many Californian languages are grammatically simpler than the infamously polysynthetic languages of the Northwest? This is the impression I've gotten from Mithun, at any rate.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

CatDoom
Avisaru
Avisaru
Posts: 739
Joined: Fri Sep 20, 2013 1:12 am

Re: How common are rhotic approximants?

Post by CatDoom »

It really depends, but in general Californian languages tend to be pretty agglutinative. In Wappo, a Yukian language, inanimate nouns usually don't take any inflections, animate nouns are marked only for number, and verbs can take one TAM suffix, a suffix for negation, and prefixes for direction and type of motion, somewhat similar to the "instrumental" prefixes found in many languages in the area.

Not far away, however, the Pomoan language Kashaya has two position classes for prefixes and fifteen for suffixes, with some pretty complex morphophonology.

The Utian languages are unusual in the region in that they have a well-developed case system; for the most part nouns in Californian languages take few inflections.

User avatar
Mike Yams
Sanci
Sanci
Posts: 32
Joined: Fri Aug 07, 2015 7:38 pm

Re: How common are rhotic approximants?

Post by Mike Yams »

Is that mostly because they're head-marking?

Edit: (Well, reading your post again, Wappo just seems less synthetic. The others I mean.)

vokzhen
Avisaru
Avisaru
Posts: 352
Joined: Sat Aug 09, 2014 3:43 pm
Location: Iowa

Re: How common are rhotic approximants?

Post by vokzhen »

Mike Yams wrote:Is that mostly because they're head-marking?

Edit: (Well, reading your post again, Wappo just seems less synthetic. The others I mean.)
I'm not really sure head-marking languages tend to be more synthetic than dependent-marking languages, though at first glance it does appear to fit. Uralic and Indo-European are generally dependent marking and don't have really extensive verbal inflection, while well-known head-marking languages like Algonquin, Iroquoian, and Salish do. On the other hand you've got languages like Chukchi, Nivkh, Eskimo-Aleut, and Muskogean, which have extensive case systems are are also inflected enough to usually be called polysynthetic. There's also a fairly large number of languages like Quechua, Turkic, Northeast Caucasian, Kartvelian, Sumerian, and Pama-Nyungen that have extensive case systems and extensive verbal inflection without generally crossing over into polysynthesis (except in very inclusionary definitions or those that count extensive suffixaufnahme as polysynthetic).

Taking a quick look at WALS, which as always assumes it's a non-biased sample, comparing marking of the clause with categories of verbal inflection:

Code: Select all

Head-marking:    0-1 0% 2-3  2% 4-5 44% 6-7 19% 8-9 23% 10-11 9% 12-13 2%
Double-marking:  0-1 0% 2-3 14% 4-5 31% 6-7 20% 8-9 26% 10-11 6% 12-13 3%
Depend-marking:  0-1 5% 2-3 27% 4-5 38% 6-7 24% 8-9  5% 10-11 0% 12-13 0%
So there's significant overlap, with 63% of head-marking, 51% of double-marking, and 62% of dependent-marking languages falling into the same range of 4-7 categories, so the "average" (mode) head- and dependent-marking languages are similar, but on "average" (median) head-marking is more synthetic than dependent-marking.

EDIT: And, after all that, I realized I wasn't answering what you were asking. Extensive case systems don't tend to overlap a lot with extensive head-marking, though it certainly happens: Georgian, Burushaski, Koasati, and Chukchi, off the top of my head, have both extensive case and polypersonal agreement. Along with Utian those probably comprise a significant number of the languages that have both.

Šọ̈́gala
Lebom
Lebom
Posts: 172
Joined: Sun Oct 27, 2013 10:58 pm

Re: How common are rhotic approximants?

Post by Šọ̈́gala »

Mandarin Chinese has a syllable final rhotic that to my ear sounds a lot like /ɹ/ in rhotic English dialects. The Chinese sound is written <r> (as in "Harbin"). On this basis, I've always assumed it is a positional allophone of initial <r>, which is a voiced retroflex sibilant. Wikipedia claims there is also an innovative rhotic realisation of initial <r>, but I don't recall noticing it. Despite the occurence of final /ɹ/ in indigenous Chinese words, I think Chinese people tend to think of final /ɹ/ in English as difficult to pronounce, to the extent that I've heard people who had developed a non-rhotic accent in English despite otherwise targeting American pronunciation.

M Mira
Lebom
Lebom
Posts: 92
Joined: Thu Mar 05, 2015 8:21 pm
Location: Taipeium, Respublica Sinarum

Re: How common are rhotic approximants?

Post by M Mira »

Šọ̈́gala wrote:Mandarin Chinese has a syllable final rhotic that to my ear sounds a lot like /ɹ/ in rhotic English dialects. The Chinese sound is written <r> (as in "Harbin"). On this basis, I've always assumed it is a positional allophone of initial <r>, which is a voiced retroflex sibilant. Wikipedia claims there is also an innovative rhotic realisation of initial <r>, but I don't recall noticing it. Despite the occurence of final /ɹ/ in indigenous Chinese words, I think Chinese people tend to think of final /ɹ/ in English as difficult to pronounce, to the extent that I've heard people who had developed a non-rhotic accent in English despite otherwise targeting American pronunciation.
I think you mixed up several phenomena:

1) There're 2~3 different <r> in Mandarin, the first is the onset <r->, pronounced [ɻ~ʐ], the second is the rhotic vowel <er>, pronounced [aɻ] in Beijing but [ɚ] in Taipei, the third is the erhua suffix <-r>, pronounced [ɻ] but interacts with the vowel and the original coda of the attached syllable. The third is predominant in northern China but rare to non-existent in the south and in Taiwan.

2) Orthographically, erhua in Pinyin is seldom seen, probably because it's still considered colloquial. Your example is a transliteration of a Manchurian toponym, and the <r> there was originally a trill [r] in Manchurian.

3) From which Wikipedia page? I don't recall any innovative pronunciations.

4) Chinese Americans fail at [ɹ] because their native languages don't have them. IIRC most US-born ethnic Chinese are predominantly from Cantonese, Hakka, or Hokkien-speaking regions where rhotic sounds are absent.

Šọ̈́gala
Lebom
Lebom
Posts: 172
Joined: Sun Oct 27, 2013 10:58 pm

Re: How common are rhotic approximants?

Post by Šọ̈́gala »

M Mira wrote: I think you mixed up several phenomena:

1) There're 2~3 different <r> in Mandarin, the first is the onset <r->, pronounced [ɻ~ʐ], the second is the rhotic vowel <er>, pronounced [aɻ] in Beijing but [ɚ] in Taipei, the third is the erhua suffix <-r>, pronounced [ɻ] but interacts with the vowel and the original coda of the attached syllable. The third is predominant in northern China but rare to non-existent in the south and in Taiwan.

2) Orthographically, erhua in Pinyin is seldom seen, probably because it's still considered colloquial. Your example is a transliteration of a Manchurian toponym, and the <r> there was originally a trill [r] in Manchurian.

3) From which Wikipedia page? I don't recall any innovative pronunciations.

4) Chinese Americans fail at [ɹ] because their native languages don't have them. IIRC most US-born ethnic Chinese are predominantly from Cantonese, Hakka, or Hokkien-speaking regions where rhotic sounds are absent.
I was basically ignoring erhua and was thinking of the [aɻ] phoneme as in 哈尔滨、耳朵、而且, etc.. I didn't know that it's [ɚ] in Taiwan, but that seems like the kind of thing that would happen. On this basis, one might exoect that Chinese people with standard mainland accents would have no trouble pronouncing rhotic sounds following [a] and Taiwanese Mandarin speakers would have no trouble with [ɚ].

"Harbin" seems to be a fully naturalised Chinese placename at this point, so I think it works as an example a rhotic coda in standard Chinese.

I don't remember which Wikipedia page it was exactly and perhaps I'm misremembering, but in any event you seem to agree that there is a rhotic variant of onset <r->. That's the one that I thought was innovative. I don't remember hearing it (except from some Taiwanese-American kids I used to know, and I wasn't sure if that was interference from their American accent).

I was thinking, not of Chinese-Americans, but of mainland Chinese people that I talked to when I lived there. I'm sure they were all fluent putonghua speakers, and maybe some of them had oddities in their realisation of [aɻ] that I failed to notice, but in general they didn't seem to have difficulty producing all the phones of putonghua.

M Mira
Lebom
Lebom
Posts: 92
Joined: Thu Mar 05, 2015 8:21 pm
Location: Taipeium, Respublica Sinarum

Re: How common are rhotic approximants?

Post by M Mira »

Šọ̈́gala wrote: I was basically ignoring erhua and was thinking of the [aɻ] phoneme as in 哈尔滨、耳朵、而且, etc.. I didn't know that it's [ɚ] in Taiwan, but that seems like the kind of thing that would happen. On this basis, one might exoect that Chinese people with standard mainland accents would have no trouble pronouncing rhotic sounds following [a] and Taiwanese Mandarin speakers would have no trouble with [ɚ].
For me, the difference sounds like that mainland Chinese pronounce <er> with their tongues moving up, while we tend to keep the tongues in place. Not sure about the effect on pronouncing English <er> or <-r> though.
Šọ̈́gala wrote: "Harbin" seems to be a fully naturalised Chinese placename at this point, so I think it works as an example a rhotic coda in standard Chinese.
I think I was confused by your use of square brackets there. <harbin> is Manchurian romanization, <hā'ěrbīn> is Mandarin pinyin. <-r> is only used to show erhua and I'm not even sure if it's sanctioned or ad-hoc usage, while the full syllable version should be written <er>.
Šọ̈́gala wrote: I don't remember which Wikipedia page it was exactly and perhaps I'm misremembering, but in any event you seem to agree that there is a rhotic variant of onset <r->. That's the one that I thought was innovative. I don't remember hearing it (except from some Taiwanese-American kids I used to know, and I wasn't sure if that was interference from their American accent).
Listened to the Wikipedia and to me [ɻ~ʐ] is a range of free variation, and fricativeness(is this a word?) is not a distinctive feature for <r->.
Šọ̈́gala wrote: I was thinking, not of Chinese-Americans, but of mainland Chinese people that I talked to when I lived there. I'm sure they were all fluent putonghua speakers, and maybe some of them had oddities in their realisation of [aɻ] that I failed to notice, but in general they didn't seem to have difficulty producing all the phones of putonghua.
I was referring to the pre-war migrants who were never educated in Mandarin, but now apparently it's not what you meant.

Are English final <r> supposed to be pronounced differently from initial <r>, like <l> do? Difference in tongue position, perhaps?

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: How common are rhotic approximants?

Post by Zaarin »

M Mira wrote:Are English final <r> supposed to be pronounced differently from initial <r>, like <l> do? Difference in tongue position, perhaps?
Not usually, though I think mine is less labialized word-finally.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

richard1631978
Sanci
Sanci
Posts: 62
Joined: Fri Oct 22, 2010 2:26 pm

Re: How common are rhotic approximants?

Post by richard1631978 »

My Chinese friend from Dandong had trouble saying the r in words like pork.

Post Reply