English /r/

[Mbwa]

I don't think this has been discussed in depth recently. Anyway, the English /r/ is kinda weird. It's usually referred to as an alveolar approximant (except I think in Scotland /r/ can be realized as a flap most of the time?), but there's other stuff going on with it that varies a lot. Catford says that it can have a retroflex articulation, "slight deep pharyngalization," and some sort of uvular thing along with that. I've also heard that English /r/ sometimes has a degree of labialization.

What can you guys say about the /r/ in your dialects? My /r/, I'm pretty sure, contains some labialization, like English /S/ can as well. I forget the term, but I recall reading that in this case the labialization is not exactly full rounding of the lips, more like compression or something like that. I can also feel some dorsal thing going on with my /r/ but I'd have a hard time pin-pointing it.

[Aurora Rossa]

For me at least, /r/ is basically a retroflex approximate with the lips in a position similar to [P] (labiodental approximant) and some pharyngeal coarticulation.

[Soap]

Mine is a retroflex with lip rounding almost identical to that of /w/. /S/ and /Z/ are also retroflex for me, but with very little lip movement, and /tS/ and /dZ/ have neither the tongue curling nor the lip rounding.

[faiuwle]

Mine is bunched - it feels like the articulation is slightly back of velar, but not quite uvular - not retroflex at all (tongue tip is pointed down), and I don't have any other retroflexes either. /r/ and /S/ are both labialized slightly less than /w/ (which isn't really fully rounded, either), but /Z tS dZ/ are not labialized at all that I can tell. The lip rounding with /r/ does feel more compressed than the lip rounding for /w/ and /S/ (/r/ is more like pursed lips). On reflection, there is really not a whole lot of difference between /r/ and /w/.

[Nortaneous]

I think there are several types of r-sounds in English:

1. Fully labiodental [ʋ]. Appears in some London(?) dialects.
2. Labiodentalized postalveolar-prevelar [ɰ̟͡ɹ̻̠ᶹ], with the labiodental articulation occasionally approaching a full fricative. (I think this is most likely before /i/, but I'm not sure.) Heard in Maryland, even from speakers who attempt to maintain GA, so it's not very marked, if at all. Might be slightly pharyngealized.
3. Prevelar [ɨ̯]. Occasionally labiodentalized. Might be characteristic of Irish English; Fionn Regan has it.
4. Slightly labialized and pharyngealized postalveolar-postvelar [ʁ̟̹͡ɹ̻̠ˤ]. This is what I have most of the time.
5. Postalveolar-pharyngeal [ʁ̠͡ɹ̺̠]. I think I've heard this referred to as characteristic of Texas, but the singer for Those Poor Bastards has it, and I think he's from Wisconsin.

There might be more information somewhere on Luciano Canepari's site, but I'm not sure how reliable it would be.

[Zumir]

The second one wouldn't be out of place in Risla's lang.

[Travis B.]

I know we have referred to it before on quite a few occasions here, but I might as well mention my /r/ here, in more detail than my signature specifies.

My /r/ has three allophones, determined by position and environment. All of its allophones have a dorsal approximant component and are pharyngealized, and none of them have any labialization. (For comparison's sake, neither do any of the allophones of my /ʃ/, /ʒ/, /tʃ/, or /dʒ/ have any labialization.)

These allophones are:

[ɹ̠͡ɰˤ] - pharyngealized coarticulated postalveolar and velar approximant - found in onsets (i.e. non-syllabic) after other coronals*
[ɰˤ] - pharyngealized velar approximant - found in onsets (i.e. non-syllabic) when not following vowels**, coronals*, or /r/
[ʁˤ] - pharyngealized uvular approximant - found otherwise, i.e. when following vowels** or /r/, in codas, or syllabic

* /t/ realized as [ʔ] does not count as a coronal here.
** In the same word, that is.

Of these, [ʁˤ] is by far the most common allophone, but [ɰˤ] is frequent enough due to frequently showing up initially and in consonant clusters to make things not sound uvular-dominated.

(I should note that it seems that there are at least a good portion of people who speak roughly the same dialect as myself where the [ɹ̠͡ɰˤ] allophone is distributed such that it is more common than it is in my speech, but I am not certain of the details of this at the present.)

[Thomas Winwood]

I can't be at all certain about it since I'm no phonetician, but from feeling the way my mouth moves I'd guess [ʋʷ], possibly with slight prevelarisation. (Intrusive R is only rounded if the previous vowel is rounded.)

e: Southern England, native speaker, probably should mention that, huh.

[sirdanilot]

I'm obviously not native, but if I talk English I think it's really nothing more than a pure [ɹ]... perhaps sliɡhtlə retroflex at that, but not even completely.

Leiden Dutch has the same phoneme for /r/ (no trilled /r/ in this variety of Dutch). It's realized as something like [͡ɹjᵞ], or alternatively a transcription similar to the 'Texas-r'.

[vec]

I'm not native and my accent at the moment is a hot mess of a trainwreak.

My English /r/ used to be pretty much [ɹ] or Ø in codas. Then I moved to America. And my accent changed, involuntarily. Now it's some sort of post-alveolar [ɹ] that's slightly velarised (maybe pharyngealised) and maybe a teensy bit labialised too – in onsets. In codas its some sort of slight retroflex with the toungue tip almost touching that little spot right behind the alveolar ridge, but slightly curled.

Weirdly, I've also developed a slight twang, despite living in New York and knowing very few people with a great deal of twang. I don't know where it comes from, but sometimes people laugh at me because I sound like I'm from South Carolina or Georgia or sum'n.

Meh.

[Mr. Z]

I'm not native either. I think I have an alveolar approximant, definitely with labialization, an maybe some sort of dorsal/radical secondary articulation. I'm pretty sure it's not velarized; it's either uvularized or pharyngealized.

[Jipí]

I'm not a native either, but my /r/ in English is probably rather consistently retroflex with labialization. I don't think I've got velarization or something in it. In post-vocalic position I got into a habit of somehow mixing vocalized /r/ [ə~ɐ] with rhotic /r/. Also in cases like <further>, where my tongue refuses [fɹ̩ðɹ̩]. I consistently learned British pronunciation in school, but now at uni had almost exclusively American teachers in my English classes, as far as the the natives go (the other two profs in the department are Germans).

[alice]

As far as I can tell, mine is a postalveolar fictionless approximant.

[Davoush]

I have [4] unless I'm trying to speak posh, in which case I have no idea how to represent it in IPA.

[spats]

Coarticulated postalveolar-postvelar approximant with pharyngealization and (word-initially only) lip-rounding.

The most concise way to express it is probably: [ɹ̙ʷ] or [ɹ̙ˠ] depending on position.

It's really a bizarre sound.

[Bob Johnson]

I hate to disappoint, but just [ɻ]~[ɻʷ] in free variation here.

As far as I can tell, mine is a postalveolar fictionless approximant.

Would that more people spoke without fiction.

[----]

My /r/ is almost always [ɹʷˤ] with heavy labialization; I could almost write it [ɹ͡wˤ] or something, except at the end of words with suffixes like -er, where it rhotacizes the vowel and leaves the pharyngealization behind, so we get [ɜ˞ˤ]. Note the loss of labialization.
Also, when it's part of a cluster, the labialization spreads to all of the consonants in the cluster.

[äreo]

My /r/ is generally [ɻʷ] initially, [ɻ] between vowels, and syllable-finally I transcribe it as [ɚ], but it's pretty light. There's definitely a distinction between my [ə] and [ɚ], but my tongue barely raises.

I think there are several types of r-sounds in English:

1. Fully labiodental [ʋ]. Appears in some London(?) dialects.
2. Labiodentalized postalveolar-prevelar [ɰ̟͡ɹ̻̠ᶹ], with the labiodental articulation occasionally approaching a full fricative. (I think this is most likely before /i/, but I'm not sure.) Heard in Maryland, even from speakers who attempt to maintain GA, so it's not very marked, if at all. Might be slightly pharyngealized.
3. Prevelar [ɨ̯]. Occasionally labiodentalized. Might be characteristic of Irish English; Fionn Regan has it.
4. Slightly labialized and pharyngealized postalveolar-postvelar [ʁ̟̹͡ɹ̻̠ˤ]. This is what I have most of the time.
5. Postalveolar-pharyngeal [ʁ̠͡ɹ̺̠]. I think I've heard this referred to as characteristic of Texas, but the singer for Those Poor Bastards has it, and I think he's from Wisconsin.

There might be more information somewhere on Luciano Canepari's site, but I'm not sure how reliable it would be.

The first one is found idiosyncratically throughout the English-speaking world AFAIK. I've heard it from kids here and in Texas - it's generally considered a speech defect.
The last one does indeed exist in Texas, and I think generally in historically solidly rhotic areas of the Midwest and South.

[Yng]

The first one is found idiosyncratically throughout the English-speaking world AFAIK. I've heard it from kids here and in Texas - it's generally considered a speech defect.

Yeah - it's very much a dialectal feature of the Southeast here, made famous by Johnathan Ross (or 'Wossy') though.

[finlay]

I think there are several types of r-sounds in English:

1. Fully labiodental [ʋ]. Appears in some London(?) dialects.
2. Labiodentalized postalveolar-prevelar [ɰ̟͡ɹ̻̠ᶹ], with the labiodental articulation occasionally approaching a full fricative. (I think this is most likely before /i/, but I'm not sure.) Heard in Maryland, even from speakers who attempt to maintain GA, so it's not very marked, if at all. Might be slightly pharyngealized.
3. Prevelar [ɨ̯]. Occasionally labiodentalized. Might be characteristic of Irish English; Fionn Regan has it.
4. Slightly labialized and pharyngealized postalveolar-postvelar [ʁ̟̹͡ɹ̻̠ˤ]. This is what I have most of the time.
5. Postalveolar-pharyngeal [ʁ̠͡ɹ̺̠]. I think I've heard this referred to as characteristic of Texas, but the singer for Those Poor Bastards has it, and I think he's from Wisconsin.

Labiodental R is kind of centred around London and associated with the Southeast, as YngNghymru says, but now occurs sporadically throughout England.

Most other Rs in the UK tend to be labialised alveolar/retroflex approximants, from what I can tell – there is nothing like the degree of velarisation on our Rs as there is in America. This is the variant I'd say I had: normally retroflex. Might be wrong in that, of course.

Tapped (and trilled) R occurs in Scotland, co-existing with the approximant R but generally not labiodental R. According to Davoush, and this sounds about right, it also occurs in Liverpool. I'm fairly sure I've heard it associated with Wales too. I have it sporadically.

[AnTeallach]

Most other Rs in the UK tend to be labialised alveolar/retroflex approximants, from what I can tell – there is nothing like the degree of velarisation on our Rs as there is in America. This is the variant I'd say I had: normally retroflex. Might be wrong in that, of course.

"Bunched" r occurs in the UK too: John Wells says he has it, and he ought to know.

Tapped (and trilled) R occurs in Scotland, co-existing with the approximant R but generally not labiodental R. According to Davoush, and this sounds about right, it also occurs in Liverpool. I'm fairly sure I've heard it associated with Wales too. I have it sporadically.

I've heard it from Yorkshire speakers too.

Mine is a laminal postalveolar approximant with some labialisation/labiodentalisation and pharyngealisation in onsets of stressed syllables; between a stressed vowel and an unstressed one it's just a postalveolar approximant. When not before vowels it tends to be absent altogether, but my dialect is such a mixed bag that there is sometimes something going on there; I've never been quite able to work out what it is (which is why I usually transcribe the variant of my speech which is fully non-rhotic).

[cromulant]

Retroflexed and labialized velar approximant.

[Shrdlu]

No native, so I use a rolliing /r/ because my brain refuses to consider anything else to be an /r/ and that makes for some funny results.

[Davoush]

The first one is found idiosyncratically throughout the English-speaking world AFAIK. I've heard it from kids here and in Texas - it's generally considered a speech defect.

Yeah - it's very much a dialectal feature of the Southeast here, made famous by Johnathan Ross (or 'Wossy') though.

I genuinely thought Jonathan Ross had some sort of speech impediment, but I've been hearing this type of /r/ a lot recently. Also I just noticed I don't have [4] in word initial position, it's something like a labialised [4] but not exactly...

And I have definitely heard a consistent [4] from some Yorkshire speakers.

[makvas]

Prevelar [ɨ̯]

I think I have this; the POA of my /r/ is between palatal and velar. But it's probably also somewhat pharyngealized (but not as pharyngealized as my coda /l/).

I've lived in the lower-midwest (OK, AR, TX, MO) all my life.