The "r" sound

[CEJ]

[quote="globaltefl"]I have found the following extremely helpful for the /r/ and /l/ problem:

1. Have the student say 00 as in boot. The lips must be and remain rounded. The tongue will automatically move back in the mouth.
2. Holding that position, have the student say /root/. The tongue will not be able to reach the alveolar ridge and will reflex into the /r/ sound.

Good luck!
Global TEFL[/quote]

So what happens if they were asked to say 'loot' or 'lute'?

[Lorikeet]

Next, I'm also arguing it's misleading--pedagogically misleading--to teach English /r/ as one sound. Consider its major appearances at the beginning of a word, in the middle of a word, and at the end (in rhotic accents). So compare [r] as in 'run', [r] as in 'correct', and [r] as in 'car' (if you say it, and even if you don't , there might an interesting phonological reason why you don't).

If we stop thinking in unified phonemes (which, I admit, are nice to have for writing systems) we see that phonetically speaking, the /r/ sound is a very confusing one. But in terms of articulation, usually there is that feature of 'retroflexion' of the tongue.

I'm sure however you teach it works for you, but in my speech, the /r/ doesn't seem to have a bunch of allophones. (Unlike the "t" sound for example, for which I have 4 different sounds.) I think I use the same "r" at the beginning, the middle, and the end, and none of them are retroflexed. (American English)

[CEJ]

[quote="Lorikeet
I'm sure however you teach it works for you, but in my speech, the /r/ doesn't seem to have a bunch of allophones. (Unlike the "t" sound for example, for which I have 4 different sounds.) I think I use the same "r" at the beginning, the middle, and the end, and none of them are retroflexed. (American English)

[/quote]
I would bet you are absolutely wrong about that, and I could show you videos of native speakers that show just how wrong the account is. It's more a matter of seeing 'gestures' at a syllable and word level involving [r] sounds in those various positions. Most people refuse to believe it until they see even at an isolated word level just how co-articulated their speech is. Retroflexion is supposed to be one key feature of an English /r/ sound, but most people don't even know what the term means. It's more at the back of the tongue than the tip. That is the whole tongue has to be drawn back enough so that the tip of the tongue is on the back of the alveolar ridge. To other people , this has meant things like tongue curling. That would be one way to get the tip of the tongue out of an /l/ or /d/ position. Once we abandon the phonemic view of pronunciation (which is really more for reading regular alphabets than anything else), we quickly see how complex our actual speech is, even if we want to analyze it at a level beneath a syllable. In other words, it's extremely difficult to turn dynamic, co-articulated spoken language into something we can look at statically in written form (hence the fictions of segments, phonemes, etc.).

[CEJ]

The following is a very good though limited account of [r] sounds, including variation within English:

http://www.hi.is/~peturk/KENNSLA/02/TOP/r.html

A gestural account would show complexity and variation well beyond a standard phoneme-allophone account (though this account does hint at that complexity). Of course the real challenge here is digesting all that in terms of possibilities and then actually teaching a learnable model to students. I say mono-phonemic accounts do a disservice to students; they are an illusion of the alphabet. The best way is to model both aurally and visually an abundance of English [r] sounds from one's own accent, realizing that mastering categories of sounds is rather like learning and revising nuances of vocabulary and grammar.

[CEJ]

[quote="Lorikeet"]
I'm sure however you teach it works for you, but in my speech, the /r/ doesn't seem to have a bunch of allophones. (Unlike the "t" sound for example, for which I have 4 different sounds.) I think I use the same "r" at the beginning, the middle, and the end, and none of them are retroflexed. (American English) [/quote]

Just focussing on this point for clarification. If your /r/ is only one /r/, how do you explain co-assimilation of vowel sounds and post-vocalic [r] in a lot of North American dialects? So called r-coloured vowels. And, what if, using a phonemic account of phonology, one said that when a New Yorker reduces a post-vocalic [r] to, say, a schwa, what if we said simply a schwa is an allophone of /r/? Certainly there is enough phonetic similarity to make the case (though I've pretty much dropped phonemics myself in terms of how I apply phonology and phonetics to TEFL).

[Lorikeet]

I'm sure however you teach it works for you, but in my speech, the /r/ doesn't seem to have a bunch of allophones. (Unlike the "t" sound for example, for which I have 4 different sounds.) I think I use the same "r" at the beginning, the middle, and the end, and none of them are retroflexed. (American English)

Just focussing on this point for clarification. If your /r/ is only one /r/, how do you explain co-assimilation of vowel sounds and post-vocalic [r] in a lot of North American dialects? So called r-coloured vowels. And, what if, using a phonemic account of phonology, one said that when a New Yorker reduces a post-vocalic [r] to, say, a schwa, what if we said simply a schwa is an allophone of /r/? Certainly there is enough phonetic similarity to make the case (though I've pretty much dropped phonemics myself in terms of how I apply phonology and phonetics to TEFL).

You'll note I was talking about "in my speech." I don't dispute the fact that many different "correct" varieties of English exist, and that many of them may indeed have different allophones for /r/. I was just commenting that from what I could tell from my own speech, I don't seem to have a different kind of "r" based on positioning in the word, and yes, it seems the /r/ I use after vowels is the same as the /r/ in the beginning and end. (The vowels, however, have slight changes when followed by an /r/.)

[Metamorfose]

Hey Lorikeet

I know you said somewhere how you pronounce your r's, but I don't know where,so, how do you pronounce it? A tap (flap)?

I ask you this because I tend to drop the r's in final position, but when there's a vowel after it I tend to use a tap rather than a retroflex to join the words, I had a teacher who said it is wrong, is there any problem with it anyway?

José

[Lorikeet]

Hey Lorikeet

I know you said somewhere how you pronounce your r's, but I don't know where,so, how do you pronounce it? A tap (flap)?

I ask you this because I tend to drop the r's in final position, but when there's a vowel after it I tend to use a tap rather than a retroflex to join the words, I had a teacher who said it is wrong, is there any problem with it anyway?

José

Sounds like you should be asking someone who speaks British English. I thought both the dropping of a final "r" and a tap in a word like "very" were British. (I don't know much about British English or Australian either, for that matter.) I don't retroflex my "r" at all, I bunch it. I use a tap (flap) for the /t/ or /d/ sounds in words like "butter," "ladder" (which I pronounce the same as "latter"), "invited" or in connecting words like "get her" or "but it".

[Metamorfose]

I see...but about the final dropping, I've heard some Americans who do drop the r in final position-- mainly in songs at least. I guess British standard will also drop the r in middle word position, when not followed by a vowel. (for example: bird).


José

[Lorikeet]

Yes, there are a lot of varieties of American English too. In the old days, "Midwestern English" was considered "standard" (whatever that meant) but things have loosened up a lot and you can hear other dialects on TV these days. I'm just not familiar enough with them to know all the intricacies. Southern speech also drops the "r", as well as some New England varieties, and New York too. They are all a little different.

[CEJ]

I'm sure however you teach it works for you, but in my speech, the /r/ doesn't seem to have a bunch of allophones. (Unlike the "t" sound for example, for which I have 4 different sounds.) I think I use the same "r" at the beginning, the middle, and the end, and none of them are retroflexed. (American English)

Just focussing on this point for clarification. If your /r/ is only one /r/, how do you explain co-assimilation of vowel sounds and post-vocalic [r] in a lot of North American dialects? So called r-coloured vowels. And, what if, using a phonemic account of phonology, one said that when a New Yorker reduces a post-vocalic [r] to, say, a schwa, what if we said simply a schwa is an allophone of /r/? Certainly there is enough phonetic similarity to make the case (though I've pretty much dropped phonemics myself in terms of how I apply phonology and phonetics to TEFL).

You'll note I was talking about "in my speech." I don't dispute the fact that many different "correct" varieties of English exist, and that many of them may indeed have different allophones for /r/. I was just commenting that from what I could tell from my own speech, I don't seem to have a different kind of "r" based on positioning in the word, and yes, it seems the /r/ I use after vowels is the same as the /r/ in the beginning and end. (The vowels, however, have slight changes when followed by an /r/.)

It's not a matter of idiolect. What I'm arguing is that all speakers are not really aware--and can't be aware of--the sound variations of 'phonemes' in their actual speech because of co-articulation. Standard phonemic-allophonic accounts don't really cover this entirely (though see the source I cited). There is no way an [r] sound could assimilate to a vowel in some sort of mutual assimilation and have the same [r] sound. All you are saying here is a certain point of articulation seems the same. If you treat the sound either segmentally or gesturally, it's very quickly obvious that they are not the same sound.

How do you position your mouth when you prepare to say the words 'row' and 'reel'? I would bet that the [r] of 'row' has much more lip rounding than one in reel.

[Lorikeet]

How do you position your mouth when you prepare to say the words 'row' and 'reel'? I would bet that the [r] of 'row' has much more lip rounding than one in reel.

I'm sorry, I tried it, and the lip rounding is the same. Yes, I even closed my eyes and did it in front of a mirror and looked. I'm not saying everyone does this--I can only say what I do. And I understand your point about people not knowing what they do when they pronounce sounds. Native speakers, for example, aren't aware of the lengthening of vowels before voiced consonants (or shortening before voiceless, however you want to say it) but can see the difference if it's pointed out.

I missed a couple of your posts earlier in this thread and will answer separately.

[Lorikeet]

Retroflexion is supposed to be one key feature of an English /r/ sound, but most people don't even know what the term means. It's more at the back of the tongue than the tip. That is the whole tongue has to be drawn back enough so that the tip of the tongue is on the back of the alveolar ridge. To other people , this has meant things like tongue curling.

Yes I agree I am one of those people. I always thought "retroflexion" referred to curling the tongue back, which I have seen in some books as the "way" to pronounce an /r/. Since I draw my tongue at the back, I explained it as "bunching." Clearly, that is not a linguistic term, and if the proper term is "retroflexion" so be it.

[Lorikeet]

The following is a very good though limited account of [r] sounds, including variation within English:

http://www.hi.is/~peturk/KENNSLA/02/TOP/r.html

A gestural account would show complexity and variation well beyond a standard phoneme-allophone account (though this account does hint at that complexity). Of course the real challenge here is digesting all that in terms of possibilities and then actually teaching a learnable model to students. I say mono-phonemic accounts do a disservice to students; they are an illusion of the alphabet. The best way is to model both aurally and visually an abundance of English [r] sounds from one's own accent, realizing that mastering categories of sounds is rather like learning and revising nuances of vocabulary and grammar.

I agree with your point about mono-phonemic accounts of sounds. I do a lot of work in that area (explaining the /l/ we pronounce initially as compared to the one we pronounce at the end, the /t/ after /s/, in initial position, in the middle, and at the end, for example.) However, I can't teach what I'm not aware of, and I really don't see/hear differences in my /r/s. I read the article you posted, and it appears my English has a "rhotic" r. The only think I noticed is that it seems the r is indeed somewhat voiceless after a voiceless consonant (try, cry, pry).

If I pronounce the words/phrases:

red
teacher
teacher in
are

I think *in my speech* they are the same. They may be very different in yours. I'm sorry you can't hear me, because then you might be able to point out the differences you are sure are there.

[CEJ]

The following is a very good though limited account of [r] sounds, including variation within English:

http://www.hi.is/~peturk/KENNSLA/02/TOP/r.html

A gestural account would show complexity and variation well beyond a standard phoneme-allophone account (though this account does hint at that complexity). Of course the real challenge here is digesting all that in terms of possibilities and then actually teaching a learnable model to students. I say mono-phonemic accounts do a disservice to students; they are an illusion of the alphabet. The best way is to model both aurally and visually an abundance of English [r] sounds from one's own accent, realizing that mastering categories of sounds is rather like learning and revising nuances of vocabulary and grammar.

I agree with your point about mono-phonemic accounts of sounds. I do a lot of work in that area (explaining the /l/ we pronounce initially as compared to the one we pronounce at the end, the /t/ after /s/, in initial position, in the middle, and at the end, for example.) However, I can't teach what I'm not aware of, and I really don't see/hear differences in my /r/s. I read the article you posted, and it appears my English has a "rhotic" r. The only think I noticed is that it seems the r is indeed somewhat voiceless after a voiceless consonant (try, cry, pry).

If I pronounce the words/phrases:

red
teacher
teacher in
are

I think *in my speech* they are the same. They may be very different in yours. I'm sorry you can't hear me, because then you might be able to point out the differences you are sure are there.

Again, it's not a matter of idiolect or necessarily dialect. It's both a phonological impossibility that they are the same (since interchanging artificially cut up segments into the other positions results in unintelligible language) and a phonetic impossibility (every time a word is said, it is said differently, though across speaker, acoustic space, and listener, there must be constructed some sort of invariablity).

Hold a mirror in front of your mouth when you make the words with the sounds. It would be more a matter of seeing you rather than just hearing you. Afterall, if sounds are really actually co-articulated features in dynamic movement, what are we actually hearing? Points of articulation? I highly doubt it. It's not a matter of teaching what you are aware or unaware of, but rather demonstrating how you actually speak.

A post-vocalic [r] that assimilates with a preceding vowel simply could not be the same as a word-initial [r] even if it shared some features (e.g., realized by the tip of the tongue getting behind the alveolor ridge, retroflexion of the tongue of some sort, etc.). We might say that they share a common feature, but as realized the overall quality could be quite different. We can see visual clues of how they sound different. If an initial [r-] in a word with more lip rounding involves more lip rounding when the [r-] is made, do you really think it sounds the same as an initial [r-] in a word that involves lip flattening (such as 'row' vs. 'real')? Changes to the vocal tract for the overall gesture results in changes in how things sound.

If your speech is rhotic in a typical North American way, clearly your post-vocalic [r]s assimilate towards the preceding vowel. That means the vowel becomes more r-like as the [r] becomes more like the preceding vowel. That is co-articulation. Would a word initial [r-] have the same sort of assimilation? Post-vocalically, the [r] 'darkens'.

I would guess that if you compared, for example, 'red' vs. 'dear' you would see that the word-final [-r] has a reduction in the lip part of the gesture.

Here are some articulatory phonology and phonetic studies that support the idea that there are near-universal allophones of /r/. In other words, we are not just talking about variation across accents or dialects, but variations within the speech repetoire of a speaker depending on the context of the sound. We might theorize an invariant feature of the sound connecting all r-gestures, but there clearly is evidence of variation as well.

http://cslu.cse.ogi.edu/nsf/isgw97/reports/alwan.html

Articulatory data of /r/ show that the vocal tract is characterized by three cavities due to the presence of two supraglottal constrictions. The primary constriction occurs in the oral cavity and the secondary constriction, in the pharyngeal cavity. The oral constriction may occur anywhere in the palatal region. The invariant feature for /r/ seems to be the existence of a large sublingual cavity anterior to the oral constriction. Inter-subject variabilities were observed in the location and the way the primary oral constriction was formed.

None of our subjects showed a truly retroflexed /r/ suggesting that the extreme form of retroflex /r/ may not be prevalent in American English. Our data also indicate that /r/ tongue shapes belong to a continuum of possible shapes created between the two `extreme' configurations, namely, the canonical retroflex and bunched varieties with a greater tendency towards the bunched configuration in the present data. As a result, the rhotic approximant in American English can be specified by a three-cavity model wherein a more anterior primary oral constriction is associated with a more superior secondary pharyngeal constriction. Furthermore, the vocal tract of a canonical retroflex /r/, which may occur in other English dialects or in other languages, can be treated as a special case of the three-cavity model, wherein the secondary pharyngeal constriction, corresponding to the anteriorly-located tongue tip-up oral constriction, is absent. Evidence supporting these observations was found in the results of an imaging study of Tamil liquids [5].

http://www.icsl.ucla.edu/~spapl/project ... /fig1.html

http://www.icsl.ucla.edu/~spapl/project ... /fig2.html

http://www.icsl.ucla.edu/~spapl/project ... /fig7.html

http://216.109.125.130/search/cache?p=a ... 1&.intl=us

http://www.linguistics.ubc.ca/isrl/ICPhS%2015%20PPr

ABSTRACTRelative timing of the gestures of /r/ was measured in
initial, final, and intervocalic positions for eight speakers
representing several dialects of English. Simultaneous
video and ultrasound were used to collect kinematic
midsagittal measurements of several events associated
with English /r/. Results show that, in initial allophones,
gestures occur front-to-back, with the lip gesture occurring
first, then the tongue body, then the tongue root. In final
position, there is little timing effect, but spatially, the lip
gesture is often obscured or reduced. Timing is not
significantly affected by resyllabification or dialect. These
results show timing patterns for /r/ that are analogous in
crucial ways to those previously observed for /l/ and /w/,
but that further support that these timing differences are
phonetic responses to perceptual recoverability factors. An
additional observation of this paper is that tongue mid
lowering is likely a tongue body raising mechanism.

4. DISCUSSION
The results of this study have shown:
1) Across dialects, gestures occur front-to-back in initial
allophones of /r/, with the lip gesture peaking first, then
the tongue body, then the tongue root. This is contrary to a
phonological view of these patterns, and supports the
hypothesis that allophonic patterns in articulatory timing
are gradient phonetic responses to the need to maintain
perceptual recoverability of all of the gestures of a
segment. There is essentially no timing offset between /r/
gestures in final and intervocalic positions, and no
substantial differences between dialects, further suggesting
that this temporal offset may be driven by more universal
phonetic factors. Thus, the perceptual recoverability model
is supported in this study.
2) Spatially, the Lip gesture is largely obscured or reduced
word-finally in all dialects. Unlike temporal offset,
however, this reduction is affected by both
resyllabification and dialect. Spatial characteristics thus
appear to behave more like language-specific phonological
properties. The results in 1 and 2 are analogous in crucial
ways to patterns previously observed for /l/ and /w/.
3) Finally, tongue mid lowering was found to pattern
closely with the more anterior tongue body gesture rather
than with the tongue root gesture, suggesting that the
tongue mid lowering is likely a mechanism to facilitate the
tongue body raising gesture.