Update: 2017-11-22 07:27 PM -0500


English Phonetics and Phonology for Burmese-Myanmar speakers


by U Kyaw Tun (UKT) (M.S., I.P.S.T., USA), Daw Khin Wutyi, Daw Thuzar Myint and staff of Tun Institute of Learning (TIL) . Not for sale. No copyright. Free for everyone. Prepared for students and staff of TIL Research Station, Yangon, MYANMAR  - http://www.tuninst.net , www.romabama.blogspot.com

-based on Peter Roach, English Phonetics and Phonology, a practical course. 2nd ed., 4th printing 1993, Cambridge University Press. For my reference, the printed book was digitized by Daw Khin Wutyi, B.Sc., TIL Research Station, Yangon, MYANMAR. Page references to the original book are shown in my text for easy reference.

index.htm |Top

Contents of this page



UKT notes :
Rhoticity across BEPS languages (& sonority scale)


Contents of this page


UKT 170908: Few realized that the Bur-Myan (Burmese-Myanmar) script is based on sound phonemic principals and is actually a phonetic script. Bur-Myan speakers of my generation are probably the last to be exposed to Bur-Myan Thin'boanGyi {n-pon:kri:}, dubbed the "spelling book" by Westerners who came across it in late 19th century. As children we thought it meant the "big black-board.

Children in the Indian sub-continent and in Myanmarpr had to sit on the floor and repeat the phoneme (speech) uttered by the instructor of the grapheme (script), until the instructor is satisfied that we had mastered the sound. It is a tedious process, but is necessary to keep the one-to-one  unchangeable system of speech to script and script to speech, which is what we meant by Akshara {ak~hka.ra}. The Akshara has nothing to do with God. This is essence of Abugida-Akshara system, which is quite different from Alphabet-Letter system. Akshara is a syllable and is pronounceable, whilst Letter more specifically the consonantal-letter is consonantal-akshara, whose inherent vowel has been killed by the Virama (viram) {a.t}. As an example we take the first akshara of our system:

Bur-Myan:  {ka.} + --> {k}
Skt-Dev:------ क + ् --> क्

In the Thin'boanGyi we have to learn how to pronounce the basic syllables which we call akshara (no sing. and plu. in Bur-Myan) according to the POA (Places of Articulation) {HTaan} in the human mouth, and how each is to be pronounced MOA (Mode of Articulation) {ka.ren:}. However, those of the younger generations right down to the present do not know almost anything of {HTaan} and {ka.ren:}.

Thus, those of our generation and older already knew phonetics. My task is to find a reliable mapping of Myanmar to IPA (International Phonetic Alphabet) for consonants, vowels, semivowels aka semi-consonants, and rimes. The intermediary script, Romabama, basically a transliteration (leading to transcription) of Bur-Myan to Eng-Lat (English-extended Latin), has been invented to serve as bridge between Myanmar and IPA. Because, the Bur-Myan reader already knows phonetics, I am giving the IPA transcriptions mainly from DJPD16 of the English words from the very beginning. However, please note that Roach is writing for those who do not know phonetics and is therefore more complicated than is really necessary.

"You probably want to know what the purpose of this course is, and what you can expect to learn from it. An important purpose of the course is to explain how English is pronounced in the accent normally chosen as the standard for people learning the English spoken in England." If this was the only thing the course did, a more suitable would have been "English Pronunciation".

UKT: English a very versatile language, is pronounced in England (the British accent) and in North America (the American English) just differentiated by " rhoticity". In Australia, the "day" becomes the "die", "basin" becomes the "bison", and your "class mate" becomes the "class mite". It is this very versatility which messes up everything when different languages of different language-groups are compared in the case of BEPS.

However, in my TIL version, I am giving the approximate pronunciations of Bur-Myan (Burmese-Myanmar), Pal-Myan (Pali-Myanmar) and Skt-Dev (Sanskrit-Devanagari) in terms of English transcriptions of IAST (International Alphabet for Sanskrit Transcription). Romabama (Burmese-extended Latin) is the intermediary script. My approximate pronunciations are based on DJPD16 . It should be noted that both Devanagari and Myanmar were derived from the Asoka script which has been dubbed the Brahmi {braah~mi} script. You can see how close Myanmar is to Asokan from the first line of the akshara-matrix.

Actually, there are eight aksharas out of 32 consonants which can be easily identified as the "same". I would like to place a high emphasis on the nasal phoneme /ŋ/ {ng} which probably was absent in Devanagari, and certainly in English. After learning Peguan dialect of Mon-Myan to some extent, I am now convinced that the akshara is nasal only as the coda-consonant.

As the onset-constant, {nga.} is not  found in Pal-Myan and also in Skt-Dev. I now classify it as a non-nasal in the onset position, and accordingly it is {gna.} (without IPA transliteration). We now have: {gna.}-onset/ {ng}-coda, and the Bur-Myan word {ngn} 'to draw in (by the tying rope)'.

UKT 171122: the only way to get rid of the trouble-some digraph ng is to change the nuclear vowel as well: {ngn} is derived from {nging}.

Devanagari nga ङ looks as if it had been "borrowed" from ḍa ड and a dot added. 

"However, at the comparatively advanced level at which this course is aimed it is usual to present the information on pronunciation in the context of a general theory about speech sounds and how they are used in language; this theoretical context is called phonetics and phonology. "

UKT: From time to time I will give the Bur-Myan equivalents of English
linguistics {Ba-a b-da.}
phonetics {d~da. b-da.}
phonology {wa-s~a.ra. b-da.}.

Though I have grouped the four languages as BEPS, two are from one linguistic group - the Tib-Bur (Tibeto-Burman: comprising Bur-Myan and Pal-Myan) and the other IE (Indo-European: comprising Eng-Lat and Skt-Dev). The commonality between the two linguistic groups is mainly due to geo-political reasons. The difference, due to the ethnic diversity (which is expected to influence the group of vocal muscles used for the production of consonants, vowels, semivowels, and rimes) between the speakers, is highly interesting.

My interest in BEPS is not only theoretical, but practical. There is a need for a transcription (speech) between Burmese and English. I am motivated by a personal reason - the need to communicate within the TUN family: between my grandsons born and educated outside Myanmar and their immediate relatives who were born and educated in Myanmar. My grandsons speak Eng-Lat but their relatives speak Bur-Myan. I, as the bilingual in Bur-Myan and in Eng-Lat, am the natural and necessary intermediary. I have already developed Romabama (Bur-Lat) and have more than a cursory interest in the backgrounds of English speakers (history, linguistic origin, politics and religion) and Burmese speakers. - UKT110218: my 76th birthday.

I had thought that Bur-Myan speakers, and Mon-Myan speakers living together for centuries in the land of Myanmarpr, using the same basic akshara, and sharing the same Theravada Buddhism, would pronounce the same Pali words the same. It came as a surprise to me to hear the difference when I started to learn online the Mon-Mon (the language of my great grandmother Daw MMa. Listen to {na.mau:boad~Da-ya.aid~Dn}, pronounced by a Mon-Myan speaker: Mon salutation<))

Because of the difference, I now have to say that Romabama transcription is not good for Mon-Myan, which belongs to Aus-Asi (Austro-Asiatic)- language group: but still it can serve as media of transliteration.

UKT: The hyoid bone is the main bone used in production of speech sounds: primarily the vowels.

The hyoid bone is not connected to any other bone and is "suspended" by the group of muscles shown in the inset. During the production of a vowel, say /i/ represented by {i} or {I}, and ई, only some of the muscles are involved. Which muscles are chosen and the tension used for each is expected to depend on the ethnicity of the speaker, his or her length of the neck and other anatomical features, and on the ear of the speaker. Thus /i/ produced by a German, and that produced by a Russian is noticeably different.

We have a similar case within the Indian subcontinent with regard to the pronunciation of highly rhotic Skt-Dev vowel ऋ. It is pronounced as a front-vowel Ri by Hindi speakers (IE group) and back-vowel Ru by Marathi and Kannada speakers (Aus-Asi group). See Wikipedia: - https://en.wikipedia.org/wiki/Richa 171122

There is no way how to teach an L2 learner to correctly emulate the speech sounds of a native speaker. It is his ear which will "teach" the L2 learner. That your ear plays a very important part in the production of speech sounds is shown by the common observation that a speaker, say an ethnic Bur-Myan, tend to pick up the "accent" of the place he had migrated to, after the lapse of a few years.

"Why is it necessary to learn this theoretical background? The same question arises in connection with grammar: at lower levels of study one is concerned simply with setting out how to form grammatical sentences, but people who are going to work with the language at an advanced level as teachers or researchers need the deeper understanding provided by the study of grammatical theory and related area of linguistics. The theoretical material in the present course is necessary for anyone who needs to understand the principles regulating the use of sounds in spoken English."

UKT: IPA Pulmonic consonants and BEPS Basic consonants

UKT 170908: The theoretical background is a real necessity in my approach, because the Myanmar akshara is a phonetic script like the IPA. Since we speakers of Bur-Myan of the older generation unknowingly knew the POA of consonants and modes of articulation of the vowels from Pal-Myan in which we say our prayers throughout our lives. We may not know the real meanings but we are familiar with the Pali sounds. What we are lacking are the sibilant sounds of English - the hissing sounds /s/ and hushing /ʃ/  sounds.

And what most non-Burmese Myanmar speakers, the native speakers of English are missing are the thibilant sounds /θ/ in spite of the presence of English words like "thin".


UKT: Jumping from Pali to Sanskrit is not as difficult as you might think especially if you know some Sanskrit words that you must have heard from the astrologers. Whether we like it or not Astrology is a part of Bur-Myan life. Sanskrit has more sibilant and rhotic sounds than English.

Keep your Burglish pronunciations - don't change them deliberately. If you want to know how a word is sounded in International English, just go online and listen. Bit by bit, your pronunciation would improve. After you have gone through my course, you and your grandchildren, can enjoy listening, "speaking", and playing with the English language. If you are a monk or a nun, you can enjoy talking to your disciples inserting English words and sentences here and there. If you are planning to go abroad to spread the word of the Buddha, my course would be the basic on which you can build your English to speak to an international audience.

"The nature of phonetics and phonology will be explained as the course progresses, but one or two basic ideas need to be introduced at this introductory stage. In any language we can identify a small number of regularly used sounds (vowels and consonants) that we call phonemes ; " [UKT ]

UKT 170909: I'm calling for a break in the paragraph to concentrate on the word consonant in our Abugida-Akshara writing system. It is entirely different from Alphabet-Letter system which is more familiar to western philologists.

On the right I've chosen 3 technical terms which are very important to us who are trying to find the underlying meanings of BEPS languages. Since all three end in -eme, I would just add a Bur-Myan transliteration {aim} as a suffix. Though I am not satisfied with my own transliteration, still would continue using it.

I need to clarify - at least for myself - the meaning of Lexeme {a.Daip~p-aim}, which belongs to the field of modern Psycholinguistics in which its precursor is "lemma". It comes to mind what I little know of Theravada Abhidamma (the Buddhist philosophy). The process of change known as lexicalisation :

Lemma (mental-thought) --> Lexeme (meaning-sound)
related to: {sait} --> {a.n}

We will come back to this topic in another section later.


To study the Skt-Dev as a spoken language, I am relying on Dr. P Rajagopal, School of Traditional  Indian Arts and Literature - SktDevGramLect-indx.htm (link chk 170908).
In her first part of lectures, - Lesson106<> (link chk 170908)
she clearly states: "The name given to these 25 consonants is . The name given to these 25 consonants is . Now, I must find out if our consonants of Abugida-Akshara system are non-pronounceable (mute) or not. I am not saying anything about the Alphabet-Letter system in which the Letter, e.g. Georgian consonant-letter თ (Tan) is mute. To make it pronounceable it must be coupled to vowel-letter ა (An). In the Alphabet-Letter system they do not have a vowel killer {a.t}, which in Abugida-Akshara system is of prime importance.

I'll now check Dr. Rajagopal with the oldest Sanskrit Grammar of 1879 that is in the TIL PDF libraries, one by W. D. Whitney:
- WDWhitney-SktGramm<> / Bkp<> (link chk 170908)
On p011, starting with sect.0031, Prof. Whitney writes:
"31. The Hindu name for 'consonant' is vyajana, 'manifester'. The consonants are divided by the grammarians into spara, 'contact' or 'mute', antaḥsthā, 'intermediate' or 'semivowel', and ūṣman, 'spirant'. First of all I'll check on on:
Consonant  vyajana which is described as "mute" spara , I digitize it as ष्पर्शा = ष्पर्शा from which I get the first sound {S~pa.} ष.




"32. Mutes. The mutes, spara, are so called as involving a complete closure or contact ( spara ), and not an approximation (cont into next page) "

Dr. Rajagopal and Prof. Whitney could have chosen either 'contact' or 'mute', but they have chosen 'mute'. I, on the other hand because of the use Virama {a.t} would have to chose 'contact'. Though the tip of the tongue is 'touching' the roof of the mouth, the stream of air from the vocal fold has to come out. Thus, I maintain consonant as akshara, or using the hyphenated form consonant-akshara of the Abugida-Akshara system is pronounceable. To the Westerners and their students in the British Indian Empire, the consonant-letter of the Alphabet-Letter system is mute.

The first part of    spara, is स्प =  स ् प , and I will have to use BEPS consonant {S~pa.}.

dharma: ध र ् म --> धर्म 
spota: ष ् प र ् श ा ः --> ष्पर


( p003cont)
for example, the vowels in the words 'pin' and 'pen' are different phonemes, and so are the consonants at the beginning of the words 'pet' and 'bet'. Because of the notoriously confusing nature of English spelling it is particularly important to learn to think of English pronunciation in terms of phonemes rather than letters of the alphabet; one must be aware, for example, that the word 'enough' begins with the same vowel phoneme as that at the beginning of 'inept' and ends with the same consonant as 'stuff'. We often use special symbols to represent speech sounds; using the symbols chosen for this course, the word 'enough' would be written (transcribed) as ɪnʌf. A list of the symbol is given on p. ix.


The first part of the course is mainly concerned with identifying and describing the phonemes of English. Chapters 2 and 3 deal with vowels and Chapter 4 with some consonants. After this preliminary contact with the practical business of how some English sounds are pronoun-(p003end-p004begin)-ced, the fifth chapter looks at the phoneme and at the use of symbols in a theoretical way, while the corresponding tape unit revises the material of Chapters 2-4.




UKT on Sphota:

I need to clarify - at least for myself - the meaning of Lexeme {a.Daip~p-aim}, which belongs to the field of modern Psycholinguistics in which its precursor is "lemma". It comes to mind what I little know of Theravada Abhidamma (the Buddhist philosophy). The process of change known as lexicalisation :

Lemma (mental-thought) --> Lexeme (meaning-sound)
related to: {sait} --> {a.n}

I need to clarify - at least for myself - the meaning of Lexeme {a.Daip~p-aim}, which belongs to the field of modern Psycholinguistics in which its precursor is "lemma". It comes to mind what I little know of Theravada Abhidamma (the Buddhist philosophy). The process of change known as lexicalisation :

Lemma (mental-thought) --> Lexeme (meaning-sound)
related to: {sait} --> {a.n}

Its parallel to the ancients is Spota  स्फोट sphoṭa "bursting, opening, spurt' of Spota-Vada sphoṭavāda of Bhartṛhari (ca. 5th CE). This theory belongs to a branch of knowledge known as Vyakarana {bya-ka.ra.Na.} (see: ) relating to the problem of speech production - how the mind orders linguistics units into coherent discourse and meaning.
See: - https://en.wikipedia.org/wiki/Lemma-psycholinguistics 161127
- https://en.wikipedia.org/wiki/Spho%E1%B9%ADa 161127
- LANGUAGE AND THOUGHT - lang-thot-indx.htm (link chk 161127)
  > Bhartṛhari's Syntax, Meaning, Sphoṭa - spho-bartri-matilal.htm (link chk 161127)
and > Sphoṭa theory of language : a philosophical analysis - spho-cwrd-indx.htm
Downloaded txt (whose origin I couldn't trace) in TIL HD-library and SD-library
- BhartrihariSphota<> / Bkp<> (link chk 170912)
  " For early Buddhists (e.g., the Theravādins ) intuition is the highest source of knowledge. This intuition (prajā) is defined as "knowledge of things as they are in themselves as distinguished from what they appear to us."
  In general, the thrust of the Buddhist criticism of the Brahmanical viewpoint seems aimed more towards discrediting the unquestioning acceptance of a handed down tradition, rather than towards the rejection of śabda as having any possibility for truth bearing ".
- There many works explaining and analyzing Sphota by different authors.
  #1. A downloaded file by K K Mishra from his Bhartrihari's Theory of Spota,  p115-121
  of his Vol13art08, is in TIL HD-PDF & SD-PDF libraries
  - KKMishra-BhartrihaiSpota<> / Bkp<> (link chk 170603)
  - UnivHumanUnity-BhartrihariThSpota<> / Bkp<> (link chk 170303)
- Dictionary of Pali-derived Myanmar words (in Bur-Myan), by U Tun Myint, Univ. of Rangoon Press, 1968. - UTM-PDMD p207










31. The Hindu name for 'consonant' is vyan/ana, 'manifester'. The consonants are divided by the grammarians into sparca, i contact' or 'mute', antahstha, 'intermediate' or i semivowel', and usman, 'spirant'. They will here be taken up and described in this order.

32. Mutes. The mutes, sparca, are so called as involving a complete closure or contact \sparca], and not an approximation





Contents of this page

UKT notes

Rhoticity across BEPS languages

From: http://courses.nus.edu.sg/course/elltankw/history/Phon/D.htm 101028

Accents of English can be either rhotic or non-rhotic. A rhotic accent generally has /r/ more or less whenever it appears in the spelling. A non-rhotic accent, however, does not have the /r/ in final or pre-consonantal positions (this is sometimes known as the post-vocalic /r/, although others use the more accurate, but perhaps more cumbersome term, the non-prevocalic /r/). What this means is that speakers of non-rhotic accents have this rule: if the <r> in the spelling does not occur before a vowel sound, dont pronounce it. (NOTE: vowel sound, not vowel letter.)

UKT: I always get annoyed when people use high sounding terms like "pre-" and "post-" . All they have to do is to say that the syllable has CV structure. Note that I simply describe the onset-consonant as C, the vowel as V, and the coda-consonant as . Thus, I will rewrite:
  "A non-rhotic accent has no /r/ either as onset or coda." and,
  "Post-vocalic /r/" means "/r/ in the coda" .

Based on the above I can describe how the anti-Buddha (I am not using the word "devil" or "evil"), Mara {maar} came to be called /{mn-nt}/ in Bur-Myan.

{maar-nt} - n. 1. Mara - the archangel of evil -- MED359

Taking the Pali and Burmese words as a combination, we get /{maar~nt}/, and appying the "repha" conversion from Sanskrit and Pali, we get /{maan~nt}/ or /{mn-nt}/.

Mara lives on the highest deva-world much higher than that of {i.kra:ming:}. The only reason why he was opposed to the Buddha, was because the Buddha has shown humans and other creatures the way out of the cycle of birth and rebirth. Mara, being the "highest" deva does not like to see his underlings - Man - to leave his sphere of influence. Other than being anti-Buddha, Mara does not incite Man to do evil. Mara should not be compared to the Christian Devil. This is my understanding after a comparative study of Buddhism and Christianity.

Here are examples of words and phrases where the <r> wont be pronounced by non-rhotic speakers:

<party pooper >
<utter nonsense and balderdash>
<Mr Carter, you are so argumentative, arent you ?>

The phenomenon of non-rhoticity can be found in some other languages as well, such as Malay. Malay in Malaysia (Bahasa Melayu, Bahasa Malaysia) is non-rhotic whilst Malay in Indonesia (Bahasa Indonesia) is distinctively rhotic.

Indeed most southern Chinese languages like Hokkien, Teochew and Cantonese do not make use of /r/ at all, whereas northern Chinese languages like Mandarin make use of it extensively, so that some Singaporean versions of Mandarin are also non-rhotic, with the non-prevocalic <r> not pronounced (eger two pronounced [@] or [3]) and <r> in other positions pronounced [l] (egren person pronounced [l@n]).

UKT: SAMPA = IPA ; [@] = /ə/ ; [3] = /ɜ/ - http://www.phon.ucl.ac.uk/home/sampa/ 110224
SAMPA (Speech Assessment Methods Phonetic Alphabet)

Turning back to English, we can say that all English accents were rhotic up until the early MnE period and non-rhoticity was a relatively late development. (Remember, spelling reflects pronunciation in the early MnE period.) What is particularly interesting about the non-prevocalic /r/ is that before it was lost, it affected the vowel preceding it. It did three kinds of things:
   (1) lengthened the preceding vowel sound;
   (2) changed the quality of the vowel sound;
   (3) caused diphthongisation.

UKT: More in the original article.

Rhotic consonant

From Wikipedia: http://en.wikipedia.org/wiki/Rhotic_consonant 110112
Unfortunately, the Wikipedia article does not mention anything about Burmese, Pali or Sanskrit not even in name!

In phonetics, rhotic consonants, also called tremulants or "R-like" sounds, are liquid consonants that are traditionally represented orthographically by symbols derived from the Greek letter rho, ρ , including Roman R and Cyrillic Р. They are symbolized in the IPA (International Phonetic Alphabet) by upper- or lower-case variants of Roman R.


UKT: The following is an excerpt from Wikipedia, http://en.wikipedia.org/wiki/Liquid_consonant 110216
   "In phonetics, liquids or liquid consonants are a class of consonants consisting of lateral consonants together with rhotics. ...
   "Liquids as a class often behave in a similar way in the phonotactics of a language: for example, they often have the greatest freedom in occurring in consonant clusters. In some languages, such as Japanese, there is one liquid phoneme which may have both lateral and rhotic allophones.
   "English has one lateral, /l/ and one rhotic, /r/, and therefore has two liquids, exemplified in words such as <led> and <red>. Most other European languages also have two liquids, corresponding to /l/ and /r/ respectively."

My guesses on the obstruction and sonority of the plosives and nasals are based on their ability to check the vowel {i.} /i/ or syllable {pi.}: {laim}, {lain}, {laing}. I'm waiting for input from my Bur-Myan peers.

It is unfortunate that the above Wikipedia article does not mention anything about Bur-Myan, nor Skt-Dev. However, in compliance with the above, we can certainly say that Bur-Myan, {ra.}, {la.}, and {La.}, and suggest that Skt-Dev र ऱ ल ळ ऴ are liquids.

Consonants clusters, in the onset and coda, are quite common in English syllables. However, in Bur-Myan, these clusters behave like medial formers such as {ra.ric} and Tavoyan {la.hsw:}. They are of interest to us because they generate "medial" sounds. Because of this, the question bothering me at present is whether to consider them to be "vowels" similar to Skt-Dev ऋ ॠ (rhotic short and long vowels), and ऌ ॡ (lateral short and long vowels). -- UKT110116.

This class of sounds is difficult to characterise phonetically; from a phonetic standpoint, there is no single articulatory correlate common to rhotic consonants.  Rhotics have instead been found to carry out similar phonological functions or to have certain similar phonological features across different languages. Although some have been found to share certain acoustic peculiarities, such as a lowered third formant , further study has revealed that this does not hold true across different languages. For example, the acoustic quality of lowered third formants pertains almost exclusively to American varieties of English. Being "R-like" is an elusive and ambiguous concept phonetically and the same sounds that function as rhotics in some systems may pattern with fricatives, semivowels or even stops in others.

The most typical rhotic sounds found in the world's languages are the following:

Trill (popularly known as rolled r): The airstream is interrupted several times as one of the organs of speech (usually the tip of the tongue or the uvula) vibrates, closing and opening the air passage. If a trill is made with the tip of the tongue against the upper gum, it is called an apical (tongue-tip) alveolar trill; the IPA symbol for this sound is [r]. If it is made with the uvula against the back of the tongue, it is a uvular trill; the IPA symbol for this sound is [ʀ]. The bilabial trill, however, is not considered a rhotic.
   Many languages, such as Bulgarian, Swedish, Norwegian, Italian, Spanish, Russian, and Dutch, use trilled rhotics. In the English-speaking world, the stereotyped Scottish rolled [r] is well known. The "stage pronunciation" of German specifies the alveolar trill for clarity. Rare kinds of trills include Czech ř [r̭] (fricative trill) and Welsh rh [r̥] (voiceless trill).

Tap or flap (these terms describe very similar articulations): Similar to a trill, but involving just one brief interruption of airflow. In many languages taps are used as reduced variants of trills, especially in fast speech. However, in Spanish, for example, taps and trills contrast, as in pero /ˈpeɾo/ ("but") versus perro /ˈpero/ ("dog"). In some English dialects, such as American and Australian, flaps do not function as rhotics but are realizations of intervocalic apical stops (/t/ and /d/, as in <rider> and <butter>). The IPA symbol for this sound is [ɾ].

Alveolar or retroflex approximant (as in most accents of English with minute differences): The front part of the tongue approaches the upper gum, or the tongue-tip is curled back towards the roof of the mouth ("retroflexion"). No or little friction can be heard, and there is no momentary closure of the vocal tract. The IPA symbol for the alveolar approximant is [ɹ] and the symbol for the retroflex approximant is [ɻ]. There is a distinction between an unrounded retroflex approximant and a rounded variety found in Anglo-Saxon and even to this day in some dialects of English, where the orthographic key is <r> for the unrounded version and usually <wr> for the rounded version (these dialects will make a differentiation between <right> and <write>).

UKT: <right> /raɪt/  and <write> /raɪt/. DJPD16 gives the same sound for both words. However, in Bur-Myan, these would be differentiated:
<right> --> {Reit}; <write> --> {wReit} : {Reit} must be pronounced as in Rakhine dialect, and <w> {wa.} sound must not be dropped.

Uvular, velar or glottal approximant or fricative (popularly called guttural r): The back of the tongue approaches the soft palate or the uvula. The standard /r/s in Portuguese, French, German, and Danish are variants of this rhotic. If fricative, the sound is often impressionistically described as harsh or grating. This includes the voiced uvular fricative, voiceless uvular fricative, voiced velar fricative, voiceless velar fricative, and velar approximant. In northern England, there used to be accents that employed the voiced velar fricative, which was called a "burr". In southern England, the velar approximant is considered a prestigious kind of lisp, though it does not occur in many other national accents. In many Brazilian Portuguese dialects, the "r" is actually realized as a voiceless glottal fricative, unless it occurs single between vowels, being so realized as a tap.


In broad transcription rhotics are usually symbolised as /r/ unless there are two or more types of rhotic in the same language. The IPA has a full set of different symbols which can be used whenever more phonetic precision is required: an r rotated 180 [ɹ] for the alveolar approximant, a small capital R [ʀ] for the uvular trill, and a flipped small capital R [ʁ] for the voiced uvular fricative or approximant.

The fact that the sounds conventionally classified as "rhotics" vary greatly in both place and manner in terms of articulation, and also in their acoustic characteristics, has led several linguists to investigate what, if anything, they have in common that justifies grouping them together. One suggestion that has been made is that each member of the class of rhotics shares certain properties with other members of the class, but not necessarily the same properties with all; in this case, rhotics have a "family resemblance" with each other rather than a strict set of shared properties. Another suggestion is that rhotics are defined by their behavior on the sonority hierarchy, namely, that a rhotic is any sound that patterns as being more sonorous than a lateral consonant [ {la.n by:} but less sonorous than a vowel. The potential for variation within the class of rhotics makes them a popular area for research in sociolinguistics.

UKT: End of Wikipedia article.


Vocalic R (German): pronounced as vowel

From: Paul Joyce, German Course, Univ. of Portsmouth. http://userweb.port.ac.uk/~joyce1/abinitio/pronounce/consonr3.html 100102

The German vocalic 'r' is so-called because it is pronounced as a vowel, not a consonant. Sometimes referred to as a 'dark schwa', vocalic 'r' is articulated with the tongue slightly lower and further back in the vowel area than the 'schwa' sound heard at the end of such German words as 'Liebe', 'Katze' and 'Ratte'.

Vocalic 'r' can only be used in certain specific situations which are outlined below. Its most common usage is in unstressed "-er" syllables at the end of German words.

Sounds 1: Vocalic 'r' in final position: 83.mp3 <))
Bruder <brother> ; Schwester <sister>; Mutter <mother>; Vater <father>

The vocalic 'r' is also used in the final position in a word when the 'r'  follows a long vowel. Listen to the following six words, all of which end with a vocalic 'r' after a long vowel.

Sounds 2: Vocalic 'r' after a long vowel: 82.mp3 <))
Tor <gate; goal> ; Uhr <clock> ;
mehr <more> ; vier <four> ;
Bier <beer> ; Chor <chorus>

Vocalic 'r' is also heard when the letter 'r' follows a long vowel but precedes another consonant. Listen to the following four words in which vocalic 'r' occurs before a following consonant.

Sounds 3: Vocalic 'r' after long vowel + before another consonant: 81.mp3 <))
Pferd <horse> ; Herd <cooker> ;
sprte <felt> ; fhrte <led>

You will also hear vocalic 'r' in the unstressed German prefixes er-, ver-, zer- and her-. Listen to the vocalic 'r' in four words containing these prefixes.

Sounds 4: Vocalic 'r' in unstressed prefixes: 84.mp3 <))
erlauben <to allow> ; vergessen <to forget> ;
zertren <to destroy) ; hereinkommen <to come in>

Distinguishing between vocalic 'r' ['r' as vowel] and consonantal 'r'  ['r' as consonant]

In the following pairs of words, the first word contains a vocalic 'r' in final position. The second word in each pair however contains a consonantal 'r'. Listen and note the distinction between the sounds that are made in each pair of words.

Sounds 5: Vocalic 'r' or consonantal 'r' ? : jnger.wav <))
jnger <younger> ; die jngere <the younger one>
Meer <sea> ; Meere <seas>
clever <clever> ; der clevere <the clever one>

Finally, listen to these words in which vocalic 'r' and consonantal 'r' occur within the same word. Note in particular how adding an '-in' suffix can change the articulation of what was previously a vocalic 'r' sound.

Sounds 6: Vocalic and consonantal 'r' within the same word: bruder.wav <))
Frankfurter (Frankfurter sausage) ; Bruder <brother>
Lehrer (male teacher) ; Lehrerin (female teacher)
Reporter (male reporter) ; Reporterin (female reporter)


Vocalic r (syllabic r)

Excerpt from: http://www.enotes.com/topic/R-colored_vowel 110215

In English

A few dialects of English, particularly GA (General American) and Ulster English, contain a vocalic R sound, equivalent to the consonantal R sound [ɹ]. In Ulster English, both long and short versions exist, conditioned by the Scots Vowel Length Rule:

/wɹ̩k/ <work> (short vowel before the vl (voiceless consonant) /k/
/kɹ̩ːv/ <curve> (long vowel before the vd (voiced consonant) /v/

This is a little different from rhotacization described below (/wɝk/, /kɝv/ as opposed to non-rhotic /wɜːk/, /kɜːv/), as /ɹ̩/ is not a rhotic vowel or even a vowel, but may be treated as a similar phenomenon in this case, because this [ɹ̩] is phonemically identical to [ɝ], just realized differently. In general, however, a syllabic r (a vocalic r) and a rhotic vowel are different concepts.

The r-colored vowels of GA are written with vowel-r digraphs. Any vowel can be used:

Stressed /ɝ/: hearse, assert, mirth, work, turkey, myrtle
Unstressed /ɚ/: standard, dinner, Lincolnshire, editor, measure, martyr

An example of an r-colored vowel written as a vowel following "r" can be found in the word <iron> /ˈaɪɚn/ .

In Sanskrit

The ancient Indian language Sanskrit possessed short and long versions of a vowel sound often referred to as "vocalic r". It is represented in Devanagari by ऋ (short form) and ॠ (long form), and in IAST transliteration by ṛ (short form) and ṝ (long form), and is thought to correspond to original vocalic "l" or "r" in PIE (Proto-Indo-European).

The grammarian Pāṇini classified this vowel as retroflex and its pronunciation is thought to have been a retroflex approximant [ɻ] in classical Sanskrit (c. 500 BC). Earlier grammarians classified its sound in the Vedic period as velar. [See my note on Vedic Sanskrit] In Middle Indo-Aryan languages, the sound developed into a short vowel, usually /i/, but sometimes /a/ or /u/ (the latter sound especially when adjacent to a labial consonant).

However, when Sanskrit words containing this sound are borrowed into modern IA (Indo-Aryan languages) such as Hindi or Nepali its pronunciation changes to [ɾɪ] (short form) or [ɾiː] (long form), leading to forms such as "Krishna" for kṛṣṇa  कृष्ण kṛṣṇa, and "Rigveda" for ऋग्वेद ṛgveda, a pronunciation that is also prevalent among contemporary pandits. In the Southern Indo-Aryan language Sinhala, vocalic r in Sanskrit words is pronounced as [ur] or [ru], depending on the phonological context.

UKT: Though the very rhotic vowels, ऋ (short) and ॠ (long), are common in Classical Sanskrit, the lateral vowels, ऌ (short) ॡ (long) were only common in Vedic. Moreover, the way the vowel signs are written:
   ऋ --> ृ ; ॠ --> ॄ , compared to
   ऌ --> ॢ ; ॡ --> ॣ , are different.
With the lateral vowels, looking at the vowel sign tells you immediately the vowel letter. But this is not the case with rhotic vowels. This reminds the case of the present-day Bur-Myan medial signs, {ra.ric}-sign and {wa.hsw:}-sign. By looking at the {ra.ric}-sign you would not know what the medial former is. This observation has led me to believe the medial formers, {ya.}, {ra.}, {la.}, {wa.}, {ha.} can be looked on as "vowels", and that I should include them in the Bur-Myan vowel table.
  {ra.} <-- {ra.ric}-sign
  {wa.} <-- {wa.hsw:}-sign
Secondly, in accommodate Sanskrit rhotic vowels, we have to introduce a highly rhotic vowel. Thus we would have the rhotic vowel-signs:
  {ra.ric}-sign and {Ra.ric}-sign .

UKT: More in the article.

Go back rhoticity-note-b

Contents of this page

End of TIL file