Update: 2011-12-31 05:36 PM +0630

TIL

English Phonetics and Phonology for Burmese-Myanmar speakers

con-vow.htm

by U Kyaw Tun (UKT), M.S. (I.P.S.T., U.S.A), Tun Institute of Learning (TIL),  http://www.tuninst.net Not for sale. Prepared for staff and students of TIL.

based on Peter Roach. English Phonetics and Phonology, a practical course. 2nd ed., 4th printing 1993, Cambridge University Press. ISBN 0-521-40718-4. pp 262 . For my reference, the printed book was digitized (type-copied) by Daw Khin Wutyi, B.Sc., TIL Computing and Language Center, Yangon, Myanmar. 2009. Page references to the original book are shown in my text for easy reference.

index.htm |Top
Eng-phon-indx.htm

Contents of this page

02.02 Consonant and vowel - vowel-letters and vowel-signs

Syllable and rime
Allophone problem

UKT notes
allophone obstruents and sonorants vowel diagram

Contents of this page
contd. p010

02.02 Consonant and vowel

The words vowel and consonant are very familiar ones, but when we study the sounds of speech scientifically we find that it is not easy to define exactly what they mean. There are sounds that are intermediate between the two and they are known as semivowels aks semiconsonants.

The most common view is that vowels are sounds in which there is no obstruction to the flow of air as it passes from the larynx to the lips. In other words a vowel can be sung continuously.

A doctor who wants to look at the back of a patient's mouth often asks the patient to say "ah" [a continuous sound]. The patient is required to "sing" this continuously] giving the doctor an unobstructed view of the back of the mouth.

But if we make a sound like ad /d/ you can feel that the air is suddenly stopped. It is the /d/ that has stopped the sound, and /d/ is therefore called a consonant. However, if we make a sound like ay /eɪ/ we are not quite sure if the sound has been stopped or not. In other words we are not sure what to call the Eng-Lat <y> . Is it a vowel or consonant? And the idea of a semi-vowel aka semi-consonant is born.

UKT: In Burmese and Indic languages, consonants are divided into two groups:
the {wag}-consonants - those that can neatly divided into tenuis, voiceless, voiced, pharyngeal, and nasal
the {a.wag}-consonants - those that are not definable

{wag} - n. 2 consonants of the alphabet group according to their POA (point of articulation) - MED2010-478 
The word <pharyngeal> is derived from <pharynx> - {l-hkyaung:wa.} - EMD2004-1014 . And, therefore, the consonant associated with this area should be called {l-hkyaung:wa. n} . I am waiting for input from my peers. -- UKT110226

 

Contents of this page

Syllable and rime

UKT: In the study of speech across BEPS languages, you will find that syllables (pronounceable) are more important than consonants and vowels which are its parts. A syllable of Eng-Lat has the canonical structure CVC and is essentially the same as that of Bur-Myanmar CV. The consonant C at the beginning of the syllable is known as the onset-consonant, V is the nucleus or peak vowel, and C (or - the killed consonant) at the end of the syllable known as the coda-consonant. The nucleus-vowel is the most important part of the syllable. The nucleus taken together with the coda is known as the rime . There can be many consonants in the onset and also in the coda: C1C2(VC3C4). However, in Bur-Myan there can be only one consonant (killed) in the coda. In recent years there have been attempts to introduce more than one killed consonant in the coda for foreign loan-words.

One problem is that some English sounds that we think of as consonants, such as the sounds at the beginning of the words <hay> /heɪ/ and <way> /weɪ/, do not really obstruct the flow of air more than some vowels do.

UKT: Roach's choice of examples, <hay> /heɪ/ and <way> /weɪ/, for this section is not suitable for our study of BEPS. In the the words <hay> /heɪ/ and <way> /weɪ/ the onset and coda are the {a.wag} or ill-defined consonants. They are approximants, and they can form medials: {ya.ping.}, {ra.ric}, {la.hsw:}, {wa.hsw:}, and {ha.hto:}. The only approximant that cannot form a medial is {a.}. They are sometimes called "semivowels", but can also be legitimately called "semi-consonants".

In <hay> /heɪ/ and <way> /weɪ/ - the examples chosen by Roach, the peak vowel /e/ is "checked" by <y> /j/. And since Eng-Lat <y> or its equivalent in Bur-Myan {ya.} is not a stop consonant, the vowel /e/ is not properly checked. Another way to look at the problem is to consider <a> and <y> forming a rime <ay> and take the nucleus as /eɪ/. This "nucleus" is not followed by any other stop consonant (plosive), and is therefore not "checked" at all.

Another problem is that different languages have different ways of dividing their sounds into vowel and consonant; for example, the usual sound produced at the beginning of the word <red> is felt is felt to be a consonant by most English speakers, but in some other languages (some dialects of Chinese, for example) the same sound is treated as one of the vowels. [{p010end}]

UKT: Roach's remark that in some dialects of Chinese, <r> is said be treated as a vowel is not surprising since {ra.} [grapheme's name is {ra.kauk} but pronounced as / {ya.kauk}/].  {ra.} /ɹ/ following {ya.} /j/ in Bur-Myan is an approximant.

In Skt-Dev, there are highly rhotic vowels (vow-letters ऋ (short), ॡ (long) which give vow-signs ृ ॄ respective). They are the equivalents of vocalic R of the German. See my note on Rhoticity across BEPS languages in Eng-phon-indx

This reminds me of 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 will have to introduce a highly rhotic vowel. Thus we would have the rhotic vowel-signs:
  {ra.ric}-sign and {Ra.ric}-sign .

This section "Syllable and rime" should be re-read after reading about obstruents and sonorants (sonority hierarchy) in my notes.

If we say that the difference between vowels and consonants is a difference in the way that they are produced, there will inevitably be some cases of uncertainty or disagreement; this is a problem that cannot be avoided. It is possible to establish two distinct groups of sounds (vowels and consonants) in another way.

Contents of this page

Consider English words beginning with the sound h; what sounds can come next after this h

The allophone problem

by UKT

By the phrase "the sound h", what does Roach mean? Such a phrase can mean two things: how a hearer hears the sound or how a speaker produces the sound. If both the hearer and the speaker belong to the same ethnic group where the people presumably uses the same set of muscles (among many), then there will be uniformity in hearing and speaking. However, when the speaker and hearer belong to different language-groups there can be no uniformity. For instance when a Burmese speaker produces the the sound h (articulating the akshara {ha.} or the medial {ha.hto:}), what another Burmese-speaker hears will be different from what an English-speaker hears. To the Burmese h is a well defined consonant with vowel like qualities, whereas to the English-speaker it is an aspirate .

aspirate - {ha.hto: n} - EMD2004-0069
UKT: The English word <aspirate> implies a very light sound similar to breathing - {l-n} - not the heavy {ha.hto: n} of Bur-Myan. This is why "h" is dropped in some English words like <honour>.

"Dropped h at beginning of words (Voiceless glottal fricative) - In the working-class ("common") accents throughout England, h dropping at the beginning of certain words is heard often, but its certainly heard more in Cockney, and in accents closer to Cockney on the continuum between that and RP. The usage is strongly stigmatized by teachers and many other standard speakers. e.g., <house> = ouse , <hammer> = ammer " -- http://www.ic.arizona.edu/~lsp/CockneyEnglish.html 110226

One word that has bothered me for a long time is <herb>. As a child before ever been to the US (before 1957), I was taught in Myanmar to say this word with an <h> something like {ha:b} . Once in the US, my good American friends made me change the pronunciation to {aa:b}. However, in Australia (in 1975), my Australian friends tried to correct me to say {ha:b}. That did really upset me, and what I said should not be repeated, and with my apology I explained to them what had happened. Now, I say {aa:b} to a Western audience, but {ha:b} to a Bur-Myan audience.

Thus to the Burmese speaker IPA /k/ and IPA /kʰ/ are different phonemes and two separate graphemes {ka.} and {hka.} are assigned to them. But to the English speaker they are allophones, i.e. the same sound - one without and the other with aspiration.
   The IPA now recognizes a phoneme, represented by the small cap ʜ, known as the voiceless epiglottal fricative which is articulated in the area between the glottal and pharyngeal. The Wikipedia , http://en.wikipedia.org/wiki/Voiced_epiglottal_fricative 090702, gives a sound clip and when I listened to it, it sounded like {hwa}. I am wondering if it could be the Burmese-Myanmar {ha.}.

To understand the allophone problem we have to look deeper down the throat. Among other things, the voice quality of a speaker depends on the many muscles controlling a unique bone in the throat - the hyoid bone which is "suspended" by many muscles. The figure on the right, "Schematic diagram of the action and location of the muscles of the hyoid complex" is from J. Laver, The Phonetic Description of Voice quality, Cambridge Univ. Press, 1980, p25.

We find that most of the sounds we normally think of as vowels can follow h ( for example e in the word <hen> /hen/ ), but practically none of the sounds we class as consonants. [UKT ]

Now think of English words beginning with the two sounds ; we find many cases where a consonant can follow (for example d in the word <bid>, or I in the word <bill>), but hardly any cases where a vowel may follow. [UKT ]

What we are doing here is looking at the different contexts and positions in which particular sounds can occur; this is the study of the distribution of the sounds, and is of great importance in phonology. Study of the sounds found at the beginning and end of English words has shown that two groups of sounds with quite different patterns of distribution can be identified, and these two groups are those of vowel and consonant. If we look at the vowel-consonant distinction in this way, we must say that the most important difference between vowel and consonant is not the way that they are made, but their different distributions. Of course, the distribution of vowels and consonants is different for each language.

UKT: I feel that reading Roach could have been easier if he had used the idea of rime in a syllable. Then the canonical structure of the syllable CVC could have been looked on as:
   C(VC) onset=1 coda=1 ,
   C(V) onset= 1 coda=0 ,
   (VC) onset=0 coda=1 .

Note that Roach has split <bid> into <bi> and <d> - splitting up the rime. This splitting is misleading for a Burmese speaker who is used to a killed consonant after the vowel. Because of this, I would split <bid> as <b> and <id>, where <id> is the rime, and, pronounce <bid> as /b/ and /ɪd/ if the vowel sound had been 'short'. However, if the vowel sound has been 'long' as in <bill>, {bi:l}, the splitting would be /biː/ and /d/ .

IPA uses the notation /ɪ/ for "short vowel" and /iː/ for "long vowel". Note the use of 'triangular colon' to indicate the long vowel, and is not the same as Bur-Myan {wic~sa.} . The IPA convention works well for Eng-Lat which has only 'short' and 'long vowel', however for Bur-Myan with three registers - the creak, the modal, and the emphatic - IPA convention does not hold. Thus I am forced to use the terms 'checked vowel' and 'free vowel' noting that a non-nasal consonant usually completely checks the vowel, nasal consonants checked the vowel partially, and approximants usually do not check at all leaving the vowel 'free'. Thus when I read Roach, I keep in mind that "long vowel" is almost the same as "free vowel", and "checked vowel" almost the same as "short'.

 

Note: /bɪd/ can be written in Burmese-Myanmar as {bd} . This type of spelling should be compared to the commonly used Bur-Myan word {hkt} meaning "era" or "age". Note the vowels in {bd} and {hkt} are both /e/ which in Romabama is represented by {} . It is this vowel that is checked by /d/ and /t/. See a vowel diagram for more information. Thus, to spell the Bur-Myan name {n-wing:}, I would use N Win , and not Ne Win which could be easily mistaken for {ni-wing:}.

There are many interesting theoretical problems connected with the vowel-consonant distinction, but we will not return to this question. For the rest of this course it will be assumed that the sounds are clearly divided into vowels and consonants.

UKT: By the above statement Roach seems to imply that the notion of "semi-vowels" which is important to Burmese-Myanmar will not be taken into consideration. Semi-vowels belonging to the approximants are important because they can form medials: {ya.ping.} , {ra.ric} , {wa.hsw:} , {ha.hto:} , and possibly {la.hsw:} .

We begin the study of English sounds in this course by looking at vowels, and it is necessary to say something about vowels in general before turning to the vowels of English. We need to know in what ways vowels differ from each other. The first matter to consider is the shape and position of the tongue. It is usual to simplify the very complex possibilities by describing just two things: firstly, the vertical distance between the upper surface of the tongue and the palate, and secondly the part of the tongue, between front and back, which is raised highest. Let us look at some examples:

1. Make a vowel like the in the English word <see> [{Bur-Myan speakers should pronounce {i:} }] and look in a mirror; if you tilt your head back slightly you will be able to see that the tongue is held up close to the roof of the mouth. Now make an vowel (as in the word <cat> /kt/) [{ {kakt} / {kt}] and notice how the distance between the surface of the tongue and roof of the mouth is now much greater. [{para-break}].

UKT: The transliteration for <cat> is represented in Burmese-Myanmar and Romabama as:


{ka.} + {akt} --> {kakt} (double "killed" consonants are not allowed in regular Burmese)
/k/ + /t/ --> /kt/

Instead of double "killed" consonants, it may be probably better to introduce another vowel sign, , (derived from {:}). The transcription for <cat> becomes:


{ka.} + {t} --> {kt} 

The problem is compounded because of the illogical way the three pitch-registers are spelled in Burmese-Myanmar:

(I am waiting for input from my Myanmar peers for the introduction of this new vowel sign. 090618)

In the Romabama transliteration, notice the inherent vowel a in {ka.}, and the checked vowel a in {ak}. The checked vowel is represented in IPA as // . The Burmese-speaker is to produce the sound {ak} /k/.

The difference between and is a difference of tongue height, and we would describe as a relatively close vowel and as a relatively open vowel. Tongue height can be changed by moving the tongue up or down, or moving the lower jaw up or down. Usually we use some [{p011end}] combination of the two sorts of movement, but when drawing side-of-the-head diagrams such as Fig.01 and Fig.02 it is usually found simpler to illustrate tongue shapes for vowels as if tongue height was altered by tongue movement alone, without any accompanying jaw movement. So we would illustrate the tongue height difference between and as in Fig.03.

UKT: Comparison of and is misleading because the former is a free vowel and the latter checked. I would rather make two sets of comparison. Ask your Bur-Myan student to place his finger on his tongue and say {i:} /iː/ and {a:}. Next ask him to say {hkt} and {hkt}. In both sets he will feel his tongue pushing his finger up. In the first set the raising of the tongue is very noticeable compared to that in the second.
   Now ask your Bur-Myan student to place his finger on his tongue again and say: {aa}, {}, {} and {i} in a sequence.  He will feel his tongue being pushed up by his tongue gradually almost without lateral movement. This sequence illustrates the front vowels with {i} in the highest-front most position.
   Now ask your Bur-Myan student to place his finger again on his tongue and say: {u}, {o} and {au}. The tongue movement is from up to down accompanied by some sort of lateral movement. This sequence illustrates the back vowels. The vowel {au} is in the lowest-back most position.

2. In making the two vowels described above, it is the front part of the tongue that is raised. We could therefore describe and as comparatively front vowels. By changing the shape of the tongue we can produce vowels in which a different part of the tongue is the highest point. A vowel in which the back of the tongue is the highest point is called a back vowel. If you make the vowel in the word <calm> /kɑːm/ {kau:m} (US) /kɑːlm/, which we write phonetically as ɑ: [{ {au:} }], you can see that the back of the tongue is raised. Compare this with in front of a mirror; is a front vowel and ɑ: is a back vowel. The vowel in 'too' [{ {u:} }] is also a comparatively back vowel, but compared with ɑ: it is close.

UKT: The examples of English words chosen by Roach can be misleading for a Bur-Myan who is used to free vowels and checked vowels, that is, those words which do not have {a.tht} consonants at the end and those which do. The vowels in <see> /siː/ and <too> /tuː/ are free vowels, whereas the vowel in <cat> /kt/ is checked.

So now we have seen how four vowels differ from each other; we could show this in a simple diagram (Fig.04). However, this diagram is rather inaccurate. Phoneticians need a very accurate way of classifying vowels, and have developed a set of vowels, arranged in a close-open, front-back diagram like Fig.04 , which are not the vowels of any [{p012end}] particular language. These cardinal vowels are a standard reference system, and people being trained in phonetics have to learn to make them accurately and recognise them correctly. If you learn the cardinal vowels, you are not learning to make English sounds, but you are learning about the range of vowels that the human vocal apparatus can make, and also learning a useful way of describing, classifying and comparing vowels. They are recorded at the end of Cassette 2. [{Unfortunately the cassettes are not available for this review.}]

 

It has become traditional to locate cardinal vowels on a four-sided figure (quadrilateral) [{see vowel diagram in my notes}] of the shape seen in Fig.05 (the design used here is the one recommended by the International Phonetic Association in 1989). [UKT ]

UKT: In the next section, English short vowels, you'll see a diagram of the vowel space which I have downloaded from www.umaniboba.ca/faculties/arts/linguistics/russell/138/sec3/vcharts.htm 090621. You can clearly see that there are only two reliable points on a vowel diagram showing the most contrastive vowels, {i} /i/ and {au} /ɑ/. The "cardinal vowels" are not the vowels of any language. They are the names of cardinal points on the IPA diagram. On comparing the IPA diagram to the vowel space, you'll see that it is actually a misrepresentation of the vowel space. However, it is still useful for discussion of vowels within one language, and for comparing languages.
   The approximate positions of Bur-Myan checked vowels within the limits of the IPA vowel quadrilateral are given in Fig.05.

The exact shape is not really important - a square would do quite well - but we will use the traditional shape. The vowels on Fig.05 are the so-called primary cardinal vowels; these are the vowels that are most familiar to the speakers of most European languages, and there are other cardinal vowels (secondary cardinal vowels) that sound less familiar. Cardinal vowel no.1 has the symbol [i] {i}, and is defined as the vowel which is as close and as front as it is possible to make a vowel without obstructing the flow of air enough to produce friction noise, {pwat teik n}; friction noise is the sort of hissing sound that one hears in consonants like s or f. [UKT ]

UKT: Since s has a problem with palatal c in the transcription of {sa.}, and f is unknown in Bur-Myan, the examples given by Roach are misleading. Instead of his examples, I would give sh as in English word <ship> /ʃɪp/ . The IPA transcription for this sound is /ʃ/. The equivalent in Bur-Myan is {thhya.}/{hya.}. Caution: the character is the Old English "thorn" which has become the digraph th in Modern English. Since, the MLC representation of /ʃ/ sound as {rha.} presents unnecessary problems from the point of view of POA (Point of Articulation), I giving {thhya.}/{hya.} for this sound. {thhya.}/{hya.} as spelled in Bur-Myan is a medial and applying an {a.t} to it would break it up. However, to include Skt-Dev श, BEPS needs to have this sound as a full-fledged consonant. My suggestion is to invent one, {sha.} which is derived from .

Cardinal vowel no.5 has the symbol [ ɑ ] {au} } - neither {au:} nor {au.},  and is defined as the most open and back vowel that it is possible to make. Cardinal vowel no.8 [u] {u}, is fully close and back, and no.4 [a] {a}, is fully open and front . [{para-break}]

UKT: The numbering of cardinal vowels is not only necessary for me at this stage, but is misleading in making mappings of Myanmar to IPA vowels. One of the main reason is Myanmar counts the vowels in clockwise direction starting from {a}, through {i} and {u}, to {au}, whereas the IPA counts anticlockwise starting from {i}, through {a} and {au}, to {u}. The second reason is because of the presence of the inherent vowel in Myanmar which does not have a dedicated grapheme but has to use {a.} for the central vowel schwa, /ə/, in words such as {a.ni}. This inherent vowel is present in all Brahmi-derived scripts including Devanagari and has been described as close to the English short "a". {a.} is also used to form rimes with killed aksharas to form rimes such as {ak}, so it is a checked vowel.

After establishing these extreme points, it is possible to put in intermediate points (vowels no. 2, 3, 6 and 7). [{para-break}]

UKT: In comparing languages such as Burmese to English, it is enough to take only three vowels into consideration: {i} /i/, {u} /u/ and {au} /ɑ/. These three form what is known as the vowel triangle.

Many students when they hear these vowels find that they sound strange and exaggerated; you must remember that they are extremes of vowel quality. It is useful to think of the cardinal vowel framework like a map of an area of country that you are interested in. Obviously, if the map is to be useful to you it must cover all the area; but if it covers the whole area of interest it must inevitably go a little way beyond that and include some places that you might never want to go to. However, it is still important to know where the edges of the map are drawn. When you are familiar with these [{p013end}] extreme vowels, you have ( as mentioned above ) learned a way of describing, classifying and comparing vowels. For example, we can say that the English vowel (the vowel in 'cat' ) is not as open as cardinal vowel no.4 [a] {aa}/{a}. (In this course cardinal vowels will always be printed within square brackets to distinguish them clearly from English vowel sounds.)

We have now looked at how we can classify vowels according to their tongue height and their frontness or backness. There is another important variable of vowel quality and that is lip-rounding. Although the lips can have many different shapes and positions [{shown above in the diagrams from John Laver}], we will at this stage consider only three possibilities. These are:

Rounded, where the corners of the lips are brought towards each other and the lips pushed forwards. This is most clearly seen in cardinal vowel no.8 [u].

Spread, with the corners of the lips moved away from each other, as for a smile. This is most clearly seen in cardinal vowel no.1 [i].

Neutral, where the lips are not noticeably rounded or spread. The noise most English people make when they are hesitating (written 'er' ) has neutral lip position.

Now, using the principles that have just been explained, we will examine some of the English vowels.

Contents of this page

UKT notes

allophone

allophone n. Linguistics 1. A predictable phonetic variant of a phoneme. For example, the aspirated t of <top>, the unaspirated t of <stop>, and the tt (pronounced d ) of <batter> are allophones of the English phoneme t. -- AHTD

UKT: For disyllabic words such as <batter> /bt.əʳ/, though it is quite common to consider the two t-s together as a "double consonant", it must be remembered that the first t can be considered to belong to the first syllable and is the "coda", and the second t to the second syllable and is therefore the "onset". Because of this I cannot agree with AHTD in calling tt an allophone of t . - 090616

Go back allophone-note-b

Contents of this page

obstruents and sonorants

UKT: The classification of sounds into obstruents and sonorants is very confusing for a person who is used to the akshara classification (sounds as well as scripts) of {wag}-nasal-{a.wag} classification. See also Sonority hierarchy in my notes in intro-voc1.htm (not included in these files).

The following definitions are from AHTD.

obstruent n. 2. Linguistics A sound, such as a stop, a fricative, or an affricate, that is produced with complete blockage or at least partial constriction of the airflow through the nose or mouth. [Latin obstruēns obstruent-, present participle of obstruere to obstruct; See obstruct ] -- AHTD

sonorant n. Linguistics 1. A voiced consonant regarded as a syllabic sound, as the last sound in the word sudden. [sonor(ous) -ant ] -- AHTD

The following are from other sources.

obstruents

UKT: We may take the aksharas of the r1, r3, r4 and r5 rows of the akshara matrix, the {wag}-consonants (exclusive of the nasals), as obstruents.
   However, the r2 consonants {sa. hsa. za. Za.} are problematical. The case of {sa.} is illuminating. It has two pronunciations: in the coda it is a stop IPA [c], whereas in the onset it is a fricative IPA [s. e.g. {thic~sa}. English <cc> is also of this type: <success> /sək'ses/ (transcription from DJPD16-515). It should be noted that since the POAs of IPA [k] and IPA [c] are close, I have suggested that the transcription could have been /səc'ses/ which calls for a palatal <c> in English. When I posted this possibility on a forum, almost all responses were that English does not have a palatal <c>, which is true if <c> has been an onset. I insisted that coda <c> could very well be [c], the case being similar to the case of {sa.}.

From: Wikipedia http://en.wikipedia.org/wiki/Obstruent download 070910

In phonetics, articulation may be divided into two large classes, obstruents and sonorants. An obstruent is a consonant sound formed by obstructing outward airflow, causing increased air pressure in the vocal tract.

Obstruents are those articulations in which there is a total closure or a stricture causing friction, both groups being associated with a noise component; in this class there is a distinctive opposition between voiceless and voiced types.

Obstruents are subdivided into stops [{plosives}], fricatives, and affricates. Obstruents are prototypically voiceless, though voiced obstruents are common. This contrasts with sonorants, which are rarely voiceless.

 

sonorant

From: Wikipedia http://en.wikipedia.org/wiki/Sonorant download 070910
UKT: Vowels are sonorants, whereas the nasals are between sonorants and obstruents (my understanding of 080313)

In phonetics and phonology, a sonorant is a speech sound that is produced without turbulent airflow in the vocal tract. Essentially this means that a sound is sonorant if it can be produced continuously at the same pitch. For example, vowels are sonorants, as are consonants like /m/ and /l/. Other consonants, like /d/ or /k/, cannot be produced continuously and so are non-sonorant. In addition to vowels, phonetic categorizations of sounds that are considered sonorant include approximants, nasal consonants, taps, and trills. In the sonority hierarchy, all sounds higher than fricatives are sonorants.

UKT: The above Wiki paragraph implies that nasal consonants (e.g. /m/) and lateral consonants (e.g. /l/) are sonorants. This means that {nga. a. Na. na. ma.} are sonorants. But, since "a sound is sonorant if it can be produced continuously at the same pitch", I cannot fully agree with my inference "{nga. a. Na. na. ma.} are sonorants", because I cannot "produce them continuously at the same pitch". But first, I will have to check with my peers.
   However, we must note that when the inherent vowel in {nga. a. Na. na. ma.} has been killed and when they occur in the coda, the "rimes" have three pitch-registers (tones) such as {a. a a:}, exemplified by {kan. kan kan:}.

Sonorants are those articulations in which there is only a partial closure or an unimpeded oral or nasal scape of air; such articulations, typically voiced, and frequently frictionless, without noise component, may share many phonetic characteristics with vowels.

The word resonant is sometimes used for these non-turbulent sounds. In this case, the word sonorant may be restricted to non-vocoid resonants; that is, all of the above except vowels and semivowels. However, this usage is becoming dated.

Sonorants contrast with obstruents, which do cause turbulence in the vocal tract. Among consonants pronounced far back in the throat (uvulars (UKT: [q, ɢ, ɴ, ʀ, χ, ʁ]), pharyngeals) the distinction between an approximant and a voiced fricative is so blurred that such sounds as voiced uvular fricative ([ʁ] and voiced pharyngeal fricative ([ʕ]) often behave like sonorants. The pharyngeal consonant (UKT: [ħ]? Can I say it is {ha.}?) is also a semivowel corresponding to the vowel /a/.

UKT: The farthest into the throat are velars {ka. ga.}. I am wondering, because of the way the Burmese-Myanmar monks recite the {ka.ma.wa}, whether {Ga.} is pronounced farther into the interior. In which case it would be a uvular. -- 070915

Whereas most obstruents are voiceless, the great majority of sonorants are voiced. It is certainly possible to make voiceless sonorants, but sonorants that are unvoiced occur in only about 5 percent of the world's languages. These are almost exclusively found in the area around the Pacific Ocean from New Caledonia clockwise to South America and belong to a number of language families, among them Austronesian, Sino-Tibetan (UKT: Burmese of Tibeto-Burman group), Na-Dene language and Eskimo-Aleut. It is notable that, in every case where a voiceless sonorant does occur, there is a contrasting voiced sonorant.

Voiceless sonorants tend to be extremely quiet and very difficult to recognise even for those people whose language does contain them. They have a strong tendency to either revoice or undergo fortition to form for example a fricative like or ɬ.

English has the following sonorant consonants: l, m, n, ŋ, r, w, j . [{The corresponding {ya.}, {ra.}, {la.}, {wa.} are {a.wag}-consonants, and {nga.} {na.} {ma.} are nasals. Note that {ya.}, {ra.}, {la.} and {wa.} are medial formers. This is because they are quite close to the high vowels in the sonority hierarchy and sometimes described as "semivowels", whereas {tha.} being closer to the plosives cannot form medials and is definitely not a semivowel.}]

UKT: Cross linguistic comparison, between Burmese and English, has brought out some interesting properties of consonants and vowels. Working with the rimes in the syllables of the type CV, has brought out that the pronunciation of C and can be quite different in both Burmese and in English. A specific example is in the <cc> digraphs of the disyllabic words such as <success> /sək'ses/ (transcription from Daniel Jones Pronouncing Dictionary, 16ed. p515). In the <cc> the first <c> belongs the syllable /sək/ whereas the second <c> belongs to the syllable /ses/. This has prompted me to suggest that "there is palatal <c> in English" but only in the coda. The rational being: POA's of velar stop /k/ and the palatal stop /c/ are so close that we have made a mistake in giving the transcription as /sək'ses/. It could very well be /səc'ses/. If we could accept this position, then we can say that the English <c> and the Burmese {sa.} are exactly the same.

Go back obst-sono-note-b

Contents of this page

vowel diagram

From Wikipedia http://en.wikipedia.org/wiki/Vowel_diagram 090622

A vowel diagram or vowel chart is a schematic arrangement of the vowels. Depending on the particular language being discussed, it can take the form of a triangle or a quadrilateral. Vertical position on the diagram denotes the vowel closeness, with close vowels at the top of the diagram, and horizontal position denotes the vowel backness, with front vowels at the left of the diagram.

The vowel systems of most languages can be represented by such a chart. Usually they are evenly distributed on the chart, a phenomenon that is known as vowel dispersion. For most languages, the vowel system is triangular. Only 10% of languages, including the English language, have a vowel diagram that is quadrilateral. Such diagrams are termed vowel trapezia or vowel quadrilaterals. German phonologists know these as, respectively, a Vokalviereck and a Vokaltrapez.

The IPA vowel chart comprises the cardinal vowels, and has the form of a trapezium. By definition, no real vowel sound can be plotted outside of the IPA trapezium, because its four corners represent the extreme points of articulation. The vowel diagrams of most real languages are not so extreme. In English, for example, high vowels are not as high as the corners of the IPA trapezium, nor are front vowels as front.

UKT: End of Wikipedia article

Go back vow-diag-note-b

Contents of this page

End of TIL file