Update: 2008-11-02 07:58 AM +0800


Burmese Grammar 1899

Formation of words (syllables):
Formation of medials


by U Kyaw Tun (UKT), Tun Institute of Learning, http://www.tuninst.net
From Burmese Grammar and Grammatical Analysis by A. W. Lonsdale, Education Department, Burma, British Burma Press, Rangoon, 1899. Copied by UKT and staff of TIL . Start: 2008 Aug.

  indx-RBM4M |Top

Contents of this page
Formation of words (syllables)
Changing the peak vowel -- ch04-1.htm
Medials -- ch04-2.htm
Formation of medials 
Syllables formed from medials in the onset
Coda consonants -- ch04-3.htm
Syllables with conjunct consonants -- ch04-4.htm
(Romabama vowels in rimes)
Pali derived syllables with coda consonants -- ch04-5.htm
Conjuncts including Kinsi {kn~si:} -- ch04-6.htm

UKT: The counterpart of Thuyathati in Hinduism is Saraswati. The original concept of Saraswati was that of a feminine water deity. Over a period of time, this riverine goddess came to be associated with all coursing and flowing energies and forces. Today, Saraswati is hailed as the patron goddess for the thought process and all creative arts.
-- http://www.indiaprofile.com/religion-culture/saraswati.htm

My study of Phonetics and Grammar on one hand and the Folk elements in Burmese Buddhism on the other and related topics, is leading me to believe that Thuyathati is the embodiment of our own voice, our ability to speak and communicate with another human being, which the ancient Pyus from which most of Myanmars have descended had worshipped as the Mother Goddess. (The reader should note that this is just a wild guess which I may abandon eventually.)

In Hindu mythology, Saraswati is supposed to be the consort of Brahma. On the right is shown the Shakti of Brahma -- the Goddess Brahmi.

Author's footnotes

UKT notes  -- note the author's Pali transcriptions: is used in place of ā , e.g., Pli for Pāli .
I am showing Lonsdale's transcriptions within Alt0171-Al0187: ..., e.g. (without slanting the character within)
Sibilant fricatives

Contents of this page
Contd from p018

Chapter IV contd.
Formation of words (syllables):
Formation of medials

28. Compound consonants. A consonant that is not combined with another consonant may be termed Simple [UKT: Basic], and a consonant combined with another may be called Compound.

UKT: A consonant combined with another consonant is a conjoined consonant or "conjunct". There are two types of conjuncts: those that are mute and those that can be pronounced. The pronounceable conjuncts are called medials, where the second consonant is always an approximant. To effect the combination, the inherent vowel of one of the consonants has to be killed using the viram or virama, which in Burmese-Myanmar is called an {a.tht} and is represented by a {tn-hkwan} (literal meaning: "flag") shaped as: . E.g.

{ka.} + + {ka.} --> {k~ka.} (mute) (vertical conjunct) ({a.tht} is hidden)
{ka.} + + {ya.} --> {kya.} (pronounceable) (note: the second consonant undergoes a change in shape) ({a.tht} is hidden).

29. Compound consonants are formed by combining simple consonants with one or more of the four letters {ya.}, {ra.}, {wa.} and {ha.} under symbolic form as shown in the following table: - [{p018end}]

{hka.} in combination with the {ya.ping.} ya-pin. symbol, is pronounced cha (see Sibilant Fricatives in my notes); and {nga.} when combined with the {ra.ric} ya-yit symbol, is pronounced precisely like {a.} nya.

30. The {ha.ht:} ha-ht: symbol merely adds an aspirated sound to an unaspirated letter. It is however, never combined with any consonant which already has its corresponding aspirate; thus {ta.~ha.} would be redundant as it is already represented by {hta.}. [UKT: I cannot accept Lonsdale position in total. To bring out the point in question, I have bolded "merely adds". It seems that Lonsdale seemed to be holding the view that the difference in pronunciation of {ta.} and {hta.} is just due to the absence or presence of aspiration. I maintain that it is more than that -- glottal features are involved. I will have to study this problem further.] UKT

This symbol, when united with {ya.} or {ra.} (not its symbol), gives the compound the power of the sibilant sh as in <ship>. The combination of {ya.ping.} ya-pin. and {ha.ht} [{p19end}] ha-ht: symbols together with {la.} is often pronounced as if written {rha.} sha [UKT: IPA /ʃ/ as in <sheep> /ʃiːp/], although its proper pronunciation is {lhya.} hlya. {hya.}/{thhya.} is also pronounced like {rha.} sha.

UKT: Because the Burmese language is non-rhotic and non-sibilant, we find it very difficult to write certain sounds such as the IPA /ʃ/. This sound is represented in two ways: {rha.} and {hya.} (note I am using the Old English letter 'thorn' instead of the modern English digraph <th> for r3c5 akshara ). Which is the correct representation has been disputed for a long time, until the MLC (Myanmar Language Commission) arbitrarily adopted {rha.} as the correct representation in the 1960s. However, the choice is very unfortunate as can be seen when Phonetics is taken into consideration.


Contents of this page

Syllables formed from medials in the onset

UKT: There are only four consonants that can form medials with other consonants, and amongst themselves. They are the approximants: {ya.}, {ra.}, {wa.} and {ha.}. These medials behave in every way like a basic consonant in forming syllables of the canonical form, CV or C1C2V. However, they can occur only in the onset, and not in the coda. In Romabama, the medial former always comes after the basic consonant, e.g., in {kya.} <k> is C1, <y> the medial-former is C2, and of course <a> is the peak vowel or nucleus. The medial formed is the {ya.ping.} medial. Similarly, in {mha.}, <m> is C1, and <h> the medial-former is C2. The medial formed is the {ha.ht:}.
   Thus, the usual way of transcription of {kha.} as <kha> implying the <h> to be a medial-former is not allowed, because <k> already has an element of aspiration and adding <h> again amounts to aspiration-over-aspiration. Therefore, the usual way of transcription of Burmese-Myanmar r1c2 akshara {hka.} as "Kh" is not allowed in Romabama.
   The maximum number of conjunct formers in the onset is found to be 3, e.g. {mhrwa} in a common word {a. mhrwa} - meaning 'twins'. You will notice that in {mhrwa} the basic akshara is {ma.} [IPA /m/] a bilabial nasal which is articulated at the most frontal position. And the vowel involved is {a} [IPA /a/] the most open front vowel. The articulators involved are the tip of the tongue and the lips two of the most versatile articulators of your vocal apparatus.

31. The vowel symbols are attached to compound consonants in the same way as they are to simple consonants. A few of these combinations are given below for the purpose of illustration: -

(1) {kya.}-medials
[UKT Note: I have taken out Lonsdale kyaw {kyau:} because it is emphatic and does not belong to this table which deals with only "short" and "long" vowels. You will see similar deletions in other similar tables.]


(2) {kwa.}-medials

UKT: See for yourself if you can fill up the missing places in the above series from the following {ka.}-syllables. Remember, to round your lips as you go from a basic akshara to the corresponding {kwa.}-medial, e.g., {ku} + (round lips) --> {kwu} .

Observe that the compound consonants formed with the {wa.hsw:} symbol are not combined with the symbols of {U.}, {U} [UKT: IPA /u/ ], {AU:}, {AU} [UKT: IPA /ɒ/], {} [UKT: IPA /o/], and {on} as there are no Burmese words which these sounds represent.

UKT: The above paragraph can be "explained" by Phonetics:
   The "compound consonants" Lonsdale is referring to are the medials formed with the consonant {wa.} (same as English <w>) (refer to the IPA and English consonant tables in ch02.htm, in which {wa.} [IPA /w/] is given as a bilabial approximant. The consonant, say {ka.} [IPA /k/], on forming the medial has its POA (place of articulation) pushed to the front toward the lips.
   The vowels that these {wa.hsw:}-medials, illustrated by {kwa.}-medials, cannot pick up are /u/, /o/ and /ɒ/ represented by Burmese-Myanmar vowels {U} /u/, {} /o/, and {AU} /ɒ/. These vowels are back vowels which are articulated way back in the throat. Because the POAs are far apart, your vocal apparatus cannot produce these sounds physically. And that is why "there are no Burmese words which these sounds represent". (I am waiting for input from Myanmar phoneticians.)
   Based on observations such as the above, I fully concur with Lonsdale when he writes, in the Preface, "The Burmese language is constructed on scientific principles, and there is no reason why its grammar should not be dealt with also from a scientific standpoint. But it may be safely said that Burmese grammar as a science has not received that attention it deserves."


(3) {mha.}-medials


(4) {mywa.}-medials


(5) {mhya.}-medials


(6) {mhwa.}-medials [{p020end}]

UKT: Chapter IV has to be subdivided into smaller files to make the display faster. Para 32 is on the next file: ch04-3.htm .

Contents of this page

Author's footnotes


Contents of this page

UKT notes

Sibilant Fricatives in Burmese

UKT: The following facts are to be rewritten into a subchapter.

" {hka.} in combination with the {ya.ping.} ya-pin. symbol", is pronounced as {hkya.}. ..."

Its pronunciation given by Lonsdale as cha is controversial. It is usual to express {hkya.} as IPA /ʧ/ or sometimes as /ʨ/ (I need confirmation on this point). In English /ʧ/ is an affricate - implying that the articulation is based on /t/ and /ʃ/. However, when I pronounce the {hkya.} carefully, I found that I do have have to start with "dental" {ta.}, meaning that  /t/ is not involved.
   The problem can be traced to the presence of the so-called allophones of /k/ - the [k] and [kʰ] (described as un-aspirated and aspirated), and the non-sibilant nature of Burmese speech.
   In English [k] is realized only when preceded by [s] as in <skin>. In other cases /k/ is always realized as [kʰ] as in <kin>. To the Western ears, the difference is only in aspiration, whereas to me at least it also involves "glottal" features. [k] and [kʰ] sounds are separate sounds and are written as separate aksharas: {ka.} (r1c1) and {hka.} (r1c2). In combination with the {ya.ping.}, we get {kya.} and {hkya.}. Both are voiceless. The voiced counterpart is {gya.}. The three form a series: {kya.}, {hkya.}, {gya.} - derived from row 1, {ka.}, {hka.}, {ga.}, of the akshara table.
   In my cross linguistic study of Burmese and Hindi, I find that the Hindi speakers writing in Devanagari (i.e. Hindi-Devanagari) list this series as r2:
Hindi-Devanagari row 2: च U091A Ca, छ U091B Cha, ज U091C Ja (ref: MS Windows Character Map). This should be compared to: 
Burmese-Myanmar row 2: {sa.}, {hsa.}, {za.} - all are alveolar sibilant fricatives.
   I am finding that the Western phoneticians (at least two with whom I have email correspondence) are under the impression that the Burmese pronunciations for {sa.}, {hsa.}, {za.} are the same as {kya.}, {hkya.}, {gya.}.
   There is one more problem with row 2. The r2c5 is occupied by {a.} (IPA: [ɲ] - palatal nasal) in Burmese-Myanmar table, and by {a.} in Pali-Myanmar table. Both {a.} and {a.} have the same sound in the onset of the syllable: IPA [ɲ] - palatal nasal. However these two characters behave differently in the coda of the syllable, e.g.
   {hky} - MLC transcription /chi/ - MEDict072
   {hkyi} - MLC transcription /chin/ - MEDict072
The difference in pronunciation is shown by the nucleus of the Romabama syllable.

"... and {nga.} when combined with the {ra.ric} ya-yit symbol, is pronounced precisely like {a.} nya."

What the above statement is stating is: {nga.} (velar nasal) on combining with the medial former {ra.} (alveolar approximant) produces a sound precisely like the palatal nasal {a.}. If you refer to the POAs of the IPA Consonants given on ch02.htm, you will see that the articulation is physically not feasible. However, if the medial former had been {ya.} (palatal approximant), the articulation is possible, i.e.

{nga.} (IPA [ŋ]) + + {ra.} (IPA [ɹ]) --> articulation not feasible
   because the POAs are far apart, however {nga.ra.ric} = is present in orthography
   with a sound precisely like {a.} and {a.} in the onset of the syllable. However, it is not
   allowed in the coda.

{nga.} (IPA [ŋ]) + + {ya.} (IPA [ j ]) --> ? (IPA [ ɲ]) - articulation feasible,
   but {nga.ya.ping.} is not present in Burmese orthography.

A partial way, to get out of this problem is to say that the pronunciation of the approximant {ra.} is not strictly alveolar, but possibly retroflex [ɻ ]. In other words, the Burmese language is so non-rotic, that not only is the trill [r] not allowed, but even the [ɹ] is not. This point should be remembered by ESL teachers teaching the Burmese students.

Go back sib-fricative-note-b

Contents of this page

End of TIL file