TIL

Devanagari Script

script-Devanag

A collection by U Kyaw Tun, M.S. (I.P.S.T., U.S.A.). Not for sale. Prepared for students of TIL Computing and Language Center, Yangon, MYANMAR.

UKT: These pages are in Arial Unicode MS font. Please remember that not all Unicode fonts are alike. Myanmar characters are in gif picture format and you do not need any Myanmar font to read it. Myanmar spellings in both Myanmar script and in Romabama and are included. Romabama spellings are within { }, regular English words are within < >, and IPA characters within / /.
   This is a collection from various sources:
• Unicode Consortium, http://www.unicode.org/versions/Unicode4.0.0/ch09.pdf
• D. Wujastyk Transliteration of Devanagari, 1996Jun25 http://www.ucl.ac.uk/~ucgadkw/members/transliteration/html/translit.html
D. Wujastyk, like most other authors did not differentiate between "alphabet" and "abugida".
• Daniel Jones Pronouncing Dictionary, 16th ed. (DJPD16), 2003
• Searching the Net with the string "abugida inherent vowel" brought up over 300 web-pages. This is a rewrite from some of those pages:
http://en.wikipedia.org/wiki/Abugida
http://maxpages.com/globalnetwork/Section_One

Devanagari script
Akshara
Inherent vowel a
Vowel killer
Consonantal aksharas (not letters)
Elusive inherent vowel a
Derivatives of inherent vowel a
Consonantal aksharas
Elusive inherent vowel a
Identification of vowels
Derivatives of inherent vowel a
Modifying characters
Conjunct Formation

Top | TIL home page
Linguistics - index

Devanagari script

Sanskrit is a language, and Devanagari is the script used in India to write Sanskrit since about 12th century. Sanskrit संस्कृत {thak~ka.ta.} or {thin~tha.ka.reik} can be written in many scripts, and in Myanmar-country , it, like Pali, is written in Myanmar-script {mran-ma sa}. Usually, few make a distinction between a language and a script, however when you are dealing with transcription of one language into another, or the transliteration of one script into another, unless you make this distinction clear, you are bound to get into misunderstandings leading to controversies. Thus, it is important to realise that in the country of Myanmar {mran-ma pré}, Myanmar {myanma sa} is the script used to write the Burmese language {ba.ma sa.ka:}.

Myanmar script is used to write not only the dialects like Yakhine {ra.hkein sa.ka:}, it is also used to write minority languages such as Karen {ka.rin sa.ka:}, Mon {mwun sa.ka:} and Shan {sham: sa.ka:}

UKT: Unicode recognises that the following languages use Myanmar as the script:
Burmese, Khamti, Mon, Shan http://www.unicode.org/onlinedat/languages-scripts.html
However, it fails to note that Pali is also written in Myanmar (country) in Myanmar (script).

When a Myanmar child goes to school to learn a language such as English, he goes to learn how to read and write. The primary goal is to write, that is, to learn the script. We say the child is learning  {in:ga.leit sa} -- meaning the way English is written. The emphasis is on the written language. Of course, he would learn the spoken language as well. The aim has NEVER been to teach the Myanmar child to speak like a native of America, Australia, Canada, or England. Please note that I have not used the word "Britain" -- I write "England". Therefore for teaching English in Myanmar, the question of which English dialect (American English, British English, etc.) to choose never arise. To the Myanmar (the people), the most important difference between American English and British English is only in the spelling -- not in the pronunciation.

The English language uses an alphabet -- the Latin or Roman alphabet, whereas Burmese language uses Myanmar script which is an abugida. The smallest unit of the alphabet is the letter which cannot be pronounced, and is therefore just a logo. On the other hand, the smallest unit of an abugida is the akshara which can be pronounced, and is therefore a syllable. The difference between an alphabet and an abugida can be illustrated by an example.

The English letter k has no sound: it must be accompanied by a vowel such as a to give it a sound. Then you have the syllable ka which can be represented by the phoneme as /ka/. The Myanmar {ka.} already has a sound because of its inherent vowel. Therefore it is legitimate to represent it as the phoneme /ka/. However, English k can neither represent {ka.} nor /ka/. For some more explanation, go to English and Myanmar syllables .

Convention used by UKT:
S-Devanagari -- Sanskrit language written in Devanagari script
B-Myanmar -- Bama language written in Myanmar script
E-Pali -- Pali written in Latin (English) script
M-Pali -- Pali written in Myanmar script.

Top | TIL home page
Linguistics - index

Akshara or the Devanagari syllable

S-Devanagari (Sanskrit in Devanagari spelling) was phonetic but with historical changes, the spelling of modern languages written in Devanagari may only be partly phonetic in the sense that a word written in it can only be pronounced in one way, but not all possible pronunciations can be written perfectly.

S-Devanagari comprises 46 to 51 akshar  or aksharas {ak~hka.ra}. The usual practice to describe the akshara as a letter is both wrong and misleading. But since, this mistake is so ingrained in our conversation, I myself might on occasions be accused of making the mistake.

By one count, Devanagari has 34 consonants, vyanjan {byi:~} , and 12 vowels, svar {tha.ra.} .

Devanagari is an abugida, and the smallest unit is the akshara.
It is both wrong and confusing to describe Devanagari as an alphabet and the akshara as a letter of the alphabet.

In an alphabet, only the vowel-letters can be pronounced, whereas the consonant-letters cannot be pronounced. Thus, in the Roman alphabet used for writing English (which we will call the English-alphabet), the consonant-letter k cannot be pronounced, whereas, the vowel-letter a can be pronounced. Though the consonantal-letter k cannot be pronounced, it is represented phonetically by /k/. The vowel-letter a can be represented as /a/ and can be pronounced. However, if k is accompanied by a vowel-letter such as a , then, it can be pronounced as /ka/.

k (no-pronunciation) + a /a/ —> ka /ka/

The akshara is a syllable because it contains an inherent vowel. The inherent vowel in the abugida of a language has the same pronunciation for all consonant-letters. However, the inherent vowel of a particular language can be different from the inherent vowel of another language. The character of an abugida is best to described as a syllable because it already has the inherent vowel a. Moreover, if the syllable has meaning it can be described as a word.

UKT:

	• Myanmar script has 44 characters similar to Devanagari aksharas. The Myanmar akshara is called an {ak~hka.ra}. Since all the 44 aksharas can be pronounced, they are true syllables. Some have meanings, and are therefore one-syllable words. e.g. the first akshara  {ka.} has the sound /ka/ and has the meaning <to dance> qualifying it as a monosyllabic word. • On the right are consonantal- aksharas presented according to the traditional array of Myanmar..
	• On the right are vowel-akharas.

Devanagari is abugida, whereas English script (properly Roman or Latin) is an alphabet. The individual character or glyph of an abugida is known as an akshara {ak~hka.ra}, which means the sound made when speaking or singing -- a root syllable of the Sanskrit language. This is to differentiate it from a letter of an alphabet. The akshara क (U0915) which is commonly transliterated as ka is pronounced as /ka/, because it has an inherent vowel very close to the English short a . On the other hand the letter k of the English alphabet has no sound of its own. The English letter k can be pronounced only when it has a vowel such as /a/ added to it, then it can be pronounced /ka/.

Top | TIL home page
Linguistics - index

Inherent vowel a

All the akshara of an abugida has the same inherent vowel a, which does not change in pronunciation, whereas the English a can have more than one pronunciation. See DJPD16 (Daniel Jones Pronouncing Dictionary 16th ed.) for pronunciation of < A> in < AE> < AEO> < AI and AY> < AU and AW>), and my notes on Myanmar < AYE>. See also some examples of change the English consonantal letters < B>, < C> (in < CC> < CCH> < CH> < CK> < CQU>) and < D>. See also phonetics < b > and < c > and also Myanmar { ba.} , { cha.} and { da.}

Because of the changing nature of the English vowels, English syllables represented by the same CV (consonant-vowel) combination can have many sounds. The following is an example (from: < A>), the CV represented by ca in <car>, <care>, and <carry> can have many sounds represented by different vowels. Even in the same word, the vowel-symbol is different for British English and for American English.

UKT: To illustrate the changing nature of English < a >, I will transcribe the above three words into Burmese-Myanmar script which will then be transliterated into Romabama.

<car> --> {ka:} . Vowel-sound: /ɑː/
<care> --> {kè:} . Vowel-sound: /eə/
<carry> --> {kè-ri} . Vowel-sound: /æ-i/

Unless an ESL (English as second language) teacher appreciates this difference in vowel-systems she will find it very difficult to teach English pronunciation to a Burmese-speaking Myanmar. That is the reason why a native-English speaker is not as effective as an ESL teacher compared to a Burmese-speaker who has received rigorous training in English pronunciation. However, since the vowel systems of Burmese and most of Indic languages are almost the same (thanks to the inherent vowel of the aksharas), an Indian-speaker (especially those from northern parts of India) has almost no problem in teaching English to the Burmese students.

Top | TIL home page
Linguistics - index

Vowel killer

UKT: The reader may prefer to read "Consonantal aksharas" first.

There are occasions when the inherent vowel of an akshara has to be "killed". The device to kill the inherent vowel is known as virama (Sanskrit) or {a.thut} (Myanmar). Though a vowel-killer is required in all abugidas, the actual sign are different in different abugidas. In S-Devanagari, it is U094D ्, and in Myanmar .

UKT: The following is based http://maxpages.com/globalnetwork/Section_One -- The author of this section does not differentiate between Sanskrit and Devanagari. I'm changing "Sanskrit" to "S-Devanagari". Failure to differentiate between language and script is the biggest head-ache for me when I read the materials on different web-sites.

The sound of an akshara does not ever get destroyed or changed. This is to be contrasted the sound of the CV (consonantal letter + vowel) syllable of an alphabet. The sound of a syllable in an abugida is essentially the sound of the akshara, and the sound of a word made of two or more syllables is the sound of the component aksharas. This is a concept which will help simplify text to speech applications with computers.

There are two aspects of non-destruction in the above explanation.
• The first one refers to the phonetic characteristics of the language, i.e., in any word, the aksharas retain their sounds.
• The second aspect of non-destruction, amazingly, is that the aksharas retain their individual meanings as well!
To give an example, consider the word guru consisting of the aksharas gu (meaning: darkness), and ru (meaning: the act of removal). When the two are taken together the meaning becomes "one who dispels the darkness of ignorance from the mind of the student" or "a teacher".

	Devanagari	ग	+	ु	→	गु
		[ga]	+	[u]	→	[gu]
	Myanmar		+		→
		{ga.}	+	{u.}	→	{gu.}
	Devanagari	र	+	ु	→	रु
		[ra]	+	[u]	→	[ru]
	Myanmar		+		→
		{ra.}	+	{u.}	→	{ru.}

On combining [gu] with [ru] we get [guru] गुरु

UKT: One should not be oversold by the idea of an akshara retaining its meaning. For example, "gu" also means a secret place, e.g. "guha" गुह  [guha]  is "one who is raised or brought up in a secret place -- Epithet for Kaarttikeya, one of the two sons of Shiva (Shivá).[kaarttikeya - shiva]" -- spelling and meaning of "guha" from   http://www.sanskrit-sanscrito.com.ar/english/sanskrit/sanskritnames.html
   The claim that an akshara retains its meaning was on the website:  http://maxpages.com/globalnetwork/Section_One , but the passage appeared under a disclaimer:
"DISCLAIMER: The Views expressed here are specific to the Samskritapriyah group and the Samskrit Education Society, Chennai. IIT Madras, has only made available the web pages as a courtesy to the group.

"The popular Sanskrit language is based on root syllables and words. Unlike the other languages of the world, every word in Sanskrit is derived from a root. It is a well accepted fact that all Indo-European languages have a common origin. On the basis of the above mentioned fact that all the words of Sanskrit are traceable to specific roots, a feature not seen in other languages, one can presume that Sanskrit is most certainly the origin."

UKT: Referring to the sentence above which I have highlighted, my question is: was the Samskritapriyah group trying to claim that Sanskrit was the "mother of all languages"?

In S-Devanagari there are 33 to 34 consonantal-aksharas (   {byi~: akhkara}; [vyañjana] in E-Pali). There are also vowel-aksharas  {thara. akhkara} to represent the vowels. However, there are only 33 consonantal aksharas in E-Pali, the number being the same in Myanmar and M-Pali. The difference in number seems to be in the way different authorities count the Semivowels ,  Sibilants and Aspirates,

UKT: Because of inherent vowel, the aksharas can be pronounced. That is, an akshara, unlike the letter of the alphabet is a syllable (aspect of pronunciation) and some have meanings making them into words (aspect of meaning). Thus,  {ka.} the first akshara in B-Myanmar means "to dance". However not all aksharas have meanings. A Myanmar example is  {ga.}, which has no meaning: it is just a syllable.

Consonant aksharas may also be rendered as half-forms ( samyuktakshar {tham~pa.yoat~ta. ak~hka.ra} -- meaning combinations) (also known as medials), which are presentation forms used to depict the initial consonant in consonant clusters. These half-forms do not have an inherent vowel. Their rendered forms in Devanagari often resemble the full consonant but are missing the vertical stem, which marks a syllabic core. (The stem glyph is graphically and historically related to the sign denoting the inherent /a/ vowel.)

UKT: There is a very large group of consonants which are derived from basic consonants. These are known as conjoined consonants or conjuncts. See consonant conjuncts. In Myanmar there are two types of conjuncts: those that can be pronounced and those which cannot be pronounced (in both cases a hidden virama {a.thut} is involved).

{ka.} + {ra.} → [ ] →   {kra.} (can be pronounced.)

{ka.} + {ka.} →  (cannot be pronounced.)

Some Devanagari consonant aksharas have alternative presentation forms whose choice depends upon neighboring consonants. This variability is especially notable for र (U0930)  (similarly for Myanmar {ra.}), which has numerous different forms, both as the initial element and as the final element of a consonant cluster. Only the nominal forms, rather than the contextual alternatives, are depicted in the Unicode code chart.

Forms of Myanmar {ra.}: ; (e.g. in ) ; in {ra.ris} for en-characters, and in {ra.ris} for em-characters

The traditional Sanskrit-Devanagari alphabetic encoding order for consonants follows articulatory phonetic principles, starting with velar (gutteral) consonants and moving forward to bilabial consonants, followed by liquids and then fricatives. ISCII and the Unicode Standard both observe this traditional order.

Both Sanskrit and Pali are studied in Myanmar particularly by Buddhist monks. However, Pali being the language of the Theravada Buddhism is learned to a greater extent by both monks and lay-persons alike. Myanmars are interested not only in Buddhism, but also in two large areas: Aryuveda medicine and predictive astrology common known as {bédin}, a word derived from {wéda.} -- {wéda.} being the collective name for Vedas.

Pali and Sanskrit used in Myanmar are in Myanmar script.  The traditional way of presenting the 33 Myanmar consonants ( {byi~:}; Pali: vyañjana) is a matrix of 7 rows and 5 columns -- the same form used for presenting Devanagari and other Brahmi-derived Indic scripts.

Note that linguists in India use phonetic terms such as "gutteral" for velar and "cerebral" for retroflex*. See An Elementary Pali Course, by Ven. Narada Thera, http://www.buddhanet.net/pdf_file/ele_pali.pdf and Devanagari Script on Achariya web-site, http://acharya.iitm.ac.in/sanskrit/lessons/Devan/cons_1.html and the related pages. These terms are now frowned upon by Western linguists.
   * http://www.campusprogram.com/reference/en/wikipedia/s/sa/sanskrit.html described this group as "Retroflex (roughly the place of articulation of English alveolars like t, but with the tongue curled back) (ṭ, ṭh, ḍ, ḍh, ṇ)" differentiating it from "Dental (tongue against teeth, like Spanish) (t, th, d, dh, n)"

The word alphabet is not usually applied to Sanskrit or other Indian languages. There is a significant difference between the notion of alphabet and the abugida (the term used by Achariya is aksharas which is ( {akhkara} in Myanmar and M-Pali)  When we think of the word "alphabet" we normally think of the letters of the language and a name given to each letter to identify it. In most languages the letters of the alphabet have names which may give a clue to the sound associated with the letter. In Sanskrit and other Indian languages and in Myanmar, there is no need to give a specific name to the letter. The sound the letter itself stands for the name for the letter. In a phonetic language reading becomes easy since the reader will be reading out the letters by uttering the sound associated with the akshara. More information on this is given in a separate section on Sanskrit and Phonetics.

In Myanmar, though there is no need to give specific names to the consonantal letters such as {ka.} and {kha.}, whenever we would like to emphasise such a letter we would say   {ka.kri:} for (meaning "the big {ka.}") and {hka.kwé:} for (meaning "the curled-up {hka.} )

Now that we know the difference between an akshara and a letter, we will use these two terms interchangeably remembering that a "letter" of an abugida has the inherent vowel a, and can be pronounced, whereas a letter of the alphabet has to have a vowel attached to it before it can be pronounced.

Devanagari characters or aksharas {ak'hkara} , like characters from many other Indic scripts can change shape on forming new characters (such as conjunct consonants) and when combined with another vowel. This is a source of confusion to students (particularly to those learning Sanskrit or Hindi as their second language) learning how to write.

On digitization a character's appearance can further be affected by its ordering with respect to other characters, the font used to render the character, and the application or system environment. These variables can cause the appearance of Devanagari characters to differ from their nominal glyphs (used in the Unicode code charts http://www.unicode.org/versions/Unicode4.0.0/ch09.pdf).

Additionally, a few Devanagari characters cause a change in the order of the displayed characters. This reordering is not commonly seen in non-Indic scripts and occurs independently of any bidirectional character reordering that might be required.

 Top | TIL home page
Linguistics - index

Consonantal aksharas

UKT: Since Devanagari is an "abugida", the unit character is an akshara, and I have captioned this section as "Consonant aksharas" instead of "Consonant letters" which would be true only for "alphabetic systems".
   This section should be read together with Consonant Conjuncts {byi:~ twè:} and {paahT-hsin.} on another page.

{byi: akhkara}; [vyañjana] in E-Pali

UKT: The transliteration of {byi: akhkara} as [vyañjana] and similar transliterations of some of words involving {ba.} into in E-Pali has been troublesome for me. In the case of , we should be getting [by-] instead of [vy-] -- remembering that is /ba/ and not /wa/. Only when the akshara is r6c4 {wa.}, should the transcription be v. The only explanation I can bring up at the present, is in the inability of some non-Tibeto-Burmese to pronounce the syllable {wa.} and {bya.} properly.

How many consonants are there in Devanagari? Devanagari has 34 consonants (vyanjan), and 12 vowels (svar). It has said that a syllable (akshar) is formed by the combination of zero or one consonants and one vowel -- http://www.campusprogram.com/reference/en/wikipedia/d/de/devanagari.html . However, if the consonant character such as {ka.} already has an inherent vowel /a/, then the previous statement (high-lighted) is to be disputed.

I have been asked many times in the US and Canada "how many consonants are there in Myanmar alphabet?" I have asked a similar question myself when I first learned Sanskrit and Pali. At that time, I did not know the difference between an abugida and an alphabet; nor the difference between an akshara and a letter. Moreover, I did not realised that, a hall mark of the Indic scripts and Myanmar is their ability to form combined consonants or conjuncts. (Myanmar can form only four types: {ya.pin.}, {ra.ris}, {wa.hswè:} and {ha.hto:}, whereas Hindi-Devanagari can form more.)

Leaving aside the formation of conjuncts, we can say that in Myanmar and in M-Pali, there are 33 aksharas to represent the basic consonants. There are 8 vowels in E-Pali (English-Pali) and M-Pali (Myanmar-Pali). In Myanmar script, the number of vowels has been increased to 12.

The consonantal-aksharas, the vowel-aksharas, and the inherent vowel a  (in fact all the aksharas) do not change in pronunciation, whereas the English a can have more than one pronunciation. See DJPD16 (Daniel Jones Pronouncing Dictionary 16th ed.) for pronunciation of < A> (in < AE> < AEO> < AI and AY> < AU and AW>), and my notes on Myanmar < AYE>. See also some examples of change the English consonantal letters < B>, < C> (in < CC> < CCH> < CH> < CK> < CQU>) and < D>. See also phonetics < b > and < c > and also Myanmar {ba.} , {cha.} and {da.}

I have already remarked that Devanagari aksharas (some consonantal-aksharas in particular) can undergo a change in shape -- the alternative presentation forms depending upon neighboring consonants. This variability is especially notable for र (U0930)  (similarly for Myanmar {ra.}), which has numerous different forms, both as the initial element and as the final element of a consonant cluster. Only the nominal forms, rather than the contextual alternatives, are depicted in the Unicode code chart.

Forms of Myanmar {ra.}: ; (e.g. in ) ; in {ra.ris} for en-characters, and in {ra.ris} for em-characters

The traditional Sanskrit-Devanagari character encoding order for consonants follows articulatory phonetic principles, starting with velar (gutteral) consonants and moving forward to bilabial consonants, followed by liquids and then fricatives. ISCII (Indian Script Code for Information Interchange) and the Unicode Standard both observe this traditional order.

Both Sanskrit and Pali are studied in Myanmar particularly by Buddhist monks. However, Pali being the language of the Theravada Buddhism is learned to a greater extent by both monks and lay-persons alike. Myanmars are interested not only in Buddhism, but also in two large areas: Aryuveda medicine and predictive astrology common known as {bédin}, a word derived from {wéda.} -- {wéda.} being the collective name for Vedas.

Pali and Sanskrit used in Myanmar are in Myanmar script.  The traditional way of presenting the 33 Myanmar consonants ( {byi~:}; Pali: vyañjana) is a matrix of 7 rows and 5 columns -- the same form used for presenting Devanagari and other Brahmi-derived Indic scripts.

Note that linguists in India use phonetic terms such as "gutteral" for velar and "cerebral" for retroflex*. See An Elementary Pali Course, by Ven. Narada Thera, http://www.buddhanet.net/pdf_file/ele_pali.pdf and Devanagari Script on Achariya web-site, http://acharya.iitm.ac.in/sanskrit/lessons/Devan/cons_1.html and the related pages. These terms are now frowned upon by Western linguists.
   * http://www.campusprogram.com/reference/en/wikipedia/s/sa/sanskrit.html described this group as "Retroflex (roughly the place of articulation of English alveolars like t, but with the tongue curled back) (ṭ, ṭh, ḍ, ḍh, ṇ)" differentiating it from "Dental (tongue against teeth, like Spanish) (t, th, d, dh, n)"

For many years now, people of the world have learnt the letters of Devanagari (Sanskrit) through equivalent Roman transliteration characters which employ special marks (known as diacritics). (UKT: refer to Romabama -- a transliteration of Myanmar abugida to English alphabet). The diacritics are based on a standardized representation for sounds followed in dictionaries. In India, the National Library at Calcutta has recommended that a single transliteration scheme be used to represent the letters across all the Indian languages, and Achariya used this scheme in its Internet pages. Students can take advantage of this when it comes to pronouncing the letters and words. A reference to this transliteration scheme is also available in these pages. The scheme is similar to the International Phonetic Alphabet representation but has some minor differences. 

UKT: Now you can go back to Vowel killer .

 Top | TIL home page
Linguistics - index

Elusive inherent vowel a

A hall-mark of an Abugida is the presence of an inherent vowel which is the same in each and every consonantal letter.

That the inherent vowel is the same, is to be contrasted with syllabaries like Katakana and Hiragana, which have one distinct symbol per possible syllable, and the signs for each syllable have no systematic graphic similarity.

The inherent vowel is described as close to the English short a, a statement which had lead me astray. The glyph a (U0061) is not given as a vowel phoneme in DJPD16. See page viii or go to 2.2 Vowels and diphthongs in my collection. It is given as a part of diphthongs /aɪ/ in English word <buy> and /aʊ/ in <now>. Moreover, when you look into how the English a is pronounced in DJPD16, you will see that the English letter a is pronounced in so many ways, that I am at a loss to pinpoint what the English a sounds like. If you follow the following links you will see what I mean:

< A> in < AE> < AEO> < AI and AY> < AU and AW>), and my notes on Myanmar < AYE>.

See also English and Myanmar syllables

Please remember that though I am a bilingual, the English that was spoken in my home (in Burma in 1930s) is not considered to be authentic English. It is generally held that every human infant has the ability to hear and articulate all phonemes of the human speech, but that ability starts to dwindle and becomes nil once the person is past the age of puberty. At my age, (I am now 70), I am no longer able to hear and articulate all the sounds of the human speech. I am totally "phoneme deaf" -- a parallel is color-blindness with respect to seeing. Like a color-blind person (who sees only shades of gray) identifying the red, yellow and green of the traffic lights with his shades of gray, I have to rely on phoneme representations.

Front Central Back   unrounded rounded unrounded rounded unrounded rounded high /i/ /y/ = /ü/ /ɨ/ = /u̶/   /ɯ/ /u/ lower-hight ɪ   /ı ̵/     /ʊ/ higher-mid /ɵ/ /o̷/ = /ö/     /ɤ/ /o/ mean-mid /E/   /ə/ /ɚ/   /Ω/ lower-mid /ɛ/ /œ/   /ʌ/   /ɔ/ higher-low /æ/       /ʌ/   low /a/   /ɑ/     /ɒ/	UKT: I have redrawn the above after DJPD16 p.xx. DJPD16 had reproduced it from IPA.
UKT: I have drawn the above table from:  http://www.ancientscripts.com/phonetics.html. Please note that some of the phonetic symbols may not be recognised by the IPA.

The above are two such representations of the vowels, and you will see that /a/ (not the English a) is described as a front vowel. What exactly are the sounds the Devanagari a अ U0905 and the Myanmar a represent?

At this stage we will have to assume that within one abugida, say S-Devanagari or B-Myanmar, the inherent vowel is the same, but such an inherent vowel is quite different from the English a . And that the dissimilarity from English a is due to the elusive nature of the English a .

Identification of Vowels

UKT: The following table on the pronunciation of Hindi-Devanagari vowels is from: http://encyclopedia.thefreedictionary.com/Devanagari

Top | TIL home page
Linguistics - index

Derivatives of inherent vowel a

UKT: Examples from Myanmar: {a.} (á -- accute) / {a} / {a:} (à -- grave)
Though these are usually described as "accent", I find it easy to look on them as separate vowels by themselves. From AHTD:
   accent n. 1. The relative prominence of a particular syllable of a word by greater intensity or by variation or modulation of pitch or tone. 2. Vocal prominence or emphasis given to a particular syllable, word, or phrase. 3. A characteristic pronunciation, especially: a. One determined by the regional or social background of the speaker. b. One determined by the phonetic habits of the speaker's native language carried over to his or her use of another language. 4. A mark or symbol used in the printing and writing of certain languages to indicate the vocal quality to be given to a particular letter: an acute accent. 5. A mark or symbol used in printing and writing to indicate the stressed syllables of a spoken word. 6. Rhythmically significant stress in a line of verse.

There are at least two derivatives of a -- the short and the long. The following is a quote:

"Vowels can be long or short. A long vowel is denoted by a colon ( : ) after the vowel. The best example in English of long vs short can be found in cases like "sad" (long) and "sat" (short). Notice how the 'a' (phonetically [æ]) sounds longer in "sad" than in "sat". So, "sad" is transcribed as [sæ:d] while "sat" is [sæt] "-- http://www.ancientscripts.com/phonetics.html

UKT: The above examples <sat> and <sad> are meaningless to a Burmese-speaking Myanmar. On transcription followed by transliteration into Burmese-Myanmar, we get {sak} and {sad} which are pronounced almost the same with the same duration.

On referring to DJPD16 for <sad> and <sat>, the vowel in both words is given as æ U00E6, the words are represented respectively as /sæd/ and /sæt/ without mentioning about the length. DJPD16 has an information panel on Length on page 313. The following is an excerpt from the said information panel.

"A term used in phonetics to refer to a subjective impression of how much time a sound takes; it is distinct from physically measurable 'duration'. Usually, however, the term is used as synonymous with duration. ... Length is important in many ways in speech: in English and most other languages, stressed syllables tend to be longer than unstressed (see RHYTHM, STRESS and WEAK FORM). Some languages have phonemic differences between long and short sounds, and BBC English is claimed by some writers to be of this type, contrasting for short vowels /ɪ e æ ʌ ɒ ʊ ə/ with long vowels /iː ɜː ɑː ɔː uː/ (though other, equally valid analyses have been put forward). However, the context in which these sounds occur must be taken into account. For example, the vowel /iː/ is said to be longer than /ɪ/ as well as having a different quality, but the vowel in <beat> /biːt/ is unlikely to be longer than the vowel in <bid> /bɪd/ as the phonetic environment in <beat> causes the vowel to be shorter. ... When languages have long/short consonant differences, as does Arabic, for example, it is usual to treat the long consonants as geminate; it is odd that this is not done equally regularly in the case of vowels. Perhaps the most interesting example of length differences comes from Estonian, which has traditionally been said to have a three-way distinction between short, long and extra-long consonants and vowels."

UKT: I have never fully grasped the difference between "short" and "long" vowels exemplified by /ɪ/ and /iː/ . I will make an attempt using the examples given in Identification of Vowels (Hindi-Devanagari vowels  http://encyclopedia.thefreedictionary.com/Devanagari ) for [ i ] as /ɪ/ and [ ī ] as /i/ . Relying on the similarity between in Hindi and Burmese, I am concluding:

इ	ई	इः
ि	ी
[ i ]	[ ī ]	[ iḥ ]
/ɪ/	/i/	/i:/
{i.}	{i}	{i:}

From the above we will conclude that the vowels can be long or short. Similar to Estonian, it is also possible for Devanagari and Myanmar to have three -- all derived from a. These I will term: short, normal, and long. However since the difference in length is a subjective impression, we will not try to compare across different abugidas. That is to say, for example, the short S-Devanagari a should not be compared to the short B-Myanmar a.

short	normal	long
अ	आ	अः
U0905	U0906	U0905 +U0903
[ a ]	[ ā ]	[ aḥ ]

UKT: It is usual to give examples of the functions of vowels by combining vowels with consonants.

Top | TIL home page
Linguistics - index

Modifying characters

http://encyclopedia.thefreedictionary.com/Devanagari

When no vowel is written, 'a' is assumed. To specifically denote the absence of a vowel, a halant (also called virama) is used.

UKT: Identification of Hindi-Devanagari consonants [ ] with IPA / / is useful in identification of Burmese-Myanmar consonants. Notice the IPA modifier used in c2 /k^h/ and c4 /g^ɦ/

	c1	c2	c3	c4	c5
r1	[ k ] /k/	[ kh ] /kh/	[ g ] /g/	[ gh ] /gɦ/	[ ṅ ] /ŋ/
r2	[ c ] /ʧ/	[ ch ] /ʧh/	[ j ] /ʤ/	[ jh ] /ʤɦ/	[ñ] /ɲ/
r3	[ ṭ ] /ʈ/	[ ṭh ] /ʈh/	[ ḍ ] /ɖ/ or /ɽ/	[ ḍh ] /ɖɦ/ or /ɽɦ/	[ ṇ ] /ɳ/
r4	[ t ] /t̪ /	[ th ] /t̪h/	[d] /d̪/	[ dh ] /d̪ɦ/	[ n ] /n̪/
r5	[ p ] /p/	[ ph ] /ph/	[b] /b/	[ bh ] /bɦ/	[ m ] /m/
r6	[ y ] /j/	[ r ] /r/	[ l ] /l/	[ v ] /v/	[ s ] /s/
r7		[ h ] /h/	ळ [ ḷ ] /ɭ /

Among these, r7c3 ळ, is not used in Hindi. The entire set is used in Marathi.

UKT: Hindi-Devanagari has three sibilants. I have already given one as r6c5 स Unicode name SA [ s ] /s/. The other two are:
•श Unicode name SHA [ ś ] /ɕ/
•ष Unicode name SSA [ ṣ ] /ʂ/

Top | TIL home page
Linguistics - index

Conjunct Formation

UKT: I have been trying out conjunct formation in Devanagari for the four conjunct formers that we find in Myanmar: {ya.} , {ra.} , {wa.} and {ha.}. The following is self-explanatory:

C + virama + ya

क    ख   ग   घ    ङ
क्य  ख्य  ग्य  घ्य  ङ्य 

च   छ    ज   झ   ञ
च्य  छ्य  ज्य  झ्य  ञ्य

ट    ठ    ड    ढ   ण
ट्य  ठ्य  ड्य  ढ्य  ण्य

त    थ   द   ध   न
त्य  थ्य  द्य  ध्य  न्य

प   फ    ब   भ   म
प्य  फ्य  ब्य  भ्य  म्य

य    र   ल   व   स
य्य  र्य  ल्य  व्य  स्य

ह   ळ    अ
ह्य  ळ्य  अ्य

ka + virama + ra

क  ख  ग  घ  ङ
क्र  ख्र  ग्र  घ्र  ङ्र

च  छ  ज  झ  ञ
च्र  छ्र  ज्र  झ्र  ञ्र

ट  ठ  ड  ढ  ण
ट्र  ठ्र  ड्र  ढ्र  ण्र

त  थ  द  ध  न
त्र  थ्र  द्र  ध्र  न्र

प  फ  ब  भ  म
प्र  फ्र  ब्र  भ्र  म्र

य  र  ल  व  स
य्र  र्र  ल्र  व्र  स्र

ह  ळ  अ
ह्र  ळ्र  अ्र

ka + virama + va

क  ख  ग   घ   ङ
क्व  ख्व  ग्व  घ्व  ङ्व

च   छ    ज   झ   ञ
च्व  छ्व  ज्व  झ्व  ञ्व

ट    ठ    ड    ढ    ण
ट्व  ठ्व  ड्व  ढ्व  ण्व

त   थ    द   ध  न
त्व  थ्व  द्व  ध्व  न्व

प   फ    ब   भ   म
प्व  फ्व  ब्व  भ्व  म्व

य   र   ल   व   स
य्व  र्व  ल्व  व्व  स्व

ह    ळ    अ
ह्व  ळ्व  अ्व

ka + virama + ha

क    ख  ग   घ   ङ
क्ह  ख्ह  ग्ह  घ्ह  ङ्ह

च   छ    ज  झ    ञ
च्ह  छ्ह  ज्ह  झ्ह  ञ्ह

ट    ठ    ड   ढ    ण
ट्ह  ठ्ह  ड्ह  ढ्ह  ण्ह

त   थ   द    ध   न
त्ह  थ्ह  द्ह  ध्ह  न्ह 

प   फ   ब   भ   म
प्ह  फ्ह  ब्ह  भ्ह  म्ह

य   र  ल    व   स
य्ह  र्ह  ल्ह  व्ह  स्ह

ह    ळ    अ
ह्ह  ळ्ह  अ्ह

Top | TIL home page
Linguistics - index
End of TIL file