An auxiliary Roman alphabet for Bengali closely follows the native orthography: the inherent a (অ) is always represented even when this is not pronoucned in Bengali or this changes its pronunciation. The latest and the perfect, especially in relation to Bengali, is the ISO 15919 Transliteration of Devanagari and related Indic scripts into Latin characters, having both print, or diacritic, system and the 7-bit ASCII system for certain chores on computers. the first line is the native character, the second is the diacritic system and the third is the system using printable characters of 7-bit ASCII (ISO 464 repertoire).
অ আ ই ঈ উ ঊ ঋ ৠ ঌ ৡ এ ঐ ও ঔ
a ā i ī u ū r̥ r̥̄ l̥ l̥̄ e ai o au
a aa i ii u uu ,r ,rr ,l ,ll e ai o au
ক খ গ ঘ ঙ চ ছ জ ঝ ঞ ট ঠ ড ঢ ণ
k kh g gh ṅ c ch j jh ñ ṭ ṭh ḍ ḍh ṇ
k kh g gh ;n c ch j jh ~n .t .th .d .dh .n
ত থ দ ধ ন প ফ ব ভ ম য র ল (ব) শ ষ স হ
t th d dh n p ph b bh m y r l (v) ś ṣ s h
t th d dh n p ph b bh m y r l (v) ;s .s s h
ড় ঢ় য় ৎ ং ঃ ঁ
ṛ ṛh ẏ (t) ;m .h ~
.r .rh ;y (t) ṁ ḥ m̐
Three rarely-used vowel characters (ৠ ঌ ৡ) are placed in the table. The character ৠ, coded ,,r, may only be used in certain Sanskrit words: pit,rr.na (পিতৄণ), maat,rr.na (মাতৄণ), bhraat,rr.na (ভ্রাতৄণ), etc. It is customary to write the words with separate characters as in pit,r,r.na (পিতৃঋণ) and maat,r,r.na (মাতৃঋণ). The same is the case with the characters coded ,l (ঌ) and ,ll (ৡ). While the first is used in only one Sanskrit root kl̥pta, (the word is mainly found in lexicons) the second is even very rarely used in Vedic Sanskrit. But these are placed in the table because these two characters used to replace the লি and লী in old manuscripts.
The character e and o are in effect long, to be coded ee and oo; their short counterparts are absent in Bangla while many other Indic languages have them. If it is de rigour to maintain uniformity in the transliteration among other Indic languages, the Bangla characters should be considered long, and coded doubled. But since there is no chance of goofing up of long and short sounds in the Bangla orthography, it is enough to use e and o, instead of ee and oo.
All the consonant characters, except the last four, subsume an inherent vowel a and this is essential to keep following each consonant unless it is followed by other vowel-markers. Absence of this inherent vowel a will assume that the consonants placed together are conceptually conjuncts and in the visual representation will, by default, try to show them as conjuncts. There may be situations where these characters, by definition, form conjuncts, but it is nonsensical to use conjuncts, or it is orthographically impossible to form such conjuncts, a colon (:) should be inserted to show the successive consonant characters with an in-between halanta or virama, which means lack of the inherent vowel.
The character va in the consonant chart should be replaced with ba because, although va and ba have different value and weightage in grammar, va is always realised as ba orthographically. This exists in the table to maintain if any orthographical equivalence table is needed among other Indic languages.
The character ৎ does not have any representation in the transliteration scheme. This character, devised and normalised by Ishvarchandra Vidyasagar, (unique to the Bengali wrting system; it is absent in any other Indic scripts) is conceptually equal to full character + halanta.
The numerals and other punctuation marks are the same as in standard Latin writing system. There are uses of extended Bangla characters, especially in dictionaries, lexicons, literary or scholarly works, some other notations are used to serve the purpose.
There are some extensions to the Bangla characters to represent words coming from or having Perso-Arabic origin. These are Bangla characters, modified by operands, which are diacritic marks. This method, also used to show foreign language sounds, is a system not widely in use, but found in lexicons and scholarly works.
The ISO draft proposal has the following characters; but there are more characters than are in the table in currency, under different schemes.
qa Perso-Arabic qaaf
_kha Perso-Arabic khaa'
.ga Perso-Arabic ghayn
za Perso-Arabic dhaal, zaa', zhaa', Daad, Dhaa'
fa Perso-Arabic faa', English f
wa Perso-Arabic waaw, English w
But there are several ways of modifying the Bangla characters to represent foreign-language sounds or scripts. A look into the details of such usage may be helpful in understanding the nature of this aspect of the Bangla script.
Revised: 5 April 2011