Marathi Roman Script Discussion

Narayan Prasad prasad_cwprs at YAHOO.CO.IN
Tue Mar 1 01:15:38 UTC 2005

VYAKARAN: South Asian Languages and Linguistics Net
Editors:  Tej K. Bhatia, Syracuse University, New York
          John Peterson, University of Osnabrueck, Germany
Details:  Send email to listserv at and say: INFO VYAKARAN
Subscribe:Send email to listserv at and say:
          (Substitute your real name for first_name last_name)

Dear Sir,

        I joined the VYAKARAN list very recently. Although Marathi is not my
mother tongue, I do use Marathi works in case required for my topic of
interest. So I would have expressed my views on the subject mentioned above.
May be, you think of some update in future. In that case I would like to
present my views.

        Thanks and regards.
        Narayan Prasad

----- Original Message -----
From: "Madhukar N. Gogate" <mngogate at VSNL.COM>
Sent: Monday, February 28, 2005 3:30 PM
Subject: Marathi Roman Script Discussion

> VYAKARAN: South Asian Languages and Linguistics Net
> Editors:  Tej K. Bhatia, Syracuse University, New York
>           John Peterson, University of Osnabrueck, Germany
> Details:  Send email to listserv at and say: INFO VYAKARAN
> Subscribe:Send email to listserv at and say:
>           (Substitute your real name for first_name last_name)
> Archives:
> Please view on full screen.
> Title --  Marathi Roman Script Discussion
> For members of [marathi-romanscript at]
> Also for general publicity.
> Madhukar N Gogate (Pune, India), Moderator, initiated discussion
> on 1 Feb 2005. It is concluded with Moderator’s this message on 28 Feb
> 2005. All messages available to members as archives. Main points are
> noted here. Email mngogate at  See M11-M12-E03-E04
> in website  Comments invited. Circulate.
> 1. Thanks to all members (about 60% were silent) for participation.
> Membership of (33) was reduced to (30) at the end. There was no voting,
> and no direct decision. What did the group achieve ? Probably better
> awareness. All members may kindly help the cause with emails to friends,
> talks at clubs, articles in newspapers etc.
> 2. Marathi uses Devanagari (DN) script. It has many merits. It should
> prosper. Emails in DN are possible but often there are font problems.
> Roman script is linear and easy for typing and word-indexing. English
> language has global importance. In future, there will be huge development
> of electronic things, using the Roman script (lipi). DN-based products may
> not be viable. So take optional Roman lipi for Marathi.
> 3. Language and script are separate entities. Kids, blind persons and
> illiterates speak a language though unable to read. Some languages are
> multi-script. We use Marathi, but also use English when needed. Similarly,
> use Roman when needed.
> 4. Roman script has only 26 symbols, which are not sufficient to denote
> various Marathi sounds. There are 4 methods to tackle the deficiency
> (- 1 -) Use digraphs, that is 2-letter combinations such as aa, ae.
> speaking aa is not a + a, and ae is not a + e. But imagine they are single
> symbols (- 2 -) Use apo = apostrophe  mark. Example - (dagad') = stone.
> (d, d') have slightly varying  sounds. (d)  has sound of th-they in
> (d' ) has sound of d-dog in English (- 3 -) Roman script is a twin script.
> English treats smalls and capitals with equal phonetic values. But Marathi
> may carry  different values. Examples -- a (u-up English) A (rAjA = king)
> d (d in dagad') D (d' in dagad'). But intermittent capitals within words
> eyesores. (- 4 -) Give up symbols of defunct sounds. Example – (r’sh’i)
> may be written as (rushi) or (rishi).
> 5. Generally, words should be spelled as per standard Marathi symbols.
> English words such as police, icecream may be easily respelled in Marathi
> polis, aaiskrim. But people may object to respelled names for reasons of
> documents and sentiment. Respelling of technical words may hurt for
> Such words and names may be started with capitals for cautioning.
Example --
> Sodium Chloride. Its sound (sod’iyam kloraaid’) may be explained.
> use capitals for starting all words in English quotations and when one is
> not sure. Example -- respell as (pepar) or write (Paper).
> 6. Usual practice of a capital at start and a dot at end of a sentence,
> deserves review. First word (with capital) of a sentence would give a
> signal that it is technical. There may be some unrespelled words, with
> capitals, at various locations in a paragraph. A single dot is not
> A triple dot (…) will easily denote the end of a sentence. Leave a blank
> space after it, before starting the next sentence. Use single dots after
> initials of names. Sentence-ending (! and ?) need two extra dots.
> 7. A change in script gives an opportunity to drop some anomalies. For
> example, when English words are expressed in DN, we drop the second
> ( l ) in (sell) and the silent (h) in (honest). We may change DN forms
> (sarakaara, uchcha, r'sh'i, madhye, duh'kha) to Roman words (sarkaar,
> utcha, rushi, madhe, dukkha). We may omit DN duality ( i - ee,  u – oo)
> but take new duality (ch - ch',  j - j')  We may take (aa) or (a’) =
> (a – arm) to distinguish from  (a) = (both a - american). We may take
> (sat + tya = sattya = truth) (pun'+ n'ya =  pun'n'ya =spiritual merit) to
> distinguish from (ty – tyaag ) (n'y - pun'yaat = In Pune). Few anomalies
> might persist. DN word ghara (=home) is usually uttered as ghar, but in
> recitals as ghara (Example -- he vishvachi maaj’he ghara). Roman word
> would be ghar (with optional sound ghara).
> 8. People will accept Roman script if it is made reasonably phonetic, if
> has only an optional status, and if it can be adopted without making any
> changes in English-typing machines available with people. All machines may
> not be latest models. Also, all people are not scholars. The script should
> be easy. Words  teraa (= 13) tel (= oil) have rhasva-dirgha pronunciation
> (e) but its symbol is common in DN. Similarly, it may be ok to cancel the
> duality of (i-ee), (u-oo) in Roman. Names Hindi, Marathi are shown in DN
> with dirgha (ee) at end, but we accept rhasva ending (i) in English.
> 9. Above points were partly discussed. Also following -- People use (a)
> both (a, aa) in Marathi names. (aa) unduly makes Roman words long and
> tedious. Make no change. But (a, aa) are needed for phonetic clarity. Use
> (a’) for (aa) might help. Symbol (u’) may be reserved for (su’lfu’r) etc.
> suggestion was about using (w) in place of (v). Not ruled out, but (w) is
> not popular due to squeezed shape (in manual typewriters, equal space to
> every symbol), its complex symbol name, and to avoid clash of dev (= God)
> with English word (dew). A suggestion was about using the stroke  \ (like
> maatraa symbol in DN) and so on. It does not ensure the (a-z) link of
> Marathi to the great electronic and information world. (a-z) is already
> for phone books. Spellings should also work vertically in puzzles. A
> suggestion was about using diacritical marks such as a horizontal bar
> (a) to denote (aa). It was used in few scholarly Sanskrit books. It is
> good but not practical, since it is absent in most PCs and machines.
> has some diphthongs = vowel after vowel. Thus -- aai = mother. Some
> training is required to read (aai) as (aa + i ) and not as (a + ai).
> 10. Marathi M12 pdf (with few English hints) in  gives
> demo of 8 Marathi proverbs. (triple dot, aa, apo, i = ee, u = oo used. No
> name, no capital). Marathi does not have sounds of (a-cat) (aw-law) except
> for few English words. Show them as (ae) (ao). Any proposal of optional
> Roman lipi for Marathi will progress for emails, websites etc if symbol
> sound relations are standardized. At present, people use arbitrary Roman
> symbols for emails. Efforts are needed to propagate a practical standard,
> using Marathi language, for Marathi people. This note in English may help
> others too. A script should look nice. Irregular English spellings can
> affect Marathi in the Roman arena. Respell them, or take them with the
> capital symbol "guards".
> --  End ---
> This article neat 2 page 40 kB attachment is available on request
> mngogate at

No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 266.5.1 - Release Date: 2/27/2005

More information about the Vyakaran mailing list