Transliteration of Urdu with Roman Alphabets
Yogesh
earlier discussion, please see the older thread "hello people".
> Afzal A. Khan wrote:
> > Yogesh wrote:
> >> Yes, we are splitting hair, and why not? The importance of correct
> >> pronunciations in a language cannot be taken lightly. On this subject,
> >> let me take this opportunity to bring out a few points on
> >> transliteration that have been on my mind.
> >> 1. Since transliteration is supposed to represent a brand new script
> >> for Urdu/Hindi, albeit with Roman alphabets, let us first dispose off
> >> any notion of following the rules that would be applicable when writing
> >> English in Roman script. For example: the use of upper case when
> >> commencing a sentence. Urdu/Hindi do not have any upper case in their
> >> alphabets, in addition, in many cases of transliteration, we
> >> specifically employ Roman upper case alphabets with entirely different
> >> connotation than the lower case. Therefore, how can we justify the use
> >> of upper case letter at he start of a sentence when the Urdu/Hindi
> >> sound does not demand it? There is no confusion when we write in either
> >> Arabic or Devanagri script, why should it be any different when
> >> employing the Roman script?
> >> 2. There is nothing wrong in adopting any transliteration scheme, so
> >> long as it is consistent in its application. Take for example the
> >> letter 'e' that in iTrans method represents the sound that we use
> >> in words like 'de' (give), 'le' (take), 'ahl-e-fan'
> >> (craftsman), 'betaab' (restless) etc. Yet I have seen 'e'
> >> employed for the sound that would be more appropriately represented by
> >> 'a' - e.g. 'lehar instead of 'lahar', 'pehar' instead
> >> of 'pahar', 'sehar' instead of 'sahar' , 'kehna'
> >> instead of 'kahna'- it is difficult to understand the rational,
> >> if any, behind it. And this being used by some very capable people
> >> proficient not only in Urdu but also in transliteration schemes,
> >> whether adopted or one of their own.
> >> 3. As to the use of 'j' instead of 'z', we find some words now
> >> being accepted with dual usage, 'nazaria' and 'najaria' etc.,
> >> by the Hindi-speaking world. But for this forum, dedicated to Urdu
> >> poetry, it still would be more appropriate to follow the lexicons. Same
> >> for 'q' vs 'k' - one has to know the correct pronunciation of
> >> the word, there is no short cut here.
> >> Regards,
> >> Yogesh
> > Sethi Saheb,
> > Your post needs a fuller reply and, right now, I am pressed for
> > time. Maybe later today.
> > Referring to your last para, "nazariya" and "najariya" are two
> > different words -- not a single word "accepted with dual usage"
> > (i.e. with both "j" and "z"). Also, "najariya" is a sort of
> > colloquial pronunciation. Wasn't there an old film with that
> > title ? And also a well-known song : "Najariya ki maari...".
> > I do hope you are NOT suggesting that "nazariya" (meaning 'point
> > of view') is also "permissible" with a 'j' -- "najariya".
> > Afzal
> Sorry for following up......
> As promised, here is a fuller reply.
> Why does anybody use Roman English while transcribing a
> text in Urdu/hindi ? In a Newsgroup where it is possible
> to have and use the Urdu/hindi font, I suppose no one would
> be required to use Roman English. The latter's use arises
> simply due to the absence or unavailability of any Urdu/hindi
> font. Any transliteration scheme should be user-friendly. And
> if posters and viewers are both comfortable with the Roman
> English (that is used in ALUP, for instance), there should be
> no objection to it. After all, it is merely a substitute for the
> original script. Is there any necessity for the users to
> follow a set of rules that may govern the use of the original
> script ? I don't think so. In the Urdu script, there are no
> capital letters. Should that mean that we must necessarily
> eschew the use of capital letters, even while using the Roman
> English alphabet ? We use this alphabet in ordinary posts too,
> that are couched in the English language. We are all familiar
> with the rules governing such usage and we always comply with
> these rules, to the best of our ability. And that compliance
> includes the use of capital letters where appropriate (such as
> in proper nouns or at the beginning of a new sentence). And,
> therefore, we find it natural and convenient to follow the same
> "rules" while couching our Urdu/hindi text in Roman English. It
> would be my submission that the rules for us to follow should be
> those of the Script we are using, rather than those of the Lang-
> uage that we are writing. In this context, I have already empha-
> sised ease of use as a cardinal principle. I daresay, if people
> found that dropping capital letters is easier to adopt, rather
> than employing them, they would readily choose to do so. And I
> for one would have no objection. But it may not be right and
> proper to insist that the rules or practices of one Script must
> be followed while using another Script altogether.
> You further state, and I quote :
> "There is nothing wrong in adopting any transliteration scheme,
> so long as it is consistent in its application."
> The difficulty is that it is virtually impossible for people
> (meaning ALUPers here) to agree to a single transliteration
> scheme. You speak of iTrans. You must also be aware of a very
> elaborate scheme devised by our own Sarwar Saheb. Except Raaz
> Saheb himself, I doubt if anyone else uses that scheme here.
> And, as far as I am aware, we find iTrans confined to the
> ISB and Giitaayan only (for the limited purpose of archiving
> lyrics). {Although UVR Saheb always claims that it has so many
> different uses !}
> I would say that the present practices that are being observed in
> ALUP are quite in order. We understand and follow each other's
> ideas and there is a healthy debate. If all of us were agreed
> on a single transliteration scheme, then your suggestion about
> "consistency in its application" would have some merit. For the
> time being at least, we can follow the time-honoured adage of
> "the greatest good of the largest number".
> Let us now take up your point about "lehar" and "lahar" etc.
> I think the very fact that people largely use "e" in such
> words (rather than "a") should suggest that "e" represents the
> most acceptable pronunciation for most people. Maybe the
> letter 'e' can be replaced in such words by the two letters
> "ai". But the single letter "a" connotes a different sound.
> I am sure you remember the Indian TV adaptation of the
> Jungle Book (that was telecast on Doordarshan some 10-12 years
> back). The title song (probably penned by Gulzar) contained a
> line : "chaDDi pahan ke phool khila hai". In ordinary use,
> the word is pronounced as "peh'n". But Gulzar deliberately
> used a diffeent (colloquial or 'dialectic') pronunciation.
> Maybe someone like Shri Laloo Prasad Yadav would say " "Hum to
> bhai, dhoti pahan ke hi amreeka jaayeNge".
> I have already dealt with para 3 of your post. If this debate
> continues, I suggest that it should be done under a different
> thread. We need not invlove Shri Kulshrestha and his poetry
> unnecessarily. I would suggest that your post and my two
> rejoinders can be included as the starting point of a new
> thread.
> With best regards,
> Afzal
Distinguished Members of the Group, aadaab 'arz hai,
This is my first contribution to your group and I would seek
your forgiveness if I unwittingly breach any procedures or
etiquette.
The discussion about the pronunciation of words written as
"pahan", "shahr","qahr" etc pronounced as "paihan", "shaihr" etc is an
interesting one. From my reading of one or two sources, it appears that
there are a couple of factors behind this phenomenon:
1) The vowel, invariably follows the consonants "h" (Lahori)
and "H" (Halwe waali He) and 'ain . In addition to the above examples,
one can cite "kahnaa", "rahnaa" etc
"mahr"(dowry),"mihr"(kindness),"muhr"(seal)
"aHmad","muHammad","iHtiraam","i'tiraaf" etc.
These are pronounced ""maihr","mehr", "mohr", "aiHmad",
"MoHammad","eHtiraam", "e'tiraaf" etc. It appears that the presence of
the h/H sound "forces" Urdu/Hindi?Farsi speakers to slightly elongate
and relax the original vowel. Afzal Sahib mentioned Shri Laloo Prasad
pronouncing "pahan" as it is written. I would like to add that, at
least Punjabis who are not necessarily affected by Urdu/Persian
pronunciation pronounce in the same fashion as Laloo Prasad Yadav! I
say this, because my grandmother used to call me by my middle name
"aHmad" and pronounced it as it is written.
2) There appears to be a vowel shift from "a" to "ai", "i" to
"e", "u" to "o".
Examples "Deevaan-i-GHaalib" is pronounced "deevaan-e-GHaalib", and
this "e" sounds almost if not totally the same as "baRii ye".
Roz-i-umed u beem" is pronounced "roz-e-umed o beem" (NB. Um(m)eed was
pronounced as umed!). This is how the transcription is in Steingass's
Persian dictionary is given. The implication of this is "waw-i-'taf"
was pronounced like a "pesh" sound and the "izaafat" was pronounced
like a "zer" sound.
In passing, could I please pose a question or two to members of ...
What book/books would people from the sub-continent have used, say in
Mirza GHalib's time to learn Farsi?
Afzal SaaHib, would it be possible for you to contact me via my e-mail
address as I need to ask you a few questions about Farsi? I would be
most grateful to you for this.
Best Regards,
Naseer
Peteborough, England.
--------------------------
Yogesh
to clarify a few points that I have tried to make. Please keep mind
that my comments are not a criticism of any individual or scheme,
rather a learning probe. I, for one, would be quite content, if all
post were in English, and the transliteration was used only for poetry.
But since some contributors prefer to use Urdu transliteration for the
text also, it would be easier and less time consuming to follow the
thoughts if there was greater consistency in the scheme.
I have no problem with anyone adopting a particular scheme or the use
of vowels as they choose. Nor am I asserting, although it would very be
desirable, that every one on this forum follow the same scheme. By
consistency I mean consistency within the post of a knowledgeable
writer. When I see two different sounds being represent by the same
symbol, I wonder how that is possible? Consulting the dictionaries only
confirms my difficulty. This is where I pointed out the conflicting use
of 'e' versus 'a'.
For example: is there any difference in the initial sounds of words
like 'kahaa', 'kahnaa', 'kahr', 'sahr', 'sahii'
etc. At least Usmani, in a new dictionary (2003 publication), does not
think so. And Muddah, in an older dictionary supports that. Therefore,
when we see among these words the use of 'e' in some and 'a' in
others, how are we to determine the correct sound for a given word.
Further, since the same writer may employ 'e' for words like
'ke', 'le' 'de' etc, we are not talking about a possible
mix up with 'ai' - which represents a much more elongated sound,
as in words like 'Khair', Khaif', 'Khaimah', etc.
Regards,
Yogesh
Naseer
For what my view is worth (as a new-comer), I do believe that it is
your regular contributors whose views are paramount. I feel it is
better to have a relatively consistent scheme to transliterate Urdu
than to have a "free for all" system. Having said this, unless all the
contributors follow one "school of thought", this is not going to be
possible. By school of thought, I mean a particular system in a well
known Urdu/Hindi language grammar book or say, the system devised by
Sarwar SaaHib. So, the aim of the contributors ought to be to strive
for vowel/consonant accuracy as far as they can.
I agree with you when you say " I, for one, would be quite content, if
all post were in English, and the transliteration was used only for
poetry". I am not used to writing Urdu in the "Roman" scheme and would
perhaps take a long time to write even a short paragraph. So, I too
would prefer to write in English. On the other hand, as this IS an
Urdu group, contributors may rightly want to write in Urdu, albeit in
transliteration. The best solution would be to type everything in the
"Inpage"'s "noori nasta'leeq" or any other desirable font, if this
becomes a reality in the future.
Now coming to my previous post. I was merely attempting to describe the
pronunciation shift that seems to have taken place. I was not
suggesting that the transliteration should reflect this change.The Urdu
and Devanaagri writing system reflects the pronunciation which was once
current. It might be better to follow Urdu/Hindi writing system (which
I know is identical as far as words like "pahan" etc are concerned).
Spelling conventions do not always reflect the change in pronunciation.
One example that I can readily think of is the American English
spelling change, which is an exception, rather than the rule.
You ask "is there any difference in the initial sounds of wordslike
'kahaa', 'kahnaa', 'kahr',
'sahr', 'sahii' etc?". Then you go on to quote two lexicons which
state that there is no difference. I believe that the "a" vowel in
"kahaa" and "sahii" is the same but Not the same as the "a" vowel in
"kahnaa", "qahr", and "saHr". In the former the "h" follows a consonant
and in the latter it follows a long vowel. In the last three examples,
it is nearer to, as Afzal SaaHib has suggested, to an "ai" sound. I
agree, transliteration with "ai" is going to confuse the issue as the
same "ai" combination is used to represent the "diphthong" in words
like "KHair" etc. So again, I would say, let's stick to the "pahan"
version knowing that it does not rhyme with "lagan"!
Aap sab kaa KHair andesh,
Naseer
Afzal A. Khan
Sethi Saheb,
There are two issues here. One relates to posts being made in our
Newsgroup. The second relates to how certain letters/words are to
be pronounced. I think we need to make a clear distinction between
the two. My own opinion is that only the first issue (as identified
above) is relevant here.
Let me start by quoting your thoughts here :
"I, for one, would be quite content, if all
posts were in English, and the transliteration was used only for
poetry."
I have absolutely no quarrel with this proposition. And it would
help resolve the "problems/difficulties" quite conveniently. BTW,
I do not think that there is any problem or difficulty at all,
so long as we understand what each one of us is trying to say.
It is no doubt true that some ALUPers write their posts or
messages in the Urdu LANGUAGE. I have deliberately written the
last letter in capitals, for the sake of emphasis. It is, of
course, their preference. And I have no quarrel with that too.
IMHO, the problem (if we can use that term) arises only when there
is a request/suggestion/demand for any sort of consistency or
uniformity in the way our ALUP postings are "scripted".
It is my humble suggestion that such a consistency
or uniformity is very difficult to achieve. And any prolonged
discussion or argument can only exacerbate the issue.
I think mention was made of iTrans and I dealt with it in a
light-hearted manner in my earlier message. I think Sethi
Saheb and UVR Saheb should both be familiar with what I have
had to say about iTrans in the pages of RMIM. So, without
dilating further on this "script", I think we can quietly
drop it from our discussions.
I also made a reference to a very elaborate system devised by
Sarwar Alam Saheb. As far as I know, he is the only ALUPer
who uses that system. That system has certain merits, no
doubt, and meets the "demands" for "consistency" almost fully.
At the same time, I had also referred to the principle of
"ease of use". The very fact that Raz Saheb is perhaps the
only one to use the said system would indicate that most other
members of our Group do not find it convenient. And if people
are not willing (or otherwise in a position) to use it, it is
futile to pursue this "rasm~ul KHat" any further.
Now we come to the other issue -- of pronunciation and spellings.
As mentioned earlier, I don't think it is quite relevant here,
except in an academic sense. The fact must be acknowledged
that Urdu/Hindustani/hindi speakers in the sub-continent come
from different areas having their own specific phonetic "idio-
syncracies". And let us not think here only of the semi-literate
rural masses. Even eminent men of letters seemingly have these
"idiosyncracies". I am given to understand that a man of almost
universally renowned stature like Allama Iqbal had a rather
"TheTh" pronunciation. In friendly conversations, he was apt
to speak of "hak" rather than "haq". If you watch any Indian
TV serials, which have a Punjabi slant (like "Shanno Ki Shaadi"),
you would know what I am talking about. Comedians in Indian
films have carved out their careers trying to portray people
with such quirks. If a guy in "Padosan" can't quite pronounce
"Qais", Kishore Kumar makes fun of him : "Kya gas nikaali hai".
People from Hyderabad pronounce "q" as "KH". South Indian
folks (read that as Tamilians) have been a favourite butt of such
comedy. This was, of course, a most unfortunate trend but,
happily, that phase is sort of over.
These phonetic differences, however, seem to disappear when one
reads, say, an Urdu book --- that is to say, a book written and
printed in the Urdu script. For instance, the word "peh~n" may
occur in the book. I may pronounce it in this particular way.
Shri L.P. Yadav may of course pronounce it in his own way.
But the point is : every reader knows exactly what he is reading,
irrespective of his pronunciation "style".
It is possible, though, that an Urdu font (like "Inpage" -
Noori Nasta'leeq etc.) may be developed in the not too distant
future and adapted for use in computer typing. And, when that
happens, all of us will be benefited. Many ALUPers must have read
Sarwar Saheb's recent Review of the book "Sehra Sehra" -- it is
available on his website. Reading the Review is a beautiful expe-
rience --- and I am not referring merely to its contents. Visually,
it is so very satisfying. I am looking forward to the day when a
corresponding technology would be available for Newsgroup postings
(through computer typing).
For the time being, therefore, I would request you to please post
your queries about specific words (and their pronunciation) as
a separate academic issue, without linking it to ALUP postings
generally or consistency of script etc.
Afzal
Naseer
It appears to me, reading Sethi SaaHib and your posts, that there is
some thing about which, I as a new-comer, am totally oblivious of.
Please forgive me, if I have said anything out of place. I think, this
discussion can be brought to an end if all the contributors accept that
absolute uniformity in transcription is an impossibility and aim to be
tolerant in everyone's efforts.
There is a book called "A Manual of Classical Persian Prosody with
chapters on Urdu, Karakhanidic and Ottomon prosody" by professor Finn
Thiesen, published in 1982 by Otto Harrassowitz.wiesbaden. From pages
182-185, it describes the Urdu vowel system and very clearly goes into
the discussion we have been having on words like "lahr" and "pahan". I
would have posted the details here, but I don't have the fonts which he
uses in the book to transcribe the vowels.
On a final point, I would be most grateful to you if you could contact
me on my e-mail address. I would like to pick your brains on a topic I
believe you may have extensive knowledge about.
With Best Regards,
Naseer
UVR
First, welcome to ALUP. I don't believe anyone did the honors of
extending you a formal welcome; it's my pleasure to be able to do
so on behalf of all of us "regulars".
Your post below was addressed to Afzal sb, but I hope you will
not take it amiss if I respond with some points of my own.
Naseer wrote:
> Afzal SaaHib, aadaab
>
> It appears to me, reading Sethi SaaHib and your posts, that there is
> some thing about which, I as a new-comer, am totally oblivious of.
> Please forgive me, if I have said anything out of place. I think, this
> discussion can be brought to an end if all the contributors accept that
> absolute uniformity in transcription is an impossibility and aim to be
> tolerant in everyone's efforts.
It is not an "impossibility".
However, strictly for the purposes of informal discussions such as
the ones we conduct on ALUP, I think (and agree with Afzal sb
and you when you say) that it is not quite necessary for every person
to transcribe Urdu using the same transliteration scheme. I think
that as long as one is able to convey what one wants to convey in
a manner that others understand, it is hardly important whether
one writes "kahnaa" or "kehnaa" or even whether one write one or
the other of these consistently in all one's posts (or even the same
post). One must be free to follow whatever transliteration scheme
one wishes to follow.
But at the same time, I also feel that there are limits to how far
this 'freedom' should be stretchable. For example, it is perhaps
equally acceptable to write "Ghazal" and "ghazal", or "ishq" and
"'ishq" (with the ' indicative of the 'ain), but should we consider
"gajal" or "ishk" also acceptable? I do not think so. So, "freedom
with restrictions" is what we should follow here. Yes, it's an
oxymoron, but it's not different from what we all practice daily.
One more thing. None of the above has any bearing on my
opinion about whether or not a uniform transliteration scheme
has its merits and/or uses in a general sense. In connection
with THAT, I certainly agree with Yogesh sb. An Urdu-English
dictionary, for example, *must* use a consistent transliteration
system throughout. Infact, while it's not a necessity for ALL
Urdu-English dictionaries to use the same or even similar
transliteration schemes, I can't help but feel that if they did,
it would be immensely useful to the dictionary-using public.
Oh, and one final thing: I too don't have any problem at all
with people writing posts predominantly in English. OR in
Urdu. Or in a mixture of these two. This is not a school, a
mehfil or a conference session. It's more like a chaupaal
or a mohalla (with emphasis placed on the 'halla' :-P).
Regards,
-UVR.
Naseer
I broadly agree with your sentiments, although I did say "absolute
conformity". That I believe can only be achieved by writing in the
script of the language in question. Having said that, perhaps, even
that may not be a possibility. Urdu writers sometimes write "lie" and
other times "liye". I am sure one could think of more examples. Even in
Devanaagri, as you know, there are "variations on the the theme". "a"
has two forms. The consonants "jha", and the consonant after "dha",
which I don't know how to transcribe with my keyboard, have two forms.
"la" has two forms and so on.
And I also agree that people should be free to write without too many
constraints ( while aiming to write as accurately as possible) but
there are certain "bounds" and gajal and ishk would be problematic.
Lexicons should set high standards, I accept. But I can not remember
where the discussion on dictionaries came in.
Aap kaa KHair andesh,
Naseer