Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Learning Arabic using Latin Alphabet

652 views
Skip to first unread message

zuhair

unread,
Aug 14, 2010, 11:39:12 AM8/14/10
to
Hi all,

The following is a link to my page on learning Arabic using Latin
alphabet.

http://sites.google.com/site/zaljohar/

It is about an a transliteration system from Arabic to Latin. I do
think it might offer great help in learning Arabic.

Adios

Johannes Patruus

unread,
Aug 14, 2010, 1:46:07 PM8/14/10
to

The main difficulty I have with your system is that it is overly
complicated, and in explaining it you employ non-standard terminology
("tunes", "grands", &c.).

Some decades ago, I devised a transliteration system of my own, an example
of which you can see here:
http://www.aliquis.plus.com/NG/q001_interlinear.png

(This particular example has the additional feature of an interlineated
translation in Latin (that is to say, the Latin language).)

Patruus

zuhair

unread,
Aug 15, 2010, 12:45:32 AM8/15/10
to

Nice, I shall examine it further to learn from it. I just took a
glance at it, there are some problems with your system as regards my
objectives. I wanted a transliteration that only utilize the ordinary
keyboard characters in English, no formatted characters are allowed.
That's why my system appears complex a little bit, but it can be
written at chats, everywhere, since it doesn't involve any formatted
characters in it. (I wrote that in my introduction).

The other matter is that your system do not copy the traditional ways
of writing in Arabic, which render it awkward medium for teaching
Arabic way of writing. Example: you use One symbol for the long vowel
"kesreht ya~"), while Arabs don't use that, however to say the truth
in my system I did that with the case of fet-hhehtalif ,but this is
because I didn't see an example of an alif preceeded by other than fet-
hhet, and I think it is a reasonable rendering, but regarding the
other long vowels I think one must copy the traditional Arabic way of
writing them using combination of "diacritic" and primary letter.

Your system doesn't have the connecting and cutting hemzeht, it
doesn't clearly shows the sun and moon letters, etc.., although I know
what you are doing really, but yet these are traditional styles of
Arabic writing and one must copy them, if he want the transliteration
system to be a teaching system which was my goal.

Although I agree in principle to a lot of what is in your system, and
it is indeed nice, but it doesn't copy Arabic writing style.

As regards my non standard terminology, it doesn't really matter, the
meaning is their. But I agree with you, it render matters a little bit
difficult.

However as I said, I shall take a more detailed look at your nice
system, to learn from it.

Thanks

Zuhair

Johannes Patruus

unread,
Aug 15, 2010, 5:09:11 AM8/15/10
to
zuhair wrote:
> On Aug 14, 12:46 pm, Johannes Patruus <inva...@invalid.invalid> wrote:
>> zuhair wrote:
>>> Hi all,
>>> The following is a link to my page on learning Arabic using Latin
>>> alphabet.
>>> http://sites.google.com/site/zaljohar/
>>> It is about an a transliteration system from Arabic to Latin. I do
>>> think it might offer great help in learning Arabic.
>>> Adios
>> The main difficulty I have with your system is that it is overly
>> complicated, and in explaining it you employ non-standard terminology
>> ("tunes", "grands", &c.).
>>
>> Some decades ago, I devised a transliteration system of my own, an example
>> of which you can see here:
>> http://www.aliquis.plus.com/NG/q001_interlinear.png
>>
>> (This particular example has the additional feature of an interlineated
>> translation in Latin (that is to say, the Latin language).)
>>
>> Patruus
>
> Nice, I shall examine it further to learn from it. I just took a
> glance at it, there are some problems with your system as regards my
> objectives. I wanted a transliteration that only utilize the ordinary
> keyboard characters in English, no formatted characters are allowed.

I agree my system does not meet this objective. (It's just something I
knocked together for my own use.)

> That's why my system appears complex a little bit, but it can be
> written at chats, everywhere, since it doesn't involve any formatted
> characters in it. (I wrote that in my introduction).
>
> The other matter is that your system do not copy the traditional ways
> of writing in Arabic, which render it awkward medium for teaching
> Arabic way of writing. Example: you use One symbol for the long vowel
> "kesreht ya~"), while Arabs don't use that, however to say the truth
> in my system I did that with the case of fet-hhehtalif ,but this is
> because I didn't see an example of an alif preceeded by other than fet-
> hhet, and I think it is a reasonable rendering, but regarding the
> other long vowels I think one must copy the traditional Arabic way of
> writing them using combination of "diacritic" and primary letter.

I use only a, i and u for short vowels, ā, ī, and ū for long vowels, and
aw and ay for diphthongs. Alif maqsūrah shares the symbol ā. And I have a
personal dislike for using 'e' in transcribing Arabic.

> Your system doesn't have the connecting and cutting hemzeht,

Hamzat al-qat' is represented by an apostrophe except at the beginning of
a word, where it is present by implication. (Indeed even English words
whose first letter is a vowel may be considered as starting with a hamza.)
An example of medial hamza is su'āl (question).

The presence hamzat al-wasl may be indicated by italicising the initial
vowel (as shown in my surat al-Fatiha example).

Note the distinction between al-ism and al-'asmā', where the elision of
hamzat al-wasl in the former is indicated by omitting the apostrophe.

(This is a tricky area, and I doubt that I have been wholly consistent.)

> it doesn't clearly shows the sun and moon letters, etc.., although I know
> what you are doing really, but yet these are traditional styles of
> Arabic writing and one must copy them, if he want the transliteration
> system to be a teaching system which was my goal.

The only phonetic assimilations I allow are in respect of the definite
article. Thus I have preferred ar-raHmān to al-raHmān.

> Although I agree in principle to a lot of what is in your system, and
> it is indeed nice, but it doesn't copy Arabic writing style.

Correct, but a system that reproduced every nuance of the Arabic writing
system would be impractical.

> As regards my non standard terminology, it doesn't really matter,

It's off-putting to the student.

And it frightens the horses!

> the meaning is their. But I agree with you, it render matters a little bit
> difficult.
>
> However as I said, I shall take a more detailed look at your nice
> system, to learn from it.

Thank you.

With regard to consonants, I prefer to avoid the conventional digraphs
(dh, gh, kh, etc.) as these give rise to ambiguities and messy horrors
like "madhhab". Among the expedients I use are capital letters for the
'emphatic' consonants (D,S,T, &c) and 'x' in place of the customary 'kh'.
I accept all of this as the lesser of two evils.

> Thanks

Good luck with your endeavour!

> Zuhair

Patruus

zuhair

unread,
Aug 15, 2010, 12:40:42 PM8/15/10
to
> I use only a, i and u for short vowels, â, î, and û for long vowels, and
> aw and ay for diphthongs. Alif maqsûrah shares the symbol â. And I have a

> personal dislike for using 'e' in transcribing Arabic.

Thanks for the clarification, yet I actually noticed that, it is a
good idea really, regarding alif maqsuoreht we do share common
grounds, although in my writing I have advanced style which is
rigorous and has these distinctions, but for the beginners they don't
need it.

The problem with symolizing long vowels by one letter is that they do
change with simple addition of a diacritic to them.

Example: huo which you wright as : hû , now you just add the
diacritic "a" which is "e" = "`" if terminal in my system, then you
will have huo` , you see just a simple addition, however in your
system it will become as: huwa, you see a lot of changes for just a
simple addition.

>
> > Your system doesn't have the connecting and cutting hemzeht,
>
> Hamzat al-qat' is represented by an apostrophe except at the beginning of
> a word, where it is present by implication. (Indeed even English words
> whose first letter is a vowel may be considered as starting with a hamza.)

> An example of medial hamza is su'âl (question).


>
> The presence hamzat al-wasl may be indicated by italicising the initial
> vowel (as shown in my surat al-Fatiha example).

nice. though against my objective.
>
> Note the distinction between al-ism and al-'asmâ', where the elision of


> hamzat al-wasl in the former is indicated by omitting the apostrophe.

Ok, I learn from you here. that's a nice point. this is equivlanet to
showing teh hemzeht when it is cutting type in my advanced style (but
not for beginners), so we are more or less equivalent here.


>
> (This is a tricky area, and I doubt that I have been wholly consistent.)

I think its OK.


>
> > it doesn't clearly shows the sun and moon letters, etc.., although I know
> > what you are doing really, but yet these are traditional styles of
> > Arabic writing and one must copy them, if he want the transliteration
> > system to be a teaching system which was my goal.
>
> The only phonetic assimilations I allow are in respect of the definite

> article. Thus I have preferred ar-raHmân to al-raHmân.

ah, Ok, I see, departing from common arabic style. but also nice.


>
> > Although I agree in principle to a lot of what is in your system, and
> > it is indeed nice, but it doesn't copy Arabic writing style.
>
> Correct, but a system that reproduced every nuance of the Arabic writing
> system would be impractical.

Well this largely depend on your goal behind developing this system,
it your goal is to teach Arabic inclusive of the details of its
writing, then I guess you must do that.


>
> > As regards my non standard terminology, it doesn't really matter,
>
> It's off-putting to the student.
>
> And it frightens the horses!

yea.


>
> > the meaning is their. But I agree with you, it render matters a little bit
> > difficult.
>
> > However as I said, I shall take a more detailed look at your nice
> > system, to learn from it.
>
> Thank you.
>
> With regard to consonants, I prefer to avoid the conventional digraphs
> (dh, gh, kh, etc.) as these give rise to ambiguities and messy horrors
> like "madhhab". Among the expedients I use are capital letters for the
> 'emphatic' consonants (D,S,T, &c) and 'x' in place of the customary 'kh'.
> I accept all of this as the lesser of two evils.

(no x cannot replace kh, also Z cannot replace dhh, these are common
problems with transliteration).


One last word about digraphs and trigraphs, I don't like them either,
but I was forced to used them, since all alternatives are either using
formatted characters, or are using upper casing, the former is not
practical at all, the later is not aesthetic, the later is not only
not aesthetic, it is actually against the roles of English writing, an
English writer would not use upper casing to change the sound of the
character, and this is a common practice, I mean the common practice
is that upper casing will not change how the changed character would
sound, and it is dangerous to implement styles that go against common
practice, let me give you and obvious example, see how you yourself
write the first suoreht of the qur~an. (qur'ân) you wright it as Surat
al-Fatihah, look you used a capital S and capital F which you ought
not to do if your system have capitalization as a letter changer,
second point which is what I was talking about you wrote h while you
must have written H in capital according to your system you must have
written it as:

sûrat al-fâtiHah.

you see what I mean, never implement a rule that goes against common
practice, Old habits die hard! also this capitalization approach make
you loose
a good practice that Latin alphabet gives you free, which is marking
nouns, and beginning of sentences, etc..., which is a practice that is
not essentially forbidden in Arabic, truly Arabic do not have
capitalization in common practice, but their is no essential point
against using it, so why deprive it this merit that Latin is giving it
for free (I mean without much effort).

multigraphs are not desirable, but yet they are common practice in
English really, and even Arabic using combinations of a diacritic and
consonant to symblolize long vowels, so this is a rule in common
practice, and to me it is
the lesser of evils.


>
> > Thanks
>
> Good luck with your endeavour!
>
> > Zuhair
>
> Patruus

Your system is all about "reasonable elegant abbreviation", which is
in reality a good idea no doubt. But I think it is a little bit
advanced for a beginner really, your system can by used by advanced
learner or actually an Arab to shorten his Latin written Arabic, but
definitely not for a beginner who wish to know how Arabic is read and
written.

Thanks a lot really for your contribution, I will make benefit of this
system of yours, especially at arriving at some abbreviation styles.

Zuhair

zuhair

unread,
Aug 15, 2010, 2:08:56 PM8/15/10
to
On Aug 14, 12:46 pm, Johannes Patruus <inva...@invalid.invalid> wrote:

just a note, there is problem with the way you write aLLah, this
subject is a little bit sensitive to Muslims and Christians Arabs as
well. The way you wrote it is how it sounds which is not altogether
appropriate.

Zuhair

Johannes Patruus

unread,
Aug 15, 2010, 2:13:42 PM8/15/10
to
>> I use only a, i and u for short vowels, ā, ī, and ū for long vowels, and
>> aw and ay for diphthongs. Alif maqsūrah shares the symbol ā. And I have a

>> personal dislike for using 'e' in transcribing Arabic.
>
> Thanks for the clarification, yet I actually noticed that, it is a
> good idea really, regarding alif maqsuoreht we do share common
> grounds, although in my writing I have advanced style which is
> rigorous and has these distinctions, but for the beginners they don't
> need it.
>
> The problem with symolizing long vowels by one letter is that they do
> change with simple addition of a diacritic to them.
>
> Example: huo which you wright as : hū , now you just add the

> diacritic "a" which is "e" = "`" if terminal in my system, then you
> will have huo` , you see just a simple addition, however in your
> system it will become as: huwa, you see a lot of changes for just a
> simple addition.
>
>>> Your system doesn't have the connecting and cutting hemzeht,
>> Hamzat al-qat' is represented by an apostrophe except at the beginning of
>> a word, where it is present by implication. (Indeed even English words
>> whose first letter is a vowel may be considered as starting with a hamza.)
>> An example of medial hamza is su'āl (question).

>>
>> The presence hamzat al-wasl may be indicated by italicising the initial
>> vowel (as shown in my surat al-Fatiha example).
>
> nice. though against my objective.
>> Note the distinction between al-ism and al-'asmā', where the elision of

>> hamzat al-wasl in the former is indicated by omitting the apostrophe.
>
> Ok, I learn from you here. that's a nice point. this is equivlanet to
> showing teh hemzeht when it is cutting type in my advanced style (but
> not for beginners), so we are more or less equivalent here.
>> (This is a tricky area, and I doubt that I have been wholly consistent.)
>
> I think its OK.
>>> it doesn't clearly shows the sun and moon letters, etc.., although I know
>>> what you are doing really, but yet these are traditional styles of
>>> Arabic writing and one must copy them, if he want the transliteration
>>> system to be a teaching system which was my goal.
>> The only phonetic assimilations I allow are in respect of the definite
>> article. Thus I have preferred ar-raHmān to al-raHmān.

>
> ah, Ok, I see, departing from common arabic style. but also nice.
>>> Although I agree in principle to a lot of what is in your system, and
>>> it is indeed nice, but it doesn't copy Arabic writing style.
>> Correct, but a system that reproduced every nuance of the Arabic writing
>> system would be impractical.
>
> Well this largely depend on your goal behind developing this system,
> it your goal is to teach Arabic inclusive of the details of its
> writing, then I guess you must do that.

The danger with this all-fuses-blown approach is that you end up with a
system which is more difficult to learn than Arabic script itself!

>>> As regards my non standard terminology, it doesn't really matter,
>> It's off-putting to the student.
>>
>> And it frightens the horses!
>
> yea.
>>> the meaning is their. But I agree with you, it render matters a little bit
>>> difficult.
>>> However as I said, I shall take a more detailed look at your nice
>>> system, to learn from it.
>> Thank you.
>>
>> With regard to consonants, I prefer to avoid the conventional digraphs
>> (dh, gh, kh, etc.) as these give rise to ambiguities and messy horrors
>> like "madhhab". Among the expedients I use are capital letters for the
>> 'emphatic' consonants (D,S,T, &c) and 'x' in place of the customary 'kh'.
>> I accept all of this as the lesser of two evils.
>
> (no x cannot replace kh, also Z cannot replace dhh, these are common
> problems with transliteration).

As a precedent for the use of x in place of kh, I offer you the first page
of the transcription table in A.S.Tritton's "Teach Yourself Arabic" -
http://www.aliquis.plus.com/NG/tritton.png

> One last word about digraphs and trigraphs, I don't like them either,
> but I was forced to used them, since all alternatives are either using
> formatted characters, or are using upper casing, the former is not
> practical at all, the later is not aesthetic, the later is not only
> not aesthetic, it is actually against the roles of English writing, an
> English writer would not use upper casing to change the sound of the
> character, and this is a common practice, I mean the common practice
> is that upper casing will not change how the changed character would
> sound, and it is dangerous to implement styles that go against common
> practice, let me give you and obvious example, see how you yourself

> write the first suoreht of the qur~an. (qur'ān) you wright it as Surat


> al-Fatihah, look you used a capital S and capital F which you ought
> not to do if your system have capitalization as a letter changer,
> second point which is what I was talking about you wrote h while you
> must have written H in capital according to your system you must have
> written it as:
>

> sūrat al-fātiHah.

You are correct. I have failed to exert as high a standard of rigour in
the title as in the text.

> you see what I mean, never implement a rule that goes against common
> practice, Old habits die hard! also this capitalization approach make
> you loose
> a good practice that Latin alphabet gives you free, which is marking
> nouns, and beginning of sentences, etc..., which is a practice that is
> not essentially forbidden in Arabic, truly Arabic do not have
> capitalization in common practice, but their is no essential point
> against using it, so why deprive it this merit that Latin is giving it
> for free (I mean without much effort).
>
> multigraphs are not desirable, but yet they are common practice in
> English really, and even Arabic using combinations of a diacritic and
> consonant to symblolize long vowels, so this is a rule in common
> practice, and to me it is
> the lesser of evils.

The problem of ambiguity remains. How, for instance, is the uninitiate to
know whether the dh of "idhā" is one consonant or two?

>>> Thanks
>> Good luck with your endeavour!
>>
>>> Zuhair
>> Patruus
>
> Your system is all about "reasonable elegant abbreviation", which is
> in reality a good idea no doubt. But I think it is a little bit
> advanced for a beginner really, your system can by used by advanced
> learner or actually an Arab to shorten his Latin written Arabic, but
> definitely not for a beginner who wish to know how Arabic is read and
> written.

The fact that you and I are working to different objectives, means that
our two end-products will naturally be different.

That's fine. That's how it should be.

> Thanks a lot really for your contribution, I will make benefit of this
> system of yours, especially at arriving at some abbreviation styles.

You're welcome.

> Zuhair

Patruus

Johannes Patruus

unread,
Aug 15, 2010, 2:48:47 PM8/15/10
to

Yes, what I have written is closer to a phonetic transcription than to a
strict transliteration. (The whole thing is an uneasy compromise between
these two approaches.)

I have just checked my version of the Throne verse, where "Allah" is the
first word, so there is no elision of the initial A -
http://www.aliquis.plus.com/NG/kursi.png

(BTW, those weird superscripted symbols represent the signs of pause (I
forget what they're called in Arabic).)

> Zuhair

Patruus

zuhair

unread,
Aug 15, 2010, 3:36:28 PM8/15/10
to

I will wright it in my way: alfarizeht , in your way it is alfarizah
(with ^ in the second a), sometimes also called " alfawasil" with ^ on
the s, in my system
alfeoassil. pleural of alfassileht.

>
> > Zuhair
>
> Patruus

zuhair

unread,
Aug 15, 2010, 3:50:38 PM8/15/10
to
> >> I use only a, i and u for short vowels, â, î, and û for long vowels, and
> >> aw and ay for diphthongs. Alif maqsûrah shares the symbol â. And I have a

> >> personal dislike for using 'e' in transcribing Arabic.
>
> > Thanks for the clarification, yet I actually noticed that, it is a
> > good idea really, regarding alif maqsuoreht we do share common
> > grounds, although in my writing I have advanced style which is
> > rigorous and has these distinctions, but for the beginners they don't
> > need it.
>
> > The problem with symolizing long vowels by one letter is that they do
> > change with simple addition of a diacritic to them.
>
> > Example: huo  which you wright as : hû , now you just add the

> > diacritic "a" which is "e"  = "`" if terminal in my system, then you
> > will have  huo` , you see just a simple addition, however in your
> > system it will become as: huwa, you see a lot of changes for just a
> > simple addition.
>
> >>> Your system doesn't have the connecting and cutting hemzeht,
> >> Hamzat al-qat' is represented by an apostrophe except at the beginning of
> >> a word, where it is present by implication. (Indeed even English words
> >> whose first letter is a vowel may be considered as starting with a hamza.)
> >> An example of medial hamza is su'âl (question).

>
> >> The presence hamzat al-wasl may be indicated by italicising the initial
> >> vowel (as shown in my surat al-Fatiha example).
>
> > nice. though against my objective.
> >> Note the distinction between al-ism and al-'asmâ', where the elision of

> >> hamzat al-wasl in the former is indicated by omitting the apostrophe.
>
> > Ok, I learn from you here. that's a nice point. this is equivlanet to
> > showing teh hemzeht when it is cutting type in my advanced style (but
> > not for beginners), so we are more or less equivalent here.
> >> (This is a tricky area, and I doubt that I have been wholly consistent.)
>
> > I think its OK.
> >>> it doesn't clearly shows the sun and moon letters, etc.., although I know
> >>> what you are doing really, but yet these are traditional styles of
> >>> Arabic writing and one must copy them, if he want the transliteration
> >>> system to be a teaching system which was my goal.
> >> The only phonetic assimilations I allow are in respect of the definite
> >> article. Thus I have preferred ar-raHmân to al-raHmân.

>
> > ah, Ok, I see, departing from common arabic style. but also nice.
> >>> Although I agree in principle to a lot of what is in your system, and
> >>> it is indeed nice, but it doesn't copy Arabic writing style.
> >> Correct, but a system that reproduced every nuance of the Arabic writing
> >> system would be impractical.
>
> > Well this largely depend on your goal behind developing this system,
> > it your goal is to teach Arabic inclusive of the details of its
> > writing, then I guess you must do that.
>
> The danger with this all-fuses-blown approach is that you end up with a
> system which is more difficult to learn than Arabic script itself!

hmmm..., yea you might be right, but still I am of the impression that
it would be
helpful to the beginners, once they master it, they can convert it to
Arabic straightforwards without much difficulty. I still think that
learning a new alphabet that is written from right to left with
strange characters to Latin people, I still think it is more difficult
than learning this script.

Even to Arabs it is useful for official purposes, I actually prefer
it to the standard way adopted by the UN, since that one use formatted
characters, which is a big headache. The one with the Capitals is
totally messy, imagine you write your name in a passport in English
with initials that are not capitalized, or imagine having capitals in
the middle of your name, too much confusion ha, or imagine you must
fill a form in English which insist that it must be filled in with in
block letters (i.e. upper case), what that system with capitals would
do then? big problem ha!

> > write the first suoreht of the qur~an. (qur'ân) you wright it as Surat


> > al-Fatihah, look you used a capital S and capital F which you ought
> > not to do if your system have capitalization as a letter changer,
> > second point which is what I was talking about you wrote h while you
> > must have written H in capital according to your system you must have
> > written it as:
>

> > sûrat al-fâtiHah.


>
> You are correct. I have failed to exert as high a standard of rigour in
> the title as in the text.
>
> > you see what I mean, never implement a rule that goes against common
> > practice, Old habits die hard!  also this capitalization approach make
> > you loose
> > a good practice that Latin alphabet gives you free, which is marking
> > nouns, and beginning of sentences, etc..., which is a practice that is
> > not essentially forbidden in Arabic, truly Arabic do not have
> > capitalization in common practice, but their is no essential point
> > against using it, so why deprive it this merit that Latin is giving it
> > for free (I mean without much effort).
>
> > multigraphs are not desirable, but yet they are common practice in
> > English really, and even Arabic using combinations of a diacritic and
> > consonant to symblolize long vowels, so this is a rule in common
> > practice, and to me it is
> > the lesser of evils.
>
> The problem of ambiguity remains. How, for instance, is the uninitiate to

> know whether the dh of "idhâ" is one consonant or two?

In my style I show alsukuon in that case, I explained that in my
Notes.

if they are two consonants you must wright it as id-hâ, if they are
one
consonant then it is idhâ.

with alsukuon playing this rule, I think the problem of ambiguity is
resolved.

zuhair

unread,
Aug 15, 2010, 3:55:24 PM8/15/10
to

Not only that, I forgot to mention that the customary way of Arabic
writing ignore the diacritics, which makes it very difficult for a
beginner from any language to understand it. Now even if we make a
website in Arabic alphabet with these diacritics in it, it would
really seem hard for a native from another language to react to it,
since this diacritic way is too crowded, that is beside he is not used
to that kind of script. so after all still my transliteration system
would be easier than all of that headache, but definitely it would
need some practice, no doubt.

> ...
>
> read more »

zuhair

unread,
Aug 15, 2010, 4:11:29 PM8/15/10
to

Furthermore, I actually mentioned the idea of incorporating the
diacritics as letters in my website, if you scroll down near the end
of the site, you'll see Arabic letters, I personally constructed
letters for diacritics, and gave an example of writing my name with
them incorporated in it. Also this Arabic font of mine is not a
continuous font, and the characters have the same shape despite their
location in the word (unlike the traditional Arabic fonts), this would
be a font that is easier for beginners to read and react with,since it
is generally similar to the English fonts, but unfortunately it is
only exists in this web-page, nobody developed such an approach to
simplify Arabic to beginner that are natives of other languages. And
still I think this font is for intermediate to Advanced level learners
only.

Zuhair

zuhair

unread,
Aug 15, 2010, 4:35:17 PM8/15/10
to
> >> I use only a, i and u for short vowels, â, î, and û for long vowels, and
> >> aw and ay for diphthongs. Alif maqsûrah shares the symbol â. And I have a

> >> personal dislike for using 'e' in transcribing Arabic.
>
> > Thanks for the clarification, yet I actually noticed that, it is a
> > good idea really, regarding alif maqsuoreht we do share common
> > grounds, although in my writing I have advanced style which is
> > rigorous and has these distinctions, but for the beginners they don't
> > need it.
>
> > The problem with symolizing long vowels by one letter is that they do
> > change with simple addition of a diacritic to them.
>
> > Example: huo  which you wright as : hû , now you just add the

> > diacritic "a" which is "e"  = "`" if terminal in my system, then you
> > will have  huo` , you see just a simple addition, however in your
> > system it will become as: huwa, you see a lot of changes for just a
> > simple addition.
>
> >>> Your system doesn't have the connecting and cutting hemzeht,
> >> Hamzat al-qat' is represented by an apostrophe except at the beginning of
> >> a word, where it is present by implication. (Indeed even English words
> >> whose first letter is a vowel may be considered as starting with a hamza.)
> >> An example of medial hamza is su'âl (question).

>
> >> The presence hamzat al-wasl may be indicated by italicising the initial
> >> vowel (as shown in my surat al-Fatiha example).
>
> > nice. though against my objective.
> >> Note the distinction between al-ism and al-'asmâ', where the elision of

> >> hamzat al-wasl in the former is indicated by omitting the apostrophe.
>
> > Ok, I learn from you here. that's a nice point. this is equivlanet to
> > showing teh hemzeht when it is cutting type in my advanced style (but
> > not for beginners), so we are more or less equivalent here.
> >> (This is a tricky area, and I doubt that I have been wholly consistent.)
>
> > I think its OK.
> >>> it doesn't clearly shows the sun and moon letters, etc.., although I know
> >>> what you are doing really, but yet these are traditional styles of
> >>> Arabic writing and one must copy them, if he want the transliteration
> >>> system to be a teaching system which was my goal.
> >> The only phonetic assimilations I allow are in respect of the definite
> >> article. Thus I have preferred ar-raHmân to al-raHmân.

Thanks for the reference. I know that, but I don't like it, for
example
how would you wright khalid, you suggest Xalid (with ^ on top of the
a)
I don't think this is appropriate at all.

Zuhair


>
>
>
> > One last word about digraphs and trigraphs, I don't like them either,
> > but I was forced to used them, since all alternatives are either using
> > formatted characters, or are using upper casing, the former is not
> > practical at all, the later is not aesthetic, the later is not only
> > not aesthetic, it is actually against the roles of English writing, an
> > English writer would not use upper casing to change the sound of the
> > character, and this is a common practice, I mean the common practice
> > is that upper casing will not change how the changed character would
> > sound, and it is dangerous to implement styles that go against common
> > practice, let me give you and obvious example, see how you yourself

> > write the first suoreht of the qur~an. (qur'ân) you wright it as Surat


> > al-Fatihah, look you used a capital S and capital F which you ought
> > not to do if your system have capitalization as a letter changer,
> > second point which is what I was talking about you wrote h while you
> > must have written H in capital according to your system you must have
> > written it as:
>

> > sûrat al-fâtiHah.


>
> You are correct. I have failed to exert as high a standard of rigour in
> the title as in the text.
>
> > you see what I mean, never implement a rule that goes against common
> > practice, Old habits die hard!  also this capitalization approach make
> > you loose
> > a good practice that Latin alphabet gives you free, which is marking
> > nouns, and beginning of sentences, etc..., which is a practice that is
> > not essentially forbidden in Arabic, truly Arabic do not have
> > capitalization in common practice, but their is no essential point
> > against using it, so why deprive it this merit that Latin is giving it
> > for free (I mean without much effort).
>
> > multigraphs are not desirable, but yet they are common practice in
> > English really, and even Arabic using combinations of a diacritic and
> > consonant to symblolize long vowels, so this is a rule in common
> > practice, and to me it is
> > the lesser of evils.
>
> The problem of ambiguity remains. How, for instance, is the uninitiate to

> know whether the dh of "idhâ" is one consonant or two?

Johannes Patruus

unread,
Aug 15, 2010, 4:45:21 PM8/15/10
to
>>>>>> I use only a, i and u for short vowels, ā, ī, and ū for long vowels, and
>>>>>> aw and ay for diphthongs. Alif maqsūrah shares the symbol ā. And I have a

>>>>>> personal dislike for using 'e' in transcribing Arabic.
>>>>> Thanks for the clarification, yet I actually noticed that, it is a
>>>>> good idea really, regarding alif maqsuoreht we do share common
>>>>> grounds, although in my writing I have advanced style which is
>>>>> rigorous and has these distinctions, but for the beginners they don't
>>>>> need it.
>>>>> The problem with symolizing long vowels by one letter is that they do
>>>>> change with simple addition of a diacritic to them.
>>>>> Example: huo which you wright as : hū , now you just add the

>>>>> diacritic "a" which is "e" = "`" if terminal in my system, then you
>>>>> will have huo` , you see just a simple addition, however in your
>>>>> system it will become as: huwa, you see a lot of changes for just a
>>>>> simple addition.
>>>>>>> Your system doesn't have the connecting and cutting hemzeht,
>>>>>> Hamzat al-qat' is represented by an apostrophe except at the beginning of
>>>>>> a word, where it is present by implication. (Indeed even English words
>>>>>> whose first letter is a vowel may be considered as starting with a hamza.)
>>>>>> An example of medial hamza is su'āl (question).

>>>>>> The presence hamzat al-wasl may be indicated by italicising the initial
>>>>>> vowel (as shown in my surat al-Fatiha example).
>>>>> nice. though against my objective.
>>>>>> Note the distinction between al-ism and al-'asmā', where the elision of

>>>>>> hamzat al-wasl in the former is indicated by omitting the apostrophe.
>>>>> Ok, I learn from you here. that's a nice point. this is equivlanet to
>>>>> showing teh hemzeht when it is cutting type in my advanced style (but
>>>>> not for beginners), so we are more or less equivalent here.
>>>>>> (This is a tricky area, and I doubt that I have been wholly consistent.)
>>>>> I think its OK.
>>>>>>> it doesn't clearly shows the sun and moon letters, etc.., although I know
>>>>>>> what you are doing really, but yet these are traditional styles of
>>>>>>> Arabic writing and one must copy them, if he want the transliteration
>>>>>>> system to be a teaching system which was my goal.
>>>>>> The only phonetic assimilations I allow are in respect of the definite
>>>>>> article. Thus I have preferred ar-raHmān to al-raHmān.

I guess you need to drum up some volunteers to try it out, otherwise we'll
never know how well it works in practice.

And I agree, of course, that the omission of the short vowel marks from
the vast majority of Arabic texts is a pain of the first magnitude for
non-Arabs.

Towards the end of Hayward & Nahmad's Grammar, they start progressively
dropping the short vowels in the Exercises so as to acclimatize the
student in easy stages. There's a link to the PDF on this page:
http://www.ghazali.org/arabic/

> Zuhair

Patruus

Johannes Patruus

unread,
Aug 15, 2010, 5:09:17 PM8/15/10
to
>>>> I use only a, i and u for short vowels, ā, ī, and ū for long vowels, and
>>>> aw and ay for diphthongs. Alif maqsūrah shares the symbol ā. And I have a

>>>> personal dislike for using 'e' in transcribing Arabic.
>>> Thanks for the clarification, yet I actually noticed that, it is a
>>> good idea really, regarding alif maqsuoreht we do share common
>>> grounds, although in my writing I have advanced style which is
>>> rigorous and has these distinctions, but for the beginners they don't
>>> need it.
>>> The problem with symolizing long vowels by one letter is that they do
>>> change with simple addition of a diacritic to them.
>>> Example: huo which you wright as : hū , now you just add the

>>> diacritic "a" which is "e" = "`" if terminal in my system, then you
>>> will have huo` , you see just a simple addition, however in your
>>> system it will become as: huwa, you see a lot of changes for just a
>>> simple addition.
>>>>> Your system doesn't have the connecting and cutting hemzeht,
>>>> Hamzat al-qat' is represented by an apostrophe except at the beginning of
>>>> a word, where it is present by implication. (Indeed even English words
>>>> whose first letter is a vowel may be considered as starting with a hamza.)
>>>> An example of medial hamza is su'āl (question).

>>>> The presence hamzat al-wasl may be indicated by italicising the initial
>>>> vowel (as shown in my surat al-Fatiha example).
>>> nice. though against my objective.
>>>> Note the distinction between al-ism and al-'asmā', where the elision of

>>>> hamzat al-wasl in the former is indicated by omitting the apostrophe.
>>> Ok, I learn from you here. that's a nice point. this is equivlanet to
>>> showing teh hemzeht when it is cutting type in my advanced style (but
>>> not for beginners), so we are more or less equivalent here.
>>>> (This is a tricky area, and I doubt that I have been wholly consistent.)
>>> I think its OK.
>>>>> it doesn't clearly shows the sun and moon letters, etc.., although I know
>>>>> what you are doing really, but yet these are traditional styles of
>>>>> Arabic writing and one must copy them, if he want the transliteration
>>>>> system to be a teaching system which was my goal.
>>>> The only phonetic assimilations I allow are in respect of the definite
>>>> article. Thus I have preferred ar-raHmān to al-raHmān.

Yes, xālid.

> I don't think this is appropriate at all.

You'll find several occurrences of "x" in the Throne verse link I gave before:
http://www.aliquis.plus.com/NG/kursi.png

"x" is also used in this way by Kristina Nelson in her book -
http://www.amazon.com/dp/9774245946

I'm not trying to persuade you to use "x" yourself, but merely to validate
my own use of it.

> Zuhair

Patruus

zuhair

unread,
Aug 15, 2010, 5:34:04 PM8/15/10
to
> >>>> I use only a, i and u for short vowels, â, î, and û for long vowels, and
> >>>> aw and ay for diphthongs. Alif maqsûrah shares the symbol â. And I have a

> >>>> personal dislike for using 'e' in transcribing Arabic.
> >>> Thanks for the clarification, yet I actually noticed that, it is a
> >>> good idea really, regarding alif maqsuoreht we do share common
> >>> grounds, although in my writing I have advanced style which is
> >>> rigorous and has these distinctions, but for the beginners they don't
> >>> need it.
> >>> The problem with symolizing long vowels by one letter is that they do
> >>> change with simple addition of a diacritic to them.
> >>> Example: huo  which you wright as : hû , now you just add the

> >>> diacritic "a" which is "e"  = "`" if terminal in my system, then you
> >>> will have  huo` , you see just a simple addition, however in your
> >>> system it will become as: huwa, you see a lot of changes for just a
> >>> simple addition.
> >>>>> Your system doesn't have the connecting and cutting hemzeht,
> >>>> Hamzat al-qat' is represented by an apostrophe except at the beginning of
> >>>> a word, where it is present by implication. (Indeed even English words
> >>>> whose first letter is a vowel may be considered as starting with a hamza.)
> >>>> An example of medial hamza is su'âl (question).

> >>>> The presence hamzat al-wasl may be indicated by italicising the initial
> >>>> vowel (as shown in my surat al-Fatiha example).
> >>> nice. though against my objective.
> >>>> Note the distinction between al-ism and al-'asmâ', where the elision of

> >>>> hamzat al-wasl in the former is indicated by omitting the apostrophe.
> >>> Ok, I learn from you here. that's a nice point. this is equivlanet to
> >>> showing teh hemzeht when it is cutting type in my advanced style (but
> >>> not for beginners), so we are more or less equivalent here.
> >>>> (This is a tricky area, and I doubt that I have been wholly consistent.)
> >>> I think its OK.
> >>>>> it doesn't clearly shows the sun and moon letters, etc.., although I know
> >>>>> what you are doing really, but yet these are traditional styles of
> >>>>> Arabic writing and one must copy them, if he want the transliteration
> >>>>> system to be a teaching system which was my goal.
> >>>> The only phonetic assimilations I allow are in respect of the definite
> >>>> article. Thus I have preferred ar-raHmân to al-raHmân.
> Yes, xâlid.

>
> > I don't think this is appropriate at all.
>
> You'll find several occurrences of "x" in the Throne verse link I gave before:
>  http://www.aliquis.plus.com/NG/kursi.png
>
> "x" is also used in this way by Kristina Nelson in her book -
>  http://www.amazon.com/dp/9774245946
>
> I'm not trying to persuade you to use "x" yourself, but merely to validate
> my own use of it.
>
> > Zuhair
>
> Patruus

Thanks a lot for the references. Yea, this use of x, looks funny
though, however it become a necessity only when one want to avoid
combination letters, however I don't see really a great problem with
using combination letters. One day I was thinking of using x instead
of sâd , really. anyhow, I guess the real pain in my system is the
combination letters, I would be eager to listen to more bad points
about it, so that I develope it further. By the way how do you wright
these formatted characters like â.

Zuhair

Message has been deleted

zuhair

unread,
Aug 15, 2010, 7:05:30 PM8/15/10
to
On Aug 15, 3:45 pm, Johannes Patruus <inva...@invalid.invalid> wrote:

>
> I guess you need to drum up some volunteers to try it out, otherwise we'll
> never know how well it works in practice.

I agree, I don't know if anybody in this group would volunteer for
such a quest.

We'll I shall try.

Call for volunteers:

Subject: To test weather the system named "The Arabic Code" presented
at the following website can work in practice, or if it is very
complex that renders it impractical?

http://sites.google.com/site/zaljohar/

Zuhair

Johannes Patruus

unread,
Aug 16, 2010, 2:56:20 AM8/16/10
to
zuhair wrote:

> By the way how do you wright

> these formatted characters like ā.

By holding down the ALT key while entering the appropriate code on the
numerical keypad, then releasing the ALT key. Some of the codes are
tabulated in this posting -
http://groups.google.com/group/alt.language.latin/msg/197752f6292c34aa

> Zuhair

Patruus

zuhair

unread,
Aug 16, 2010, 7:47:20 AM8/16/10
to
On Aug 16, 1:56 am, Johannes Patruus <inva...@invalid.invalid> wrote:
> zuhair wrote:
> > By the way how do you wright
> > these formatted characters like â.

>
> By holding down the ALT key while entering the appropriate code on the
> numerical keypad, then releasing the ALT key. Some of the codes are
> tabulated in this posting -
>  http://groups.google.com/group/alt.language.latin/msg/197752f6292c34aa
>
> > Zuhair
>
> Patruus

Gee, that's nice. Thanks.

Zuhair

zuhair

unread,
Aug 16, 2010, 8:04:09 AM8/16/10
to
On Aug 16, 1:56 am, Johannes Patruus <inva...@invalid.invalid> wrote:
> zuhair wrote:
> > By the way how do you wright
> > these formatted characters like â.

>
> By holding down the ALT key while entering the appropriate code on the
> numerical keypad, then releasing the ALT key. Some of the codes are
> tabulated in this posting -
>  http://groups.google.com/group/alt.language.latin/msg/197752f6292c34aa
>
> > Zuhair
>
> Patruus

Thanks again. You mean the number pad the one on the corner of the
keyboard, I tried it, it works, its nice!! however it doesn't work in
Microsoft word and sometimes it gives a different symbol on Microsoft
Front page, for instance
ALT+131 will be the symbol given for male sex instead of â. However it
fairly work on Notepad, Word pad, Yahoo and Google mailbox, groups and
websites, which is fairly satisfactory.

Perhaps this will change my mind about those formatted characters.

Zuhair

Johannes Patruus

unread,
Aug 16, 2010, 8:39:34 AM8/16/10
to

In MS Word (at least in the traditional versions before they completed
redesigned (or fubar'd) the interface), you select "Symbol" from the
"Insert" menu, and from there you can insert any character you like and
assign it its own shortcut key.

BTW1 - The font I favour for transliterated Arabic is "Comic Sans MS"
(size 12), which commends itself rather well to my
several-degrees-less-than-20/20 vision.

BTW2 - Tips for "arabicizing" MS Word are to be found in Dr. Madhany's
article:
http://www.uga.edu/islam/arabic_windows.html

> Zuhair

Patruus

zuhair

unread,
Aug 16, 2010, 11:22:54 AM8/16/10
to

Thanks.

zuhair

unread,
Aug 16, 2010, 12:26:22 PM8/16/10
to
On Aug 16, 1:56 am, Johannes Patruus <inva...@invalid.invalid> wrote:
> zuhair wrote:
> > By the way how do you wright
> > these formatted characters like â.

>
> By holding down the ALT key while entering the appropriate code on the
> numerical keypad, then releasing the ALT key. Some of the codes are
> tabulated in this posting -
>  http://groups.google.com/group/alt.language.latin/msg/197752f6292c34aa
>
> > Zuhair
>
> Patruus

Still I am of the opinion, that writing combination letters is much
easier, and with alsukuon separating them to avoid confusion, the
system is as precise as using singlet letters, although one might fall
into errors by not putting alsukuon were it should be.

By the way my system really do not differentiate between the two types
of Hemzeht, this is left for advanced learners to figure out, there is
absolutely no need to involve beginners in like matters, and in
reality nowadays we Arabs are not anymore using the terminal
diacritics, and many words in everyday speach are terminated by sukuon
actually, in reality the use of terminal short vowels is a practice of
the past, and presently they are only used for official purposes, it
is not a common daily practice at all, and with this fading away,
discriminating these two types of hemzeht will actually have much less
importance than before.

I am still of the idea (I might be wrong of course) that the
transliteration that I've made is actually more practical for teaching
purposes to beginners, it is simple, and it does copy the essential
stuff of Arabic a beginner need to master.

This transliteration system of mine is a simplified one, it doesn't
not aim to wright Arabic in the most perfect way using Latin. Its' aim
is to wright Arabic in Latin in a handy manner without any formatted
characters, nor confusing Capitalization, it uses combination without
lose of precision, and it doesn't involve the beginner with Arabic
stuff that is irrelevant to beginners. I personally think it would
greatly facilitate teaching Arabic to people used to Latin alphabet,
and it serve as a very good initial introduction to Arabic,
furthermore it is readily convertible to Standard way of writing
Arabic (with the only exception of the hemzeht types). The other
systems I saw thus far are either clumsy or difficult to write, and
even to read.

Regards

Zuhair

Johannes Patruus

unread,
Aug 16, 2010, 1:04:00 PM8/16/10
to

May your endeavours prove successful and beneficial.

Patruus

zuhair

unread,
Aug 16, 2010, 2:16:09 PM8/16/10
to

~a!miyn

0 new messages