Talk:Arabic alphabet/Archive 1

From Wikipedia, the free encyclopedia
Jump to: navigation, search

For users needing assistance with Arabic script, please add requests at Wikipedia:WikiProject Arab world/Requests for Arabic script.

Talk:Arabic alphabet/from the French Wikipedia - temp page moved into talk namespace, as per policy.


Jeem in IPA

Currently the page says jeem can be pronounced either as ʒ or dʒ. However, shouldn't the dʒ be linking to Voiced palato-alveolar affricate instead of referring to d and ʒ separately? —Preceding unsigned comment added by (talk) 23:17, 11 March 2010 (UTC)

Map - "Worldwide use of the arabic script"

I interpret the second key on map to indicates countries where arabic script and adapted script, may be used (even if not official - eg- India and China are shown light green on map). The accompanying paragraph to map refers to Indonesia and Malaysia, but both omitted as light green on map. Article on Jawi script also says South Philippines, Sth Thailand, Brunei & Singapore may use this script. Jawi is official script in Brunei. ````

Arabic keyboards

In experimenting with the localization features of Windows XP, I see that there isn't a single "Arabic keyboard". For example, the "Arabic - Iraq" keyboard has ذ ّ in the key to the left of the Qwerty '1'/'!', while "Arabic - Saudi Arabia" has < >. There are other more important changes between those two such as the location of vowel marks (I don't know the right terminology) in the position of the uppercase Qwerty keys X C V.

I haven't experimented with the other keyboard layouts beside Iraq, Egypt and Saudi Arabia to see how many other national variants exist.

I'm not at all sure how to reflect that in the article. I made a small note in the article, but it deserves to be explained in more detail.

Bill Smith 18:30, 27 December 2006 (UTC)

Miscellaneous threads

This is a terrible article on what is arguably a very important subject. Its structure and language are both dense and pedantic, and concepts that require special clarification are inadequately developed. Too many subjects are discussed in tandem - there should be a much clearer demarcation of each example. As well, the arabic glyphs are currently in a format much too small to be comprehended. It is at present nearly opaque to the linguistic non-specialist, which is hardly appropriate for Wikipedia.

The above comment, I have to say, still seems extremely cogent to me. On the principal of Boldness, I've rewritten the introductory paragraphs, which I hope are now less repetitive-- but yes, the rest of the article could benefit from some work as well. Herbivore 19:23, 7 January 2007 (UTC)
Upon revisiting, I've rewritten the Structure section, in hopes of making the question of different letter forms clearer to the novice. Herbivore 03:02, 9 October 2007 (UTC)

Can anyone contribute to a list of languages commonly written in Arabic alphabet? I know Urdu is, and of course Arabic. I think the entry should also include a note that the Qur'an is written in the Arabic alphabet.

turkish was former written in arabic until some political reforms. --Elian
Didn't Ataturk decide that the Roman alphabet worked better with the Turkish language, than the Arabic alphabet does? -- Zoe
Seems to have been one of the practical reasons. Latin alphabet easier to learn etc. But there were also political reasons, orientation more toward europe and secularization. IMO the reform was mainly destined to mark a clear break with all (Islamic) traditions of the Ottomane Empire. --Elian
Due to the lack of notation for vowels (only 3 distinct vowels in arabic, see below) and the fact that short vowels are normally not noted, the arabic alphabet and conventions are not too well suited for languages where vowels play a more important role than in Arabic. (My 0.02 cents.) That may be one of the main non-political reasons. --FvdP

::: I should say that arabic is not restricted to the three vowels ( A,o,I) as many people would think, there are (Tashkeel) which is a vital element in the language (Dhamma, fat'haa, Kasrah), basically, arabic is easily adapted to any other language, especially, that there are new (not commonly used) letters, for instance, P can be written in arabic like the second letter in the arabic alphabet (Baa') except with three dots beneath instead of one. I disagree with the statement, that arabic is a difficult language, and isn't flexible. Once you cross the pronounciation barrier, then it is matter of some vocabulary and grammar, and you can speak the language fluently, it is just like any other language really, but to completely master the language,it'll require a lifetime, but we don't need to master a language in order to speak it, do we ??

Sort of, but the objection was spurious insofar as that could easily be remedied: look at its adaptation to Uighur or Kurdish! - Mustafaa 17:29, 15 Apr 2004 (UTC)
I addition to the possibility of modifying the orthography (as Mustafa mentioned), if a language really requires diacritics then they should be made obligatory for that language :) I think Greek ideally requires extensive use of diacritical marks, which are often omitted. No? --Alif 18:27, 13 July 2005 (UTC)
Modern Greek only requires an accent mark, and a diaresis to indicate diphthongs. I don't believe they're normally left out, except maybe in newspaper headlines. In Classical Greek, diacritics indicated aspiration and pitch accent. I'm not sure how good an analogy this is to Arabic vowels, though it's on the right track: you might create more confusion by mixing up Arabic vowels than mixing up Greek accent. kwami 01:08, 2005 July 14 (UTC)

I think the table of Unicode codes should go elsewhere, something like "Unicode codes from the Arabic alphabet". It clutters the article, not everyone will be interested in this technical matter.

I may do the move in a few days if no one protests.

FvdP 20:56 Oct 11, 2002 (UTC)

Seconded. I have limited internet access, though, and cannot keep a computer for long enough to do such a thing. Scott Paeth 05:27, 21 January 2007 (UTC)

I have started a merge of the Arabic examples from the German Wikipedia Karada

From the German version (crudely translated):

There are only three vowels: A, i and u. Long vowels are indicated by Alif, Ya and Waw. Short vowels marked by Fatha (A - diagonal line over the letter), Kasra (i - diagonal line under the letter) and Damma (u - a kind of small 9 over the letter). If no vowel follows, a Sukun (a small circle over the letter) is set.

Is this correct? Karada

Yes, however, this follows German pronounciation. a is spoken like in "thus", in most modern dialects more like "action" or the German "ä", i "this", u "look" or "roof", in dialect it resembles more an "o". Maybe confusing for English readers. --Elian
Yes, but short vowels and sukun are normally not marked at all. FvdP

You need to do the HTML Unicode in decimal, not hexadecimal. Using hexadecimal reduces even further the number of browsers on which it will work. --Zundark 21:17 Jan 5, 2003 (UTC)

Done.. --Gabbe 18:04 Jan 16, 2003 (UTC)

The French Wikipedia article really is much better than this one. Does anyone want to start the process of translating the French article to English? -- The Anome 10:52, 9 Aug 2003 (UTC)

See Arabic alphabet/from the French Wikipedia for a start on this translation process. Anyone want to translate a paragraph? -- The Anome 13:32, 9 Aug 2003 (UTC)

Can anyone tell me how to write articles using Arabic script? I have absolutely no idea on how to do it! ThaGrind 02:48, 21 Dec 2003 (UTC)

Huh? Please clarify. Jeru 17:18, 3 Apr 2004 (UTC)

Took the liberty to change ta to dal د in the vowels section for clarity. There was no reason to use ta, write it incorrectly and mislead the surfers. Jeru 17:18, 3 Apr 2004 (UTC)


There's a good list of languages using the Arabic script in the Arabic language article, for some reason. I'll see if I can find them a place here. - Mustafaa 18:00, 15 Apr 2004 (UTC)

"When the moors and arabs ruled Spain"

When the moors ruled Spain they ruled Spain and Portugal and there was no spanish language. People spoke Latin in several variations, possibly dialectal, all called "Vulgar Latin". The emergence and separation of the national languages in the Iberian Peninsula happened for the most part in the last 1000 years. -- 03:13, 24 Jul 2004 (UTC)

And the end of Moorish rule happened 500 years ago... - Mustafaa 09:37, 24 Jul 2004 (UTC)

Never mind that. The language spoken in Granada was not spanish in the current understanding of the concept, i.e., the language that developped in the kingdom of Castille, was later influenced by arabic when Castille expanded south and then spread to vast territories throughout the world. The languages spoken in Granada, and before that in the rest of the iberian territories under arabic rule, was in part arabic, especially in administration and religious issues, but mostly a romance language, very influenced by arabic, called "mozarabic". That is (was) a language on its own, different from all the languages currently spoken in the Iberian Peninsula. 04:27, 25 Jul 2004 (UTC)

Well, Mozarabic language was certainly used - but are you sure Spanish wasn't used as well towards the end? For instance, describes Aljamiado as "lengua castellana escrita con grafía árabe" (in 1462, that is). - Mustafaa 01:39, 29 Jul 2004 (UTC)

Hum... I forgot about aljamia. Aljamia was a rather marginal use of the arabic script to write not only castillan (and not spanish, mind you - it's the same thing nowdays, but the term is more recent than the aljamias), but also catalan/aragonese and portuguese amongst the moorish communities that remained in the iberian territories conquered by the christian kingdoms. So, OK, you could say that the arabic script was used to write castillan, but in that case the list is misleading: Not only the usage of the script to write mozarabic was far more important, but also there are other aljamias, not only the castillan one. Furthemore, it wasn't used "when the moors ruled Spain", but by the moors that remeined when they no longer ruled Spain, precisely the other way around (remember that the last moorish communities were expelled from the Peninsula (or converted by force, together with the jews) only in early XVII century). I still think it should be changed. -- 17:25, 8 Aug 2004 (UTC)

I've changed the entry to reflect this discussion. Let me know what you think. Were there any Portuguese aljamiado manuscripts? - Mustafaa 23:46, 9 Aug 2004 (UTC)

Well, to tell the truth I'm not sure, since different authors have different understandings of what are exactly aljamiado texts. I know that most of what's known as portuguese aljamía is written in safi, not in portuguese. I'm not sure about the existence or not of aljamia in portuguese. In any case, there's a book that might answer that question (Lopes, David - Textos de Aljamía Portuguesa, Imprensa Nacional, Lisboa, 1897), but I do not have access to a copy. The title translates as "Texts of Portuguese Aljamía", BTW.
Regarding the entry, you could add aragonese to spanish. Unlike portuguese, aragonese was definitely aljamiado... in fact, the earliest example of aljamía known is written in aragonese, according to what I've found. Other than that, it's fine. -- 01:34, 14 Aug 2004 (UTC)

Thanks! What was Safi? By the way, you know anyone can edit here - why not get a user account and start writing some articles yourself? You certainly seem knowledgeable on the subject, and it sounds interesting... - Mustafaa 08:05, 14 Aug 2004 (UTC)

Argh! What did I write above? I should have been sleeping! I totally misread my sources!
Safi is not a language - it's a town in western Morocco that was under portuguese rule for a time (until 1541). That (the place and the time) is where the examples of portuguese aljamía come from.
It is true, though, that different authors have different understandings of what aljamía really is and that it leads to some discussions about the validity of calling aljamiado to these texts. They really were written in portuguese, though. -- 15:21, 15 Aug 2004 (UTC)

--- Persian, Dari, Farsi are all the names for one language i.e. Persian language. In the list these terms were used superfulously, so I edited it. --Mani1 16:14, 18 Aug 2004 (UTC)

Persian and Farsi may mean the same thing, but Dari is NOT the same language as Farsi.

Aljamía was and still is in essence, the language of the Moorish descendants of Al Andalus, and thus it remains our language, the language of the Muslim mudejáres and moriscos!

(Al Fatiha)

(Çora l'alfátaha)-in aljamía

Nel nomme d'Allah, l'Arrahimo, l'Arrahimano.

1. Alhamdanzas ad Allah, arrabí dellos aílemos.

2. L'Arrahimo, l'Arrahimano.

3. Malico d'Alyamidino.

4. A bós t'albudamos, ed a bós anestainamos.

5. Alhedenáde-mos por l'alçerado almostaquino.

6. L'alçerado d'ellos que hádes almetado.

7. Non allos que amerraban. Ed allos albideros.

Xucrá, Ah'med Birzali!


hi can you show me the arabic alphebet then can you translate it for me? I have this really big project due on ancient islam tomorrow. this would really make my night a lot less hard than it already is. thank you alyssa

Image:Arabic alphabet.png seems to include the alphabet with corresponding transcriptions. I'm not sure what else you're looking for, but you could try asking at Wikipedia:Reference desk. Cheers, -- Hadal 05:18, 8 Oct 2004 (UTC)

Primary Letters

As a person who knows aboslutely nothing about Arabic, I find the 'Primary Letters' table to be quite useless. The text is too small to discern the characters, even when I view text in the largest size. Can something be done about this? RealGrouchy 17:13, 21 Nov 2004 (UTC)

I second this. Ideally, some helpful soul who knows Arabic would contribute images of the letters, because (as demonstrated by the Allah example) some browsers cannot be trusted to render the letters properly. Thanks. --Doradus 03:33, Dec 13, 2004 (UTC)
Something had already been done about this; the big image right up near the top. I would definitely not support removal of the Unicode; even if many browsers still can't handle it, it's the recognized standard, and in a few years they all will. - Mustafaa 17:06, 14 Dec 2004 (UTC)
But the big table only shows the letters in isolation, not joined with other letters. I'd be happy if all variants of all letters were included in the table. --Doradus 19:39, Dec 14, 2004 (UTC)
Ah. I see your point. - Mustafaa 11:07, 15 Dec 2004 (UTC)
All right, you guys. I checked all the Unicode characters myself, and they are all correctly inputted. However, my browser refuses to display them all (Internet Explorer 6.0). Since I do read, write and speak the language, I did not need clearer and larger images. When I have some time, I'll see what I can do about it. Cbdorsett 08:54, 14 Feb 2005 (UTC)
Some letter when they are in the initial position they will be isolated (hence they can never have Medial form).

For those 8 letters marked with no [Medial] or [Initial]: while they don't have a [Medial] form, they do have an [Initial] form which is the same as the [Isolated] form. Should we change the various tables to reflect that? Wael Ellithy (talk) 10:49, 23 July 2008 (UTC)

I can see how that is a bit confusing since any letter can come initially in a word, but the meaning, as I understand it, of the table is initial in a string of connected characters, not in a word. So, a letter that cannot connect to the following letter to the left cannot be initial in a connected string of characters, it can only be final (attached to the letter to the right), or isolated. Maybe that could be made clearer somehow, but not knowing much about it, I'll leave it alone for now.StephenHudson (talk) 18:13, 23 July 2008 (UTC)
That's correct. The conventional use of the terms "initial", "medial", and "final" here is relative to the position in a connected string, not to position in the word. —Largo Plazo (talk) 21:54, 23 July 2008 (UTC)

Wrong Unicode codepoints

Unlike some of the commenters above, I can see all Arabic letters in the text correctly (maybe because I'm using Mozilla SeaMonkey on Linux, and have installed a number of additional fonts, including some for Arabic, and maybe also because my favourite Wikipedia skin is Cologne Blue, which uses larger text than the default Monobook). All the shapes in the table are correct (except of course insofar as isolated and final shapes of non-joining letters are also listed respectively as "initial" and "medial"), however some of the letters are listed with identical Unicode codepoints numbers (for contextual shapes relating to different letters), which cannot be correct. I don't have the time to correct that section now, but the reference PDF pages are for common codepoints and for "Presentation forms B" (i.e., distinct codepoints for the various contextual shapes). — Tonymec (talk) 06:42, 16 May 2009 (UTC)

Transcription used on the net

The Romanization I see most often in Internet use, uses digits in addition to letters, for example 3 for `ayn and 7 for Haa', based on visual similarity of the characters. I'd like information on it. Wikipedia has information on similar ASCII keyboard-friendly Internet transliterations for other non-Roman alphabets, for example Greeklish and Volapuk_encoding.

I deleted the "addition" of the numeric symbols for two reasons:
  • There is no uniform agreement about which symbol should represent which Arabic letter. As far as I know, these numerals are only used in do-it-yourself language books.
They're used extensively online (DALnet, for example, join any Arabic-speaking channel), and there is a good deal of agreement on the most common ones: the ones I added. There's slight regional/national variation, like, some people use an 8 for qaf, but for the most part, 23679 are a de facto standard. — J’raxis 19:02, 2005 Feb 15 (UTC)
  • The editor took the opportunity to change some of the Unicode symbols from decimal to hexadecimal, which many browsers (including mine) do not support, and in the process evidently tried to change what was originally represented.
I didn't just change them to hexadecimal; what I was primarily doing was changing letters that had combining diacritics after them with single characters that included the diacritics. The combining diacritics looked like a mess in the browser I was using, some crooked, some not "combining" at all, so I thought, for example, using "H with dot below" was better than "H" followed by "dot below."
I only used the hexadecimal because that's what I had (U+nnnn entities) on the Unicode chart I was using at the time. I didn't know there were still some browsers unable to handle the hex entities so I didn't take the time to convert them to decimal.
I wonder if it would be a good idea for the MediaWiki code itself to decimalize entities either on input to the database (might confuse editors as what they typed in isn't what they got back) or at the very least upon output when the Wiki markup is parsed to HTML.
J’raxis 19:02, 2005 Feb 15 (UTC)
Great idea - how can I get involved in that? Cbdorsett 13:26, 16 Feb 2005 (UTC)
I'm not a MediaWiki coder, so I'm not sure, but here's Wikipedia's project site:
J’raxis 19:46, 2005 Feb 18 (UTC)
If the editor truly feels that the article can be improved by the addition of the numerical symbols, I suggest creating a page that deals with the various transliteration schemes, and adding an appropriate link.
I also rewrote the paragraphs about transliteration so that they are clearer. I disagree that the DIN standard should be used above all others in this article, as its symbols do not show up on Internet Explorer 6.0 (Microsoft's latest) and are not as accessible to readers who do not know the Arabic alphabet. Cbdorsett 09:35, 14 Feb 2005 (UTC)

Since the digit representations are widely used and standardized, they should go back in the table or text. The table currently has as many as 4 transliterations per Arabic letter, some of which are used much less frequently than the digits. Also, the digits are used online where it is not obvious how to find a guide to their meaning, unlike printed academic works which usually include notes about the transliteration they use, if there is any doubt that readers will understand. --JWB 21:13, 22 Jun 2005 (UTC)

The table also doesn't identify which transliterations of a letter are used by which transliteration system. DIN-31635 and SATTS are also stubs with no info. So effectively there is no info on transliteration systems at all, just a listing of the union of several systems, with emphasis on less used ones. --JWB 23:32, 23 Jun 2005 (UTC)

Edits by beginners

We all thank you for your new-found enthusiasm for the language, but please, please, please be sure of what you write, before you change something.

Someone just removed the dagger alif from 'allah' and replaced it with a fatha. A first-year student, perhaps, who has just learned about the diacritics for the short vowels?

If I get some time, I'll try to clarify the section about the diacritics - the 'harakat' - but in the meantime, PLEASE do not 'fix' anything on your own unless you KNOW you are right.

Thanks. Cbdorsett 15:31, 21 Feb 2005 (UTC)

Arabs Achtung! I totally share Cbdorsett's conscern, but I would like to generalize his message to include Arabs as well. Being an Arab doesn't mean that you are expert on the alphabet and/or the language. Hakeem.gadi (talk) 10:50, 27 March 2008 (UTC)

some notes by

The above is complete bullshit. "Alif maqsura" is a grammatical concept in the Arabic language; it is not a letter form. It can be represented by either a dotless ya (ى) or an alif. Furthermore, the dotless ya does not always mean "alif maqsura". The dotless ya is used in Arabic in all four forms. The problem, which can be traced to the stupidity of Unicode, is that a) "character" is not so obvious; and b) letterform shape is distinct from context. E.g. a so-called "initial form" can be used in the middle of a word: قائل for example.

[Well I can't say anything about where if came from or what it meant originally, but I can tell you that the alif letterform serves numerous functions *in the Arabic* writing system; other systems that use Arabic letterforms may do things differently. More to the point, it is *not* a consonant in modern Arabic (or in any Arabic so far as I can tell). Why not ask the locals? They will tell you that, where alif seems to indicate a long vowel, in fact it indicates the doubling of a short vowel, which is not quite the same thing. In other words, alif has *no primary phonological value* whatsoever, ever.]

  • If you don't delete this paragraph, I will. Cbdorsett 10:04, 6 Apr 2005 (UTC)

Vowel marks

I just went around lk-ing "vowel point", and in i think one case, "vowel mark" to Niqqud.

In this article the section "Ligatures" says

The latter is a work-around for the shortcomings of most text processors, which are incapable of displaying the correct vowel marks for the word "Allah". Compare the display below, which depends on your browser and installed fonts:

I made no lk to the Hebrew word Niqqud, but is there an equivalent lk available or needed? Should this instance of "vowel mark" lk to Abjad?
--Jerzy (t) 20:55, 2005 Mar 24 (UTC)

Yes - harakat. - Mustafaa 03:13, 25 Mar 2005 (UTC)

In the section "Vowels" is the following paragraph:
"Note that when the acute-shaped fatḥa which denotes a short a is added on top of a geminated consonnant (i.e. after a šadda), the fatha accent takes a vertical shape to make the composition more distinctable from the tanwiin vowel sign fatḥatan (which marks a /-an/ ending with indeterminate nunation in fully vocalized texts, see below). For an example, see the encoded ligature for ʻAllah above."
Which is unfortunately, incorrect. A šadda may be accompanied by fatḥa, damma or kasra(written below the šadda or below the letter) of which all may be single or tanwiin(nunated). A single fatḥa above a šadda is not written in a vertical shape. This "vertical shape" is the alif ḫanjariyyah or "dagger alif" which indicates an unwritten alif, as found in allāh or ar-raḥmān where the alif in both of these words is denoted with a ḫanjar rather than a full alif character). 21:36, 29 March 2006 (UTC)

Help with Transliteration

Can anyone tell me what the following Arabic phrase would be written in Latin is? The phrase in question is, ما الخليفة أين?

This is not an intelligible phrase! --Alif 22:02, 3 Apr 2005 (UTC)
Well, that would explain why it won't translate. The phrase I want, in English is, 'Where is the Caliph?' Kaiser Matias 02:00 Apr 4 2005 (UTC)
That would be أين الخليفة 'ayna lkhalîfah? - Mustafaa 20:30, 5 Apr 2005 (UTC)

Thanks for the help, I really appreciate it. Kaiser Matias 02:02 Apr 5 2005 (UTC)

o in arabic?

Reading the passage about sukun I encountered the use of o in the examples. I have seen this letter also in other transliterations, the majority of transliteration tables do not mention it however, telling about just three vowels - a,i and u. Could this be briefly clarified here too? 21:55, 6 Apr 2005 (UTC)

O is sometimes used in transliterations for damma instead of u: Mohammad, Muhammad. U follows academic transliteration styles (Encyclopedia of Islam, MELA, etc.), o follows how it sounds in some vernaculars.

alt transliterations

Just made a couple edits (cells misaligned, non-IPA diacritics in the IPA column, etc.). Also a couple alternate transcriptions. Since the under-dot diacritic doesn't work for all browsers, I added the unicode letters with the under-dot as a second variant. Some people have complained about one and some about the other in the comments above, so hopefully by offering both, at least one version will work for everybody. kwami 21:21, 2005 May 24 (UTC)

Persian Alphabet + Gim

I think that the Persian Alphabet article should not be redirected here, and requireds a dedicated article, since the Persian alphabet has additional letters and different sounds. Likewise, there should be a note in the article indicating that Geem is the Egyptian pronounciation for Jeem, as it is known in the rest of the Arab world, and in Fusha and Classical Arabic. DigiBullet 19:14, 25 May 2005 (UTC)

It's also Geem in Yemen.
There are lots of adaptations of the Arabic alphabet, so it's nice to have them all grouped together here for easy reference. But you're right, lots of languages using the Latin alphabet have their own alphabet pages. I imagine it's just that no one's gotten around to making a special page for Persian. I'd do it myself, but I'm afraid I'd mess up the more subtle details. kwami 20:19, 2005 May 25 (UTC)
I gree that Persian alphabet needs a page of its own. The Arabic alphabet, like the Latin, is the basis for many writing systems, but this doesn't mean that the English alphabet is the same as Italian or German.
Also, the Egyptian (unvoiced) geem is not exclusively an Egyptian phenomena, it is the pronounciation of many Arabic speaking peoples, including some Yemenis like Kwamikagami mentioned above, and as mentioned in teh article itself. --Alif 16:48, 6 Jun 2005 (UTC)
To the best of my knowledge Yemenis pronounce jim as j. They do however pronounce qaf as g. The Egyptian pronunciation of jim is mostly restricted to Africa, though Egyptian media dominance has made it more recognizable. Jim is pronounced sometimes as y in parts of Yemen and the Gulf. -- 22:51, 10 November 2005 (UTC)

IPA transcription

First of all, is خ really a fricative-trill combination? I thought it was a pure uvular fricative.

Second, ɢ for ق is simply not within the recorded variability of MSA; in modern Arabic, it's q or g, though in Sibawayh's time it may have been ɢ.

Third, I've never heard of غ being realized as ɣ, but maybe that's just me. - Mustafaa 21:03, 3 Jun 2005 (UTC)

The IPA Handbook, which uses al-Shaam Arabic (one speaker from Safad/Beirut/Damascus) as a sample language, says "/x/ is accompanied by uvular trill", but treats it as a fricative. It looks like the research was done at the University of Kuwait. We should probably list other realizations of this phoneme (in Cairo, isn't it just [x]?), but al-Shaam is a prestige dialect, so I think we should keep it.
As for غ, the symbol is supposed to be a gamma: is it the wrong symbol, or are you saying that غ sould be an approximant, as g in Spanish? (I thought I'd always heard a fricative.) And for ق, please correct it; I don't know what I'm talking about! kwami 21:53, 2005 Jun 3 (UTC)
I believe both خ and غ are uvular rather than velar fricatives, but if the IPA handbook says otherwise, I guess that's a more reliable source... - Mustafaa 22:14, 3 Jun 2005 (UTC)
I'd always just assumed they were both velar because they don't affect vowels the way the uvular qof does (at least not in Cairene). It's pretty common for all uvulars to have such an effect - Yup'ik, for example, though that's hardly confirmation. The Handbook just lists the ghayn as velar in the chart, and never mentions it again, which suggests by omission that there's nothing special about it. And the xaa has uvular accompaniment. Another possibility is that they're velar, but further back than English velars. Such sounds are very often called "uvular" in the literature even when they don't involve the uvula. A lot of "uvulars" in the American Pacific Northwest are actually back velars, for example. But of course in other dialects they may very well be true uvulars. kwami 19:50, 2005 Jun 6 (UTC)
Interesting. In Algerian pronunciation, they do both affect vowels the way qaf does. Maybe this is the effect of the Berber substratum... - Mustafaa 20:52, 6 Jun 2005 (UTC)
We might want to confirm, but that's good reason to suspect that the frics are uvular in Maghrebi Arabic. I'd say put them in as alternate pronunciations if you're comfortable with it. (If that's your native language, you should be able to see it in a (dentist's?) mirror, or maybe just feel it.) kwami 23:10, 2005 Jun 6 (UTC)
They do so too in some Khaliji and Iraqi accents, specially where the madd (long vowel a) is made to sound like how Farsi speakers would pronounce the long vowl in 'Iran'; slightly to the O. Does this only occure in Arabic when a long vowel a is follows the xaa or ghayn? What are your examples, Mustafa? In Qur'anic recitation both velar and uvelar sounds occure, as I recall. --Alif 11:56, 9 Jun 2005 (UTC)

I believe they are uvular -- (talk) 01:30, 8 October 2008 (UTC)

Abgadi order

I propose resorting the letters table to reflect the Abgadi order rather than the current Hega'i. This will help those who want to compare Arabic letters to their countrparts in other alphabets. Addtionally, the image that has the letters' names and shapes depicts the hega'i order, so having the other will be a benefit, not to mention it being the original and the one used in numbering. --Alif 16:54, 6 Jun 2005 (UTC)

Abjadi order is very much less widely used than Hija'i order; maybe a separate article linked from here would be a better place for such a table? Also, we should note that the Abjadi order used in the Maghreb and West Africa is slightly different to the normal one. - Mustafaa 18:59, 6 Jun 2005 (UTC)


my browser [latest ie] doesn't display a lot of symbols unless they are surrounded by {{IPA|}}, so i added that. i know it's a hack; wikipedia really needs to fix this itself. [anyone know where i can suggest this/complain?]

The problem is in not using a modern browser. With IE, the best you can do is a hack. It's a well known issue; it either isn't "fixable" in Wikipedia, or else isn't worth the trouble. kwami

also, some of the IPA renderings of the letters were wrong or strange. most obviously was /x/; trying to represent a "uvular trill offglide" that may be peculiar to one particular dialect, but is certainly not representative, is strange, at best. also, ayn is glottalized only in Iraq and Kuwait.

Benwing 06:36, 11 July 2005 (UTC)

Ayn is not 'glottalized' but a glottal in the al-Shaam standard, at least according to the IPA handbook. In Cairene it sounds glottal to my ear, but I don't know for sure. It doesn't seem to be a pharyngeal approximant, however (and certainly not a fricative!). As for the 'strange' /x/, that wasn't an offglide. The IPA tie bar means that the sounds are pronounced simultaneously. See the discussion above. As is, there is an odd and probably artificial assymmetry between kh and gh. kwami 07:22, 2005 July 11 (UTC)

you are right, "offglide" isn't right. i did see the discussion above but i didn't look quite closely enough. however, no authorities i can find indicate that there is an accompanying trill, so i don't think this belongs unless we can show that it is *widespread* not just in one particular dialect, "prestige" or not. [the authorities disagree on whether it's velar or uvular and indicate it may vary dialectally.]

"glottalization" is not wrong as it can mean "glottal accompaniment" [see wikipedia article]. as for ayn, i've heard quite a lot of egyptian and moroccan arabic, and there is no glottal accompaniment in it. if there were, it would be obvious -- you'd hear a closure. what you're probably hearing is creaky voice. cf. Anatole V. Lyovin "An Introduction to the Languages of the World": "[ayn] is phonetically a creaky-voice, pharyngealized resonant (there does not seem to be much air friction if there is any at all) that mimics the following vowel, or if such does not exist, the preceding one". actual glottal closure occurs only in Iraqi and Kuwaiti Arabic. Benwing 21:46, 11 July 2005 (UTC)

Okay, might want to add an under-trema to the notation then. Your explanation of the glottalization sounds reasonable. However, in the Handbook description, it was not glottal accompaniment, but pharyngeal accompaniment - equivalent to the emphatic consonants. kwami
"glottal accompaniment" meaning a glottal closure in addition to the pharyngeal constriction that is used elsewhere. Benwing 03:24, 20 July 2005 (UTC)

Geem, and shadda on final yaa

Macrakis wrote: Use 'j' consistently for ج (g is Egyptian); final shadda on yaa (-iyyun) is usually transliterated ī Unlike what the common belief, g is NOT an exclusively Egyptian phenomena. This has been clarified elsewhere. As such, we should be able to use g and j interchangeably, and still be correct.

On the otherhand I'm not sure whether we should follow the practice of transliterating the final yaa which has shadda on it as ī, even if it is indeed common, as this leads to confusion due to omitting the marking of a grammatical feature, namely AlNisba[h]

--Alif 11:05, 19 July 2005 (UTC)

About g: I do understand that it is not exclusively Egyptian. However, the article should be internally consistent in its usage. Otherwise, the reader may think that they're two different phonemes, or phonetic variants within a single dialect.

Transliterating -iyyun as -yy is as far as I know completely non-standard. Though -iyy and -īy might be more logical than -ī, I don't think they're standard usage. Anyway, there's no particular need in an English encyclopedia to make the Nisba nature of the suffix transparent. --Macrakis 16:45, 19 July 2005 (UTC)

I've seen both -iyy and -īy used, mostly in more technical contexts. - Mustafaa 22:40, 20 July 2005 (UTC)
Anyway, there's no particular need in an English encyclopedia to make the Nisba nature of the suffix transparent.
But is does affect pronunciation, at least in classical Arabic. --Alif 21:39, 27 July 2005 (UTC)

No. G and J cannot be used interchangeably. I am tired of this. You need to see Arabic how Arabs see it! Not how you would like to see it. An accent of arabic is Not Arabic. If you are transliterating an accent usage, then G may work with certain dialects, notoriously, the Egyptian. If you are transliterating text, then you use J.-- (talk) 16:48, 27 February 2010 (UTC)

Mystery character set for Arabic

I have a document, found on the Net, which appears to be in ISO8859-6, except that there are some "extra" letters. One, in particular, I think is supposed to represent a "g" sound (which is of course not available in the standard Arabic alphabet). Is there a standard superset of ISO8859-6, perhaps used for Farsi, that includes more letters? (specifically, the word "Grendizer" appears to be written in Arabic, and all letters appear to translate okay except for the "g")

You can compare ISO 8859-6 with some other Arabic script encoding schemes here. ISO 8859-6 doesn't appear to resemble other encodings very much. They could be using private/unused space, which is plenty. And no, there's no standard superset of ISO 8859-6 for Farsi/Persian. --jonsafari 15:14, 20 August 2005 (UTC)
The way I remember it from ages ago the g in Grendizer was realized as a ghayn.

Abjadi order

The abjad order doesn't really "preserve" the older Phoenician/Hebrew/Aramaic alphabet order, since the Abjad order was pretty much created after the Arabic alphabet had already been expanded to 28 letters, by matching each Aramaic letter with a corresponding letter of the fully consonant-dotted Arabic alphabet (with six Arabic letters left over at the end). If the Abjad order PRESERVED the older order, then there would be no Arabic letter corresponding to the Aramaic letter semkat (samekh) in the Aramaic alphabet ordering (i.e. 15th position), since no Arabic letter is derived from Aramaic samekh. AnonMoos 01:01, 6 September 2005 (UTC) 05:01, 6 September 2005 (UTC)

I understand your argument, but it would be nice to have a solid reference on it; what is your source? --Macrakis 15:27, 15 September 2005 (UTC)

Suggestion: Computer-related information must appear only in dedicated sections

I would expect a clear division between sections talking about the Arabic script itself (as it has been used on paper) and sections talking electronic-data related issues. The computer-related sections should refer back to the normal paper sections discussing corresponding computer-related issues.

I do not know what the best structure of the article would be:

  • either there are computer-related subsections of the primary sections spread all over the article (but clearly marked as being computer-related)
  • or all computer-related stuff moved to one big section at the end (into the existing Arabic alphabet#Computers and the Arabic alphabet).

I'm uncomfortable with reading Unicode codes when I'm just reading about the script. For example:

and perhaps other.

I would move this information out of the main sections.

Best regards,--Imz 22:21, 22 October 2005 (UTC)

I was thinking the same thing. I find the Unicode stuff useful, but it's certainly not what a casual user interested in the alphabet (especially in learning the alphabet) would care about. --Saforrest 02:35, 4 May 2006 (UTC)

Arabic alphabet and other languages: transcribe vs. write

In the initial paragraph, I chnaged transcribe other languages to write other languages, simply because transcribe appeared to me an inappropriate special term. From Merriam-Webster:

2 a : to represent (speech sounds) by means of phonetic symbols

Then, I thought perhaps there was some significance in the word. Perhaps, it was meant that the way they adopted it for other languages was similar to transcribing what they heard, not trying to respect some other sides of the other language's structure. But anyway, I'm not an expert, and that was not clear from the text. So, if someone has the required knowledge, you could write about it more somewhere in the article.--Imz 19:14, 24 October 2005 (UTC)

The meaning I generally associate with "transcribe", especially in a context like this, is to translate written text from one alphabet to another. For example, I might "transcribe" my first name to Arabic as "ستيفن". --Saforrest 03:41, 29 June 2006 (UTC)

Hamza rules

the hamza rules page is left unedited. i've found this nice thread. someone should form it well and edit Arabic alphabet (writing of the hamza)

Arabic Characters Table

I don't know much about Arabic, but I feel that the 2 tables and 1 image are redundant. Both the "Primary Letters" and "Other Characters" tables are useful enough, but the image is redundant. Not knowing anything about Arabic, I didn't want to delete it unless it wasn't really useful. --Limetom 04:50, 28 November 2005 (UTC)

I agree. Not only is the image redundant and, it is not searchable. The only plus I can see is for browsers/OSes that can't render Arabic text correctly, but I feel the cons outweigh the pros. In any case, I'm removing it. chet the gray (talk) 07:46, 15 April 2008 (UTC)

Additional letters?

I've encountered usage of "extended" letters to transliterate foreign sounds (such as p, g, v) in Arabic. Could someone clarify this issue? -- 18:30, 7 December 2005 (UTC)

I don't know about p and v, but "g" is not a foreign sound in Arabic - it's quite common, and is represented by the 5th letter of the alphabet - geem. --Gene_poole 03:07, 8 December 2005 (UTC)
The [v] sound is sometimes written with the letter fa, but with three dots over it, rather than one. I think that using the Persian [p] letter (ba with three dots under it, rather than one) is rarer. As for a [g] sound, [g] is the ordinary pronunciation of the letter jim among Egyptians (but the standard Arabic pronunciation of this same letter is with a "j", and a common Syrian pronunciation is with a "zh"). Non-Egyptian Arabs occasionally use a letter jim with three dots under it to transcribe a [g], but I'm not sure how widely that would be understood (the Persian use of this same symbol is to transcribe a "ch" or "tsh" sound). AnonMoos 15:15, 13 December 2005 (UTC)
beh with three dots below is pretty much automatically understood to mean P. feh with three dots above is equally understood to mean V. G is transliterated using standard geem in Egypt, and ghayn in the sham. geem with three dots inside indicates a zh voice in Egypt and sometimes called a zheem (e.g. as in garage = geem - reh - aleph - zheem)

Alphabet vs. script

This is a question for native English speakers: Does the opening sentence »The Arabic alphabet is the script used for writing in the Arabic language.« make sense in English? Can an alphabet be a script, or are these two different terms? Judging on the use in my native Slovenian, I would call »a script« a set of defined symbols used to write a language, and »an alphabet« a common, or defined, order of these symbols. Consistent with this, we are talking about Latin script, but about English alphabet, and, for instance, about different positions of letters Ä and Ö in German and Swedish alphabets. Judging from the usage in Wikipedia, though, I would say that the terms »alphabet« and »alphabetic writing system« are used more loosely and perhaps interchangeably in English. Please comment on this. --Peterlin 15:55, 5 January 2006 (UTC)

Well actually this isn't an alphabet. An alphabet represents vowel sounds while an abjad only represents the consonants. This isn't a pure abjad, but that's how it would be best described. I'm not sure why this is under Arabic alphabet. It should be moved. --LakeHMM 03:53, 6 January 2006 (UTC)
Maybe because Alphabet is the traditional and commonly-used term, while "abjad" in that particular sense is a rather recently-coined esoteric technical linguistics term. AnonMoos
I don't think that's a very good reason. It's incorrect, and Wikipedia should, of course, be correct. --LakeHMM 00:30, 7 January 2006 (UTC)
Dude, it's a "consonantal alphabet", if that will salve your conscience. There's no need for Wikipedia to be so pedantic that it impedes understanding. AnonMoos 02:15, 7 January 2006 (UTC)
Dude, that's kind of an oxymoron. I mean, there's a reason Chinese alphabet redirects to Chinese character and not the other way around. Calling it an abjad and linking to the definition of an abjad doesn't impede understanding, it aids it. My conscience isn't at issue here. Let's stay on topic. --LakeHMM 04:00, 7 January 2006 (UTC)
regarding the original question, "script" is used synonymously with "writing system", and alphabets are a subset of these. The Chinese script isn't an "alphabet", being ideo-/logographical, that's not an issue here. If the Arabic alphabet isn't considered an "alphabet" in the strict technical sense, that's still what it is commonly called. Yes, it is an abjad, but we won't move the article to Arabic abjad or Arabic consonantal alphabet because of that. An abjad is, in common usage, considered an alphabet, because it maps phonemes to symbols. In contexts where pedantry is required, it can be identified as an abjad, or if that term is considered too opaque, a consonantal alphabet. dab () 13:53, 7 January 2006 (UTC)
Thank you for your help. I realize, however, that my question probably wasn't clear enough, since no answer actually answered what I hoped I had asked. So I will try to rephrase it. I believe I understand the difference between an alphabet, an abjad and an abugida, and I also know that Arabic is an abjad, even though professor Daniels considers it shouldn't be called an abjad, since it actually has characters for some vowels (alif, for instance). My question was not, however, whether the article should be renamed from Arabic alphabet to Arabic abjad or even Arabic impure abjad, but whether Arabic script would be a more appropriate term for it. Wikipedia applies this term to Indic scripts, e.g., Gujarati script, Gurmukhi script, Malayalam script, Tamil script or Telugu script, to name a few. The titles Gujarati alphabet, Gurmukhi alphabet, Malayalam alphabet, Tamil alphabet and Telugu alphabet exist as redirections for the above mentioned articles. In Arabic (and numerous other articles) the redirection goes the other way round. --Peterlin 08:40, 10 January 2006 (UTC)
but the question whether the Arabic script is an alphabet is related to your question. We seem to use "script" for abugidas, and we seem to allow "alphabet" for abjads. That makes sense, since abjads are traditionally called alphabets. The difference between "abjad" and "alphabet" would be difficult to translate into Arabic, since I understand "abjad" is simply the Arabic generic term for "alphabet" (ar:أبجدية -- is there a difference between abjad, abjadi and abjadiyyah?). It wouldn't be false to move the article to Arabic script, but clearly the same would then apply for Hebrew script, Nabatean script, Syriac script, Proto-Canaanite script etc. These are all commonly called "alphabets", and by MoS I argue that this is where we should keep them. dab () 11:25, 10 January 2006 (UTC)
My Arabic dictionary translates Abjad as "alphabet". Abjadi is the adjective form "alphabetical". Abjadiya is the feminine singular (and inanimate plural) form of the adjective, which could theoretically be used as an abstract noun (though I'm not sure what this abstract noun would mean). AnonMoos 19:03, 11 January 2006 (UTC)
yes, the point is that the ar-wiki article on "Alphabet" is on "abjadiyyah", and not on "abjad", so I wonder whether "abjad" is even in use anymore... dab () 19:48, 11 January 2006 (UTC)
P.S. see Talk:Elative (gradation) commenting on an article you originated. AnonMoos 17:37, 12 January 2006 (UTC)
I'm going to WP:BE BOLD and insert the term abjad into the introductory sentence to clarify things. Warmest Regards, :)—thecurran Speak your mind my past 08:04, 19 January 2010 (UTC)

Abjadi orders tabulated

With the browser I was using (Mozilla Firefox 1.0.7) the two sets of lines demonstrating the Abjadi orders in Arabic and roman letters were misaligned, so I decided to put them in tables. It turned out Firefox was also being too clever and rendering them right to left, which put them in the wrong order. So these are now both in the right order and properly aligned.

I may have inadvertently introduced some errors into the order (I am not familiar with Arabic myself). I would appreciate it if someone who knows what they're doing would check this over.

Hairy Dude 17:48, 25 February 2006 (UTC)

Is there a better way besides an asterisk?

Currently the section titled Numerals has the following table:

Standard numerals
٠ 0
١ 1
٣ 3
٤ 4
٥ 5
٦ 6
٧ 7
٨ 8
٩ 9
East Arab numerals
۰ 0
۱ 1
۲ 2
۳ 3
۴ 4
۵ 5
۶ 6
۷ 7
۸ 8
۹ 9

When I first read this and saw an asterisk near the ٢, I thought that meant it was part of the letter. However, in the paragraph below it has this: "*Standard form of number 2 in Egypt is slightly different". Isn't there a better way to do this? Kirbytime 19:39, 26 March 2006 (UTC)

Hopefully you'll find the new table addresses this issue. –jonsafari 20:19, 5 August 2006 (UTC)

Arabic term for "Arabic alphabet"

What is the term for the Arabic alphabet/abjad in the relevant languages? The North Indian script is called Devanāgarī, the East Asian is called Hanzi in Chinese and Kanji in Japanese and Hanja in Korean and Hán tự in Vietnamese. How do the users of this alphabet refer to it? Could someone please add that information to the article? -leigh (φθόγγος) 05:49, 27 March 2006 (UTC)

الابجديه العربيه is what Google's machine translator gives me. -Fsotrain09 02:31, 25 September 2006 (UTC)

Order of numeral systems?

This strikes me as a bit odd:

Unlike Arabic alphabetic characters, Arabic numerals are written from left to right.

Isn't this sort of a matter of perspective? We think of reading a number from left to right starting with the highest digit because that's our experience as English speakers and that's how our language works, but in older days they used constructions like "four-and-twenty blackbirds" (this is still used in German as vierundzwanzig) and so "42" might have been quite a natural way to write "twenty-four" in English had things happened differently.

With this in mind, we could say that Arabic numerals are read from right-to-left, by reading the ones-place first (e.g. 316 is "6 and 10 and 300"), and it's the European languages which had things backwards. I guess this only makes sense if the way numbers are spoken in Arabic sounds somewhat like "four-and-twenty", though. --Saforrest 02:32, 4 May 2006 (UTC)

I agree. I thought that the whole reason columns of numbers have to be justified on the right side, backwards from columns of text, was that we'd inherited this system from a language that wrote right to left. Could somebody who knows how to speak numbers in Arabic comment on which digits of a multidigit number are actually pronounced first? EbonyTotem 01:22, 9 August 2006 (UTC)
I don't speak Arabic, but for 11-99 it usess the "four and twenty" order. Hundreds, thousands and above precede, I think, although this could also be a later development. --JWB 03:18, 9 August 2006 (UTC)
JWB is right about the order. But in any case, Arabs do generally write numbers from left to right; it's not a matter of perspective, if they're writing ٧٤٦٥٨ they start with the ٧! So for example if you are writing a telephone number in Arabic and the number is 092 353484, you'll begin on the left with the zero, just like in the Latin alphabet. Palmiro | Talk 20:10, 4 September 2006 (UTC)

farsi yeh

I have changed the sound of farsi yeh, form a to i (which is true for Egypt and Iran), but as Urdu is mentioned as a language using Farsi yeh as well, could someone confirm that for Urdu. Or we have to differentiate. Template:U+06D6

Arabic phrase needed

Hi all -- looking for some kind soul to contribute the Arabic for Qisas Al-Anbiya at the article of the same name. If anyone here can do it that would be great. Many thanks! --Bookgrrl 03:15, 16 August 2006 (UTC)

Looks like Al-Qisas and Al-Anbiya are at User:Grenavitar/sura. Putting them together: القصص الأنبياء
Deleting the first definite article: قصص الأنبياء
--JWB 03:27, 16 August 2006 (UTC)
You are so quick!! Thanks (this place is so amazing...) --Bookgrrl 03:40, 16 August 2006 (UTC)


The Arabic alphabet isn't "the script used for writing in the Arabic language" any more than the Latin alphabet is the script used to write the Latin language. The Latin alphabet article doesn't ignore J, U, W, or even Ŋ, so why does this article not make any mention of letters like چ? --Ptcamn 09:57, 16 August 2006 (UTC)

What are you talking about? Giim is certainly listed in the article. --Gene_poole 23:58, 16 August 2006 (UTC)
Count the dots. Giim (ج) has one, che (چ) has three. --Ptcamn 02:19, 17 August 2006 (UTC)
These are legitimate concerns. After all, the Latin Alphabet article almost immediately mentions variations, while here I really only find vowel variations, à la Mater lectonis. The ikiroid (talk·desk·Advise me) 19:43, 29 August 2006 (UTC)
This article combines the general Arabic script and the specific writing system for the Arabic language, the Arabic alphabet. For example, Urdu, Persian, Arabic, and Uyghur all use forms of the Arabic script, but each has their own specific implementation and pronunciations (eg. how ﻕ and ﺽ are pronounced). Having an article that discusses both the Arabic script and the specifics of how it's used for the Arabic language would be like having an article about both Cyrillic and the Russian alphabet (or OCS), or the Latin script and how the Romans pronounced the letters. In following the conventions of most parallel articles, I think we should have an Arabic script article discuss the general features of the the writing system not specific to the Arabic language, akin to Cyrillic script or Devanagari, and have Arabic alphabet cover the specifics of how the script is used for the Arabic language (eg. that ﻕ is pronounced [q], that ﺽ is pronounced [dˁ], etc.) –jonsafari 00:03, 30 August 2006 (UTC)
I absolutely agree. Arabic alphabet should have all the details of the implementation of the Arabic script for the Arabic language. Arabic script should not be language-specific because doing so tends to marginalize the languages which are not Arabic. This article should be as international as possible. I think this issue is why the article was tagged with {{globalize}} --Cbdorsett 08:33, 8 January 2007 (UTC)
One can also argue that the Latin letters J, C, X can be pronounced differently according to the language. But nevertheless in the Arabic script there are even more issues on the vowel sounds and what they represent- e/a in Arabic is ae in Farsi and a in Urdu; i in Arabic and Urdu is e in Persian, etc. Also, there are articles about each specific version of the Arabic alphabet, like Persian alphabet and Urdu alphabet, so I'd think its safe to focus on Arabic in this article. Mar de Sin Talk to me! 18:45, 24 September 2006 (UTC)
You'll notice in the Latin alphabet article that it clearly distinguishes between the graphemes and their usage in Latin. The specific implementation of the Latin alphabet to Latin is covered in a separate article: Latin spelling and pronunciation. It only touches on it as a historical note in the Evolution section, as would be understandable here too. –jonsafari 01:30, 25 September 2006 (UTC)

Unicode problem

Some people (like me) can't see some unicode symbols (sadda, sukum, harakat). Isn't a good idea to create some .png to resolve this problem? Lemke -- 00:50, 8 September 2006 (UTC)

the problem being on your end, I think it would be a good idea for you to fix your browser / install unicode fonts instead. dab () 15:02, 3 October 2006 (UTC)
I think the problem is here on the page. There are tons and tons of transliterations using the old ArabDIN encoding instead of Unicode. See the Abjadi Order section.

ʼ b ǧ d h w z

--Cbdorsett 09:32, 8 January 2007 (UTC)

Writing Arabic letters separately for easy distinguishing

I think that writing Arabic words without ligatures and joining letters will make reading Arabic text much easier. Why do not perform this reform at least in home? Example:

الله - word Allah (A LLAH) written with joining - difficult to read, appears optically as three separate symbols - A and A are separated, but L, L, and H are joined into one symbol.

ا ل ل ّ ه - word Allah (A L L A H) written without joining - easy to read, appears optically as five separate symbols - A, L, L, A and H are fully separated.

But writing that way would be like writing Japanese text in hiragana only. It's technically easier to read for learners, but you'll not find any authentic text written that way, except in various books written for those learning to read the language. The conventional way has always been natural Arabic script (ie with ligatures) plus a romanisation. Rhialto 00:34, 30 October 2006 (UTC)
Actually, I think it was better first. In the second case (not joined) you have five charachters, but it is one word - it makes sense to join them, it takes less space so you can see more words in a glance, it can be seen and read faster and there is little chance of having words that are too long (if such thing exists in Arabic). --Maha Odeh 11:46, 9 July 2007 (UTC)

Hebrew is written with distinct letters, but why Arabic is not written that way, at least since modern times? Why Arabs never didn't performed this reform?

Hebrew was never written cursively (with systematically connected letters) in the main line of the history of the development its alphabet, but Arabic borrowed a cursive form of the Aramaic alphabet basically from the beginning. AnonMoos 03:51, 28 November 2006 (UTC)
Personally, I don't think I'll be able to read it if the letters were not joined, I'll keep reading the letters seperately. Anyway, there are many benifits of writing this way, the most evident of which is that the words are much smaller in size, hence you use up less pages when writing.
Besides, what would happen to caliagraphy if that disappeard?!! It's just a matter of getting used to, everyone else is comfertable with joining the letters - assuming the Arabs feel really affectionate about their alphabet, why didn't anyone else do that? It seems to me they are comfertable too. Just a thought.

"In the hotel"

Why the particular phrase "in the hotel" used as the example phrase in the infobox? AnonMoos 15:51, 7 November 2006 (UTC)

Arabic and Hebrew letters

I have noticed that several pages for Arabic alphabet letters, such as tāʼ, ṣād, and ṭāʼ, redirect to a page titled for the equivalent Hebrew letter. I think this might be inappropriate. What does anyone else think? 16:34, 15 November 2006 (UTC)

Well, not quite. They redirect to pages for the common Semitic letters, and the titles of these pages are supposed to be the Phoenician names of the letters. It is true that many of these pages have more information on the Hebrew letter than on other letters; that should be corrected by adding more sections to those articles. --Macrakis 19:26, 15 November 2006 (UTC)

Bi-directional order

I know Arabic is said to be displayed "right-to-left" though is technically bidirectional because, I think, some string are displayed LTR (ie numbers). So is a sentence displayed right to left character by character (CBA), word by word (ghi def abc) or concept by concept (ccc ddd aaa bbb). It seems like a basic question but I can't find the answer. —The preceding unsigned comment was added by Gfannick (talkcontribs) 16:33, 4 January 2007 (UTC).

It's right-to-left by character and word, except for numbers. That is, if the sentences "I am 42. He is 18." were written in "Arabic" order, it would be ".18 si eH. 42 ma I" Hope this helps! Herbivore 04:24, 8 January 2007 (UTC)

Take a look at Bi-directional text for more information on the question. Cbdorsett 08:17, 8 January 2007 (UTC)

I think it's worth noting that as we borrowed the numerals from Arabic and not the other way around, in actuality we are the ones that use bi-directional text. This is why when you justify numbers, you must do so on a right margin instead of on the left, as would be expected with the left-to-right ordering of our script:


One might make an argument against this by saying that we read numbers in the order they're written, but this is hardly universally true in European languages -- German, for example, reads 27 as "7 and 20", and this was common enough in English that in older texts instances of it can still be found. Granted, no one says 7 and 20 and 100, but still, it's worth thinking about. 01:02, 24 February 2007 (UTC)

I must point out that in Arabic, numbers (at least the smaller ones) are read right-to-left, with the text. --Xyzzyva (talk) 21:40, 17 March 2008 (UTC)

Proposed changes January 2007

In an effort to improve this article, I'd like to suggest the following changes:

This should universalize the content sufficiently to remove the {{Globalize}} tag.

  • Convert all of the ArabDIN 31635 characters representing alif and ‘ain so that they render correctly in common browsers such as Internet Explorer 6.0.
  • Combine the two alphabet tables into a single table. This should be done as a text exercise, not cutting and pasting images from the large graphic that came from the German Wikipedia.
  • Combine the two explanations of the four letter forms into a single explanation.
  • Briefly explain the usage of abjad and alphabet in the head section.
  • Move the section on collation order (now called Abjadī order) to after the presentation of the alphabet, and change the title to Collation order.
  • Increase the size of the characters displayed in the collation order section.
  • Move the "modified letters" section to the pages for Arabic Alphabet and Persian alphabet, and correct the IPA rendition of ta’ marbuta.
  • Move the long description of diacritics to Arabic alphabet, with a very short summary section and an appropriate {{main|}} tag.
  • Delete the section on numerals, but leave a link at the bottom of the page.
  • Fix the map entitled "Worldwide use of the Arabic alphabet" - it should indicate Brunei, Malaysia, Indonesia, Pilipinas, Pattani and Singapore. Maybe it could also mark the countries where the alphabet was used in the past?
  • The section Computers and the Arabic Alphabet should be shorter summaries, with references to the appropriate main articles where the topics are fleshed out in full.
  • PLEASE find a better text to illustrate the script. Fi'l-funduq ("in the hotel") is just too prosaic. I'll check my own photos to see if any are suitable.
  • We should have a References section for all the footnotes to appear in. We have a lot of unsupported assertions now, and this situation should be remedied ASAP.
  • Add an appropriate reference to [[Category:Arabic-derived alphabets]]

--Cbdorsett 06:35, 11 January 2007 (UTC)

Article title

Should the article be titled "Arabic alphabet"? By definition, it's not an alphabet, but an Impure Abjad. Therefore, it should be "Arabic abjad," or at least "Arabic script." Any thoughts? The ikiroid (talk·desk·Advise me) 19:53, 25 February 2007 (UTC)

WP uses the most common English name for things. "Abjad" vs. "alphabet" is a useful distinction when discussing writing systems, but pedantic and unhelpful as a title. "Arabic script" would make sense, but I see no compelling reason to change. --Macrakis 20:20, 25 February 2007 (UTC)
That was previously discused up above in the #Alphabet vs. script section... AnonMoos 21:46, 25 February 2007 (UTC)
Oh, I went over the talkpage looking for this topic and somehow missed that. Still, the fact that "alphabet" is technically (and perhaps politically) incorrect is a well enough reason to change the wording to "script." Its definition as an "alphabet" seems a bit elementary. True, "abjad" is not mainstream, but "alphabet" is wrong, so "script" works out best. The ikiroid (talk·desk·Advise me) 22:29, 25 February 2007 (UTC)
I'm sorry, but alphabet wasn't "incorrect" until one person started inventing new terms ("abugida") and rather arbitrarily and whimsically changing the meaning of old words ("abjad") rather less than 20 years ago. It's nice that linguists recently have some shiny new terminology to play around with, but that doesn't all of a sudden make the meaning of the word "alphabet" -- as it has been used for many centuries -- somehow now allegedly "incorrect"[sic]. AnonMoos 08:06, 28 February 2007 (UTC)
The creation of new terms and new meaning for existing terms is a feature of language evolution that has been going on for quite a long while, not just the last 20 years, and not just the last few centuries. Daniels's coopting of "abjad" for a new use and coining "abugida" was neither arbitrary nor whimsical - it served a definite purpose, and since the introduction of these innovations, linguists have apparently found the distinction to be useful. I suggested changing the title of this article to Arabic script because all the titles of all the other writing system articles use the term "script" instead of "alphabet." Where the term "alphabet" is used, the articles relate to the adaptation of a particular script to a particular language. Cbdorsett 08:55, 28 February 2007 (UTC)
These things just don't work on "Internet time" -- a book published in 1996 has stirred up some discussion in certain specialist circles, but it has hardly served to significantly change the accepted definition of common words in general use, like "alphabet". Ten years is not sufficient to decide that. AnonMoos 09:22, 28 February 2007 (UTC)
I'm with Cbdorsett on this one: Arabic script is probably more used in any case: it gets around 805,000 hits on Google whereas Arabic alphabet gets around 288,000. It's more generic, and cannot be accused of being incorrect. Drmaik 13:56, 28 February 2007 (UTC)

Noinclude wiki markup at top of article

Can someone remove the "</noinclude>" wiki markup from the beginning of the article? I cannot even find the markup code in the article code. Very strange . . . can someone help please? Thanks! JeffreyN 23:39, 2 April 2007 (UTC)


Hello, I was hoping to find a little information about wasla on this page, but I can't find anything about it. Would anyone be kind enough to add it? I am referring to the thing on top of the ʼalif in this character: ٱ

-- 14:08, 13 May 2007 (UTC)

If you mean this ء so it's called Hamza not wasla..and it cuold be written in defferent shapes according to spelling and its position in the word like this: ء أ ؤ ئ orـئـ. --Lubna 10:16, 28 June 2007 (UTC)
Actually, Wasla or Hamzt-ul-Wasl is a character that looks like a tiny little sad ص , and which indicates that the letter alif does not have a glottal stop sound (and does not have a long vowel sound either). It's not really used except when writing full diacritics... AnonMoos 21:09, 28 June 2007 (UTC)

Pronunciation of 'r'

I remember this article earlier mentioning that the r is generally rolled /r/, except for in religious Quranic recitation. This is clearly true from hearing Quranic recitation, but this article should mention whatever consonant(s) are used to replace /r/.

Would you please let me know where that has been mentioned?
I am sure that the Quranic /r/ is a rolled one. I mean by "rolled" similar to the [spanish R] or [Italian R], it's an Alveolar trill consonant, not like the American /r/.
Please play the sound sample on the right hand section of this page Alveolar trill.
Opensourceit 19:19, 7 July 2007 (UTC)

Missing letter

this chart is missing a letter, the seen. even when I scroll down and specifically click on the seen it still takes me to the page about the sheen! I need to refer to the seen in another article and I cant find the page. I don't mind refering to something else but I specifically need to refer to the seen distinguishing it from the sheen and the saad, so this page only makes the reader more confused. --Maha Odeh 11:50, 9 July 2007 (UTC)

Yes, the chart needs to be fixed. Who do we notify? Cbdorsett 12:05, 9 July 2007 (UTC)
Indeed. However, the sheen article does have a small section at the bottom about this issue, distinguishing sin from shin. Drmaik 12:10, 9 July 2007 (UTC)
Yes, I read the sheen Article and noticed that. But the chart is still missing a letter from the alphabet even if they both direct to the same page. --Maha Odeh 12:31, 9 July 2007 (UTC)

you can thank Epson291 (talk · contribs) -- he successfully crippled the template in April. I've reverted to the previous version (you can do this too, it's a wiki!) dab (𒁳) 08:37, 24 July 2007 (UTC)

Unicode for "Allah"

The article says,

Unicode has a special glyph for the ligature llāh, the post-vocalic form of Allāh (“God”). U+FDF2 ARABIC LIGATURE ALLAH ISOLATED FORM

I don't believe that this is correct. I am pretty sure that, in fact, this particular glyph is actually supposed to be the whole of "Allah", not just "llāh"; all the Windows Arabic fonts are simply done incorrectly. -- 02:04, 20 July 2007 (UTC)

[1] says you are right -- another clear blunder on the part of the Unicode Consortium :( dab (𒁳) 16:47, 22 July 2007 (UTC)
Unicode Consortium does not make Windows Arabic fonts, Microsoft or its contractors do. --JWB 22:20, 22 July 2007 (UTC)
you misunderstand. The blunder is the Consortium's, the non-compliance of the font designers is the workaround. Arabic typography needs a "llah" glyph, it doesn't need an "allah" glyph. dab (𒁳) 18:04, 23 July 2007 (UTC)
None of the ligature codepoints are needed and Unicode's philosophy actively discourages their use. Ligatures are included only for back compatibility with legacy character encodings. --JWB 18:20, 23 July 2007 (UTC)
I know this. I said "Arabic typography needs a 'llah' glyph", I didn't say "Unicode needs a 'llah' character". It is still a blunder to define a ligature that is totally unneeded, and instead leave out one that is desired. dab (𒁳) 06:59, 24 July 2007 (UTC)
If it's a blunder, it is the blunder of whatever legacy encoding the ligature was inherited from, not of Unicode Consortium, which you did say. Personally, I can't tell if an "allah" ligature instead of "llah" is a worse idea; "allah" is used as an isolated legend with distinct typography far more than other compounds of "llah", which would have to appear along with other text. --JWB 07:18, 24 July 2007 (UTC)
look, I don't think this is a very fruitful discussion. What "legacy encoding" is Arabic Presentation Forms-A even based on? I have been trying to find the proposal for this character but did not succeed. It is annoying that these Unicode charts do not refer to the original proposal giving a rationale. It is often impossible to second-guess the intention behind the character. Regarding Allah, doesn't it give you pause that all font designers (not just Microsoft) silently draw a "llah" glyph at this codepoint, not an "Allah" glyph? That's because people need a llah glyph, and have no use for an Allah glyph. Why? Because the alif doesn't take part in the ligature. It is perfectly trivial to place an alif before the "llah" glyph. Quite apart from that, I daresay your claim that isolated "Allah" occurs more often is mistaken (even if it would matter). dab (𒁳) 07:35, 24 July 2007 (UTC)
Googling actual usage [2] and usage with preceding alif [3] it looks like the character is most often used as decoration, and is meant to read "Allah" either with or without added alif. Is there any case where "-llah" is part of a word or phrase and does not have a preceding alif? --JWB 14:03, 24 July 2007 (UTC)
Further information: Allah and Al-

In Arabic orthography, the Al of Allah is (unetymologically) not treated as an article. This means that the alif is not written unless pronounced. Now the alif is only a prosthetic vowel, the word is phonologically just /l:a:h/. The alif is only added when nothing else precedes. Now this would not need to be Unicode's problem. They could just have not offered an "Allah" glyph. But since they decided they did want to offer one, it's a pity they got it wrong. Conscientious font designers will just design a ligature for lam+lam+shadda+superscript_alif+ha للّٰه‎ (while FDF2 is properly أللّٰه‎), and this would solve the problem if common word processors were able to deal with ligatures. Microsoft Word still (!!) cannot cope with ligatures, but it is to be hoped this will change in a year or so. Once this happens, font designers' dependence on what the UC did or did not choose to encode will finally be over. dab (𒁳) 14:40, 24 July 2007 (UTC)

Bismillah has the alif. Googling the phrase quoted there gives 56300 hits; removing the alif and googling gives only 34600 hits. --JWB 18:57, 24 July 2007 (UTC)
you are right. It's an alif al-wasl. The point isn't that the ailf is seldom used, the point is that it doesn't enter the ligature, and can be added as an unconnected glyph to the llah ligature. Alhamdulillah doesn't have the alif. dab (𒁳) 06:21, 25 July 2007 (UTC)
I've never contested the obvious point that it is unconnected. You also "daresay"ed (daresaid?) the claim that isolated Allah occurs much more than the no-alif phrases. Alhamdulillah with the ligature gets only 76 hits. [4] And it would hardly look good to have part of a phrase (llah or lillah) rendered with the proper ligature while the rest is rendered without proper ligatures. --JWB 07:25, 25 July 2007 (UTC)
yes. My bold estimate that alif-less forms will be more frequent is quite unrelated to this debate. Fwiiw, I get 1.7M hits for -الله لله and 10.2M hits for الله- لله. Let's say I take back that claim and settle for non-committal "the alif-less form is in frequent use". ولله alone gets 5M hits. dab (𒁳) 08:07, 25 July 2007 (UTC)
Oddly, those searches give me 1.4M and 2.15M respectively. Also, the second search seems to be finding the sequence "allah lillah" rather than pages using "lillah" but not "allah". Finally, all three strings are expressed with individual letters and not the "llah" codepoint, so they will be rendered either consistently all without ligatures or consistently all with ligatures, depending on the renderer. "Wallah" using the ligature codepoint [5] gets only 23 results.
I guess I would sum up: ligature codepoints are not needed for Arabic prose (especially when renderers become able to produce appropriate ligature glyphs from individual letter codepoint input), but the "Allah" ligature codepoint seems to have found use as a symbol or logo (and usually without added alif, so I'm guessing some fonts or renderers actually do display the ligature codepoint as "Allah") rather than as a part of connected Arabic text.
As far as the origin of the ligature codepoint(s), I too wish that I could easily find records of how UC made the decision. --JWB 09:14, 25 July 2007 (UTC)
I tend to agree with your summary. It is a truism that no ligature codepoints are needed, at all, if you have a layout engine that can deal with a font's ligature definitions. Regarding "some fonts", I am not sure: it appears to render as llah out of the box on Windows (but maybe not on Arabic editions of Windows??) Concerning our frequency counts, I suppose this should be done with a reasonable corpus (the Quran or various Hadiths), not the internet. dab (𒁳) 10:39, 25 July 2007 (UTC)

[6] says: The Unicode-conformant fonts, on the other hand, are: SIL's Scheherazade, Adobe Arabic (distributed with the [7] Middle-Eastern version of the latest Adobe Reader 7), Arial Unicode MS, and Arabic Typesetting (distributed with VOLT and with Microsoft Office Proofing Tools 2003).

I tried searching the Qur'an and did not get "llah" without alif - maybe you can find something I am doing wrong: [8] [9]

But my main point was about current usage. Character coding of the Qur'an is not popular anyway - Qur'an notes It is extremely difficult to render the full Qur'ān, with all the points, in computer code, such as Unicode. --JWB 11:20, 25 July 2007 (UTC)

thanks, the link is very enlightening. You are right, Quranic orthography seems to use ٱ rather than omitting the alif (?). I think we more or less understand this now and should put a note in the article. dab (𒁳) 11:51, 25 July 2007 (UTC)

Arabic Presentation Forms-A

let's see what we can collect on these "word ligatures" at U+FDFx:

  • U+FDF2 ﷲ‎: "ALLAH ISOLATED FORM" -- الله. nice to have an 'isolated form', dear UC, but we would prefer a non-isloated form.
  • U+FDF7 ﷷ‎: "ALAYHE" عليه
  • U+FDF8 ﷸ‎: "WASALLAM" وسلم
  • U+FDFA ﷺ‎: "SALLALLAHOU ALAYHA WASALLAM" صلى الله عليه وسلم
  • U+FDFB ﷻ‎: "JALLAJALALOUHOU" جل جلاله
  • U+FDFC ﷼‎: Rial ريال
  • U+FDFD ﷽‎: Basmala

FDF5 is the صلعم "peace be upon him" interjection. FDF7, FDF8, FDF9 are required to typeset صلى الله عليه وسلم and FDFA appears to repeat the entire phrase. FDFB jalla jalaluhu جل جلاله appears to be a pious exclamation "majestic be his splendour" or something. I would really be interested in what "legacy encoding" we are looking at here. I doubt there is one. This is Unicode 1.1, when the UC was still enthusiastic about encoding ligatures. Now they decided to stop adding ligatures, and we are stuck with the useless "allah" ligature. The sensible course for the UC would be to deprecate FDF2 (since prescribed and actual usage have diverged beyond reparation), and introduce an actual "llah" ligature e.g. at FDFE. dab (𒁳) 08:10, 24 July 2007 (UTC)

The question at this point is, is there any notable font that does comply and presents an "isolated Allah" glyph? I do not seem to have such a font. dab (𒁳) 15:07, 24 July 2007 (UTC)

online use of the ﷲ character

Find sources: "ﷲ" – news · newspapers · books · scholar · HighBeam · JSTOR · free images · free news sources · The Wikipedia Library · NYT · WP reference

I get 119,000 google hits for the character. That's not very much compared to the 180M for الله. There seem to be almost no pages in Arabic. There seems to be a fair number of Turkish pages, but in most cases, the character is apparently just used for decoration ( etc.)

  • perhaps notable is dv:ﷲ (Dhivehi Wikipedia) which has the Allah article located at that character.
  • this page uses the character correctly (Unicode-compliant) in the takbir.
  • this article appears to use it for llah.

dab (𒁳) 11:07, 25 July 2007 (UTC)

Please help to identify the letter/sound

From my childhood I remember a comment about one letter "this letter is for the sound made by giaurs when they drink too much wine" (or something like that). What letter it may be and who is the author of this witticism? Is this remark famous enough to be included in the article about the letter? I would guess it is about Ayin. Any comments? `'Míkka 23:04, 21 August 2007 (UTC)


Why is Zāīn given as "Zāī" in the table? Badagnani 05:35, 14 September 2007 (UTC)

زاي is certainly the most common name of the letter in Arabic, as far as I'm aware (but should be transcribed Zāy, not Zāī). -- AnonMoos 10:32, 14 September 2007 (UTC)
As far as I'm aware, Zai (or zay, don't see any difference) is much more common. --Maha Odeh 11:22, 14 September 2007 (UTC)

In which nations are you speaking of? Our own article gives the Arabic name of this letter as "zain" or "zayin." Badagnani 00:10, 15 September 2007 (UTC)

a question of WP:CITE. My Standard Arabic textbook (Schulz-Krahl-Reuschel) gives zāy. --dab (𒁳) 07:52, 15 September 2007 (UTC)

Removal of letters

What is the consensus on this undiscussed blanking? Are these extra letters never used to write the Arabic language, even when writing Persian names or phrases? 01:02, 25 September 2007 (UTC)

well, there are various 'extra' letters used around the Arab world, varying according to region. But the 'keheh' is widely used in North Africa, esp. in handwriting and children's books. People consider it just to be kaf. I'm not sure how that would be best handled... Drmaik 05:41, 25 September 2007 (UTC)

Perhaps I should have added my reasoning here, sorry. My main reasons for removing them from that table were 1) as far as I know, they don't fall into the sequence of primary letters as given in the table, and 2) their presentation did not match the rest of the table (the name and translit. columns were empty and the phonemic values were not correctly given). Perhaps the 'keheh' could be added to the modified letters table below. 'Tcheh' could also be added there, but may require more discussion since it apparently represents different sounds in different places (Talk:Arabic_alphabet#Persian_Alphabet_.2B_Gim).StephenHudson 16:59, 27 September 2007 (UTC)

I asked "Are these extra letters never used to write the Arabic language, even when writing Persian names or phrases?" You didn't answer that. We must exercise due deliberation (and discussion) here. You didn't. Please fix this situation. Badagnani 17:02, 27 September 2007 (UTC)

Hamza Hate and Abjad??

Why is the letter hamza missing from the listing of letters in the Arabic alphabet on this page? In all Arabic academic works it is considered a full letter not a diacritic or anything else. The only dispute is pretty much whether in some given Arabic dictionary they decide to place hamza at the beginning or end of the alphabet! Also could someone explain all this abjad abujadiyyah craziness? Its not clear from the respective pages what on earth people meant seeing as in Arabic 'abjad' just means alphabet. An alif may be vowel or seat for hamza, a yaa' a consonant or a vowel, etc with waaw. At least to a speaker of Arabic, 'abjad' has no distinctive nature w.r.t. 'alphabet', are people just being pretentious? Brought to you by the Save the Hamza Campaign, and Bored Arabic Students United 22:58, 11 October 2007 (UTC)

Since hamza mostly occurs as a diacritic above or below alif, or above waw, or ya (and rarely occurs on its own except word-finally), it's not going to be exactly on the same level as letters which occur without positional restrictions.
Abjad in the specific English-language meaning "consonantal alphabet" is a recent neologism of technical linguistic terminology (a somewhat poorly-chosen neologism, in my opinion). AnonMoos 03:11, 12 October 2007 (UTC)

Arabic printing press

Should mention in that section that the Ottoman Empire had a specific policy of forbidding printing on its territory of works using the Arabic alphabet, or aimed at Muslims, until the 18th-century. AnonMoos (talk) 06:06, 13 December 2007 (UTC)

Do you have a source for this? Was there a policy, or was Arabic-script typography simply underdeveloped compared to calligraphy? cf. Haralambous --Macrakis (talk) 22:04, 10 February 2008 (UTC)
Francis Robinson attributes this to the emphasis on orality; that oral transmission was considered more reliable than written (let alone printed):
"current scholarship is unsure about why Muslims rejected printing for so long-indeed, it is a problem that seems not to have been seriously studied.... My own feeling, however, is that the origin of the negative Muslim response to printing lay much more deeply than this. The problem was that printing attacked the very heart of Islamic systems for the transmission of knowledge; it attacked what was understood to make knowledge trustworthy, what gave it value, what gave it authority."
Francis Robinson, "Technology and Religious Change: Islam and the Impact of Print" Modern Asian Studies 27:1:229-251 (Feb. 1993) at JSTOR
He doesn't mention any legal prohibition, though Ibrahim Muteferrika (qv)'s patent (1729) excluded religious works. --Macrakis (talk) 03:17, 11 February 2008 (UTC)
Bayezid II says "Bayezid is also responsible for certain self-inflicted intellectual wounds in Islamic civilization, such as the outlawing of all printing in Arabic and Turkic, a ban lasting in the Islamic world until 1729."; although I'm not sure how good the citation for it is. — Hex (❝?!❞) 15:40, 21 January 2009 (UTC)

How many primary letters?

In most of the article, the figure given is 28, but one place says 29. I guess the bone of contention is the Hamza. Is the hamza considered a primary letter of the Arabic alphabet, or not? FilipeS (talk) 18:03, 10 February 2008 (UTC)

Traditionally, Arabic is said to have 28 letters; other signs used in writing are considered diacritics (harakat or otherwise). Sometimes lam-alif is counted as a 29th letter. Hamza is not considered a letter, but a diacritic "seated" on waw, ya, or alif (where it may be under or over), or standing alone. Similarly, ta marbuta and alif maqsura are considered forms of base letters, not independent letters. In the context of computer encodings such as Unicode, the analysis may be a bit different, and the five forms of hamza, ta marbuta, and alif maqsura may be treated as letters on their own. The article could be clearer about all this. --Macrakis (talk) 18:53, 10 February 2008 (UTC)
Interesting quote: "The Arab grammarians were at pains to establish the number of letters in their alphabet, but they have not always agreed among themselves." T F Mitchell, Writing Arabic. (talk) 03:21, 24 February 2008 (UTC)
Yeah, interesting all right. Very clever way to put down the authorities who actually speak the language. Cbdorsett (talk) 09:19, 25 February 2008 (UTC)
Grammarians and linguists, ancient or modern, native speakers or not, may (and do!) analyze the same facts in different ways. This is not a put-down. And though native speakers are of course the "authorities" on actually using the language, their linguistic analyses are no better or worse a priori than anyone else's. As for the specific question of the number of letters (as opposed say, to the number of phonemes), well, that is really more a convention than anything else, with little or no effect on linguistic theory.
For example, further up in this talk page, you mention the analysis of long vowels as doubled short vowels. This is one way to do it, but not the only one or necessarily the most perspicuous one. --Macrakis (talk) 13:37, 27 March 2008 (UTC)

Need pronunciation for ص

Information about the pronunciation of ص should be added at Tsade. Badagnani (talk) 21:22, 7 April 2008 (UTC)

Need pronunciation for ط

Information about the pronunciation of ط should be added at Teth.

Arabic Wordlist

Arabic Wordlist ordered by frequency (in Arabic Wikipedia) [10] -- (talk) 09:49, 5 June 2008 (UTC)

Gaf needs to be added

Why is gaf (گ), which is used in Moroccan Arabic, not mentioned in this article? Badagnani (talk) 04:42, 16 June 2008 (UTC)

Also . Badagnani (talk) 08:05, 16 June 2008 (UTC)

Becouse it's not Arabic; it may be used in dialects but the standard alphabet does not include them. --Maha Odeh (talk) 05:52, 17 June 2008 (UTC)

This article isn't about the Arabic language but the Arabic alphabet. Users will come here looking for the standard letters, as well as the modified letters, which are certainly encountered in Arabic writings, whether in transliteration, Maghrebi names, etc. However, the latter can't be found anywhere else at Wikipedia because there is no other article collecting them. In order to have the best reference possible, we need to collect them. Badagnani (talk) 05:56, 17 June 2008 (UTC)

I was in fact talking about the alphabet not the language. Dialects are actually not written formally and such letters are not part of the alphabet, modified or not. Even the Maghribi Arabic Language Academy (one of the regulators) do not include them (check the official website).
Also, it's not accurate that these letters are not found anywhere at Wikipedia, the letters are actually from the Persian alphabet which is based on the Arabic one but not identical. See Perso-Arabic script. --Maha Odeh (talk) 07:25, 17 June 2008 (UTC)

Some are used in Persian, some are used in transliterating words from English and other languages, some are used for various Pakistani and South Asian langauges, some are used for West African languages, and some are used for non-Semitic Maghrebi languages. Badagnani (talk) 07:48, 17 June 2008 (UTC)

Additional modified letters

Additional modified letters used in the Arabic alphabet for other languages and transliterating foreign words have been added. Can someone put them into the grid so that we have a complete listing, properly formatted? Most of the letters have their own Wikipedia articles. Many thanks for this. Badagnani (talk) 20:52, 16 June 2008 (UTC)

I don't think they can go in the grid because they aren't part of the canonical Arabic alphabet and have no place in its order, even though they obviously do within the adapted alphabets that use them. It wouldn't make any more sense to synthesize them all into a single ordering than it would be for an article on the Roman alphabet to include ll and ñ from Spanish and þ from Icelandic and ø from Danish and rh from Welsh. In fact, what about ö? That's a separate letter in Hungarian, and it comes right after o—but it's also a separate letter in Swedish, where it's the last letter, after z, å, and ä. I imagine the same problem prevails among the scripts that are based on Arabic. —Largo Plazo (talk) 21:20, 16 June 2008 (UTC)

They are, in fact, listed at List of Latin letters. Badagnani (talk) 21:32, 16 June 2008 (UTC)

Not in the same grid! And as I noted, the order is arbitrary. For example, the o variants I cited aren't placed at the end as they are in Swedish. —Largo Plazo (talk) 11:33, 17 June 2008 (UTC)

It's a start. Our project is characterized by constant improvement. My HTML skills aren't such to allow for me to create such a grid. However, now all the letters are there and ready to be explained for our readers, so we have a complete, thorough article and don't fail to discuss any of them. Badagnani (talk) 17:15, 17 June 2008 (UTC)

That's good. However, let me stress that improving articles isn't the same is improving the things that the articles are about. It isn't up to you to improve the Arabic alphabet! And it's the job of the article not to present it as something other than what it is. Having a separate listing of letters that supplement the standard Arabic ones in languages that use the Arabic alphabet as their base is great, but don't confuse their status as such as meaning that they are part of the Arabic alphabet. —Largo Plazo (talk) 03:10, 18 June 2008 (UTC)

Many of the characters I added already have their own Wikipedia articles, which are in the category "Arabic letters," much like some Latin letters are not used in the Latin or English languages. Of these, three are used in the Arabic language for transliteration of foreign wards, particularly in the Levant. The characters, as modifications of Arabic letters used in the Arabic script both in the Arabic language as well as applied to other languages, should be presented to our readers, in an easily findable and central place, wherever that may be (even if it means splitting them out, except for the three letters used in Arabic transliteration, into their own article, like the Latin characters article mentioned above). Badagnani (talk) 03:13, 18 June 2008 (UTC)

it might be a good idea to make a seperate page for all the letters derived from the Arabic alphabet, similar to List of Latin letters, which only lists the possible variants. --Maha Odeh (talk) 06:22, 18 June 2008 (UTC)

That's very possible, if we can get someone who can do a box showing all the forms of each letter. I presume we would keep the three letters sometimes used in Arabic for transliterating V, P, and CH? Badagnani (talk) 06:25, 18 June 2008 (UTC)

Availability of sorting in software

I removed the sentence that someone else had tagged for a citation earlier today, claiming that support for the abjadi collation order isn't found in major software. I went found [11] and found that while this is true for most of the specific software I looked into that have any Arabic collation at all (Windows XP, SQL Server, MySQL), it isn't true for Oracle, which provides both collation options. —Largo Plazo (talk) 16:07, 11 July 2008 (UTC)

problem with medial and final forms

I printed out the table of letters for future reference, and then I realized that many of medial and final forms are showing incorrectly! Letters like dal, ra, zay, sin, and others don't have the little tail connecting them to the previous letter in the medial (if it exists) and final forms. E.g.

Contextual forms Name Translit. Phonemic Value (IPA)
Isolated Final Medial Initial
dāl d /d/

YMMV, but for me, the dal in the 'final' column there has a hook to the left but no connector to the right.

I investigated further (using Character Map) and found that this is because major fonts on my Windows XP machine have the wrong glyph for certain Unicode points, e.g. U+FEAA "Arabic letter dal final form"! For example, New York Times has it wrong (without connector), but Arial Unicode MS has it right (with connector). I checked on another Windows XP machine, and in both Firefox 3.0 and IE 7. Same behavior. Bummer! This means that probably a lot of users, in fact probably most users, who visit the Arabic page, are seeing the wrong medial/final forms for many letters. Bummer! (What glyphs do you see? What type of machine are you on?)

Well, I wouldn't put the onus on Wikipedia to make up for apparent mistakes in these Windows fonts; but I noticed that this problem doesn't occur in certain other articles. For example, the article for Dāl shows the final shape just fine:

Position in word: Isolated Final Medial Initial
Glyph form: د‎ ـد‎ ـد‎ د‎

It does this by using the template [Arabic alphabet shapes].

Question: should the Arabic Alphabet article be using a similar template, which seems to succeed in not showing wrong glyphs?

Secondary question: notice that the template (and therefore the Dal article) has a different convention for order of contextual forms: initial, medial, final, compared to the order in this article (which is the reverse). One can argue about which order is more intuitive, but it would be nice to have them consistent, so you don't always have to be checking column headers.

Primary Letters table - General Unicode & Contexual Isolated Forms are identical

In the Primary Letters table (see below), the two columns General Unicode and Contexual Isolated Forms have letters that are identical apart from the number below them. I suggest we merge the 2 columns together and have the both numbers listed below each letter. Arjun G. Menon (talk · mail) 10:45, 16 December 2008 (UTC)

The characters display the same in browsers which are able to display both, but they're different characters in the Unicode character set, serving separate functions... AnonMoos (talk) 12:05, 16 December 2008 (UTC)

Primary Letters table

Contextual forms Name Translit. Phonemic Value (IPA)
Isolated Final Medial Initial
ʾalif ʾ / ā various, including /aː/
bāʾ b /b/
tāʾ t /t/
ṯāʾ /θ/
ǧīm ǧ (also j, g) [ʤ] / [ʒ] / [ɡ]
ḥāʾ /ħ/
ḫāʾ (also kh, x) /x/
dāl d /d/
ḏāl (also dh, ð) /ð/
rāʾ r /r/
zāī z /z/
sīn s /s/
šīn š (also sh) /ʃ/
ṣād /sˁ/
ḍād /dˁ/
ṭāʾ /tˁ/
ẓāʾ /ðˁ/ / /zˁ/
ʿayn ʿ /ʕ/
ġayn ġ (also gh) /ɣ/ (/g/ in many loanwords)
fāʾ f /f/
qāf q /q/
kāf k /k/
lām l /l/, ([lˁ] in Allah only)
mīm m /m/
nūn n /n/
hāʾ h /h/
wāw w / ū /w/ / /uː/
yāʾ y / ī /j/ / /iː/

Edits to gemination and nunation

I've tried to make these two sections more user-friendly, as I think the normal person coming to this article will be a beginner-learner of Arabic. PiCo (talk) 05:06, 25 June 2009 (UTC)

Link addition proposal

I would like to propose a link to a free course of the Arabic Alphabet:

The course is not made by me (albeit it is hosted on my website). I think it genuinely is a useful resource. —Preceding unsigned comment added by Lyzazel (talkcontribs) 16:50, 1 October 2009 (UTC)

Underlining in linked letter names overwrites subscript diacritics

Where the names of the letters are wikilinked, the underlining means that underbars to the letters disappear (and underdots etc become very indistinct); "tāʾ" and "ṯāʾ" are seen to be different - tāʾ and ṯāʾ render identically (in my browser, anyway). Not sure what a solution is - possibly something like ṯāʾ (article), although that's very messy.Pseudomonas(talk) 09:34, 19 October 2009 (UTC)

Article name

Since the Arabic writing system is technically not an alphabet (it is an abjad), wouldn't it be more appropriate to move this article to Arabic script? GSMR (talk) 18:19, 2 December 2009 (UTC)

Add the parent to the light-green column on the right side

The parent is Egyptian hieroglyphs>>Proto-Sinaitic in parent and systems form like all letter systems on wikipedia, so that, it should be like this:

Parent : Egyptian hieroglyphs systems: Proto-Sinaitic

        Proto-Canaanite alphabet 
        Phoenician alphabet 
        Aramaic alphabet 
        Arabic abjad 

Thank you very much,


According to the language map ("Worldwide use of the Arabic alphabet"), Pakistan is a country "where the Arabic script is the only official orthography". However, English is an official language of Pakistan, so doesn't that mean that the Latin alphabet must also have official status? (talk) 14:36, 22 March 2010 (UTC).