The Arabic alphabet is the script used for writing the Arabic language, which is the language of the Quran, the holy book of Islam. The alphabet's influence spread with that of Islam and it has been, and still is, used to write many other languages from families unrelated to the Semitic languages, such as Persian and Urdu. (See fuller list below.)
In order to accommodate the phonetics of other languages, the alphabet has been adapted by the addition of letters and other symbols. The alphabet presents itself in different styles such as Nasta'līq, Thulthī, Kūfī and others (see Arabic calligraphy), just like different fonts for the Roman alphabet. Superficially, these styles appear quite different, but the basic letterforms remain the same.
Structure of the Arabic alphabet
The Arabic alphabet is written from right to left and is composed of 28 basic letters. Adaptations of the script for other languages such as Persian (Farsi) and Urdu have additional letters. There is no difference between written and printed letters; the writing is unicase (i.e. the concept of upper and lower case letters does not exist). On the other hand, most of the letters are attached to one another, even when printed, and their appearance changes as a function of whether they connect to preceding or following letters. Some combinations of letters form special ligatures.
The Arabic alphabet is an "impure" abjad - since short vowels are not written, though long ones are - so the reader must know the language in order to restore the vowels. However, in editions of the Quran or in didactic works a vocalization notation in the form of diacritic marks is used. Moreover, in vocalized texts, there is a series of other diacritics of which the most modern are an indication of vowel omission (sukūn) and the lengthening of consonants (šadda).
Presentation of the alphabet
The Arabic alphabet can be transliterated and transcribed in various ways. The preferred method in this document will be DIN-31635. Alternatives belonging to other standards are indicated after the oblique bar.
Notice that the horizontal-line diacritic above the long vowels is often replaced by a circumflex, because it happens to be easier to type in many keyboards.
A transliteration from Arabic must allow the reconstruction of the original Arabic letters, so it shows the characters which are not pronounced or which are pronounced as others. A phonemic transcription indicates only the pronunciation. See below for more details. The phonemic transcription (somewhat simplified here) follows the conventions of the Continental version of the International Phonetic Alphabet: for more details concerning the pronunciation of Arabic, consult the article on Arabic pronunciation.
SATTS, the Standard Arabic Technical Transliteration System, is a US military standard transliteration of Arabic letters to the Latin alphabet.
|Stand-alone ||Initial ||Medial ||Final ||Name ||Transliteration ||Phonetic Value (IPA) |
|ﺀ ||أ ؤ إ ئ ٵ ٶ ٸ, etc. ||hamza ||ʾ / ’ ||[ʔ] |
|ﺍ ||— ||ﺎ ||ʾalif ||ā ||[æː] |
|ﺏ ||ﺑ ||ﺒ ||ﺐ ||bāʾ ||b ||[b] |
|ﺕ ||ﺗ ||ﺘ ||ﺖ ||tāʾ ||t ||[t] |
|ﺙ ||ﺛ ||ﺜ ||ﺚ ||thāʾ ||th ||[θ] |
|ﺝ ||ﺟ ||ﺠ ||ﺞ ||jīm ||j / g / ǧ ||[ʒ] / [ʤ] / [g] |
|ﺡ ||ﺣ ||ﺤ ||ﺢ ||ḥāʾ ||ḥ ||[ħ] / [ɦ] |
|ﺥ ||ﺧ ||ﺨ ||ﺦ ||khāʾ ||kh ||[χ] |
|ﺩ ||— ||ﺪ ||dāl ||d ||[d] |
|ﺫ ||— ||ﺬ ||dhāl ||dh ||[ğ] |
|ﺭ ||— ||ﺮ ||rāʾ ||r ||[r] |
|ﺯ ||— ||ﺰ ||zāy ||z ||[z] |
|ﺱ ||ﺳ ||ﺴ ||ﺲ ||sīn ||s ||[s] |
|ﺵ ||ﺷ ||ﺸ ||ﺶ ||shīn ||sh / š ||[ʃ] |
|ﺹ ||ﺻ ||ﺼ ||ﺺ ||ṣād ||ṣ ||[ṣ] |
|ﺽ ||ﺿ ||ﻀ ||ﺾ ||ḍād ||ḍ ||[ḍ] |
|ﻁ ||ﻃ ||ﻄ ||ﻂ ||ṭʾ ||ṭ ||[ṭ] |
|ﻅ ||ﻇ ||ﻈ ||ﻆ ||ẓāʾ ||ẓ ||[ğ̣] |
|ﻉ ||ﻋ ||ﻌ ||ﻊ ||ʿayn ||ʿ / ‘ ||[ʔ] |
|ﻍ ||ﻏ ||ﻐ ||ﻎ ||ghayn ||gh / ġ ||[ɣ] |
|ﻑ ||ﻓ ||ﻔ ||ﻒ ||fāʾ ||f ||[f] |
|ﻕ ||ﻗ ||ﻘ ||ﻖ ||qāf ||q / ḳ ||[q] |
|ﻙ ||ﻛ ||ﻜ ||ﻚ ||kāf ||k ||[k] |
|ﻝ ||ﻟ ||ﻠ ||ﻞ ||lām ||l ||[l] |
|ﻡ ||ﻣ ||ﻤ ||ﻢ ||mīm ||m ||[m] |
|ﻥ ||ﻧ ||ﻨ ||ﻦ ||nūn ||n ||[n] |
|ﻩ ||ﻫ ||ﻬ ||ﻪ ||hāʾ ||h ||[h] |
|ﻭ ||— ||ﻮ ||wāw ||w / ū ||[w] / [uː] |
|ﻱ ||ﻳ ||ﻴ ||ﻲ ||yāʾ ||y / ī ||[j] / [iː] |
Letters lacking an initial or medial version are never tied to the following letter, even within a word. As to ﺀ hamza, it has only a single graphic, since it is never tied to a preceding or following letter. However, it is sometimes 'seated' on a waw, ya or alif, and in that case the seat behaves like an ordinary waw, ya or alif.
This chart does not contain an exhaustive list of all the variant transliterations that European orientalists have invented. There are also numerous local variations on pronunciation, even when speaking the standard, literary language (Fusha)
|Stand-alone ||Initial ||Medial ||Final ||Name ||Trans. ||Value |
|ﺁ ||— ||ﺂ ||ʾalif madda ||ʾā ||[ʔaː] |
|ﺓ ||— ||ﺔ ||tāʾ marbūṭa ||h or t / Ø / h / ẗ ||[a], [at] |
|ﻯ ||— ||ﻰ ||ʾalif maqṣūra ||ā / ỳ ||[aː] |
|ﻻ ||— ||ﻼ ||lām ʾalif ||lā ||[laː] |
The ʾalif maqṣūra, commonly using Unicode 0x0649 (ى) in Arabic, is sometimes replaced in Persian or Urdu, with Unicode 0x06CC (ی), called "Farsi Yeh". This is appropriate to its pronunciation in those languages. The glyphs are identical in isolated and final form (ﻯ ﻰ), but not in initial and medial form, in which the Farsi Yeh gains two dots below (ﯾ ﯿ) while the ʾalif maqṣūra has neither an initial nor a medial form.
Writing the hamza
Initially, the letter ʾalif indicated an occlusive glottal, or glottal stop, transcribed by [ʔ], confirming the alphabet came from the same Phoenician origin. Now it is used in the same manner as in other abjads, with yāʾ and wāw, as a mater lectionis, that is to say, a consonant standing in for a long vowel (see below). In fact, over the course of time its original consonantal value has been obscured, since, ʾalif now serves either as a long vowel or as graphic support for certain diacritics (madda or hamza).
The Arabic alphabet now uses the hamza to indicate a glottal stop, which can appear anywhere in a word. This letter, however, does not function like the others: it can be written alone or on a support in which case it becomes a diacritic:
- alone : ء ;
- with a support : إ, أ (above and under a ʾalif), ؤ (above a wāw), ئ (above a yāʾ without points or yāʾ hamza).
The details of writing of the hamza are discussed below, after that of the vowels and syllable-division marks, because their functions are related.
The only compulsory ligature is lām+'alif. All other ligatures (yaa - mīm, etc.) are optional.
Some fonts include a (Salla-llahu 'alayhi wasallam) glyph:
Muslims normally use this phrase after any mention of the prophet Muhammad.
Fonts also include a special glyph for the word "li-llah", which means "to God."
Combined with the letter 'alif, it becomes Allah:
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:
Arabic short vowels are generally not written, except sometimes in sacred texts (such as the Quran) and didactics, which are known as vocalised texts. Occasionally short vowels are marked where the word would otherwise be ambiguous and cannot be resolved simply from context.
Short vowels may be written with diacritics placed above or below the consonant that precedes them in the syllable. (All Arabic vowels, long and short, follow a consonant; contrary to appearances: there is a consonant at the start of a name like Ali — in Arabic ʾAlī — or a word like ʾalif.)
Long "a" following a consonant other than hamzah is written with a short-"a" mark on the consonant plus an alif after it (ʾalif). Long "i" is a mark for short "i" plus a yaa yāʾ, and long u is mark for short u plus waaw, so aā = ā, iy = ī and uw = ū);
Long "a" following a hamzah sound may be representend by an alif-madda or by a floating hamzah followed by an alif.
In an un-vocalised text (one in which the short vowels are not marked), the long vowels are represented by the consonant in question (alif, yaa, waaw). Long vowels written in the middle of a word are treated like consonants taking sukūn (see below) in a text that has full diacritics.
For clarity, vowels will be placed above or below the letter د dāl so it is necessary to read the results [da], [di], [du], etc. Please note, د dāl is one of the six letters that do not connect to the left, and is used in this demonstration for clarity. Most other letters connect to ʾalif, wāw and yāʾ.
|Simple vowels ||Name ||Trans. ||Value |
|دَ ||fatḥa ||a ||[a] |
|دِ ||kasra ||i ||[i] |
|دُ ||ḍamma ||u ||[u] |
|دَا ||fatḥa ʾalif ||ā ||[aː] |
|دَى ||fatḥa ʾalif maqṣūra ||ā / aỳ ||[aː] |
|دِي ||kasra yāʾ ||ī / iy ||[iː] |
|دُو ||ḍamma wāw ||ū / uw ||[uː] |
|tanwiin letters: |
|ً, ٍ, ٌ ||used to produce the grammatical endings /an/, /in/, and /un/ respectively. ً is usually used in combination with ا (اً) or taa marbuta. |
Syllabation signs and others
ّ shadda marks the gemination (doubling) of a consonant; kasra (when present) moves to between the shadda and the geminate (doubled) consonant.
An Arabic syllable can be open (ended by a vowel) or closed (ended by a consonant).
- open: CV[consonant-vowel] (long or short vowel)
- closed: CVC (short vowel only)
When the syllable is closed, we can indicate that the consonant that closes it does not carry a vowel by marking it with a sign called sukūn, which takes the form "°", to remove any ambiguity, especially when the text is not vocalised: it's necessary to remember that a standard text is only composed of series of consonants; thus, the word qalb, "heart", is written qlb. Sukūn allows us to know where not to place a vowel: qlb could, in effect, be read /qVlVbV/, but written with a sukūn over the l and the b, it can only be interpreted as the form /qVlb/ (as for knowing which vowel to use, the word has to be memorised); we write this قلْبْ.
You might think that in a vocalised text sukūn is not necessary, because the lack of vowel after a consonant might be signalled by simply not writing any mark above it, so قِلْبْ would be redundant. That is not so because such a convention ("lack of any vowel mark means lack of vowel sound") does not exist: k + u + t + b may indeed be read "kutib". Such a rule would make sense if everybody writing a vowel mark were forced to write all vowel marks in the same word, and that is not the case. In fact, you may write as many or as few of the vowel marks as you like.
In the Qur'an, however, all vowel marks must be written: there, sukuun over a letter (other than the alif indicating long "a") indicates that it is pronounced but not followed by a short vowel, while the lack of any sign over a letter (other than alif) indicates that the consonant is not pronounced.
Outside of the Qur'an, putting a sukuun above a yaa' which indicates long ee, or above a waaw which stands for long oo, is extremely rare, to the point that yaa with sukuun will be unambiguously read as the diphthong ai (as in English "eye") and waaw with sukuun will be read au (as in English "cow").
So, the word zauǧ, "husband", can be written simply zwǧ : زوج (which might be also read "zooj" if such a word existed); or with sukūn زوْجْ which is unambiguously "zowj"; or with sukūn and vowels: زَوْج.
The letters mwsyqā (موسيقى with a ʾalif maqṣūra at the end of the word) will be read most naturally as the word "mooseekaa" ("music"). If you were to write sukuuns above the waaw, yaa and alif, you'd get وْسيْقىْ, which looks like "mowsaykay". (note that an ʾalif maqṣūra is an alif and never takes sukūn.
You cannot place a sukuun on the final letter j of "zawj" even if you don't pronounce a vowel there, because fully vocalised texts are always written as if the ighraab vowels were in fact pronounced, and this word can never have a sukuun as an ighraab. Let's take the sentence "ahmad zawj sharr", meaning "Ahmed is a bad husband". The theoretical pronunciation with the ighraab vowels is "ahmadun zaujun sharrun". Interestingly, regardless of the fact that most people say "ahmad zauj sharr", you cannot write the mark for sukuun over that j; you either leave it markless, or use the mark for "un". By the same token, you can leave the final r of this sentence either completely unmarked or topped with a shadda plus "un", but a sukuun never belongs there, regardless of the fact that the only correct pronunciation of "sharrun" at the end of an utterance is "shar".
There are two kinds of numerals used in Arabic writing; standard Iran, Pakistan and India. In Arabic, these numbers are referred to as "Indian numbers" (أرقام هندية). In most of present-day North Africa, the usual Western numerals are used; in medieval times, a slightly different set (from which, via Italy, Western "Arabic numerals" derive) was used.
|Standard numerals |
|٠ ||0 |
|١ ||1 |
|٢ ||2 |
|٣ ||3 |
|٤ ||4 |
|٥ ||5 |
|٦ ||6 |
|٧ ||7 |
|٨ ||8 |
|٩ ||9 || |
|EastArab numerals |
|۰ ||0 |
|۱ ||1 |
|۲ ||2 |
|۳ ||3 |
|۴ ||4 |
|۵ ||5 |
|۶ ||6 |
|۷ ||7 |
|۸ ||8 |
|۹ ||9 |
In addition, the Arabic alphabet can be used to represent numbers, a usage rare today. This usage is based on a different and older (abjad) order of the alphabet: أ ب ج د ه و ز ح ط ي ك ل م ن س ع ف ص ق ر ش ت ث خ ذ ض ظ غ. (In the Maghreb and West Africa, a few of these letters are transposed. See the article on abjad. ) أ alif is 1, ب ba is 2, ج jim is 3, and so on until ي ya = 10, ك kaf = 20, ل lam = 30, ... ر ra = 200, ..., غ ghayn = 1000. This is sometimes used to produce chronograms.
The Arabic alphabet can be traced back to the alphabet used to write the Nabataean dialect of Aramaic, itself descended from Phoenician (which, among others, gave rise to the Greek alphabet and, thence, to Etruscan and Latin letters.). The first known text in the Arabic alphabet is a late fourth-century inscription from Jabal Ramm (50 km east of Aqaba), but the first dated one is a trilingual inscription at Zebed in Syria from 512. However, the epigraphic record is extremely sparse, with only five certainly pre_Islamic Arabic inscriptions surviving, though some others may be pre_Islamic. Later, dots were added above and below the letters to differentiate them (the Aramaic model had fewer phonemes than the Arabic, and some originally distinct Aramaic letters had become indistinguishable in shape, so in the early writings 15 distinct letter_shapes had to do duty for 28 sounds!) The first surviving document that definitely uses these dots is also the first surviving Arabic papyrus (PERF 558), dated April 643, although they did not become obligatory until much later.
Yet later, vowel signs and hamzas were added, beginning sometime in the last half of the sixth century, roughly contemporaneous with the first invention of Syriac and Hebrew vocalization. Initially, this was done by a system of red dots, said to have been commissioned by an Iraq, Hajjaj ibn Yusuf: a dot above = a, a dot below = i, a dot on the line = u, and doubled dots gave tanwin. However, this was cumbersome and easily confusable with the letter-distinguishing dots, so about 100 years later, the modern system was adopted. The system was finalized around 786 by al-Farahidi.
Arabic alphabets of other languages
Arabic script is not used solely for writing Arabic, but for a variety of languages. In each language, it has been modified to fit the language's sound system. There are phonemes not found in Arabic, but found in, for instance, Persian, Malay and Urdu. For example, the Arabic language lacks a "P" sounding letter, so many languages add their own "P" in the script, though the symbol used may differ between languages. These modifications tend to fall into groups; so all the Indian and Turkic languages written in Arabic tend to use the Persian modified letters, whereas West African languages tend to imitate those of Ajami, and Indonesian ones those of Jawi. The script in which the Persian modified letters are used, is called Perso-Arabic script by the scholars.
The Arabic alphabet is currently used for:
- Hausa for many purposes, especially religious (known as Ajami);
In the past, it has also been used to represent other languages:
- Fulani, known as Ajami;
- Sanskrit has also been written in Arabic script, though it is more well known as using the Devanagari script - the same script used for writing the Hindi language.
- Tatar (iske imlâ) before 1928 (changed to Latin), reformed in 1880's, 1918 (deletion of some letters);
Computers and the Arabic alphabet
The Arabic alphabet can be encoded using several character sets, including ISO-8859-6 and Unicode, in the latter thanks to the "Arabic segment", entries U+0600 to U+06FF. However, neither of these sets indicate the form each character should take in context. It is left to the rendering engine to select the proper glyph to display for each character.
When one wants to encode a particular written form of a character, there are extra code points provided in Unicode which can be used to express the exact written form desired. The Arabic presentation forms A (U+FB50 to U+FDFF) and Arabic presentation forms B (U+FE70 to U+FEFF) contain most of the characters with contextual variation as well as the extended characters appropriate for other languages. These effects are better achieved in Unicode by using the zero width joiner and non-joiner, as these presentation forms are deprecated in Unicode, and should generally only be used within the internals of text-rendering software, when using Unicode as an intermediate form for conversion between character encodings, or for backwards compatibility with implementations that rely on the hard-coding of glyph forms.
Finally, the Unicode encoding of Arabic is in logical order, that is, the characters are entered, and stored in computer memory, in the order that they are written and pronounced without worrying about the direction in which they will be displayed on paper or on the screen. Again, it is left to the rendering engine to present the characters in the correct direction, using Unicode's bi-directional text features. In this regard, if the Arabic words on this page are written left to right, it is an indication that the Unicode rendering engine used to display them is out-of-date. For more information about encoding Arabic, consult the Unicode manual available at http://www.unicode.org/
This article contains major sections of text from the very detailed article Arabic alphabet/from the French Wikipedia, which has been partially translated into English. Further translation of that page, and its incorporation into the text here, are welcomed.