Arabic Alphabet Table

Because Arabic is an abjad - a consonant-first script descended from the same Phoenician family as Hebrew - and its 28 letters cover the consonants plus the three long vowels (alif ā, waw ū, ya ī), which double as the consonants ʾ, w and y. The short vowels exist as marks, not letters: fatha (a), damma (u) and kasra (i) ride above or below the consonant they follow. Everyday text omits them because native readers recover the vowels from grammar and context - the word مكتوب is unmistakably maktub (written) without marks. Teaching editions, the Quran and children's books add every mark (with sukun marking the absence of a vowel and shadda doubling a consonant), so the same sentence can be printed bare or fully vocalized depending on its audience.

Alif (ا), dal (د), dhal (ذ), ra (ر), zay (ز) and waw (و) connect only to the letter before them, never to the one after. The consequence is visible in every word that contains them: دار keeps its alif and dal at full width with a break between segments, while a same-length word of joining letters like كتاب flows as one connected ribbon. The rule matters for three practical reasons: handwriting (your pen lifts after these six), typography (ligatures and justified stretching stop at these letters), and reading (a word with two non-joiners shows two or three disconnected chunks - that is correct, not sloppy). Every other letter joins on both sides, and lam plus alif additionally fuse into the one mandatory ligature, لا.

Because the base skeletons predate the dots. The script's ancestors wrote 22-ish consonant shapes for a language with more contrasts, so the early Islamic era standardized dot marks to split each skeleton into distinct letters: the shallow bowl covers ب b, ت t and ث th (one, two, three dots below) and also ن n and ي y in other configurations; the deep bowl is ج j, ح ḥ and خ kh (dot below, none, above). The earliest dated Arabic inscriptions (like Zabad, 512 CE) are mostly dotless, dots appear irregularly in the 7th century, and the papyrus PERF 558 (643 CE) already shows them working - readers of dotless text were expected to disambiguate from context alone, which is why the Quran pushed dot conventions hard.

Every letter carries a numeric value from the old order of the alphabet (abjadī, not the modern kindergarten order): alif 1, ba 2, ja 3, dal 4, ha 5, waw 6, zay 7, ha 8, ta 9, ya 10, and onwards in tens and hundreds up to ghayn at 1000. Arabic mnemonics the whole sequence as eight nonsense words (abjad, hawwaz, hutti, kalaman...). The system is the ancestor of the Hebrew gematria style of letter arithmetic and is still alive in niche jobs: numbering list items and book chapters, marking historical years (a phrase whose abjad sum spells the year), and encoding short religious formulas. The modern Arabic-language order you learn first in school is alphabetical by shape, not by value - the table keeps to the school order but flags each letter's abjad value so both systems sit side by side.

No - they are hamza (ء), the glottal stop, wearing carriers. Hamza is a consonant sound with no seat of its own: at a word's start it rides alif (أ or إ, depending on the vowel), and mid-word it takes the carrier that matches the neighboring vowel - ؤ after u, ئ after i, and bare or on alif near a. The 28-letter count counts alif once and hamza zero times. Two other frequent suspects are also not separate letters: ة (ta marbuta, "tied ta") is ha's shape with two dots, the feminine ending that sounds t in construct state and h in pause; and ى (alif maqsura) is a final-form ya written without dots, sounding ā. Charts that count 29 or 32 letters are quietly including these contextual variants - this table keeps the 28 and explains the rest here.