Persian Script: What the Abjad Records

Persian is normally written in a right-to-left abjad: consonants and long vowels are represented much more consistently than the three short vowels. That description is more useful than saying that Persian “leaves out vowels,” because the writing still records enough structure—letters, long vowels, word spacing, affixes, and conventional spellings—for a reader to identify words systematically.

This page explains that basic contract between spelling and reading. It does not ask you to master every sound yet; the detailed sound system begins with Persian vowels, while the mechanics of reading complete words continue in Reading Your First Unpointed Words.

An abjad, adapted for Persian

The Persian script descends from the Arabic script, but Persian orthography is its own system. It has 32 letters, including پ pe, چ che, ژ zhe, and گ gâf, and it uses Persian conventions for letters such as ک and ی. Calling it “Arabic with four extra letters” hides the more important fact: Persian speakers use the shared script to encode Persian words, Persian morphology, and Persian pronunciations.

Ordinary Iranian Persian text is unpointed (written-standard; all modern registers). It records the consonantal frame of a word and usually records the long vowels â, i, u with ا or آ, ی, and و. The same letters ی and و can also represent the consonants y and v, and و can occur in spellings where its value is o. Context and learned word patterns resolve those possibilities.

In these words, the written ا, و, and ی make the long vowels visible. English spelling also requires readers to interpret letters in context—compare the vowel spellings in food, good, and blood—but Persian makes a different bargain: short vowels are normally absent rather than represented by ambiguous full letters.

The three short vowels are normally unwritten

Modern Iranian Persian has three short vowels, romanized here as a, e, o. Learning materials and dictionaries can show them with fathe, kasre, and damme, but ordinary books, messages, news sites, forms, and signs generally omit those marks (written-standard; all registers).

The spelling گل contains the consonants گ and ل. Its familiar reading gol “flower” supplies the short o from lexical knowledge. The same principle gives del “heart” from دل and mard “man” from مرد.

The vowel marks are not “missing letters” that a writer forgot. They are optional annotations outside normal spelling. A marked learning form such as گُل simply exposes information that fluent readers ordinarily recover without the mark. Fully marked forms are useful (pedagogical) and in this guide’s hidden audio field, but a fully marked newspaper would look unusual.

Some unpointed strings genuinely have more than one possible reading. کرم may be read as kerm “worm,” kerem “cream,” or karam “generosity,” depending on the sentence. English has comparable written homographs such as lead and read. A surrounding phrase normally decides immediately; when it does not, a writer may add a vowel mark or rephrase.

What spelling does preserve

Persian spelling carries more than a row of consonants. Four layers work together:

The ZWNJ is invisible but structurally important. In می‌روم, it keeps the prefix می‌ attached as part of one written word while preventing an unwanted cursive join. In کتاب‌ها, it does the same for the plural suffix. This is not decoration: searching, copying, line breaking, and automated processing all depend on the underlying characters. The dedicated spaces and ZWNJ page develops the convention.

Direction and connected shapes

Words and sentences advance from right to left. Within that line, most letters connect cursively to eligible neighbors, so a letter’s appearance changes according to its surroundings. The underlying letter does not change: ب in isolated, initial, medial, and final contexts is still one letter, not four alphabet entries.

For example, the word شب shab “night” is typed as ش followed by ب. The shaping engine displays the connected word in right-to-left order. You should learn the contextual shapes as predictable handwriting-like variants, not memorize each word as a picture. The next pages treat directionality and joining forms separately.

Persian ک and ی can look nearly identical to Arabic-script alternatives in some fonts, but their Unicode code points differ. The visual pair is ک (Persian kaf, U+06A9) versus ك (Arabic kaf, U+0643), and ی (Farsi yeh, U+06CC) versus ي (Arabic yeh, U+064A). Iranian Persian text uses the Persian code points. This affects search and data matching even when the glyphs appear close.

Common Mistakes

1. Reversing the underlying letter order

2. Putting spaces between letters because the script looks unfamiliar

3. Substituting Arabic code points for Persian ones

4. Replacing the ZWNJ with an ordinary space

5. Inventing full letters for omitted short vowels

Key Takeaways

• Persian runs right to left and uses contextual cursive shapes, but each letter has one underlying identity.
• Consonants and long vowels are normally visible; short a, e, o are normally recovered from vocabulary and context.
• Unwritten short vowels are a systematic feature, not forgotten letters and not a license to guess randomly.
• Spaces, ZWNJs, Persian ک and ی, and conventional spellings are meaningful parts of standard Iranian orthography.