Word-counting an Arabic paragraph looks the same as English — split on whitespace, count the pieces — but the details are trickier. Arabic uses different punctuation (؟ ، ؛), joins particles like ال, و, ف, ب, ك, ل directly onto the word they modify (so والكتاب is arguably two words and one root), and has bidirectional layout that can throw off cursor-based counters.

What this tool counts

• Words — split on whitespace and Arabic punctuation.
• Characters — with and without whitespace.
• Unique tokens — after optional prefix/suffix stripping.
• Approximate unique roots — with the built-in stemmer.

The (simple) stemmer

Toggle 'stem to roots' to strip common Arabic prefixes (ال, و, ف, ب, ك, ل) and suffixes (ة, ات, ون, ين, ه, ها) before counting unique tokens. It's a rule-based approximation — not a full morphological analyser like Farasa or MADAMIRA — but it groups الكتاب / كتاب / وكتاب / كتابه / كتابات under كتاب, which is usually enough for readability audits, keyword density on Arabic SEO, or 'is this essay too repetitive' checks.

When to use it

• Arabic SEO — keyword density on landing pages and blog posts.
• Homework / essay length — Arabic word counts on Google Docs vary by version.
• Content moderation — quick pattern of over-used words.
• Translation invoicing — many Arabic translation contracts price per word.