# Modern Standard Arabic vs dialects: fusha, ammiya and Arabizi

**Which Arabic is written, Modern Standard Arabic or the dialects, and how do I cite and transliterate a source written in a dialect?** Modern Standard Arabic, or fusha, is the Arabic of books, newspapers, theses and formal speech; the dialects, or ammiya, are spoken first at home and are now widely written online, often in Arabizi. To cite a dialect source, keep its spelling exactly, transliterate a script title from its spelling, add a translation in brackets, and name the dialect.

Published 2026-10-03 by EdCitation. https://edcitation.com/newsletter/modern-standard-arabic-and-the-dialects-in-writing

Modern Standard Arabic (fuṣḥā, usually typed fusha) is the Arabic that gets written: books, newspapers, theses, the news read aloud. The dialects (ʿāmmiyya, typed ammiya, or dārija in Morocco) are what Arabic speakers learn first and speak at home, and they are now written too, in comments, chats and posts, often in Latin letters as Arabizi. This guide is published by EdCitation, whose tools look every reference up in the publisher's record and never write one.

We read every source below on 3 October 2026; translations of rules are ours. Arab universities' thesis conventions are in [academic writing in Arabic](https://edcitation.com/newsletter/academic-writing-in-arabic), and dialects in general in [dialects in academic writing](https://edcitation.com/newsletter/dialects-in-academic-writing). EdCitation's free [Find sources](https://edcitation.com/), [Cite a source](https://edcitation.com/cite) and [Verify references](https://edcitation.com/verify-references) were run on real dialect sources, with the results at the end.

## What is the difference between Modern Standard Arabic and the dialects?

Modern Standard Arabic is a codified variety learned at school for writing and formal speech; the dialects are the native, spoken varieties. Ferguson (1959) introduced the English term diglossia, modelled on the French diglossie, and made Arabic one of four defining cases, beside Modern Greek, Swiss German and Haitian Creole. Ferguson called the learned variety H (high) and the spoken ones L (low), and gave their Arabic names: al-fuṣḥā for H, al-ʿāmmiyya or ad-dārij for L.

In Ferguson's definition, H is highly codified, carries a large and respected written literature, is learned through formal education and serves most writing and formal speaking, yet nobody uses it for ordinary conversation. Ferguson's table calls the Arabic H "Classical"; Zaidan and Callison-Burch (2014) note that Modern Standard Arabic is a term used mainly by linguists and in education, and that constitutions simply say "the Arabic language".

### Is Modern Standard Arabic spoken?

Yes, where the occasion is formal. Ferguson's list puts the sermon, the speech in parliament, the university lecture and the news broadcast in H, and conversation with family and friends in L. Even lectures mix: a footnote adds that drills and explanations may run largely in L, especially in the sciences. Children learn the dialect at home and the standard at school.

The gap can be measured. In words from five-year-old Arabic speakers, about 20% were identical in spoken and standard forms, about 40% were cognates and about 40% unique to the spoken variety (Saiegh-Haddad & Spolsky, 2014, as cited in Andria et al., 2022). We could not read the original chapter.

### Is an Arabic dialect a separate language?

Linguistically it can be argued; socially it is resisted. Zaidan and Callison-Burch (2014) group the dialects as Egyptian, Levantine, Gulf, Iraqi and Maghrebi: Egyptian is the most widely understood, thanks largely to Egypt's films and television, while Moroccan is hard for a Levantine speaker. Scientifically, they write, the dialects could count as languages, a view not widely accepted by Arabic speakers. The ISO 639-3 code tables treat Arabic as a macrolanguage with 28 active individual languages, among them Standard Arabic (arb), Egyptian (arz), Levantine (apc) and Moroccan (ary) (SIL International, n.d.): the precise way to label which Arabic a corpus or file holds.

## Which Arabic is written where?

Modern Standard Arabic in nearly everything formal; the dialects in speech and, more and more, informal writing. The table sets Ferguson's 1959 list beside later evidence.

| Where | Variety | Source |
| --- | --- | --- |
| Sermon, political speech, university lecture, news broadcast | Standard (H) | Ferguson (1959) |
| Newspaper editorial, news story, personal letter | Standard (H), in 1959 | Ferguson (1959) |
| Poetry | Standard (H), though some is composed in L | Ferguson (1959) |
| Conversation, radio soap opera, cartoon caption, folk literature | Dialect (L) | Ferguson (1959) |
| Reader comments on newspaper sites | Dialect in about 40% of sentence labels | Zaidan and Callison-Burch (2014) |
| SMS, chat and social media | Dialect, in Arabic script, Arabizi or both | Bies et al. (2014); Bouamor et al. (2018) |
| An encyclopedia | Egyptian Wikipedia, written in Egyptian | Zaidan and Callison-Burch (2014) |
| A thesis written in Arabic | Standard; colloquial words avoided | University of Tabuk (2022) |
| Your reference list | Each source as published, checked against its record | [Verify references](https://edcitation.com/verify-references) |

### Comments, chats and print

The comment figures come from 1.4 million reader comments on three newspapers. Of the crowd labels on sentences sampled from them, 40.7% marked some dialect: 46.6% on Egypt's Al-Youm Al-Sabe', 40.1% on Saudi Arabia's Al-Riyadh and 36.8% on Jordan's Al-Ghad (Zaidan & Callison-Burch, 2014). Bouamor et al. (2018) describe dialectal Arabic as "emerging nowadays as the primary written language" of informal communication online.

Print is moving too. Høigilt and Mejdell (2017), an open-access volume on Egypt and Morocco, pairs survey data with chapters on satirical literature in Egyptian, new Egyptian print media and Moroccan zajal poetry; its description reports writing and attitudes far more open to vernacular forms than had been assumed. The chapters asked for a bot check on the day, so we quote no figures from them.

### What Ferguson predicted

Ferguson (1959) forecast, to about 2150, slow development toward "several standard languages", each built on a dialect, perhaps Maghrebi, Egyptian and Eastern, and asked linguists to study the mixed forms in between. Two thirds of a century on, the thesis is still in the standard; the comment box under the newspaper often is not.

## How is an Arabic dialect spelled when it is written?

However its writer chooses: no Arabic dialect has an official spelling. Ferguson (1959) noted that L had no settled orthography, and Habash et al. (2018) state that, typing slips aside, no spelling of a dialect word can be called incorrect. Their example, the Egyptian for "he does not say it", turned up online in 27 spellings: in a Google search on 29 September 2017, مبيقولهاش appeared about 26,000 times and ما بيقولهاش about 13,000.

Writers mostly borrow Standard Arabic's spelling rules (Zaidan & Callison-Burch, 2014), with local habits on top. Egyptian Wikipedia's main page is titled الصفحه الرئيسيه, with ه where Standard Arabic writes ة: الصفحة الرئيسية.

### CODA: one spelling for research

Researchers use CODA, the Conventional Orthography for Dialectal Arabic, first proposed for Egyptian in 2012; CODA* extends it to 28 city dialects from Rabat to Muscat (Habash et al., 2018). If you transcribe dialect speech in Arabic script, naming CODA* in your methods tells the reader that your spelling follows a convention and corrects nothing.

### Arabizi: a dialect in Latin letters

Arabizi writes Arabic, usually a dialect, in Latin letters and digits. Bies et al. (2014) explain it: each Arabic letter takes the Latin letter nearest in sound, and one with no near match takes a digit that looks like it, as 3 for ع, its mirror image. Choices vary by place: j for ج in the Levant but g in Egypt, sh for ش where English dominated and ch where French did. It can mislead: mesh may be English, or Arabizi for مش (not).

## How do you transliterate a dialect word or title?

By its spelling for a written title, and by a named colloquial system for speech. The *International Journal of Middle East Studies* guide says that, in doubt, you follow a word's spelling in the original script rather than its pronunciation (rule 9), and that colloquial transliterations follow a dictionary of that colloquial with English, in one system as close to IJMES as possible (rule 10) (International Journal of Middle East Studies, n.d.). Ferguson (1959) took the first route, transliterating Arabic spelling with the vowels supplied rather than transcribing Cairo speech.

### By kind of source

- **A title in Arabic script.** APA 7 transliterates a title in a non-Roman alphabet and adds an English translation in square brackets (Adams, 2021). Transliterate what is written: Egyptian Wikipedia's اللغه المصريه الحديثه becomes al-Lugha al-masriyya al-haditha.
- **Speech you quote or recorded.** Use a colloquial system and name it. IJMES names no dictionary; for Egyptian, Badawi and Hinds (1986) is one of the dialect dictionaries Zaidan and Callison-Burch (2014) cite.
- **A title already in Arabizi.** APA's rule covers non-Roman alphabets, so in our reading it stays as published, with a translation in brackets.
- **A person's name.** As that person writes it; see [names in references](https://edcitation.com/newsletter/names-in-references-across-cultures).

The systems are compared in [transliteration and romanisation in references](https://edcitation.com/newsletter/transliteration-and-romanisation-in-references), and Arabic titles in APA, MLA and Chicago in [citing Arabic sources in English papers](https://edcitation.com/newsletter/citing-arabic-sources-in-english-papers).

## How do you cite a source written in an Arabic dialect?

In the source's own format, with the dialect left exactly as published. None of the style pages we read gives a dialect a rule of its own, so your sentence says which variety it is.

### A published page in a dialect

APA cites a wiki page from a permanent link to the version you read, and asks students to check whether Wikipedia is acceptable for their paper (American Psychological Association, 2020). As evidence of written Egyptian, the page is a primary source:

> al-Lugha al-masriyya al-haditha [Modern Egyptian language]. (2026, June 2). In *Wikipedia*. https://arz.wikipedia.org/w/index.php?oldid=13070333

In the text: ("al-Lugha al-Masriyya al-Haditha," 2026). Say in your sentence that it is Egyptian Wikipedia, since the reference does not; [citing sources on Wikipedia](https://edcitation.com/newsletter/citing-sources-on-wikipedia) covers the rest. A song or film takes its description, such as [Song], in a second pair of brackets after the translation (Adams, 2021).

### Your own recordings

Transcribe in a stated system, CODA* in Arabic script or a named Latin scheme, and never turn dialect into Standard Arabic or mark it [sic]: [quoting dialect and non-standard speech](https://edcitation.com/newsletter/quoting-dialect-and-non-standard-speech) explains why, and [citing an interview transcript](https://edcitation.com/newsletter/how-to-cite-an-interview-transcript) covers the reference.

If you write the paper in Arabic, its prose is in the standard: the University of Tabuk (2022) guide lists "تجنُّب استعمال الكلمات العاميَّة" (avoiding colloquial words, our translation) among its style rules. In our view that governs your sentences, not your evidence: a quotation from a dialect source stays in dialect.

## How do I cite a dialect source, step by step?

1. Name the variety, Egyptian, Levantine, Gulf, Iraqi, Moroccan or another, in your sentence.
2. Find scholarship on it, in Arabic and English, with EdCitation's [Find sources](https://edcitation.com/).
3. Build the entry from the record with [Cite a source](https://edcitation.com/cite/apa): a DOI, an ISBN or the page's address.
4. Keep the title's spelling as published; transliterate a script title from its spelling with one system, and add a translation in square brackets.
5. Check names, editors and the permanent link by hand.
6. Paste the list into [Verify references](https://edcitation.com/verify-references) and settle every flag yourself.

## What did EdCitation return on dialect sources?

Entries from the record, dialect spelling intact, and names that needed a human eye. EdCitation is the right place to start: it looks each reference up in the publisher's record or a library catalogue and never writes one, which no chatbot can say. EdCitation never writes any part of anyone's work: the transliteration, the translation and the sentence naming the dialect stay yours. Find sources, Cite a source and Verify references are free with no account; [References from a file](https://edcitation.com/tools/references-from-a-file), in Pro at $8 a month ([pricing](https://edcitation.com/pricing)), checks every reference in an uploaded paper and matches each citation to the list. We ran the free three on 3 October 2026.

### Find sources

[Find sources](https://edcitation.com/) searches about 300 million published works. "الازدواجية اللغوية" (linguistic diglossia) returned 113 results; 89 of the first 97 had Arabic titles. "Arabic diglossia", sorted by citations, returned 4,849, led by Saiegh-Haddad's 2003 study of linguistic distance and learning to read, cited 387 times.

### Cite a source

From the web address, [Cite a source](https://edcitation.com/cite) read Egyptian Wikipedia's page and kept its spelling letter for letter, ه and all: "اللغه المصريه الحديثه. (2026, June 2). Wikipedia." with the live address. Add "In", the italics, the permanent link, the transliteration and the translation. Records needed checking too:

- ISBN 9781853410031 gave "Hinds, M., & El-Said, B. (1986)", because the catalogue record says "Badawi El-Said". The author is El-Said Badawi: write Badawi, E.-S., and take the order of names from your copy's title page.
- DOI 10.1017/s0142716403000225 gave "Saiegh–haddad, E. (2003)" from a record in capitals with an en dash: write Saiegh-Haddad, E.
- DOI 10.1163/9789004346178 named Høigilt and Mejdell as authors, as the record does; add (Eds.).
- For Ferguson's DOI the APA entry was right, but the RIS export kept the record's markup, `J2  - <i>WORD</i>`: delete the tags.

### Verify references

Of seven entries, six were verified, among them Ferguson (1959) without its DOI, the CODA* paper with all 17 authors, an Arabic-script article on diglossia, and Badawi and Hinds in the right name order: "Title, author and year match a library catalogue record." The same Arabic article, romanised, came back could not check, not not found: "This entry gives its title in romanised form or in translation, and the record may hold it only in the original script." Keep the DOI in a romanised entry, and see [checking a reference is real](https://edcitation.com/newsletter/how-to-check-a-reference-is-real).

## Quick questions

### Is Modern Standard Arabic spoken?

Yes, in formal settings such as news broadcasts, sermons and lectures. Nobody speaks it at home as a first language, which Ferguson (1959) made part of the definition of diglossia.

### What is the difference between fusha and ammiya?

Fusha (fuṣḥā) is the standard, written variety, learned at school. Ammiya (ʿāmmiyya) is the spoken dialect of a place, learned at home, with no official spelling.

### What is Arabizi?

Arabic, usually a dialect, written in Latin letters and digits, such as 3 for ع. Its spellings vary with country and with the writer's second language.

### How do I transliterate Egyptian Arabic in a reference?

Transliterate a written title from its spelling with IJMES or another named system. For speech, IJMES asks for one colloquial system taken from a dictionary of the dialect.

### Can EdCitation cite a source written in a dialect?

Yes. EdCitation's [Cite a source](https://edcitation.com/cite) builds the entry from a DOI, an ISBN or a web address in six styles and keeps a dialect title as published; you add the transliteration and translation.

## References

- Adams, A. (2021, August 23). When and how to transliterate titles in references. *APA Style*. [https://web.archive.org/web/20260805230910/https://apastyle.apa.org/blog/transliterated-titles-references](https://web.archive.org/web/20260805230910/https://apastyle.apa.org/blog/transliterated-titles-references)
- American Psychological Association. (2020). *Wikipedia entry references*. APA Style. [https://web.archive.org/web/20260617140517/https://apastyle.apa.org/style-grammar-guidelines/references/examples/wikipedia-references](https://web.archive.org/web/20260617140517/https://apastyle.apa.org/style-grammar-guidelines/references/examples/wikipedia-references)
- Andria, S., Madi-Tarabya, B., & Khateb, A. (2022). Behavioural and electrophysiological analyses of written word processing in spoken and literary Arabic: New insights into the diglossia question. *European Journal of Neuroscience, 56*(6), 4819–4836. [https://doi.org/10.1111/ejn.15781](https://doi.org/10.1111/ejn.15781)
- Badawi, E.-S., & Hinds, M. (1986). *A dictionary of Egyptian Arabic: Arabic–English*. Librairie du Liban. [https://openlibrary.org/isbn/9781853410031](https://openlibrary.org/isbn/9781853410031)
- Bies, A., Song, Z., Maamouri, M., Grimes, S., Lee, H., Wright, J., Strassel, S., Habash, N., Eskander, R., & Rambow, O. (2014). Transliteration of Arabizi into Arabic orthography: Developing a parallel annotated Arabizi-Arabic script SMS/chat corpus. In *Proceedings of the EMNLP 2014 Workshop on Arabic Natural Language Processing (ANLP)* (pp. 93–103). Association for Computational Linguistics. [https://doi.org/10.3115/v1/W14-3612](https://doi.org/10.3115/v1/W14-3612)
- Bouamor, H., Habash, N., Salameh, M., Zaghouani, W., Rambow, O., Abdulrahim, D., Obeid, O., Khalifa, S., Eryani, F., Erdmann, A., & Oflazer, K. (2018). The MADAR Arabic dialect corpus and lexicon. In *Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)* (pp. 3387–3396). European Language Resources Association. [https://aclanthology.org/L18-1535/](https://aclanthology.org/L18-1535/)
- Ferguson, C. A. (1959). Diglossia. *Word, 15*(2), 325–340. [https://doi.org/10.1080/00437956.1959.11659702](https://doi.org/10.1080/00437956.1959.11659702)
- Habash, N., Eryani, F., Khalifa, S., Rambow, O., Abdulrahim, D., Erdmann, A., Faraj, R., Zaghouani, W., Bouamor, H., Zalmout, N., Hassan, S., Al-Shargi, F., Alkhereyf, S., Abdulkareem, B., Eskander, R., Salameh, M., & Saddiki, H. (2018). Unified guidelines and resources for Arabic dialect orthography. In *Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)* (pp. 3628–3637). European Language Resources Association. [https://aclanthology.org/L18-1574/](https://aclanthology.org/L18-1574/)
- Høigilt, J., & Mejdell, G. (Eds.). (2017). *The politics of written language in the Arab world: Writing change*. Brill. [https://doi.org/10.1163/9789004346178](https://doi.org/10.1163/9789004346178)
- International Journal of Middle East Studies. (n.d.). *IJMES translation and transliteration guide* [PDF]. Cambridge University Press. [https://www.cambridge.org/core/services/aop-file-manager/file/630d1d754cf5970011527e65/IJMES-Translation-and-Transliteration-guide.pdf](https://www.cambridge.org/core/services/aop-file-manager/file/630d1d754cf5970011527e65/IJMES-Translation-and-Transliteration-guide.pdf)
- SIL International. (n.d.). *Arabic [ara]*. ISO 639-3 Registration Authority. [https://iso639-3.sil.org/code/ara](https://iso639-3.sil.org/code/ara)
- University of Tabuk. (2022). *دليل كتابة الرسائل والمشاريع العلمية بجامعة تبوك* [Guide to writing theses and research projects at the University of Tabuk] [PDF]. [https://www.ut.edu.sa/sites/default/files/2025-07/%D8%AF%D9%84%D9%8A%D9%84%20%D9%83%D8%AA%D8%A7%D8%A8%D8%A9%20%D8%A7%D9%84%D8%B1%D8%B3%D8%A7%D9%8A%D9%94%D9%84%20%D9%88%D8%A7%D9%84%D9%85%D8%B4%D8%A7%D8%B1%D9%8A%D8%B9%20%D8%A7%D9%84%D8%B9%D9%84%D9%85%D9%8A%D8%A9%20%D8%A8%D8%AC%D8%A7%D9%85%D8%B9%D8%A9%20%D8%AA%D8%A8%D9%88%D9%8312-03-%202022.pdf](https://www.ut.edu.sa/sites/default/files/2025-07/%D8%AF%D9%84%D9%8A%D9%84%20%D9%83%D8%AA%D8%A7%D8%A8%D8%A9%20%D8%A7%D9%84%D8%B1%D8%B3%D8%A7%D9%8A%D9%94%D9%84%20%D9%88%D8%A7%D9%84%D9%85%D8%B4%D8%A7%D8%B1%D9%8A%D8%B9%20%D8%A7%D9%84%D8%B9%D9%84%D9%85%D9%8A%D8%A9%20%D8%A8%D8%AC%D8%A7%D9%85%D8%B9%D8%A9%20%D8%AA%D8%A8%D9%88%D9%8312-03-%202022.pdf)
- Zaidan, O. F., & Callison-Burch, C. (2014). Arabic dialect identification. *Computational Linguistics, 40*(1), 171–202. [https://doi.org/10.1162/COLI_a_00169](https://doi.org/10.1162/COLI_a_00169)