Romanized Arabic in English Texts — Part 4: Omitting, Capitalizing, and Alphabetizing the Definite Article

AElfwine Mischler

As an editor and indexer in Cairo, I often work on materials containing Arabic terms and names. The Arabic definite article, usually romanized as al-, although such a small word — only two letters, alif lam — often presents problems for writers and editors of English texts.

In Part 3, I talked about:

• assimilating with the following letter
• merging the article
• elliding the vowel

In this part, I talk about more difficulties with the definite article:

• dropping the article in names
• capitalizing
• alphabetizing

Dropping the Article in Names

Most newspapers and trade books drop the article from surnames when the surname alone is used on subsequent mention of a person. For example, Bashar al-Assad on first mention, and Assad on subsequent mention. If this is the style of the publisher for whom you are writing or editing, be consistent, but note that in scholarly works, styles often call for the article to not be dropped. In those instances, Bashar al-Assad is al-Assad on subsequent mention.

My Egyptian colleagues have often complained to me about styles that drop the definite article from names. They feel that the article is an integral part of their names. One colleague said that her family name El-Naggar was a Muslim family name and that Naggar was a Christian family name. Dropping the article was changing her identity. Some contemporary people spell their family name as one word without the hyphen (sometimes with camel capping, as Nobel Prize–winner Mohamed ElBaradei spells his surname) and thus ensure that the article is not dropped.

Capitalizing

My Egyptian colleagues also were adamant that the definite article in their names be capitalized, even when it was hyphenated to the main part of their name. However, most styles of romanization do not capitalize al- at the beginning of a name unless it comes at the beginning of a sentence or bibliography entry. Some styles capitalize al- only at the beginning of a sentence and not elsewhere.

When the definite article comes at the beginning of a book or journal title, styles vary. Some publishers lowercase the article in all cases, even when it comes at the beginning of a citation. Others capitalize the article if it is the first word in a title, but not elsewhere in the title.

For that matter, capitalization of other words in titles also varies among publishers. Some capitalize only the first word and proper nouns within the title, but not adjectives derived from proper nouns. The Chicago Manual of Style, 17th ed., for example, uses sentence style: “Capitalize only the first word and any proper nouns. This practice applies to titles of works as well as to names of journals and organizations” (Sec. 11.80). The International Journal of Middle East Studies (IJMES) says to follow English capitalization rules, which I would interpret to include capitalization of adjectives derived from proper nouns.

Once again, my advice to authors and copyeditors is to know what the publisher wants, be meticulous in creating your style sheet, and keep a sharp eye out for inconsistencies.

Alphabetizing

Arabic names beginning with the definite article are conventionally alphabetized by ignoring the definite article. Thus, for example, al-Nahhas is alphabetized under N. How the article is handled in alphabetical lists again varies from one publisher to another. Some will keep the article in its place but ignore it (a style often preferred by indexers), while others will detach it and add it at the end preceded by a comma: Nahhas, Mustafa al-. Indexes containing Arabic names should carry a note to that effect at the beginning to direct readers, especially in trade books. The convention is well known in scholarly circles, but a note is still useful for nonspecialist readers who might search the index.

Perhaps I have a streak of rebelliousness in me, or I am just influenced by my Egyptian colleagues who consider the article in their surnames to be integral to their identities. As more Arabs write their surnames as one unhyphenated unit, will the convention of ignoring the article in alphabetizing change? When I index a book with modern Arabic names, I itch to alphabetize all the names beginning with an article — whether it is attached or not — under A or E, depending on how the person spells his or her name. Otherwise, how will index users know (or remember if they have already read the book) that So-and-so spells his name as one word but What’s-her-name does not? Although it resulted in a long list of names under A in the index, I was quite pleased when the editor of The Oxford Handbook of Arab Novelistic Traditions requested that I not discount the article in names. However, book titles beginning with Al- had the article moved to the end of the entry.

One final note. The word for “clan,” Āl, is often written without the macron in trade books and newspapers, and might be confused for the definite article and wrongly attached to the following word with a hyphen. It appears in the name of the ruling family of Qatar, Āl Thānī, and also in the name of one sura of the Qur’an, Āl ʿImrān. If you should come upon such a name, it should be alphabetized under Al.*

*Heather Hedden, “Arabic Names,” in Indexing Names, ed. Noeline Bridge (Medford, NJ: Information Today, 2012).

Ælfwine Mischler (www.MischlerEditorial.com) is an American copyeditor and indexer in Cairo, Egypt, who has been the head copyeditor at a large Islamic website and a senior editor for an EFL textbook publisher. She often edits and indexes books on Islamic studies, Middle East studies, and Egyptology.

Romanized Arabic in English Texts — Part 3: Spelling the Definite Article

by Ælfwine Mischler

As an editor and indexer in Cairo, I often work on materials containing Arabic terms and names. The Arabic definite article is usually romanized as al-, but the vowel is sometimes written as e (especially common in Egyptian names) or u. Although it is such a small word — only two letters, alif lam — it often presents problems for writers and editors of English texts.

In this essay, I talk about these elements:

• assimilating with the following letter
• merging the article
• elliding the vowel

In Part 4, I will discuss these difficulties:

• dropping the article in names
• capitalizing the article
• alphabetizing names and words with the article

Assimilating with the Following Letter

Years ago when I joined the staff of a large Islamic website, it did not have a style guide, so I set out to write one in consultation with the heads of several departments. It was not easy because the website had a broad range of intended audiences and levels of formality between departments, and for technical reasons we could not use diacritics (which I felt were inappropriate for most of the audiences anyway). The Arabic definite article was the source of many arguments, which I lost. The books I now work on use the style that I prefer, so I am not constantly cringing as I edit.

The arguments were about what to do with lam, the letter that is usually written as l in English. Half the letters in Arabic are shamsiya letters (“solar” letters) and half are qamariya (“lunar” letters). If lam comes before a solar letter, it is assimilated to the letter following and is known as lam shamsiya (“solar lam”). “The sun” in Arabic, al-shams, is actually pronounced ash-shams. If lam comes before a lunar letter, it is pronounced as usual and is known as lam qamariya (“lunar lam”). “The moon,” al-qamar, is pronounced as it is spelled.

Most scholarly books and trade books ignore the lam shamsiya and do not show assimilation. To my mind, this is best for the average reader, who will perhaps recognize al- as a morpheme but be confused by its variants. The assimilation should be shown when the correct pronunciation is important, such as in transcribing poetry, prayers, or Qur’an. Authors of Islamic books might insist on showing the assimilation in all cases. If you are an author, you should, of course, check the publisher’s guidelines and discuss them with your editor if you have any disagreement. If you are a copyeditor and your author has shown assimilation of lam and the managing editor is OK with it, be sure it is done consistently.

In romanization, the l of the definite article assimilates with the following letters, with or without diacritics: t or th, d or dh, s or sh, z, r, l, n.

Merging the Article

The article is usually romanized as al- in scholarly texts, but individuals may write the vowel differently in their names, and the article may merge with the preceding word. A common Arabic male name consists of Abd (or ʿAbd) [ʿ 02bf] (slave) plus one of the names of God: for example Abd al-Aziz (or ʿAbd al-ʿAzīz) [ʿ 02bf, ī 012b] “slave of the Almighty.” An individual with such a name might spell it with Abdal or (more often) Abdul or Abdel as the first part, and the second part might be attached to the first with a hyphen or closed up. Thus, someone named Abd al-Aziz might spell his name Abdal-Aziz, Abdal Aziz, AbdalAziz, Abdalaziz, Abdul Aziz, Abdul-Aziz, AbdulAziz, Abdulaziz, Abd el-Aziz, Abdel Aziz, Abdel-Aziz, AbdelAziz, or Abdelaziz. The name Abdallah (or Abdullah) “slave of Allah” is often spelled as one name.

My experience has been that people with Arabic names who grow up in a country that uses the Latin alphabet are consistent in spelling their names, but people who grow up in a country that uses the Arabic alphabet are often inconsistent in romanizing their names. This can be a problem for researchers — those who publish under multiple spellings will not get all the credit they should, and those who are looking for a particular person have to search multiple spellings.

Your job as an editor is to check that the spelling of an individual’s name is consistent, even if two people with the same Arabic name spell their names differently. A carefully prepared style sheet is essential for this. As I mentioned in Part 1, your task is easier when editing scholarly works that use diacritics (where ʿAbd al-[name] is used for historical names), but, depending on the style guide, names of people from recent centuries may or may not be transcribed using those rules and thus may be variously romanized.

Elliding the Vowel of the Article

In Arabic script, some conjunctions and prepositions are inseparable from the following word, and in most transcription systems these are shown with a hyphen: bi-, wa-, li-, la-, etc. The vowel of the definite article is not pronounced. Whether and how this ellision is shown in transcription varies from one system to another, giving writers and editors one more thing to watch for.

The International Journal of Middle East Studies (IJMES) gives the following examples in its guidelines: “fī al-ʿirāq wa-miṣr” (in Iraq and Egypt; fī is not an inseparable prefix in Arabic script) but “fī miṣr wa-l-ʿirāq” (in Egypt and Iraq). However, the Brill Encyclopaedia of Islam 3 differentiates between prefixes that keep the alif or delete it in Arabic script, and gives these examples in its Instructions for Authors: “wa-l-kitāb, fī l-masjid, Muḥyī l-Dīn, bi-l-kitāb, but lil-masjid.” Yet another transcription system shows the ellision with an apostrophe: wa-’l-kitāb, fī ’l-masjid. In this case, the author and copyeditor must also ensure that the symbol for hamza (ʾ) is not used where an apostrophe should be.

Part 1 of this series discusses the reasons for various spellings of Arabic names and terms, and Part 2 discusses some other challenges that authors and copyeditors might have. Part 4 will provide more discussion of the definite article.

Ælfwine Mischler is an American copyeditor and indexer in Cairo, Egypt, who has been the head copyeditor at a large Islamic website and a senior editor for an EFL textbook publisher. She often edits and indexes books on Islamic studies, Middle East studies, and Egyptology.

Romanized Arabic in English Texts — Part 2: Other Challenges for Editors

By Ælfwine Mischler

I have edited many articles and books with Arabic names and terms. Because of my language skills, I can often answer questions about romanized Arabic that colleagues ask in e-groups or to me directly. For example, a copyeditor once complained about the variant spellings her author was using, such as Kamal and Kamel, and asked which way she should fix them. Editor, beware! There are a lot of pairs or triplets of Arabic names that you might think are the author’s careless spelling when in fact they are different names. Don’t be too quick to correct, but do query.

Kamal and Kamel (or Kamil) are different names, Kamāl and Kāmil, respectively. Another common pair is Salah and Saleh (or Salih), which are Ṣalāḥ and Ṣāliḥ, respectively. There are several names romanized with a-m-r in English, but not all are from the same root: Amr or ʿAmr; Ammar or ʿAmmar; Amir or Ameer; Aamir or Amir (see image below). The last is rare but I came upon it while doing a quality control edit: there was a caliph known as al-Aamir. That one is hard to spell without diacritics; the double a is ugly, but Aamir and Amir have different meanings and should be distinguished.

Arabic names

And be aware that in addition to there being multiple ways to spell Muhammad, there is another name that you might think is a typo but is not: Muhannad.

What Does All This Mean for a Copyeditor?

If the text is academic English, romanization with diacritics will eliminate the ambiguities between similar names and terms. However, you do have to keep a careful watch for mistaken variants, such as an author forgetting a macron or a dot under a letter. In my experience, software to catch inconsistencies does not catch them if they involve special characters with diacritics.

Whether or not the author uses diacritics, keep a detailed style sheet — which you should do in any case. Record the first instance of every romanized name or term, and check every subsequent instance of it against your record. If there is any variation, correct it if you know enough Arabic to check it and do so, or else just flag it for the author to check and put the variant spelling on the style sheet. There is always the possibility that there are in fact two different but similar names or terms.

Dealing with terms and with names from the classical era of Islam is easier in that there should not be variation within a text. A name such as Yūsuf might be spelled Yusuf without diacritics, but if there are multiple people with the same name in the text, the name should be spelled the same way for all of them.

However, the same name from the current or recent centuries might have various spellings for the reasons I gave in Part 1. Without diacritics, Yūsuf could be spelled, for example, Yusef, Youssef, Yousef, or Yousuf. Of course, if there is a common or preferred spelling of a particular person’s name — especially if that person wrote his or her name in the Latin alphabet — that is the spelling that should be used.

When you are editing, if the variant spellings of a name are clearly referring to the same person but you do not know which spelling is correct or preferred, keep a record of them and query the author. If they are not referring to the same person, the “variants” might in fact be two different names, as I noted above. If you are editing a memoir, history, or other material where there are several people with the same name but different spellings, make a note in your style sheet to identify each person (“sister of the author,” “financial minister,” etc.).

Names with Abu

Another source of apparent inconsistency is names that in Arabic change the final vowel in the genitive. Many names are formed with Abu (Abū) plus another name, for example Abu Bakr (Abū Bakr) and Abu Taleb (Abū Ṭālib). In Arabic, the nominative Abū changes to Abī in the genitive, but in English the nominative ending u/ū should be retained. Untrained translators often keep the Arabic genitive ending i/ī when the name follows a preposition — for example, “She gave the money to Abi Bakr” rather than “to Abu Bakr” — but I correct Abi to Abu.

In most transcriptions an exception is made when Abu is preceded by ibn (“son”) or bint (“daughter”). Then the genitive is kept because the full name in Arabic will always have the genitive: Asma’ bint Abi Bakr, Ali ibn Abi Taleb. If you see bint Abi [something] or ibn Abi [something], you can keep Abi. If your author has consistently kept Abu after ibn or bint, query it. Some publishers might prefer to keep the nominative form of the name in all cases.

Virgules in Transcriptions

Another colleague asked about the use of a virgule in a transcription. Her author had followed the translation of a term with the romanized Arabic term followed by a virgule and more romanized Arabic. What did it mean and what should she do with it? For example, the author had written “companion (ṣāḥib/aṣḥāb)” and “word (kalimah/-āt).”

Fortunately for the editor, she did not have to do anything. The author was presenting the singular and plural forms of the words. Arabic has more than ten plural forms, so this is often necessary. In the first case the word has a broken plural in which letters are inserted, and in the second case the word has a regular feminine plural and only the ending is shown. This would be understood by the book’s intended audience. If you see a similar use of virgules when you are editing, it should not worry you.

Splitting Headaches

When you are proofreading, keep an eye out for Arabic names and terms that are split and hyphenated at the end of a line. It is best to not divide them except after the definite article al- or ibn. The letter pairs dh, gh, kh, sh, and th can represent the end of one syllable and the beginning of another, or they can be digraphs (two letters representing one sound), in which case they must not be split. If either letter of the above pairs has a dot or other diacritic, the word can be divided between the letters if necessary, according to New Hart’s Style, but The Chicago Manual of Style says the word can be divided only if both letters have a dot. Vowel digraphs (ou, oo, ee, aa) and diphthongs (ai, ay, aw, au) must never be split, and words must not be split before or after hamza, which is represented by an apostrophe or ʾ.

In Part 3: Spelling the Definite Article and Part 4: More on the Definite Article, I discuss some of the editing questions raised by al-, the Arabic definite article.

(For the first essay in this series, see: Romanized Arabic in English Texts — Part 1: Sources of Variations.)

Ælfwine Mischler is an American copyeditor and indexer in Cairo, Egypt, who has been the head copyeditor at a large Islamic website and a senior editor for an EFL textbook publisher. She often edits and indexes books on Islamic studies, Middle East studies, and Egyptology.

Romanized Arabic in English Texts — Part 1: Sources of Variations

by Ælfwine Mischler

As a native English speaker and editor in Cairo, I am often asked how to spell a name or Arabic word “in English,” meaning with the Latin alphabet and for an English-speaking audience. A child’s name on a birth certificate, a name and address on a visa application, Islamic terms in a web article or book. English-speaking copyeditors frequently joke or complain about the multiple spellings of Arabic names (“twelve ways to spell Muhammad”), and other difficulties might appear in a manuscript.

In this and subsequent essays, I explain why these differences occur, what you as an author or editor need to know, and how to use special characters if you (or your publisher) choose to include them.

Romanize, Spell, Transcribe, and Transliterate

Transcribe and transliterate are often used interchangeably, but if Wikipedia rather than Merriam-Webster is to be believed, transliterate is to represent a word letter by letter from one alphabet to another, whereas transcribe is to represent the sounds of a language. Arabic does not write short vowels, so (according to Wikipedia) the Arabic كتب (“to write”) should be transliterated as ktb, which does not give enough information to pronounce it. A transcription would show the vowels: kataba. Some transcriptions, especially for linguistic studies, use the International Phonetic Alphabet (IPA) to show the precise pronunciation. I do not have expertise in such transcriptions and do not discuss them.

For these essays I avoid using transliterate. I usually use transcribe for romanizing Arabic in more-precise ways using diacritics, such as in an academic text, and spell for romanizing without diacritics, such as in documents, newspapers, and trade books.

My focus is on Arabic represented by Latin letters in an English text, as that is my area of expertise. Speakers of other languages that use the Latin alphabet, such as German and French, will have their own ways of romanizing Arabic.

Why So Many Spellings?

As noted earlier, there are often multiple Latinized spellings of the same Arabic word. The four primary reasons for this are as follows:

1. Several Arabic phonemes don’t exist in English. Among these are pairs that when spoken sound very similar to non-Arabic speakers. In less-precise English spelling, these pairs are usually represented by the same letter, but in more-precise transcription the emphatic consonant is shown with a dot or other diacritic underneath (e.g., ḥ or ḩ versus h). Some other phonemes may be romanized with a digraph or diacritics (e.g. kh or ḫ).
2. The short vowels are not written in Arabic. Their pronunciation and romanization can vary across dialects.
3. Arabic names and terms are used by Muslims in many countries, and the spelling of phonemes varies across languages. For example, the Arabic letter shīn (ش) is written by English speakers as sh. French speakers render the same letter as ch, and Malaysians as sy. The letter jīm (ج) is written as j in English if it is pronounced as “soft g” (see below), but it will be spelled as dj where French influences the spelling. The Arabic ḍāḍ (ض), an emphatic letter, is often spelled dh by South and Southeast Asians, so they spell the month of fasting as Ramadhan.
   A doubled consonant in Arabic changes the meaning of a word, but sometimes names are romanized with doubled consonants to prevent a mispronunciation in English. The name Yāsir (ياسر) does not have a double consonant in Arabic, but a common spelling in English is Yasser because with a single s the name would likely be pronounced “Yazer.”
4. Another source of variation is that a few Arabic letters are pronounced differently in different dialects. The letter jīm (ج) is pronounced as English “hard g” in Cairo and northern Egypt. (Gamal Abdul Nasser’s first name begins with jīm.) Meanwhile, the letter qāf (ق) — an emphatic consonant pronounced in classical Arabic something like a k but with the tongue touching the palate farther back than for k — is pronounced as English “hard g” in some dialects. In the more popular academic transcription systems it is written as q, but many people use k in spelling their names (therefore not distinguishing qāf [ق] and kāf [ك]). In Cairo, qāf is usually pronounced as a glottal stop, but in romanized place names it is written as k, leading uninitiated tourists and expats to pronounce the neighborhood Dokki in a way that might confuse taxi drivers. The letter qāf is also one reason the world had such trouble spelling al-Gaddafi (al-Qaddafi, al-Qadhdhafi) (القذافي), the ousted Libyan leader. The Libyan dialect pronounces qāf as “hard g.”

So How Do You Spell…?

When I am asked how to spell an Arabic name or term in English, I usually ask what is being written (e.g., an academic paper, a trade book, a letter, a journal article) and who is the audience. If you’re deciding the spelling of your child’s name or writing something for a general reader, my answer will be different than if you are writing an academic paper.

If you are writing an academic paper, several different systems for romanizing Arabic exist. If you are writing or editing a scholarly text, you will have to follow the publisher’s preferred method. The Chicago Manual of Style recommends ALA-LC Romanization Tables: Transliteration Schemes for Non-Roman Scripts and the IJMES system of the International Journal of Middle East Studies. There are others.

If you are completing a form or writing for a general reader, I usually recommend a simplified spelling without using diacritics. If you are writing Islamic materials for a general audience, you might prefer to use diacritics for Islamic terms. Editors have asked me about spellings when their non-Arabic-speaking authors apparently took materials from various sources that used different systems of romanizing. Whether you choose to use diacritics or not, be consistent in spelling or transcribing terms.

Place names and personal names are more difficult. Names of recent and living people will unlikely be transcribed following an academic system, but there are exceptions. I indexed a book on Arabic literature in which the names of authors who had only published in Arabic were transcribed with diacritics, even if their works were well known in translation. Thus Nobel Prize–winning Naguib Mahfouz was written as Najīb Maḥfūẓ.

If recent and living people have a preferred spelling for their name, use that. If a personal name or place name appears in the news or is otherwise well known, use that spelling. If there is variation between news sources, choose one and stick to it.

In Part 2: Other Challenges for Editors, I discuss some other features of Arabic that may cause problems for editors and writers who are unfamiliar with the language.

Ælfwine Mischler is an American copyeditor and indexer in Cairo, Egypt, who has been the head copyeditor at a large Islamic website and a senior editor for an EFL textbook publisher. She often edits and indexes books on Islamic studies, Middle East studies, and Egyptology.