Arabic words can appear very different when written in Roman letters.

The Arabic word العربية may be represented as al-ʻArabīyah, al-ʿArabiyya, al-ʿarabīya, or simply al-Arabiya. القاهرة can appear as al-Qāhirah, Al Qāhirah, Cairo, or a dialect-influenced spelling. محمد is commonly written Muhammad, but established variants include Mohammed, Mohamed, Mohammad, and Mehmet in other linguistic traditions.

These forms do not all follow the same rules.

Some systems attempt to preserve every Arabic letter. Others prioritize Modern Standard Arabic pronunciation, geographic-name consistency, library cataloging, academic readability, or convenient typing without specialist characters.

The major Arabic romanization traditions include:

• ISO 233
• ISO 233-2 simplified transliteration
• ALA-LC Arabic romanization
• BGN/PCGN geographic-name romanization
• DIN 31635
• IJMES scholarly transliteration
• Simplified popular and media spellings
• Informal Arabic chat alphabets

No single system is best for every purpose.

A reversible scholarly transliteration may be ideal for a manuscript catalog but inconvenient on a road sign. A simplified spelling may be readable in a news article but incapable of reconstructing the Arabic source. A pronunciation-oriented form may require vowels that are not written in the original text.

The correct system depends on the language variety, source text, audience, institutional standard, and amount of acceptable information loss.

Quick Comparison

ISO 233:1984 remains the published stringent standard for transliterating Arabic characters into Latin characters, although ISO states that it is expected to be replaced by the developing ISO 233-1. ISO 233-2:1993 provides a separate simplified system specifically for the Arabic language.

What Is Arabic Romanization?

Arabic romanization is the representation of Arabic-script text with Latin letters.

Depending on the system, the process may represent:

• Visible Arabic letters
• Standard Arabic pronunciation
• Classical grammatical forms
• A regional spoken pronunciation
• A conventional personal name
• An officially standardized geographic name

This distinction matters because Arabic is normally written as an abjad. Its main letters primarily represent consonants and long vowels, while short vowels are usually omitted or shown only with optional marks. Unicode describes Arabic as normally unvocalized and encodes its vowel and pronunciation marks as combining characters placed above or below base letters.

As a result, Arabic romanization is often more than character substitution.

The unvocalized sequence كتب may potentially represent different words when vowel marks and grammatical context are absent:

• كَتَبَ — kataba — he wrote
• كُتِبَ — kutiba — it was written
• كُتُب — kutub — books

A converter cannot determine the correct Latin form from the three consonant letters alone without understanding the word, sentence, or source.

Transliteration vs Transcription in Arabic

Arabic romanization may be transliteration-oriented or transcription-oriented.

Transliteration

A transliteration system follows the written Arabic characters.

Its goals may include:

• Distinguishing every Arabic letter
• Preserving spelling
• Supporting reverse conversion
• Maintaining consistent bibliographic records
• Allowing structured machine processing

A strict transliteration should not silently add information that the source does not contain.

If short vowels are absent, the system may need a dictionary, grammatical analysis, or a fully vocalized reference before producing a complete readable form.

Transcription

A transcription represents pronunciation.

It may:

• Supply omitted short vowels
• Reflect assimilation
• Follow regional pronunciation
• Ignore letters that are not pronounced
• Represent sounds rather than exact spelling

For example, the written definite article ال is structurally al-, but its l assimilates in pronunciation before a sun letter:

• الشمس may be pronounced ash-shams
• الرجل may be pronounced ar-rajul
• النور may be pronounced an-nūr

A spelling-oriented system may retain al-shams, while a pronunciation-oriented system may use ash-shams.

Hybrid Systems

Most practical standards are hybrids.

ALA-LC follows Arabic spelling closely but supplies established vowels and grammatical endings needed for usable catalog records. BGN/PCGN is based on fully pointed Modern Standard Arabic but applies special geographic-name rules and sometimes reflects the assimilation of the definite article. IJMES usually follows source spelling for scholarly terms while simplifying personal and place names for editorial readability.

Why Arabic Is Difficult to Romanize

Arabic presents several challenges not encountered in simple alphabet-to-alphabet conversion.

1. Short Vowels Are Usually Omitted

Arabic has three basic short vowels:

• fatḥah: a
• kasrah: i
• ḍammah: u

These are normally represented with optional marks:

• ـَ
• ـِ
• ـُ

Most everyday Arabic writing omits them. Unicode notes that Arabic is usually written without full vocalization, while the BGN/PCGN geographic-name standard warns that reliable romanization often requires knowledge of the fully pointed spelling.

A tool that receives مدرسة may need to know whether the intended reading is madrasah, madrasa, or a dialectal alternative. The source letters alone do not show every vowel.

2. One Letter Can Have Several Functions

The letters و and ي can represent consonants, long vowels, or parts of diphthongs:

The ALA-LC table explicitly distinguishes these contextual functions and gives examples such as dalw, yad, ṣūrah, īmān, kitāb, aw, and ay patterns.

3. Arabic Contains Sounds Without Simple English Equivalents

Several Arabic consonants do not have one obvious Latin-letter representation.

Examples include:

• ح
• خ
• ص
• ض
• ط
• ظ
• ع
• غ
• ق

Systems solve this problem with:

• Dots below letters
• Macrons below letters
• Digraphs
• Modifier letters
• Apostrophe-like symbols
• Simplified approximations

For example, خ may appear as:

• kh
• ḫ
• ẖ

The letter ع may be represented by:

• ʻ
• ʿ
• ‘
• An apostrophe
• Nothing in a simplified spelling

These symbols are not interchangeable in a strict technical system.

4. Hamza Has Several Written Forms

Hamza represents a glottal stop and can appear:

• Independently: ء
• Above alif: أ
• Below alif: إ
• Above wāw: ؤ
• Above or connected with yāʾ: ئ
• With maddah: آ

Unicode notes that hamza handling is structurally complex because it can appear as an independent element, a combining mark, or part of a precomposed character.

Many romanization standards omit word-initial hamza but represent medial or final hamza.

ALA-LC examples include:

• أسد → asad
• مسألة → mas’alah
• مؤتمر → mu’tamar

The ALA-LC rules omit initial hamza and represent medial or final hamza with a distinct sign.

5. Hamza and ʿAyn Must Remain Distinct

Hamza ء and ʿayn ع are different Arabic consonants.

Scholarly systems commonly distinguish them as:

• Hamza: ʾ, ʼ, or ’
• ʿAyn: ʿ, ʻ, or ‘

For example:

The BGN/PCGN rules explicitly warn that the symbols for hamza and ʿayn must be carefully distinguished, while IJMES requires authors to preserve both characters in scholarly terms and most names.

Using a plain keyboard apostrophe for both destroys the distinction and can create search, sorting, and conversion problems.

6. The Definite Article Changes in Pronunciation

The Arabic definite article is written ال.

Before a moon letter, the l is pronounced:

• القمر → al-qamar
• الكتاب → al-kitāb
• البحر → al-baḥr

Before a sun letter, the l assimilates to the following consonant:

• الشمس → pronounced ash-shams
• الرجل → pronounced ar-rajul
• النور → pronounced an-nūr
• الصحراء → pronounced aṣ-ṣaḥrāʾ

Systems disagree about whether the Roman form should follow spelling or pronunciation.

ALA-LC always writes the article as al-, even before sun letters, and joins it to the following word with a hyphen. Its official examples include al-kitāb al-thānī, al-ittiḥād, and al-ḥurūf al-abjadīyah.

BGN/PCGN normally reflects assimilation in geographic names and writes the article as a separate element without a hyphen, producing examples such as Al Baḩrayn, Ar Rawḑah, As Sulaymānīyah, and Ash Shām.

IJMES generally preserves al- rather than changing it to match sun-letter pronunciation, but applies its own editorial rules for capitalization, elision after prefixes, names, titles, and accepted English forms.

7. Tāʾ Marbūṭah Changes According to Grammar

The letter ة, called tāʾ marbūṭah, often appears at the end of feminine nouns and adjectives.

Its romanization depends on the system and grammatical context.

In ALA-LC:

• It is normally written h in a standalone or definite form.
• It becomes t in a construct or iḍāfah relationship.
• It may become tan in an adverbial form.

Examples include:

• رسالة → risālah
• وزارة التربية → Wizārat al-Tarbiyah
• مرآة الزمان → Mir’āt al-zamān

These rules reflect grammatical structure rather than simple letter replacement.

IJMES uses a different editorial convention. It normally renders Arabic tāʾ marbūṭah as a, but uses at in an iḍāfah construction.

BGN/PCGN may use ah, at, āh, or āt depending on position, the preceding vowel, and whether the word occurs in an iḍāfah construction.

Thus, the same Arabic word may end in -ah, -a, -at, or another form under different systems.

8. Shaddah Represents Consonant Doubling

The mark ّ, called shaddah, indicates a doubled consonant.

Examples include:

• معلّم → muʿallim
• محمّد → Muḥammad
• مكّة → Makkah
• سنّة → sunnah

Because shaddah is often omitted in unvocalized writing, a converter may need lexical knowledge to determine whether a consonant should be doubled.

Failure to preserve doubling can change meaning or produce a nonstandard name.

9. Alif Maqṣūrah Resembles Dotless Yāʾ

The letter ى, called alif maqṣūrah, occurs at the end of certain words and represents a long ā sound.

Examples include:

• موسى → Mūsá, Mūsā, or Musa, depending on the system
• على → ʿAlī when it is the name علي, but ʿalá/ʿalā for the preposition على
• هدى → Hudá or Hudā

ALA-LC commonly uses an acute accent in forms such as ilá and Mūsá, while other scholarly systems prefer ā. BGN/PCGN also assigns a special form to alif maqṣūrah.

A converter must distinguish final ى from final ي and identify the actual word.

10. Dialects Differ from Modern Standard Arabic

Arabic includes many regional spoken varieties whose pronunciation differs from Modern Standard Arabic.

The letter ج may be pronounced approximately as:

• j in many standard and regional contexts
• g in much Egyptian Arabic
• Another affricate or fricative realization elsewhere

ق may be pronounced as:

• q
• g
• A glottal stop
• Another regional sound

The Arabic definite article, vowels, diphthongs, and consonant clusters also vary.

BGN/PCGN bases its Arabic geographic system as far as possible on fully pointed Modern Standard Arabic precisely because regional and idiosyncratic pronunciations make uniform geographic-name results difficult.

A dialect transcription should therefore be labeled by variety, such as Egyptian Arabic, Levantine Arabic, Gulf Arabic, Iraqi Arabic, or Moroccan Arabic. It should not be presented as though it were a standard transliteration of written Arabic.

Major Arabic Romanization Systems

ISO 233

ISO 233 is a stringent international transliteration standard for Arabic characters.

Its main objective is precise conversion between writing systems rather than easy pronunciation for general readers. It assigns distinctive Latin characters or marks to Arabic letters so that source-character differences can be preserved.

ISO 233:1984 remains published and current pending its expected replacement by the developing ISO 233-1. The standard was last reviewed and confirmed in 2017.

Characteristic ISO-style forms include:

The exact standard uses highly specific Unicode characters and distinctions, including special treatment of hamza and its carriers.

Advantages of ISO 233

ISO 233 is suitable when a project needs:

• Strict character differentiation
• Structured bibliographic data
• Reversible conversion
• Comparative script analysis
• Scholarly documentation
• Machine processing

Limitations of ISO 233

It is less suitable for:

• Tourism
• Public signs
• General news
• Users without specialist keyboards
• Systems that strip combining marks
• Readers expecting familiar English spellings

A strict ISO result should generally be displayed beside the original Arabic because the Latin symbols are not self-explanatory.

ISO 233-2

ISO 233-2:1993 provides a simplified transliteration system specifically for Arabic.

It reduces some of the complexity of the stringent 1984 standard and is intended to facilitate the handling of bibliographic information such as catalogs, indexes, and citations.

The simplified system is easier to use but necessarily preserves less information than a fully stringent transliteration.

It should be selected when:

• A recognized international framework is needed
• Full ISO 233 precision is impractical
• Bibliographic usability matters more than perfect reversibility

A system should always record whether it uses ISO 233 or ISO 233-2, because “ISO Arabic romanization” alone is ambiguous.

ALA-LC Arabic Romanization

ALA-LC is the Arabic romanization system approved by the American Library Association and the Library of Congress for bibliographic work.

The current Library of Congress collection identifies the Arabic table as the 2012 version, and the main romanization tables page was updated on May 11, 2026.

ALA-LC is designed for:

• Library catalogs
• Author and title records
• Authority control
• Bibliographies
• Collection searching
• Consistent citation

Characteristic ALA-LC mappings include:

ALA-LC preserves the definite article as al-, even before sun letters, and retains long-vowel marks. It also contains detailed rules for hamza, tāʾ marbūṭah, alif maqṣūrah, prefixes, grammatical constructions, capitalization, and vowel inference.

ALA-LC Examples

ALA-LC is precise and useful for catalog matching, but it is not intended as a simple travel-pronunciation system.

BGN/PCGN Arabic Romanization

The BGN/PCGN Arabic system is used for geographic names by the United States Board on Geographic Names and the United Kingdom’s Permanent Committee on Geographical Names.

The system was adopted by the BGN in 1946 and by the PCGN in 1956. The current published presentation was checked for validity and accuracy in November 2022. It applies to official geographic-name romanization in a specified set of Arabic-language territories, while several North African and other countries use official Roman-script sources instead.

The system is based as far as possible on fully pointed Modern Standard Arabic.

Examples in the official table include:

BGN/PCGN uses diacritics for several consonant distinctions but also uses familiar digraphs such as kh, sh, th, and dh. The system may use a middle dot when a sequence such as dh, kh, sh, or th must be distinguished from two separate Arabic consonants.

Advantages of BGN/PCGN

It is appropriate for:

• Official maps
• Gazetteers
• Geographic databases
• Government place-name records
• Cross-border geographic communication

Limitations

It should not automatically be used for:

• Personal names
• Book titles
• Academic terminology
• Dialect transcription
• Countries using official established Roman-script place names

A city’s internationally established English name may also differ from its systematic BGN/PCGN form.

DIN 31635

DIN 31635 is a German standard for transliterating Arabic-script alphabets used for Arabic, Persian, Ottoman Turkish, Kurdish, Urdu, and Pashto.

The current edition is DIN 31635:2011-07. DIN states that its Arabic and Persian provisions are based largely on scholarly recommendations adopted by the International Congress of Orientalists and that its tables are close to ISO 233, while differing in some additional rules.

DIN 31635 is widely recognizable in Arabic and Islamic studies because it uses forms such as:

This system is compact and structurally clear, especially for readers familiar with academic transliteration.

Advantages

• Distinguishes Arabic consonants precisely
• Common in German and European scholarship
• Suitable for dictionaries, critical editions, and linguistic work
• Closely related to established Oriental-studies practice

Limitations

• Requires specialist diacritics
• Not intended as an intuitive English pronunciation guide
• May differ from ALA-LC or IJMES editorial forms
• Should not be mixed casually with simplified name spellings

IJMES Transliteration

The International Journal of Middle East Studies uses a modified scholarly transliteration system for Arabic, Persian, and Turkish.

IJMES distinguishes between technical terms and proper names.

Its current guidance requires:

• Full diacritics for technical Arabic terms
• Preservation of ʿayn and hamza
• No ordinary diacritics on personal names, place names, organizations, or titles
• Accepted English spellings for prominent figures and familiar places
• Lowercase al- except where ordinary sentence rules require capitalization
• Source spelling rather than colloquial pronunciation in most scholarly contexts

Examples of accepted English words such as mufti, jihad, and shaykh are written without diacritics, while specialized terms may appear as ʿashāʾ or similarly fully transliterated forms.

The IJMES consonant chart uses:

• ث → th
• ح → ḥ
• خ → kh
• ذ → dh
• ش → sh
• ص → ṣ
• ض → ḍ
• ط → ṭ
• ظ → ẓ
• ع → ʿ
• غ → gh
• ء → ʾ

IJMES is often a practical choice for English-language Middle East studies because it preserves important Arabic distinctions without forcing full diacritics into every personal and place name.

Comparing the Systems

The precise output depends on vocalization and grammatical context, but the following table illustrates common stylistic differences.

The table should be treated as an overview rather than a substitute for each system’s full contextual rules.

Diacritics Used in Arabic Romanization

Arabic romanization uses several kinds of marks.

Macron

A macron marks a long vowel:

• ā
• ī
• ū

Examples:

• كتاب → kitāb
• كبير → kabīr
• نور → nūr

Removing the macron merges long and short vowels.

Dot Below

A dot below commonly identifies emphatic or pharyngeal consonants:

• ḥ
• ṣ
• ḍ
• ṭ
• ẓ

These are distinct Arabic letters, not stylistic pronunciation marks.

For example:

• س → s
• ص → ṣ

and:

• د → d
• ض → ḍ

Removing the dots makes different source letters look identical.

Macron Below or Line Below

Some systems distinguish consonants with a line below:

• ṯ
• ḏ
• ḫ

Other systems use digraphs instead:

• th
• dh
• kh

Caron

DIN-style scholarly systems may use:

• š for ش
• ǧ for ج

Modifier Letters

Special modifier characters represent hamza and ʿayn:

• ʾ
• ʿ
• ʻ
• ʼ
• ‘
• ’

A reliable digital system should use the exact Unicode character required by its standard rather than substituting a visually similar punctuation mark.

Popular Spellings vs Technical Romanization

Technical romanization and established English spelling are different concepts.

Examples include:

The common English form may be:

• An exonym
• A historical spelling
• A personal preference
• A form influenced by French or another European language
• A dialect-based pronunciation
• A passport or institutional spelling

A technical converter should not automatically replace an established personal or official form.

Personal Names

Arabic personal names require more than a letter table.

A name may contain:

• The definite article
• A patronymic such as ibn or bin
• A teknonym such as Abū
• A family name
• A tribal or geographic nisbah
• An honorific
• A multiword religious name

For example:

عبد الرحمن may appear as:

• ʿAbd al-Raḥmān
• Abd al-Rahman
• Abdul Rahman
• Abdurrahman
• Abdelrahman
• Abderrahmane

These forms can reflect technical standards, spoken assimilation, regional spelling traditions, or personal documentation.

Use the following priority:

1. The person’s official Latin spelling
2. The person’s stated preference
3. Established professional usage
4. An institutional style guide
5. A generated romanization only when no established form exists

The generated scholarly form should be stored as an alias rather than presented as a correction of someone’s identity.

Geographic Names

Arabic geographic names are especially complex because pronunciation and spelling vary across regions.

A place-name database may need to store:

• Official Arabic name
• Fully pointed Modern Standard Arabic form
• Official national Latin spelling
• BGN/PCGN form
• UN or other international form
• Conventional English exonym
• French or historical spelling
• Local dialect pronunciation

BGN/PCGN applies its Arabic system to specified territories but relies on official Roman-script sources for several countries, including parts of North Africa where established French-influenced spellings are common.

For example, the Moroccan city فاس is widely known as Fes or Fez, rather than only by a theoretical Modern Standard Arabic transliteration. The correct public form depends on the authority and intended audience.

Arabic Script Is Used for Languages Other Than Arabic

The Arabic-derived script is also used for languages including:

• Persian
• Urdu
• Pashto
• Sindhi
• Kurdish
• Uyghur
• Jawi Malay
• Several African Ajami traditions

These languages contain additional letters and different pronunciation rules.

A Persian word should not be converted with an Arabic-language table merely because it uses Perso-Arabic script. ISO publishes a separate Persian transliteration standard, and ALA-LC maintains separate tables for Persian, Urdu, Pashto, Sindhi, and other languages.

For example, the letter پ represents p in Persian and Urdu but is not part of the basic Standard Arabic alphabet. Persian also uses چ, ژ, and گ.

Language detection is therefore essential.

Informal Arabic Chat Alphabet

Informal online communication sometimes represents Arabic sounds with Latin letters and digits.

Common conventions include:

Examples might include:

• عربي → 3arabi
• حبيب → 7abib
• سؤال → so2al

This is not one standardized romanization system. It varies by region, dialect, platform, and user.

Arabic chat spelling may be useful as a search alias or sociolinguistic feature, but it should not be presented as ISO, ALA-LC, BGN/PCGN, or formal academic transliteration.

Common Romanization Errors

Adding Vowels Without Evidence

Unvocalized Arabic may support several readings.

A converter should not invent a confident full pronunciation without:

• Dictionary evidence
• Grammatical context
• A vocalized source
• A verified name record
• Human review

Confusing Arabic With Persian or Urdu

The script may look similar while the language and letters differ.

Using One Apostrophe for Everything

Hamza, ʿayn, quotation marks, and an ordinary apostrophe are not the same character.

Dropping Diacritics From the Only Stored Form

Store the complete standard form first. Generate an ASCII or simplified copy separately.

Reflecting Sun-Letter Assimilation in a Spelling-Based System

ALA-LC requires al-shams, not ash-shams, because it preserves the written article. BGN/PCGN or a pronunciation guide may behave differently.

Treating Tāʾ Marbūṭah as Always H

Its form changes according to the standard and grammatical construction.

Ignoring Shaddah

Consonant doubling can be linguistically meaningful and may be required in established names.

Applying Modern Standard Arabic to a Dialect Name Without Warning

A locally pronounced place or personal name may differ from its formal standard reading.

Replacing Official Names Mechanically

A passport spelling, company name, or famous conventional form should usually be preserved.

Mixing Standards

A form such as al-ʿArabīyah may combine one system’s ʿayn symbol with another system’s ending or article rules. Mixing is acceptable only when declared as a house style.

Choosing the Right Arabic Romanization System

Best Practices for Romanization.org

A dependable Arabic romanization tool should provide more than one unlabeled result.

Preserve the Original Script

The Arabic source must remain the authoritative record.

Identify the Language

The tool should distinguish:

• Modern Standard Arabic
• Classical Arabic
• A regional Arabic variety
• Persian
• Urdu
• Pashto
• Kurdish
• Another Arabic-script language

Show Vocalization Status

Label the input as:

• Fully vocalized
• Partially vocalized
• Unvocalized
• Automatically inferred

Name the Exact Standard

Use precise labels:

• ISO 233:1984
• ISO 233-2:1993
• ALA-LC Arabic 2012
• BGN/PCGN Arabic 1956, revised presentation 2019 and checked 2022
• DIN 31635:2011-07
• Current IJMES style

Separate Output Types

A useful result could include:

Show Inferred Vowels

Any short vowel not present in the source should be visibly marked as inferred or dictionary-derived.

Distinguish Hamza and ʿAyn

Use correct Unicode modifier letters and explain the difference.

Preserve Diacritics

Do not silently convert:

• ḥ to h
• ṣ to s
• ḍ to d
• ṭ to t
• ẓ to z
• ā to a
• ī to i
• ū to u

A simplified option can remove them, but the complete form should remain stored.

Support Alternative Article Styles

A comparison view should show whether the selected system uses:

• al-shams
• ash-shams
• Al Shams
• al-Shams

Record Provenance

Store:

• Source text
• Language
• Vocalization status
• Standard
• Standard version
• Dictionary or reference source
• Conversion date
• Manual corrections

Frequently Asked Questions

What is the standard system for Arabic romanization?

There is no single universal system. ISO 233 serves stringent transliteration, ALA-LC serves library cataloging, BGN/PCGN serves geographic names, and DIN 31635 and IJMES serve important academic contexts.

Why are short vowels missing from Arabic writing?

Arabic normally writes consonants and long vowels in the base text, while short vowels can be added as optional marks. Everyday Arabic is usually unvocalized.

Can Arabic be romanized automatically?

Yes, but fully readable romanization of unvocalized text may require morphological, syntactic, and lexical analysis. Research on automatic bibliographic romanization describes the task as requiring Arabic phonology, morphology, and sometimes semantics.

Is Arabic romanization reversible?

A stringent transliteration can be substantially reversible, but a pronunciation-oriented or simplified form usually is not. Missing short vowels, merged letters, dialect forms, and omitted diacritics reduce reversibility.

What is the difference between ʿ and ʾ?

ʿ normally represents ʿayn ع. ʾ normally represents hamza ء. They are separate Arabic consonants.

Why is Muhammad spelled in so many ways?

Different spellings reflect romanization systems, European-language conventions, regional pronunciation, historical usage, and personal preference.

Should it be Quran, Qur’an, Qurʾan or al-Qurʾān?

The appropriate form depends on style:

• Quran is a common simplified English form.
• Qur’an uses a punctuation apostrophe.
• Qurʾan preserves hamza with a specialist modifier.
• al-Qurʾān is a fully marked scholarly form with the article and long vowel.

Follow the selected publication or institutional standard consistently.

Is Cairo a romanization of القاهرة?

Al-Qāhirah is a systematic Arabic romanization. Cairo is the established English exonym.

Why is the article sometimes al- and sometimes ar-, ash- or an-?

The article is written ال, but its l assimilates in pronunciation before sun letters. Spelling-oriented systems may retain al-, while pronunciation-oriented systems may reflect assimilation.

Why does ة become h, a or t?

Its representation depends on the standard and grammatical position. It may be silent or vowel-like at the end of a phrase and pronounced t in a construct relationship.

Should personal names include diacritics?

Usually follow the person’s official or preferred spelling. Scholarly systems may show a technical form separately, but should not overwrite the established identity form.

Is BGN/PCGN appropriate for every Arabic place name?

No. It applies to specified geographic-name contexts, while some countries use official Roman-script forms instead.

Is Arabic chat alphabet a formal standard?

No. It is an informal collection of regional digital-writing conventions.

Conclusion

Arabic romanization is not a single letter-conversion problem.

It requires decisions about:

• Written spelling
• Unwritten short vowels
• Standard or dialect pronunciation
• Hamza and ʿayn
• Long vowels
• Emphatic consonants
• The definite article
• Sun-letter assimilation
• Tāʾ marbūṭah
• Shaddah
• Grammatical endings
• Personal preferences
• Geographic-name authorities
• Technical character support

The principal systems serve different goals.

ISO 233 emphasizes stringent international transliteration and character differentiation.

ISO 233-2 provides a simplified system for Arabic-language documentation.

ALA-LC supports library catalogs, authority control, and bibliographic access.

BGN/PCGN standardizes Arabic geographic names for mapping and government use.

DIN 31635 provides a precise academic framework widely associated with Arabic and Islamic studies.

IJMES balances scholarly accuracy with readable editorial practice.

A reliable Arabic romanization platform should never present one unexplained Latin form as the only correct result. It should preserve the Arabic script, identify the language and standard, mark inferred vowels, distinguish technical from conventional names, and explain what information has been omitted.

The most useful question is not:

How is this Arabic word written in English letters?

It is:

Which system, pronunciation, authority, and purpose should this Latin form represent?