A Cyrillic name can appear very different when converted into Roman letters.

The Russian surname Ельцин may be represented as Elʹcin, Elʹt͡sin, or Yel’tsin. Чайковский can become Čajkovskij, Chaĭkovskiĭ, or Chaykovskiy. The Ukrainian capital Київ may appear as Kyiv, while older international sources often use Kiev.

These differences are not merely spelling preferences. They come from romanization systems designed for different purposes.

Three of the most important standards are:

• ISO 9, a script-oriented international transliteration standard
• ALA-LC, a collection of language-specific systems used in library cataloging
• BGN/PCGN, systems and agreements developed for foreign geographic names

ISO 9 prioritizes consistent character conversion across Cyrillic alphabets. ALA-LC prioritizes bibliographic access, differentiation, and cataloging consistency. BGN/PCGN prioritizes standardized, usable spellings for maps, gazetteers, government documents, and geographic databases.

No system is universally best. The correct choice depends on whether the text is being prepared for scholarship, library metadata, mapping, public communication, historical research, identity records, or machine conversion.

Quick Comparison

ISO 9:1995 covers Cyrillic characters used by both Slavic and non-Slavic languages. The current published standard has a 2024 amendment, while a third edition was progressing through ISO’s approval process in July 2026. The draft retains broad coverage, describing 118 Cyrillic characters and character-plus-diacritic forms.

ALA-LC publishes separate tables for Russian, Ukrainian, Belarusian, Bulgarian, Serbian, Church Slavic, and non-Slavic languages written in Cyrillic, among others. BGN and PCGN likewise maintain language-specific systems and agreements rather than treating Cyrillic as one uniform alphabet.

Cyrillic Is a Script, Not a Single Alphabet

It is common to speak of “the Cyrillic alphabet,” but Cyrillic is more accurately a script family used by multiple languages.

Russian, Ukrainian, Belarusian, Bulgarian, Serbian, Macedonian, Mongolian, and many languages of the Caucasus and Central Asia use different subsets or extensions of Cyrillic. Unicode explicitly notes that Russian, Ukrainian, and Serbian use different collections of letters within the Cyrillic script.

Some letters occur in one language but not another:

• Ukrainian uses Ґ, Є, І, and Ї
• Belarusian uses Ў
• Serbian uses letters such as Ј, Љ, Њ, Ћ, and Џ
• Macedonian uses Ѓ, Ѕ, Ј, Љ, Њ, Ќ, and Џ
• Mongolian Cyrillic adds Ө and Ү
• Non-Slavic languages may contain additional modified Cyrillic characters

Even shared letters may represent different sounds or receive different Latin equivalents.

For example, the letter Г is usually romanized as g in Russian, but as h in the current Ukrainian national and BGN/PCGN system. Ukrainian also has Ґ, romanized as g.

This leads to the first essential rule:

Identify the language before selecting a Cyrillic romanization system.

A Russian table should not automatically be applied to Ukrainian, Bulgarian, Serbian, or a non-Slavic language merely because the text uses Cyrillic characters.

Romanization, Transliteration and Transcription

Cyrillic conversion can follow either writing or pronunciation.

Transliteration

Transliteration maps written characters from Cyrillic into Latin characters.

A transliteration system may aim to ensure that:

• The same Cyrillic letter receives the same Latin equivalent
• Different Cyrillic characters remain distinguishable
• The original text can be reconstructed
• Conversion can be automated consistently

ISO 9 is principally a transliteration standard.

Transcription

Transcription represents pronunciation rather than preserving every letter distinction.

A transcription may change according to:

• The source language
• Word position
• Neighboring letters
• Actual pronunciation
• The target reader’s language

Some practical systems for names and public communication contain transcription-like features.

Hybrid systems

ALA-LC and BGN/PCGN are best understood as practical transliteration frameworks with different levels of language-specific and contextual treatment.

ALA-LC generally remains closely tied to written characters because libraries need consistent access points. BGN/PCGN systems may introduce contextual spellings intended to make geographic names more readable and useful to English-language government users.

What Is ISO 9?

ISO 9 is the International Organization for Standardization’s standard for transliterating Cyrillic characters into Latin characters.

ISO 9:1995 replaced the 1986 edition and expanded the standard’s scope beyond Slavic Cyrillic to include Cyrillic characters used by non-Slavic languages. Its unified table includes 118 individual characters or characters carrying diacritics.

The current published framework consists of:

• ISO 9:1995
• ISO 9:1995/Amd 1:2024

A proposed third edition, ISO/DIS 9, passed the Draft International Standard consultation stage in July 2026 and was approved to proceed toward the Final Draft International Standard stage. Until a replacement is formally published, ISO 9:1995 with its amendment remains the published standard.

ISO 9’s Core Approach

ISO 9 is script-oriented rather than designed as an English pronunciation guide.

Its characteristic Russian mappings include forms such as:

The use of individual Latin letters with diacritics reduces ambiguity. A single Cyrillic character can usually be represented by one distinct Latin character rather than a digraph such as zh, kh, or shch.

ISO 9 Examples

These forms are systematic, but they may not be intuitive to readers unfamiliar with Slavic Latin alphabets.

For example:

• c represents Cyrillic Ц, not an English k or s
• č represents Ч
• š represents Ш
• j represents Й
• ŝ represents Щ

ISO 9 should therefore be treated as a conversion standard, not as a self-explanatory pronunciation guide.

Advantages of ISO 9

ISO 9 is especially useful when a project needs:

• Consistent conversion across many Cyrillic alphabets
• Precise differentiation of source characters
• Machine processing
• Round-trip transliteration
• Scholarly comparison
• Multilingual archival data
• Stable identifiers independent of English spelling conventions

Because the system is built around character distinctions, it can preserve information lost by more practical systems.

Limitations of ISO 9

ISO 9 may be unsuitable for a general English-speaking audience because:

• Many users do not know letters such as č, š, ŝ, ž, â, or û
• Specialized characters can be difficult to type
• The result may not resemble established international names
• Script-wide consistency does not necessarily reflect language-specific pronunciation
• Public authorities may require a national system instead

For example, Čajkovskij is systematic, but the established English name Tchaikovsky is more familiar. Neither should automatically replace the other in every context.

What Is ALA-LC Romanization?

ALA-LC Romanization Tables are transliteration schemes approved jointly by the American Library Association and the Library of Congress.

They are designed primarily for:

• Bibliographic descriptions
• Author and title access points
• Authority records
• Catalog searching
• Alphabetical arrangement
• Citations
• Data exchange between libraries

The Library of Congress maintains separate tables for individual Cyrillic-script languages. Its current list includes Belarusian, Bulgarian, Church Slavic, Russian, Serbian, Ukrainian, and a broader table for non-Slavic languages written in Cyrillic. Earlier versions are retained for reference but are not recommended for current cataloging.

ALA-LC Russian Mappings

The current Russian ALA-LC table is identified as the 2012 version.

Selected mappings include:

The table also covers obsolete Russian letters removed during the 1918 orthographic reform and directs catalogers to the Church Slavic table for other historical characters.

ALA-LC Examples

These spellings preserve distinctions useful in library records but can appear unfamiliar outside cataloging.

For example:

• Й becomes ĭ, not y or j
• Ц is represented by a tied t͡s
• Ю and Я are treated as i͡u and i͡a
• The soft and hard signs remain visible through prime-like marks

Why ALA-LC Uses Diacritics

Libraries must distinguish similar titles, authors, and names reliably.

Replacing every specialized character with ordinary ASCII sequences may produce collisions. Diacritics help preserve:

• Different source letters
• Historical characters
• Soft and hard signs
• Distinctions across related Cyrillic alphabets
• More dependable reverse identification

The Library of Congress revised its Russian table partly to improve reversibility and consistent searching.

ALA-LC Is Language-Specific

Unlike a universal script-wide conversion, ALA-LC uses separate rules for different languages.

For example:

The current Ukrainian table distinguishes Г as h and Ґ as g, reflecting Ukrainian rather than Russian orthography.

Advantages of ALA-LC

ALA-LC is usually the best choice for:

• North American library cataloging
• Bibliographies tied to library authority records
• Archival descriptions
• Scholarly citations using library conventions
• Matching existing catalog records
• Distinguishing historical Cyrillic spellings

Limitations of ALA-LC

ALA-LC may not be appropriate for:

• Road signs
• Tourist maps
• Passports
• Public geographic databases
• General news writing
• Users without diacritic support
• Search interfaces that fail to normalize specialized characters

A library form is an access and cataloging convention. It should not automatically replace an official place name or a person’s preferred Latin spelling.

What Are BGN/PCGN Systems?

BGN/PCGN refers to romanization systems and agreements jointly developed or approved by:

• The United States Board on Geographic Names
• The United Kingdom’s Permanent Committee on Geographical Names for British Official Use

Their purpose is to establish standardized Roman-script spellings for foreign geographic names written in non-Roman scripts or Roman alphabets containing special characters.

BGN/PCGN forms are used primarily for:

• Maps
• Gazetteers
• Geographic information systems
• Government publications
• Foreign-name databases
• Diplomatic and defense information
• Geographic intelligence

The NGA Geographic Names Server is the official repository of foreign geographic-name spellings approved by the U.S. Board on Geographic Names. It stores current names, variant spellings, historical forms, and non-Roman-script names.

BGN/PCGN Is Not One Universal Cyrillic Table

BGN/PCGN maintains separate systems for languages including:

• Russian
• Ukrainian
• Bulgarian
• Macedonian
• Mongolian Cyrillic
• Avar
• Kabardian
• Ossetian
• Other Cyrillic-script languages

Some are original BGN/PCGN systems. Others are agreements adopting a national romanization standard after sufficient domestic implementation.

BGN/PCGN Russian

The BGN/PCGN Russian system was adopted by the BGN in 1944 and by the PCGN in 1947.

Selected mappings include:

The system uses contextual forms for Е and Ё. Е is written ye initially, after vowels, and after Й, Ъ, or Ь; otherwise it becomes e. Ё follows a similar pattern with yë and ë.

BGN/PCGN Russian Examples

These forms are more immediately usable for English-language maps and government publications than diacritic-heavy ISO 9 forms.

However, they are not always perfectly reversible. A sequence such as shch may represent Щ or an actual sequence of Ш followed by Ч. The BGN/PCGN Russian rules include special separator conventions for unusual character sequences where extra differentiation is necessary.

Russian Comparison Table

The differences become clearest when the three systems are placed side by side.

The comparison reveals three design philosophies:

• ISO 9 uses distinctive Latin letters and diacritics.
• ALA-LC uses library-oriented digraphs and specialist marks.
• BGN/PCGN uses practical English-readable sequences and contextual spelling.

Ukrainian Romanization

Ukrainian demonstrates why Russian romanization rules cannot be applied to every Cyrillic language.

The BGN/PCGN 2019 Agreement for Ukrainian adopts the Ukrainian national system introduced in 2010 and replaces the earlier BGN/PCGN 1965 system.

Important mappings include:

Examples include:

The system’s documentation warns that it is not recommended for reverse transliteration. Several Latin digraphs can represent either one Cyrillic letter or a sequence of multiple letters, and the soft sign is omitted.

This explains why Kyiv is not merely a stylistic update of Russian-derived Kiev. It represents the Ukrainian form Київ under Ukrainian rules.

Bulgarian Romanization

The BGN/PCGN 2013 Agreement adopts Bulgaria’s national system, which was officially established by state decree in 2009. It replaced the earlier BGN/PCGN 1952 system.

Several Bulgarian mappings differ sharply from Russian:

Examples include:

• България → Bulgaria
• София → Sofia
• Щипка → Shtipka
• Търново → Tarnovo

The agreement permits Bulgaria as an exception for the country name and writes final -ия as -ia, as in Sofia.

A Russian table would incorrectly convert Bulgarian Щ as shch and Ъ as a hard-sign mark. Bulgarian requires its own language-specific rules.

Macedonian and Other Cyrillic Languages

BGN/PCGN’s 2013 Macedonian agreement uses forms such as:

• Ж → ž
• Ѓ → ǵ
• Ќ → ḱ
• Ц → c
• Ч → č
• Ш → š
• Љ → lj
• Њ → nj
• Џ → dž

This reflects Macedonian orthographic conventions rather than English-style Russian romanization.

Mongolian Cyrillic requires mappings for Ө and Ү, while Caucasian languages may contain complex Cyrillic sequences representing ejective or other consonants. BGN/PCGN therefore maintains dedicated tables for Mongolian, Avar, Kabardian, Ossetian, and other languages.

ISO 9 vs ALA-LC vs BGN/PCGN by Use Case

Academic Linguistics and Script Comparison

Recommended starting point: ISO 9

ISO 9 is appropriate when the project needs systematic comparison across several Cyrillic alphabets.

Typical uses include:

• Comparative linguistics
• Script inventories
• Historical orthography
• Multilingual text alignment
• Digital humanities
• Reversible transliteration

A project should state the exact ISO edition and any national implementation used.

Library Cataloging

Recommended system: ALA-LC

Use the applicable ALA-LC table when creating or matching records in institutions that follow Library of Congress cataloging practice.

Do not use the generic Russian table for every Cyrillic-script publication. Select the table corresponding to the language of the text.

Maps and Geographic Names

Recommended system: relevant national or BGN/PCGN standard

For geographic names, start with:

1. The officially standardized local name
2. The applicable national romanization
3. The current BGN/PCGN agreement or system
4. Established international variants where appropriate

The GNS stores approved, variant, historical, and native-script names, demonstrating that geographic-name management normally requires multiple linked forms rather than one destructive replacement.

General Publishing

Use the form that is:

• Officially used by the person, place, or organization
• Established in the target language
• Consistent with the publication’s style
• Clear to the intended reader

For example:

• Use Tchaikovsky when discussing the composer for general English readers.
• Use Čajkovskij in an ISO 9 dataset.
• Use Chaĭkovskiĭ in an ALA-LC-style bibliographic context.
• Use Chaykovskiy when applying BGN/PCGN Russian rules mechanically.

Personal Names

A person’s official or preferred Latin spelling should normally take priority.

A Russian name may appear in a passport, professional profile, immigration record, or publication under a spelling that does not match any one technical standard.

Store separately:

• Original Cyrillic name
• Official Latin spelling
• Generated ISO 9 form
• ALA-LC form
• BGN/PCGN form
• Historical aliases

Search and Databases

A search system should index multiple forms.

For Чайковский, useful aliases may include:

• Чайковский
• Čajkovskij
• Chaĭkovskiĭ
• Chaykovskiy
• Tchaikovsky
• Chaikovsky

The display name and the search aliases should not be forced into the same field.

Diacritics, Digraphs and ASCII Forms

The three systems make different compromises.

ISO 9: More Diacritics, Fewer Digraphs

Examples:

• Ж → ž
• Ч → č
• Ш → š
• Щ → ŝ

Advantages:

• Compact
• Distinctive
• Easier to decode systematically

Disadvantages:

• Less familiar to English readers
• Harder to type on basic keyboards
• May be stripped by poor software

ALA-LC: Specialist Diacritics and Digraphs

Examples:

• Ж → zh
• Й → ĭ
• Ц → t͡s
• Щ → shch

Advantages:

• Established in library practice
• Preserves useful distinctions
• Works with authority-controlled records

Disadvantages:

• Contains specialist combining marks
• Can be difficult to display correctly
• Not designed for public signage

BGN/PCGN: Practical Digraphs

Examples:

• Ж → zh
• Х → kh
• Ц → ts
• Ч → ch
• Ш → sh
• Щ → shch

Advantages:

• Mostly typeable with basic Latin letters
• Familiar in English-language geographic names
• Suitable for maps and databases

Disadvantages:

• Long letter sequences
• Greater ambiguity
• Reverse conversion may be unreliable

Common Problems in Cyrillic Romanization

Applying Russian Rules to Ukrainian

This can turn Ukrainian Г into g instead of h, confuse И with І, and mishandle Є, Ї, and Ґ.

Treating Every Cyrillic Х as Kh

Russian BGN/PCGN uses kh, but Bulgarian’s national system uses h.

Assuming Щ Has One Universal Equivalent

Depending on the language and system, Щ may become:

• ŝ
• shch
• sht
• Another language-specific form

Removing Diacritics Without Recording the Original

Forms such as Č, Š, Ž, Ĭ, and Ė may encode distinctions required by the selected system.

Confusing Transliteration With Translation

Москва becomes Moskva through romanization. Moscow is the established English name, not a direct transliteration.

Replacing Established Names Mechanically

Tchaikovsky, Khrushchev, Gorbachev, and other familiar forms may not match ISO 9, ALA-LC, or BGN/PCGN exactly.

Ignoring Historical Spelling

Older Russian texts may contain letters removed in 1918. Church Slavic and historical tables may be required instead of a modern Russian table.

Assuming ASCII Conversion Is Reversible

Replacing č with ch, š with sh, and ŝ with shch may create collisions with genuine letter sequences.

Best Practices for a Cyrillic Romanization Tool

Detect the Language, Not Only the Script

A tool should identify or request:

• Russian
• Ukrainian
• Belarusian
• Bulgarian
• Serbian
• Macedonian
• Mongolian
• Another Cyrillic-script language

Preserve the Original Cyrillic

The source text should remain the authoritative record.

Name the Exact Standard

Avoid a generic label such as “Cyrillic romanization.”

Use:

• ISO 9:1995 with Amendment 1:2024
• ALA-LC Russian 2012
• ALA-LC Ukrainian 2011
• BGN/PCGN Russian 1947
• BGN/PCGN Ukrainian 2019
• BGN/PCGN Bulgarian 2013

Show Multiple Results

A comparison tool might display:

Mark the Intended Use

Label results as:

• Reversible transliteration
• Library romanization
• Geographic-name romanization
• Established conventional name
• Simplified ASCII form

Preserve Version Metadata

ISO 9’s active revision process and the replacement of older Ukrainian and Bulgarian BGN/PCGN systems demonstrate why the standard name alone is insufficient. Store the edition, agreement year, and conversion date.

Support Historical and Variant Names

Search should connect current, historical, conventional, and system-generated forms without treating them as interchangeable display names.

Frequently Asked Questions

What is ISO 9?

ISO 9 is an international standard for transliterating Cyrillic characters used in Slavic and non-Slavic languages into Latin characters. The currently published base standard is ISO 9:1995, supplemented by Amendment 1:2024.

Is ISO 9 the official system for every Cyrillic-language country?

No. Countries may use national romanization systems for passports, geographic names, signs, and administrative records.

What is ALA-LC romanization used for?

ALA-LC tables are used primarily for library cataloging, bibliographies, authority records, and access to materials written in non-Latin scripts.

What is BGN/PCGN used for?

BGN/PCGN systems establish standardized Roman-script spellings for foreign geographic names used in U.S. and British official contexts.

Why does ISO 9 use Č while BGN/PCGN uses Ch?

ISO 9 favors a distinctive single Latin character for a single Cyrillic character. BGN/PCGN generally favors digraphs that are more accessible to English-language users.

Why is Й written differently?

In Russian:

• ISO 9: j
• ALA-LC: ĭ
• BGN/PCGN: y

Each form reflects a different institutional design.

Why is Ц written c, t͡s, or ts?

The three systems represent the same Cyrillic character differently:

• ISO 9 uses compact c
• ALA-LC uses the tied form t͡s
• BGN/PCGN uses the practical digraph ts

Is BGN/PCGN reversible?

It depends on the language and system. The Ukrainian 2019 agreement explicitly warns against relying on reverse conversion.

Which system should be used for Russian books?

Use ALA-LC when following Library of Congress cataloging practice. Use ISO 9 for a systematic international transliteration project. Preserve the source’s own form in quotations and historical citations.

Which system should be used for place names?

Use the officially standardized national name or the applicable BGN/PCGN agreement. Retain historical and conventional aliases for search.

Is Moscow a romanization of Москва?

Moskva is a direct romanization. Moscow is the established English exonym.

Is Kiev incorrect?

Kyiv is the current Ukrainian national romanization of Київ. Kiev is a historically established English form influenced by Russian. Historical sources should not be silently rewritten, but current references generally use the official Ukrainian form.

Can one converter handle every Cyrillic language?

Only when it identifies the language and applies the correct table. A single Russian mapping table is not sufficient for all Cyrillic text.

Conclusion

ISO 9, ALA-LC, and BGN/PCGN represent three distinct approaches to Cyrillic romanization.

ISO 9 is a broad international transliteration standard. It emphasizes systematic character correspondence across Slavic and non-Slavic Cyrillic alphabets and commonly uses Latin letters with diacritics.

ALA-LC is a family of language-specific library systems. It supports catalog searching, authority control, bibliographic identification, and the consistent description of Cyrillic materials.

BGN/PCGN develops or adopts language-specific systems for geographic names. Its outputs are designed for maps, gazetteers, databases, and official government use.

For the Russian name Чайковский:

• ISO 9 gives Čajkovskij
• ALA-LC gives Chaĭkovskiĭ
• BGN/PCGN gives Chaykovskiy
• General English convention gives Tchaikovsky

All four forms serve different purposes.

The correct approach is not to select one universal spelling and discard the rest. A reliable reference should preserve:

1. The original Cyrillic form
2. The source language
3. The exact romanization system
4. The standard’s edition or agreement year
5. The official or established public spelling
6. Historical and searchable aliases
7. Any information lost in simplified conversion

Cyrillic romanization works best when it is transparent. Users should be able to see not only the Latin result, but also which rules produced it, which audience it serves, and whether the original Cyrillic spelling can be recovered.