Transliteration
Transliteration is the process of representing written text from one writing system by using the characters of another writing system. It is commonly used to represent names, places, historical documents and foreign-language words in a form that can be read by people familiar with a different alphabet or script.
Transliteration is distinct from both translation and transcription. Translation conveys the meaning of words from one language into another, whereas transliteration seeks to represent their written forms. Transcription, on the other hand, seeks to represent the sounds of spoken language. Although the three processes can overlap in practical applications, they have different purposes.[1]
Definition and purpose
The word transliteration describes the conversion of characters from one script into another according to an established system of correspondences. The resulting text represents the original writing without translating its meaning.
For example, the Greek word Φόβος can be transliterated into the Latin alphabet as Phobos. Similarly, Δεῖμος becomes Deimos. These forms allow readers who do not read Greek to recognize and refer to the original names without translating them into English.[2]
Transliteration is particularly useful in:
- Scholarly publications dealing with ancient and modern languages.
- Historical documents and the study of original sources.
- The cataloguing and indexing of books and manuscripts.
- The representation of personal names and geographical locations.
- International communication and the exchange of information between computer systems.
- The preparation of dictionaries, linguistic studies and reference works.
Transliteration can also make information searchable across different writing systems. The Unicode Consortium identifies searching, indexing and the handling of names in international databases as important applications.[2]
Transliteration, transcription and translation
These three processes should be distinguished.
Transliteration converts written characters from one script into another. Its primary concern is the written form of the original text. Ideally, a reader familiar with the transliteration system should be able to reconstruct the original spelling.
Transcription represents the sounds of a language using the writing system or phonetic symbols of another. It is concerned primarily with pronunciation rather than the precise spelling of the original text. Phonetic transcription can use the International Phonetic Alphabet (IPA), which provides symbols for distinguishing speech sounds.[1]
Translation conveys the meaning of a word, sentence or complete text into another language. It does not seek to preserve the original spelling or pronunciation.
For example:
| Process | Greek original | English representation |
|---|---|---|
| Transliteration | Φόβος | Phobos |
| Translation | Φόβος | Fear |
| Transcription | Φόβος | fo-boss - A phonetic representation of its pronunciation |
The distinction is important in historical research. A transliteration of an ancient name does not necessarily establish how that name was pronounced in its original period. Likewise, an English translation may convey the meaning of an ancient passage while giving no indication of its original wording.
Romanization
Romanization is a particular form of transliteration in which a non-Latin writing system is represented using the Latin alphabet. It is widely used in representing languages that employ Greek, Cyrillic, Arabic, Hebrew, Chinese, Japanese and other writing systems.
There are numerous Romanization systems, and more than one may exist for the same language. The choice of system can depend on the intended audience, the requirements of scholarly research, national conventions or international standards.[3]
Examples include:
- Greek: Φόβος → Phobos; Δεῖμος → Deimos.
- Russian: Москва → Moskva. The familiar English form Moscow is an established English name rather than a strict character-by-character transliteration.
- Japanese: しんばし can be represented as shimbashi under the Hepburn system or sinbasi under the Kunrei system.
- Chinese: 北京 → Beijing, using the Pinyin Romanization system.
The examples demonstrate that transliteration is not always a matter of replacing every character with one corresponding Latin letter. A single original character may be represented by a combination of letters, and different systems may make different choices.[2]
Chinese Romanization
Chinese provides an important example of the distinction between alphabetic transliteration and phonetic representation. Chinese characters do not constitute an alphabet in which each character corresponds to a single alphabetic sound. The Pinyin system represents the pronunciation of Chinese characters using Latin letters.
The Library of Congress uses Pinyin as the basis for its ALA-LC Romanization of Chinese. The system follows standard Chinese pronunciation and takes account of characters that can have different readings in different contexts. Its cataloguing rules omit tone marks, although tone marks are used in other applications of Pinyin.[4]
Methods and standards
Transliteration systems are developed for different purposes. Some seek to reproduce the original spelling as accurately as possible, while others give greater attention to pronunciation, readability or established conventions.
A system may employ:
- Direct correspondences between individual characters.
- Combinations of Latin letters to represent a single original character.
- Diacritical marks to distinguish otherwise similar characters.
- Special punctuation to identify syllable boundaries or distinguish sounds.
- Context-dependent rules where the representation of a character depends on its position or linguistic use.
Diacritical marks are particularly important in scholarly transliteration. They can distinguish letters that would otherwise become indistinguishable when represented in an ordinary Latin alphabet. Their omission can therefore introduce ambiguity or make it impossible to reconstruct the original spelling accurately.
Reversibility
An important property of some transliteration systems is reversibility. A reversible system allows a reader to reconstruct the original written text from its transliterated form, using the prescribed rules.
A fully reversible system should avoid representing two different original characters by the same combination of target characters. Otherwise, the original spelling may not be recoverable.
Reversibility is especially important in scholarly research, historical documentation and the exchange of bibliographic information. The Unicode Consortium identifies completeness, predictability and reversibility among the desirable properties of transliteration systems, while acknowledging that these objectives can sometimes conflict with established conventions and ease of pronunciation.[2]
The United Nations Group of Experts on Geographical Names also requires an acceptable degree of reversibility when considering systems for international standardization of geographical names.[3]
Not every commonly used transliteration is reversible. Some simplified systems omit diacritical marks or merge distinctions between characters. Such representations can be convenient for ordinary readers but may be unsuitable for precise scholarly work.
International standards
A number of organizations have developed transliteration systems to promote consistency in international publications and records.
- International Organization for Standardization (ISO): ISO 9:1995 establishes a system for transliterating Cyrillic characters into Latin characters, covering Slavic and non-Slavic languages written in Cyrillic. The standard was amended in 2024.[5]
- Library of Congress and American Library Association (ALA-LC): These organizations maintain Romanization tables for the representation of numerous non-Latin writing systems in bibliographic and cataloguing records.[6]
- United Nations Group of Experts on Geographical Names (UNGEGN): This body considers and recommends standardized Romanization systems for geographical names, with the objective of ensuring consistency in international usage.[3]
The existence of several standards is important because there is no single universally accepted system for transliterating every language. A person's name or geographical location may therefore appear in different Latin spellings in different publications and catalogues.
Problems and limitations
Transliteration is not always straightforward. Writing systems differ considerably in the way they represent language, and there is not necessarily a direct correspondence between the characters of two scripts.
Some languages use writing systems in which vowels are not routinely represented in the written text. Arabic and Hebrew, for example, commonly omit short vowels in ordinary writing. A transliterator may therefore have to consult the context, historical pronunciation or other linguistic evidence when deciding how to represent a word accurately. This can make the process more complicated than simply substituting one character for another.[7]
Other difficulties include:
- Different historical pronunciations: Ancient languages may have undergone substantial changes in pronunciation. Modern pronunciation cannot automatically be assumed to represent the pronunciation of an earlier period.
- Multiple transliteration systems: Different scholars, countries and publishing organizations may use different systems for the same language.
- Loss of distinctions: Simplified transliteration can omit sounds, vowel lengths, accents or other features of the original writing.
- Ambiguous characters: A character may have several pronunciations or meanings depending on its context.
- Technical limitations: Older computer systems and catalogues were sometimes unable to display the original writing system or its diacritical marks.
Consequently, scholarly publications should identify the transliteration system being used, particularly when dealing with ancient languages, historical names or texts whose exact original spelling is important.
Historical and scholarly importance
Transliteration has particular importance in the study of historical documents. Many ancient texts survive in writing systems that are unfamiliar to modern readers. Transliteration permits these texts to be discussed, quoted and indexed without requiring every reader to possess the ability to read the original script.
In the study of ancient Greek, Hebrew, Aramaic and other languages, transliteration is frequently used alongside the original characters and an English translation. This enables researchers to examine the wording of the source while making it accessible to a wider audience.
It is important, however, not to confuse a transliteration with the original document itself. A transliteration is a representation made according to a particular system. It may be accurate and systematic, but it remains dependent on the decisions and conventions used in producing it.
For the preservation and study of original sources, the original text should therefore be retained wherever possible. A transliteration can then serve as a supplementary representation, allowing readers to examine the written form without losing access to the original characters.
See also
References
- ↑ 1.0 1.1 David Crystal, A Dictionary of Linguistics and Phonetics, 6th edition (Malden, Massachusetts: Blackwell Publishing, 2008), entry "Transliteration"; see also publisher information.
- ↑ 2.0 2.1 2.2 2.3 Unicode Consortium, "Unicode Transliteration Guidelines".
- ↑ 3.0 3.1 3.2 United Nations Group of Experts on Geographical Names, "Romanization Systems".
- ↑ Library of Congress, "Chinese: ALA-LC Romanization Tables", Rules of Application.
- ↑ International Organization for Standardization, ISO 9:1995, Information and documentation — Transliteration of Cyrillic characters into Latin characters — Slavic and non-Slavic languages.
- ↑ Library of Congress, "ALA-LC Romanization Tables".
- ↑ Karin C. Ryding, "Arabic transcription/transliteration/romanization," in Arabic: A Linguistic Introduction (Cambridge University Press, 2014), pp. 145–148.