Transcribing Braille Documents in Foreign Languages*

If a foreign language book cannot be located through contacting a braille library or printing house, then either the library will braille it for the student or it will be done at the educational institution or other transcription services. Before deciding to transcribe the entire textbook into braille, it is necessary to check with the professor to see which specific sections are covered. In many foreign language courses, one textbook is used for the first three levels, which are taken over a few terms. Also, students may need to reference the glossaries or verb tables at the back of the book. If the student already has a braille dictionary for the foreign language, this may be sufficient.

Braille Overview

The origin and basic construct of braille characters is well known and documented on a variety of web sites, including many of those cited in this document. Because of its physical nature as a tactile medium, there are inflexible constraints on the size of the characters and the size of the individual dots. The number of different characters that can be formed is finite and specific.

Braille is not a language. Braille is a set of symbols that, given appropriate equipment, can be written down and read. Like the Latin alphabet, it can be used for any number of languages. Many of the individual braille symbols have several different meanings that are determined only by the context, or relative proximity to surrounding characters and what those surrounding characters are. A print equivalent of this type of situation is letters s, p and o. In most fonts, it is difficult or impossible to tell whether they are upper or lower case when standing alone; placed in surrounding text, the meaning is instantly apparent. Consider the upper case “I”. Is it a Roman number or the word I? Or is it simply the upper case letter I? This can only be determined by the context of surrounding text.

There is no print equivalent to braille contractions. Contractions are formulated and used according to rules, like spelling, and cannot be changed by the individual user and still be considered correct or even readable. Contractions make reading more efficient by shortening the word, and also serve to save some space. Some contractions consists of a single character (or a pair of two characters) that stands for an entire word, or a commonly used combination of letters used as part of many words, like tion, ity, ance, sh, th, etc. Some contractions are a single alphabet letter and others are letters preceded by special indicator symbols that have no print equivalent.

Some whole words are contracted as a single alphabetic letter; other whole words are represented by a single braille character that has no print equivalent. The “and” contraction is an example. The word “and” is always shown in braille by a single character that has no print equivalent. Words such as sand, band and abandon, are also shown in braille with the “and” character, so that sand and band are written as two characters instead of four and abandon with five characters instead of seven.

Other braille contraction symbols mean a specific whole word when standing alone or a letter combination when used within a word. These symbols also have no print equivalent. The ch symbol is a single braille symbol that means the word child only when standing alone. In the word childish, only the ch is shown by this symbol and the rest of child, the ild, is spelled out letter for letter. There are several other type of braille contractions, some of which are punctuation when appearing at the end of words, but letter-combination contractions when appearing in the middle of a word. These do not relate to foreign language braille; the use of contraction symbols in foreign language braille is discussed later on.

Languages in the Latin Alphabet

All the foreign languages that in the Latin alphabet, with the exception of Dutch, use accented letters in addition to the conventional alphabet. Sighted students have probably seen such letters in foreign words and in general reading. Such letters are easy to identify and remember because they look exactly like English letters with a visual embellishment and they look the same in every language that uses them.  The é acute, for instance, looks the same when used in Italian as it does in French and Spanish.

Latin alphabet foreign languages are transcribed in uncontracted braille, letter by letter, for two reasons. First, the rules for using contractions apply only to English braille and second, contraction characters are used to designate the special accented letters used in the foreign language. This is necessary because there is no way to change the physical appearance of a letter in braille. In print, an accent mark can simply be typographically added to the accented letter. Since this is not possible in braille, the accented letter itself is treated as a unique character that includes both the letter and the accent. It is not necessary for the braille reader to have a visual description of the accented letter, but it is critical that each be identified by name. These letters are usually identified as letters followed by the type of accent they carry, i.e., é (e acute); à (a grave); î (i circumflex); ç (c cedilla); ñ (n tilde); and so forth.

The braille symbol for é (e acute) tends to be the same for all the languages that use it, but this is not always the case. Both Turkish and French use the â (a circumflex), but the braille symbol used for it in French is different from the one used for it in Turkish.  The same braille symbol used for the ä (a umlaut) in German is used for the â (a circumflex) in Turkish, where the ä (a umlaut) does not occur.

A list of each braille symbol and the identification of the accented letter it represents must be given at the beginning of every volume of a foreign language transcription. This is standard procedure.

Dutch, like English, uses no accented letters except for those that occur in foreign words. (There is provision for such letters in Dutch braille.) Dutch is simply brailled letter for letter.

“Nu is het duidelijk, dat het duratieve aspect in de feitelijke handeling moeilijk tot zijn recht kan komen”.

Reading this in braille would be about the same as reading it in print. Even though there are no accented letters, no contractions are used because this is not English.

Of the critical need languages, only Turkish uses the Latin alphabet with several accented letters. The regular braille letters are used with contraction characters designating each accented letter.

Learning the braille code for Latin alphabet languages is essentially the same experience for blind readers as sighted readers. The sighted reader, who knows how to read English, can appreciate the visual appearance of the various accented letters, learn their names, and become accustomed to recognizing them.

The blind reader understands that English contractions are not used. When a contraction symbols is encountered the reader knows this means a particular accented letter that can be identified by referring to the list of symbols at the beginning of the braille volume. This symbols list gives the blind reader the means to identify each accented letter and to have the same access to the material as the sighted readers. As long as the reader can name the accented letter, there is a common means of communication with the teacher and the others in the class. The visual appearance is not necessary because the blind reader does not use visual appearance in order to read.

Languages Not Using the Latin Alphabet 

The visual appearance of non-Latin alphabet letters is of no concern to the blind reader, although braille readers with low vision may choose to learn it as in the case examples provided earlier in this informational guide. Languages that have braille code provisions also have Latin names or designations for the letters. When such identifiers are paired with their corresponding braille symbols, the braille reader can refer to the print letters when conferring with the teacher and other students.

It is not necessary for the non-Latin alphabet letters to be transliterated into Latin-letter equivalents but it is necessary that non-phonetic notations systems be modified to be phonetic. This has been done with languages that have provision for braille symbols. The critical need languages all have braille provisions, as do Hebrew and Greek. Languages that are read in columns and/or right to left are brailled left to right in rows top to bottom.

Learning the braille codes for the foreign language under study presents the same issues for blind and sighted students. An unfamiliar alphabet for sighted students may be more of a challenge for them than it is for blind students because the physical appearance of the letters is new. For braille readers, it all looks like braille. The symbols are all familiar. Braille readers are long accustomed to symbols assigned to various usages depending on context and the issue of learning new symbol assignments in foreign language is a familiar task. A list of each braille symbol and the identification of each letter of the non-Latin alphabet it represents must be given at the beginning of every volume of a foreign language transcription.

Print to Braille Conversion

In the United States, English print to braille conversion using braille “translation” software can yield accuracy on the word level that approaches 100%. It is possible for a person who does not know braille to produce decent, usable and useful braille. This is particularly critical in educational settings, where students need braille in a timely manner in order to keep current with their studies. However, like any other computer tool, such as spell checkers, this is not foolproof. The quality of automatic braille conversion results varies.

Issues relating to print to braille conversion are even more critical in the area of foreign language braille because the reader is actually in the process of learning the language itself and the complexity of the material may make the conversion process less reliable. Duxbury DBT software, for example, says it converts many foreign languages, including Arabic, Persian/Farsi, Russian and Turkish. While not all critical need languages are listed, Duxbury provides a contact email and assurance that they can work with users to “translate” languages that are not listed on their website.

The term translation is often used in connection with software that automatically converts print text into braille. This is a slightly misleading term because, as already stated, braille is not a language. What is being translated are the contractions. Modern braille software can and does “translate” print into braille using contractions, rather than just letter for letter. It is useful to understand this process as conversion rather than translation.

Automatic conversion is an invaluable tool and resource, but it is not a panacea and it does not eliminate the need for skilled intervention from a knowledgeable person when quality braille is the desired outcome.

Foreign language braille presents unique quality control issues that require a certain amount of skill and specialized knowledge from the braille provider. Ideally, the braille provider is a certified braille transcriber. In the case of Latin-letter foreign languages, it is generally not necessary for the transcriber to know the foreign language because the alphabet is familiar and it is easy to compare braille with the print for accuracy. However, in the case of non-Latin letter alphabets, knowledge of the language is necessary.

Print to Braille Software Programs Developed Abroad

Software developed in other countries for native speakers of the languages may also be options depending on if the functional menus are also in the foreign language or available in English for the disability service provider or beginning language student to understand. Contact information for each country’s braille authorities or blind organizations for learning more about their braille systems and available conversion programs is found online from the Royal National Institute of Blind People or Mobility International USA.

The IIT Madras software, which may be available to U.S. students, can be used for students taking Hindi or other languages of India. It “has taken a phonetic approach to representing Indian language text and so it is quite easy to convert the text prepared using the Multilingual editor into braille codes. Just a simple table look up procedure is all that one would require and the program converts text in the vernacular (a .llf file as prepared with the editor) into appropriate braille codes for use with an embosser connected to a computer. It is possible for one to prepare a bilingual braille document where Bharati braille applies to text in Indian languages while standard [contracted] translation applies to text in English.”

In Japan, “a beta version of the screen reader (BRLTTY Plus) for Linux expanded the functions available for Japanese braille output. The Japanese braille ‘translation’ was able to display high-definition Japanese braille using the EXTRA for Linux, automatic braille ‘translation’ engine developed by Jun Ishikawa. The software is compatible with almost all braille displays currently used in the world.”

Libbraille, a computer shared library that makes it possible to easily develop braille displays, also provides a free online conversion tool for foreign languages from unicode text to braille. Unicode is a computer industry standard allowing computers to consistently represent and manipulate text expressed in most of the world’s writing systems developed in tandem with the Universal Character Set standard. It is possible to select the braille alphabet for many foreign languages, including Arabic, Bharati-Devanagari, Korean, and Russian. It is free software and open source, and provides downloads and how-to instructions.

All of the above programs conversions need to be verified for accuracy before embossing. Someone who knows the foreign language and braille transcription should check the documents and include a list of each braille symbol and the identification of the accented letter it represents at the beginning of every volume.

Location of Instruction: What are the Rules?

In discussing foreign language braille, English tends to get lost, but it is a major factor in most foreign language braille transcriptions. Foreign language instruction material typically has considerable amounts of English in it, especially at the beginning, to present vocabulary, to explain the lesson, for exercise directions and for translation exercises. In these cases, the English is contracted. This is discussed further below.

In the United States, braille standards and rules are set by BANA (Braille Authority of North America), a voluntary consortium of some 16 organizations of and for the blind, representing braille readers, transcribers and educators. The relevant rules books are:

• NBA Interim Manual for Foreign Language Braille Transcribing available from National Braille Association (http://www.nationalbraille.org)
• Braille Formats: Principles of Print to Braille Transcription 1997 available from American Printing House for the Blind (http://www.aph.org)
  

The foreign languages rules specify the braille symbols to be used and the procedures for presenting instruction material, usually textbooks. Procedures for presenting such things as exercises, vocabularies and translations are given. Braille Formats covers general rules for all textbooks and anything not covered specifically for foreign languages in the Interim Manual.

English adjacent to the foreign language is not an issue for sighted readers because the switch from one language to the other can be seen at a glance, particularly in the case of non-Latin alphabets. This is obviously not the case in braille and these transitions are potentially awkward. Ample provision for this is made in the Interim Manual.

Many languages are covered and the symbols designations and identifying names are generally in accordance with those used in the countries where the language is spoken. In cases of variance, a single standard is given. For instance, punctuation usage differs among the various Spanish-speaking countries. In the United States, conventional English punctuation (used by some Spanish-speaking countries too, but not all) is used.

When using software, such as Duxbury, to convert text, all three Spanish entries in their translation tables are designed for native speakers in countries where Spanish is the national language. A Spanish book in the United States would be produced using the English/American table, with Spanish as a secondary language.

In the case of languages not covered in the Interim Manual (like Arabic), it is recommended that the braille symbols designations be researched and when determined, used in conjunction with the other general provisions the Interim Manual and Braille Formats.

Sometimes the student will travel to the country where the language is spoken to take courses. The braille codes for the various countries outside North America are provided by the designated organizations or entities in that country, a central agency or commission for the blind, or a university perhaps. If braille for foreign language study is provided locally, it would no doubt conform to the braille code of that country, rather than BANA provisions. However, the English part of the transcription may or may not be an issue. In order to follow the lessons and fully participate in the instruction, the English part of foreign language instruction must be high quality braille and that means fully contracted English, where appropriate.

On the other hand, if the foreign language braille is done according to BANA rules, it doesn’t impact instruction. If the braille corresponds to the print foreign language correctly, it doesn’t matter which braille symbol is used for the period or the question mark as they already match. All that is needed is the exact braille equivalent to the print instruction material.

Converting Foreign Language Braille into Print

If a student writes homework assignments that need to be turned in for grading to a sighted foreign language instructor, then certain steps need to be taken to ensure the proper symbols unique to the language appear properly (as this usually is graded). This is possible to do using uncontracted, 8-cell computer braille code characters and using ASCII characters. For step by step instructions on how to do this and the ASCII character numbers for the accented symbols in commonly used languages, go to the online presentation by clicking here.

To avoid the process of computer translation of braille to print, or relying on an underwriter, some students may use a regular keyboard instead of a braille display to type up their test or homework (and a screen reader described in the previous section to ensure it is typed correctly). In order to find the appropriate characters or symbols in the foreign language, student will need to first select the language in Microsoft Word, under the Tools menu, choose Language, and then select the language. Full instructions, including where to go if the foreign language is not shown in the menu, is available online by clicking here.

Next, the student adapts a standard American English keyboard to type in the foreign language and learns the new key patterns. A braille labeler could be used to mark the new keys. A sample of foreign language keyboards for all the critical need languages is easily found through an Internet search. Free foreign language fonts can be downloaded online.

*Thanks to Joanna Venneri (see how to reach her below) for assistance in writing portions of this section.