Writing a source a translator can work from
A translator is one of a document's readers, and an English source either gives that reader what it needs to carry the meaning into another language or leaves the translator to guess and to query. Writing a translatable source means treating the translator as an audience, giving the translator a glossary of the terms that could be misread, and writing so that meaning survives outside its English setting: without culture-bound expressions, with relationships and referents stated, and with context for words that lose their sense when a string is isolated.
The translator as one of the document's readers
By 2013, medical writers were increasingly expected to consider translators as a further audience when producing documentation for overseas markets, because companies sought readers from many nations and cultures. Traditionally, the readers of medical documentation were patients and health care professionals learning how to perform a process or use a product. Translation gives speakers of other languages access to the information they need to understand ideas or operate devices, and documents that are easy to translate contribute to that access. Viewing the translator as one of a text's readers can help a writer meet the translator's informational needs, and it can be coupled with writing strategies that can improve the readability of texts in general as well as facilitate translation.
Ambiguity is highly problematic in translation, and it often takes one of two forms: the intended meaning of a word is unclear, or the context the word appears in makes its meaning uncertain. "We need a set" leaves the translator to determine which meaning of set is intended, and gives too little context to determine it. The translator might then guess at the meaning, which threatens the quality of the translation, especially when the translator is unfamiliar with the topic. Or the translator might query the writer, and time is spent exchanging messages. Each further meaning-related question takes more time, and while waiting for an answer the translator may be unable to continue with part or all of the document.
Writers who address translation-related issues while creating documents can contribute to the quality of the translation and to the speed with which translated materials reach an overseas market. The strategies that remove ambiguity and specify meaning rest on two parts: glossary-building and writing for translation.
The glossary for each document
An essential part of facilitating translation is a glossary for each document, giving translators definitions of potentially problematic terms. Glossaries remove much of the guesswork and questioning in translation, can contribute to the quality of a translation, and speed the translation of a document. The more information a glossary provides, the better.
A glossary entry should have four parts: the term, its part of speech, its definition, meaning the sense intended in this document, and a sentence showing how the term is used in the source text. An entry for set might read: set (noun): a collection or a group of items. Example: These six scalpels are part of a set of tools used by the surgeon when performing this procedure. The four-part format limits assumptions about what the term could mean. Marking set as a noun lets the translator dispense with the meanings set has as a verb, the definition focuses on one meaning, and the example sentence helps the translator confirm that meaning and see how the term is used in the document being translated. Once a glossary entry limits a multiple-meaning term to one definition, the writer must confine the use of that term to the sense the glossary gives.
Two types of term should be defined. The first is the commonly used word with several meanings, such as set, whose multiple meanings can cause problems as the translator works out what the writer intended. The second is the technical or industry-specific term essential to the context of the document, such as thrombus rather than blood clot, or lead for the wire connecting a pacemaker to the heart. Translators might not be familiar with such terms, and even translators familiar with a field might not know all of its technical terms: many health care professionals recognise defibrillator, and some of them might not recognise sinus rhythm, P waves or R waves.
Glossary entries should be ordered alphabetically rather than in the order the terms appear in the text. Order of appearance seems to save the translator time, but it makes a term hard to find again: on its second or third mention, the translator must read through a series of entries to find it.
Writing for translation
Writing for translation reduces ambiguity and removes culture-specific references, so that a text can be translated more quickly and more effectively than it otherwise could. Its focus on removing ambiguity is associated with effective writing in general, and it can improve the quality of the source text as well as facilitate its translation. It involves three general principles: avoiding culture-specific expressions, indicating relationships, and focusing on singular meaning.
Avoid culture-specific expressions. Translators often rely on a literal reading of a text, so expressions whose nonliteral meaning the reader must interpret can be problematic. Use text that can be read literally, such as the rain is falling rapidly and in great quantity, in place of an idiom with a culture-specific meaning, such as it's raining cats and dogs, and avoid metaphoric expressions such as a bear of a task, whose implied meaning relies on a particular cultural understanding.
Indicate relationships. Include the words that make clear how items in a sentence or a paragraph relate to each other. Every pronoun should have a clear antecedent: "This is a problem" leaves the translator asking what this refers to, and an inserted noun, as in "This failure to correctly calibrate the device is a problem", settles it. Vague pronouns such as it, this and they may sound clear in English while it is sometimes difficult to infer what they refer to, so make subjects and antecedents explicit. Parallel structure makes clear how many items a list names: "You need a pacemaker and lead" can be taken as a single combined unit, in the way a block and tackle is, while "You need a pacemaker and a lead" names two separate devices.
Focus on singular meaning. Alongside the glossary, each word must have only a single meaning throughout the text; a word used for two or more ideas leaves the translator constantly querying the writer to confirm the intended meaning each time it appears.
Writing for translation often requires reflection while writing and continual consciousness of how a text is being written, which can initially impede a writer's flow and style. Once learned, its principles can reduce the time and money spent on translation and improve the quality of the source text.
Source features that read well in English and still hinder translation
Some source features read perfectly well in English and still create problems downstream, because even well-written English can hide assumptions and ambiguity.
- Unpack noun chains whose relationships are unclear. In study drug administration compliance assessment, the relationship between the words is not explicit, and the target language may need it worded differently. Not every noun string is a problem, but where a clearer way to express the same meaning exists, use it.
- Check phrasal verbs. Bring up a concern, come off a treatment and rule out infection sound natural in English and may even feel clear and patient-friendly, but they can be difficult to translate consistently, because they are idiomatic and depend on context.
- Supply context for words with several meanings, especially in tables, headings, labels, buttons and lists of strings. Positive, control and screening may be clear in a sentence and ambiguous when isolated. In English, screening can refer to clinical trial eligibility, an initial assessment in a hospital or a public-health initiative such as cancer screening, and Portuguese uses different words for those meanings, so the translator needs the context to choose the right term.
- Specify the intended form of address. English leaves formal or informal, singular or plural, and gender open, and other languages need clear information about them.
- Allow room for text expansion in buttons, labels, diagrams, interfaces and brochures. Some languages need more space, and a layout already tight in English will generate layout problems in translation.
- Identify metaphors, humour and everyday examples, which are specific to a locale and may need to be localised or transcreated. These concerns do not call for generic or lifeless content; they call for flagging such items and giving the translation provider as much context as possible. Avoid expressions that are avoidably regional or local.
Readability and translatability
Readable does not always mean translatable. Plain language can help with complexity and with the relationships between concepts and wording, reduce jargon and make content easier to understand, but it can also create problems for translation where it relies on idioms, vague everyday expressions or cultural assumptions. Readability checks tend to reward short words, short sentences and a low reading grade, and they do not capture all the ambiguity a source can carry, so a sentence can score well and still be difficult to translate. A text for translation should be explicit, context-rich, simple in structure and as easy as possible to transfer to other languages and contexts, which asks for clarity about meaning, context and terminology as well as plain language.