Accent, dialect and standard language
People often use “accent” and “dialect” loosely. An accent refers to how words are pronounced. A dialect usually includes vocabulary and grammar as well as pronunciation, and the difference between a dialect and a separate language is often a matter of politics and history as much as linguistics. Every speaker has an accent, and most speakers use some regional forms, even if they also know the standard written language.
Subtitles are written text, and written languages usually have a standard form. That creates a built-in tension: speech varies from place to place, while a subtitle is normally in standard spelling. The question is how much of the regional flavor, if any, should reach the screen. Regional dialects can be an essential part of a character’s identity, and sometimes the point of a scene.
The main approaches
1. Standardize. The subtitle writes the line in standard language and does not mark the regional features. This is the most common approach, because it keeps the text easy to read and keeps reading speed manageable. The accent remains audible in the soundtrack, and the viewer who hears it gets the flavor that way.
2. Mark lightly. The writer keeps a few regional words or short expressions to suggest where a character is from, while leaving the grammar and spelling mostly standard. This can give color without a heavy reading load. The choice of words has to be consistent: if a character uses a regional term in one scene, readers expect it to reappear.
3. Use phonetic spelling. The writer changes the spelling to imitate the sound, for example dropping letters or writing a word as it is pronounced. Some printed fiction does this extensively. In subtitles it is used sparingly, if at all, because unusual spellings take longer to read and the viewer has to decode them while following the picture.
4. Replace with another dialect. When translating, the writer may use a regional form in the target language to stand for the regional form in the source. This is risky. A dialect carries associations of class, education and region, and these differ from place to place, so mapping one onto another can produce a misleading impression, or even unintentional comedy.
Why phonetic spelling is used with care
Viewers read subtitles quickly, and every non-standard word slows them down. Phonetic spelling also tends to be inconsistent, since there is no agreed way to write a sound that the standard alphabet does not cover. And it can read as mockery. Spelling out one group’s speech while writing everyone else’s in standard form can imply that certain speakers are less educated or less serious. Writers who consider this use an even hand: if a feature is marked for one character, it should be justified by the story and not by a stereotype.
Accessibility is a further reason for caution. Viewers with reading difficulties, or those learning the language, may struggle with unusual spellings. The relevant needs are discussed in the articles on learning disabilities and new language speakers. Clear text helps those viewers, and it helps everyone else too.
Dialect in translated subtitles
When subtitling from one language into another, the translator typically has to decide how much of a regional register to carry over. Some practical considerations:
- If the source dialect is a plot point, such as a character being teased for his accent, the translator may need to explain or mark it, perhaps by having another character refer to it.
- If the dialect only provides background color, standardizing is usually safe.
- If the source dialect is hard even for native speakers to understand, the translator must first make sure the meaning has been correctly understood. This may require help from a speaker of that variety.
- Slang attached to a region is part of the same problem as other fixed expressions. The choices for these are described in idioms and slang.
Translators also consider the audience. A subtitle for a regional audience that shares the dialect may keep more local flavor than one meant for an international release.
Same-language subtitles and captions
When a program is subtitled in the same language, for example for viewers who are hard of hearing or who find an accent difficult, the same questions arise in a different form. Most style guides advise writing what is said in standard spelling, and adding a short note if the speech is heavily accented or unclear. Some guides allow showing a few distinctive words as spoken when they matter to the story. Verbatim versus edited approaches vary between broadcasters, so a project should follow its stated style rules.
Machines and accents
Automated systems can have more trouble with some accents than others, and the reasons depend on the training data they were built from. The difficulty is discussed in accent recognition. The effect on subtitles is that a draft produced automatically may contain more errors for some speakers, and these errors can look like random wrong words. Reviewing such drafts is much easier when the reviewer knows the dialect, and the lines for the speakers in question deserve extra attention. Human reviewers are the safeguard. Claims that an automated tool handles every accent well should be treated with caution.
Practical tips for subtitle writers
- Decide the approach for the whole project at the start, and write it into the style notes.
- Keep the treatment of each character consistent from one scene to the next.
- If a regional word is kept, make sure the context makes it understandable.
- Avoid spellings that might make a viewer pause.
- Ask someone from the relevant region to check sensitive or ambiguous cases.
- Where the dialect is the subject of a scene, take the time to discuss it with the director or client.
There is no single right answer. The balance between honoring how people really speak and keeping the text easy to read depends on the program, the audience and the purpose, and thoughtful professionals disagree on where to draw the line.