Ensure that subtitles provide equivalent information


The content of subtitles should provide the same information and meaning as the original speech, within the constraints of readability and timing. Non-speech text such as signs may also include important information.

Directions and techniques

For speech, try to provide a full and accurate translation (high priority)

As far as possible, directly translate everything that is spoken without simplifying it. This is not always possible because people generally speak much faster than written speech can be read. It is often therefore necessary to edit the speech in order to shorten it. This should be done without altering its meaning or ‘flavour’. Some things to consider are:

  • Do no remove entire sentences. Instead, remove parts of a number of sentences. This makes it more likely that all of the meaning will beretained.
  • Avoid removing the names of people being addressed. These are often important for following the plot.

Do not censor speech (high priority)

Do not censor speech in order to remove language or ideas that may appear crude or offensive. This is unnecessary and condescending. If the audible content is deemed suitable for its intended audience, it is also suitable for members of that audience who are reading a translation. The subtitler’s comfort level when translating sexual or disturbing language for example should not be a factor.

Identify individual sources (high priority)

If a scene contains a number of different speakers, it should be clear which subtitle belongs to which speaker. This can be done using labels, colour or positioning.

Subtitles can be labelled by adding a prefix giving the speaker’s name or identifying the source. This works if the names are certain to be known to the viewer or if it is obvious, from the name, which speaker is indicated. It may only be necessary to add the label to the first subtitle from that speaker. Subsequent subtitles from that speaker may be obvious, or the speaker may be represented by colour after that.

Different colours can be used to identify different speakers over a period of time if the speaker are clearly identified initially.

The position and justification of subtitles can be used if the speakers maintain their positions on the screen. Left justified subtitles for a speaker on the left, right justified for a speaker on the right and centred for a speaker in the middle.

Translate any important written text that appears in the programme (high priority)

The contents of any important textual items displayed within the programme should also be translated. This may include for example, signs, newspaper headlines, computer screen displays or handwritten notes.