In the legal world, it is often necessary to translate a large amount of documents for a big case or understand the gist of something quickly. It is no surprise that machine translation (MT) is frequently mentioned when considering these business needs. Sure, an MT engine can translate more quickly than humans—think tens of thousands of words overnight. Also, there is no doubt that MT can get translation done more cheaply; with no human intervention, the only cost is that of the engine. But, with no human touch, important legal implications or key pieces of information can be missed.
With MT, it is important to discuss how to achieve the desired level of output quality. Quality is not exactly black or white, and not everyone defines it the same way. But generally, we discuss and implement one of two levels of post-editing—light or full—to bring raw MT up to the required level of quality.
This involves performing as few modifications as possible to the raw MT output in order to make the translation understandable, factually accurate and grammatically correct.
Light post-editing tasks include:
- Correcting grammar and spelling mistakes only if they affect the meaning;
- Rewriting confusing sentences partially or completely;
- Fixing machine-induced mistakes; and
- Deleting unnecessary or extra translation alternatives generated by the engine.
The localized text needs to convey the meaning of the source concepts correctly. Only major errors (incomprehensible content) and critical errors (that may incur legal consequences or use the wrong terminology) are covered. The resulting content might sound robotic or just a little bit ‘off’, yet it is fluid enough for a reader to understand the meaning.
This level of light editing is not easy to achieve: naturally detail-oriented linguists literally have to force themselves to skip over ‘minor’ errors and limit their work; their job is to achieve the stated quality level and nothing more.
The key phrases for light post-editing are ‘factual correctness’ and ‘good enough.’
Full post-editing, a slower and more in-depth process, must produce absolutely accurate translations that consistently use correct terminology, have the appropriate tone and style, have no stylistic inconsistencies and variations and are free from any grammatical mistakes. After this edit, the translation should read as if a professional linguist translated it.
Full post-editing tasks include all of the light post-editing tasks plus:
- Checking terminology against approved terminological resources to make sure it is consistent and appropriate;
- Cross-referencing translations against other resources;
- Making syntactic modifications in accordance with practices for the target language;
- Producing stylistically consistent, fluent content;
- Applying correct formatting and tagging; and
- Correcting ALL grammatical errors, typos, punctuation issues and spelling mistakes.
The expectation is high: full post-edited content should be as fluent as human translation in all aspects. Therefore, content must meet the quality criteria defined by the client for human translations.
But a quick caveat: the effort to achieve human-level quality from MT output may exceed the effort to have it translated by a linguist in the first place.
Getting it right
Is it as simple as choosing one of the two levels of post-edit? Not quite. There are plenty of clients who want human-level quality as well as the speed and low cost of light post-editing. In addition, quality can differ from project to project, client to client, language to language. In order for legal departments to fulfill their quality needs, this has to be discussed and defined beforehand.
If you are interested in machine translation for legal content, contact us to discuss the levels of quality that can be achieved by post-editing and the best-fit process for your needs.