Fluency and adequacy
Fluency and adequacy are the two classic dimensions of translation quality. Fluency is whether the output reads naturally and correctly in the target language, judged without the source. Adequacy is whether it preserves the meaning of the source, judged against it. A translation can be strong on one and weak on the other, which is why both are assessed.
How it works
Fluency is rated by reading only the target text: is it grammatical, idiomatic, natural? Adequacy is rated by comparing target with source: is all the meaning there, nothing added or lost? Separating them is diagnostic. Fluent but inadequate output is the dangerous case, since it reads perfectly while saying the wrong thing, exactly the failure mode of modern machine translation.
Most quality frameworks, including MQM, decompose into finer categories, but fluency and adequacy remain the underlying split.
How SourceTarget uses it
SourceTarget's composite quality score reflects both dimensions, and weights them by content type: fluency counts for more on campaigns, adequacy and terminology for more on legal and medical work. Human review concentrates on adequacy, since fluent but inaccurate output is the main risk with modern engines.