This happy union can be summarized in a syllogism:

Machine translations are the more accurate the simpler the source texts are. The main purpose of Clear Language is to make texts simple. As a result, texts written in Clear Language undergoing machine translation generate very accurate translations.

But let's try to articulate this concept a bit.

What characterizes a written, or rewritten text as Clear Language? These texts are never ambiguous, use short sentences with linear and predictable syntactic structures, basic and unambiguous vocabulary, and contain no run-on sentences or colloquialisms subject to interpretation. So by their very nature they clear the field of all the difficulties that machine translation normally struggles to overcome. All stumbling blocks for machine translation, based on statistical models or even artificial neural networks, no longer exist. Here, then, translation time goes down and quality goes up.

The future of machine translation seems to be already written, as market trends confirm that more and more companies are relying on it to reduce the cost of producing multilingual texts and to have materials ready in a very short time. In some cases they rely on a human translator for proofreading (post editing) and greater assurance of accuracy. In other cases, daringly, they don't. But is there a way to reduce this risk margin?

An effective strategy to reduce the risk of suboptimal output even in the absence of human review is special attention to input texts. The economic advantage is mathematical: while the cost of revising the output must be multiplied by the number of target languages, the intervention on the input texts to machine translation is single. Incidentally, texts in the source language also take advantage and are more readable, but from the perspective of translation costs, this is only a side effect.

For target languages, on the other hand, there is a true multiplicative effect.