To translate scientific literature is the easiest part of translation. In fact, a lot of papers don't even get translated because everyone can read them after learning 100-200 words from the respective language. I can tell you that most of my time goes by reading scientific papers in languages that are not my mother tongue.

The question discussed in this thread is _not_: Would it be beneficial to have all scientific papers in one language? It is: Would it be beneficial to have all literature, discussions, jokes, poems and songs in one language. A question, that Newton's work and the amount of NSF funded papers in 2015 has absolutely no relation to.

Do you really think you can understand the concepts in a scientific paper knowing just 200 words of the language they are written in?

In my field of expertise, which is mathematics, this is certainly possible. My French and Russian is bad, to say the least. I doubt I have a vocabulary of more than 200 words in each, but I have no problems reading, for example, EGA [1]. Also there are entire journals in the field publicising translations of articles, which seems to contradict your claim that bleeding-edge scientific concepts are non-translatable.

[1] https://en.wikipedia.org/wiki/%C3%89l%C3%A9ments_de_g%C3%A9o...

Mathamatics is not like other fields - it isn't even really a science.

The non-mathamtical language used in mathamatics papers is easily translated, but understanding the maths is impossible without the required background and it can't be translated into any natural human language. Maths uses a synthetic symbolic language which mathematicians around the world share allowing them to communicate - in the natural sciences we use English instead.