> And, crucially, it is not supported by neuroscience [3]. It seems that there are distinct brain areas involved specifically in mathematical thinking outside o

> And, crucially, it is not supported by neuroscience [3]. It seems that there are distinct brain areas involved specifically in mathematical thinking outside of those that are language-specific, which would mean that mathematical thinking is more than language [3]. I would not construe a small study of 30 people (15 math experts) as having the backing of the field of neuroscience. The steel man position of mathematics-as-language would be something along the lines of "Mathematics is mostly a language + some other stuff". The paper in question only shows there is "some other stuff"---but that is also true of almost every non-contrived use of language. The paper shows that the other stuff for sentences about history and nature appear differently in the brain than the other stuff for sentences about math, but that is far from demonstrating that math on the whole is not mostly a linguistic phenomenon. Personally, I think case studies of famous mathematicians or physicists discussing their thinking process (e.g. Einstein) is more convincing evidence against the hypothesis than fMRI studies.