Given that he's tested for lots of languages, shouldn't he be applying something like bonferroni correction to the Granger causality results? Wouldn't this have

Given that he's tested for lots of languages, shouldn't he be applying something like bonferroni correction to the Granger causality results? Wouldn't this have the effect of making them all insignificant?