When seriously looking at the world's linguistic diversity, languages are more different than often assumed. We might use the same names over and over to describe particular structures in different languages, but their forms and functions are always different. In this talk, I will start from the assumption that constructions are always language-particular, and thus that languages are in principle incomparable (a problem well-known in construction grammar).
Still, I will argue that it is possible to compare languages, when we accept that language comparison is crucially different from language- particular analysis. Specifically, language comparison might be 'besides the point' for individual languages, because the comparative view has to be kept constant. However, from this perspective, then, the internal organisation of individual languages will turn out to be the key to compare languages. Given a well-defined and invariable comparative perspective, the actual comparison of languages can be almost completely relegated to the summation over many individual language-particular analyses.