When doing palaeolinguistic analysis, only words with very specific meanings are relevant. There is no basis to claim that words for 'mountain' would testify for the mountainous homeland, or that words for 'ice' and 'snow' would testify for northern or mountainous homeland. There are probably no...