Module 4: Hungarian, Catalan, Welsh

The problem is that words consist of two sides, form and meaning, and either of them may change. So it might be that the word in Proto-Indo-European for salmon was not used to refer to salmon, but to another kind of fish or to fish in general, and that would seriously weaken our argument. In the discussion above, it has been assumed that Proto-Indo-European is known, but, of course, it isn't. It has to be reconstructed from the present languages and from what we know of their development. Again, there is at least one major trap when it comes to reconstructing languages. Take the example of the word telephone. It is similar in English, German and Basque, both in form and meaning. Why would it nevertheless not be permissible to base our argument on words like this?

  » Solution