This assumption is too extreme to be necessary. The author needs only to assume that some populations have different communication needs. He doesn’t need to assume that all populations have different communication needs.

The author’s argument is that international trade itself will eventually lead to the creation of new dialects. So the behavior of people not engaging in international trade is irrelevant.

The author’s argument is about what happens when trade leads to a universal language. He doesn’t have to assume that this is the only possible way a universal language could arise.

The author’s argument is about the creation of dialects for a new language; whether dialects of older languages persist is irrelevant. The author could believe that both kinds of dialects will coexist.

Suppose that this weren’t the case, and populations would not vary in their communicative needs. The author’s basis for believing that new dialects will arise because of trade would then vanish. Consequently, the author must assume this is true.