As also pointed out in some other threads: it's less that it's dispositive evidence, and more that it corroborates. If you have what seems like a good linguist

As also pointed out in some other threads: it's less that it's dispositive evidence, and more that it corroborates. If you have what seems like a good linguistic theory backed by language phylogenetics or other linguistic-only evidence, and then you do some DNA work and it lines up, that tends to make the linguistic theory stronger, because it at least verifies that humans were moving in the patterns suggested by the language drift. But it's not a "strong signal" by itself, because there are many ways for languages to disperse other than within family groups.