Retroviruses aren't simply retrotransposons. Most also have a coding region for an envelope protein. You also have to assume incorporation of the same retrovirus with the same envelope protein in the same place in independent events: and bear in mind retroviruses mutate far faster than their hosts. In fact, we do see evidence of incorporation of new insertions of some of the retroviruses by reinfection events (there was a paper in PNAS last year on this) and they don't go into the same site. There is even one element (K103) that most humans have, but that some sub-Saharan Africans have lost. That would tend to argue against functional significance.
Another problem is that the HERVs seem to have been incorporated a long time ago, and we've been slowly losing them. If they have a functional role or confer some advantange, then why were they only incorporated in a single burst, and why do they seem to be lost over time? Why are they apparently descended from only a small number of clades?
Oohhhhhh...
A real debate?