None of the premises are examined; they're all assumed. Clearly, as we all agree the argument is unsound at least one of them (including those implied but not delineated) must be false, and it's not particularly important which. What Morendil asked for, more or less, was a rational argument against private homosexuality.

Obviously, no unsound argument should be stable under reflection, but from the point of view of Classical Logic this seems to satisfy the requirements.

If you'd like it more formally, I'll write out all the premises in full and come up with a cost/benefit analysis / natural deduction proof - but it wouldn't help answer the request, because we're not discussing whether or not private homosexuality is bad, but whether there are any (close enough to) rational arguments for the other side.