It has been a while since I looked at this literature (since shortly before I got married, as it happens) but my conclusion at the time was that it is not all selection and that the underlying causal factors are two: First, as detailed with great (nay, perhaps even creepy) joy by the female authors of this book, wives cause husbands to do fewer earnings-reducing fun things like drinking heavily. Second, casual observation suggests that (many) single men spend a lot of time and effort looking for sex. Post-marriage, in most cases, that time and effort is devoted at least in part to work.
I should note that my theory has a testable hypothesis, which is that married men who cheat a lot will have a lower, or perhaps zero, wage premium. It also has more straightforward implications that could be tested using time use data. If you write a paper on this, please thank me and send it along. I would like to read it!