An often neglected aspect of these standard assumptions is that they are *sufficient* not necessary conditions. Perfect information, for example, might not exist, but that doesn't mean there won't be an equilibrium capturing all the potential gains from exchange.
These statements may well form a set of conditions that are sufficient but for the most part not necessary to support Mankiw's conclusions. You could say the much same foe some list of statements in most intellectually mature movements. With Mankiw, I'd go even further and say that if these statements don't have to be true; they just have to approximate reality to a sufficient degree in order to make his case.
But we're talking about something slightly but significantly different. In this context, these assumptions are part of the arguments that Mankiw is making. It is, of course, possible for invalid argumnts to lead to correct conclusions and you can have a trivially valid argument that starts with a false premise, but (putting aside those old logic lessons about the implications of the existence of unicorns) you can't have a valid and meaningful argument based on false assumptions.
It's important to put this in context. Mankiw is arguing that, in addition to being immoral, a return to Clinton era tax rates would cause a sharp drop in productivity and economic growth. It is possible that he's right, but there is considerable historical evidence and any number of counterarguments (many by Nobel Prize winners) that contradict his conclusions. Under those circumstances, I think the burden of proof should rest with the guy who's saying this time it will be different.