The point of all of that modeling is that despite the high child mortality peasant families who want manageable levels of fertility (which would be slow population growth in the aggregate, but they’re not concerned about the aggregate) have to engage in a meaningful amount of fertility control (beyond any infanticide or exposure, which is, again, baked into our infant mortality estimates). The precise mix of methods and techniques is going to vary by culture and region and we are able to glimpse these only very imperfectly, but we can be sure some degree of fertility control was always going on because we don’t see the sort of runaway population expansion implied by a maximum fertility model. Instead, given the high mortality estimates we laid out in part II, we would expect around five or six live births per woman, rather than the c. nine implied by a maximum fertility model. If we were to tweak the variables of our mortality regime (generally downward), we would also be lowering the expected number of births, something that may in part explain the late/late early modern European marriage pattern.