Humans have been fully modern genetically speaking for around 200,000 years. Since then, different races have developed, but the genetic differences between the races are very small indeed, and we are all very much the same species.
What we have seen since then is what you might term cultural evolution. This is all that separates us from our ancestors 200,000 years ago. Put a modern baby with them or take a baby from them and put it with us and it will grow up to be a fully functioning member of its society.
During this period of cultural evolution, social mores such as sexual taboos have grown around the successful (evolutionarily speaking) patterns of behaviour that are our evolutionary heritage. These social mores may become strong enough to suppress behaviour that previously would have gone unsuppressed and led to the eventual failure of that individual's gene line or even the downfall of their group. Cultural evolution takes the place of biological evolution and in doing so allows different (you might say deviant) 'human natures' to reproduce successfully by having their destructive behaviour suppressed by taboo.
200,000 years in a relatively long-lived creature like humans is a very short time period evolutionarily speaking, but long enough, I'd have thought, for this process to be showing visible effects. If true (and it's a theory I'm formulating as I go along, so I'd be interested to hear criticism of it), then it is saying the direct opposite of Freud. Far from being introduced to suppress destructive behaviour, taboos in fact allow the urge to destructive behaviour to be passed on as the 'deviants' are no longer, or at least not as effectively, weeded out by natural selection. One might expect human nature, under the weight of cultural rules, to begin to diverge from a 'natural' propensity to obey these rules, which would presumably lead to ever more strictly enforced rules, which would produce ever greater divergence.