The ongoing revolution in molecular genetics has led many to speculate that one day we will be able to change the expression or phenotype of numerous complex traits to improve ourselves in many different ways. The prospect of genetic enhancements has generated heated controversy, with proponents advocating research and implementation, with caution advised for concerns about justice, and critics tending to see the prospect of genetic enhancements as an assault on human freedom and human nature. Both camps base their arguments on the unquestioned assumption that the science will realize either their dreams or nightmares. In this paper, I show that their beliefs are based upon two fundamental mistakes. First, they are based upon an unwarranted reliance in a genetic determinism that takes for granted that the traits that we might most want to enhance, like intelligence, aggression, shyness, and even athletic ability, can be causally directed by specific genes. In so doing, character descriptions are reified to be concrete and discrete entities, in this case, genes. Second, they have accepted on faith that there is, or will be, a science to translate their hopes or worries into reality when, in fact, that is unlikely to occur because of the irreducible complexity of phenotypic expression.