Several influential psychologists have attempted to estimate to what extent human happiness levels are directly controlled by genes by comparing the happiness levels of identical twins raised apart. If we discover that the happiness levels of identical twins raised apart tend to be closer than the happiness levels of fraternal twins raised apart, this is taken as evidence that average happiness levels are largely controlled by genes. However, if it turns out that identical twins' happiness levels tend to be substantially (...) different, as different as fraternal twins raised apart, it has been argued that this implies that the environment largely determines happiness levels. I contend that this interpretation of the data rests on a set of questionable, closely related assumptions: a) that pairs of identical twins raised apart are raised in substantially different environments; b) that genetically identical twins aren't identical in other pertinent ways; and c) that average levels of affect that are correlated with certain sets of genes in the environments the twin studies were conducted in will be correlated with those genes in other relevant environments because those genes govern average affect levels. In this paper, I will show that these assumptions are false and explain how to properly go about investigating the contingent causal links between genes and happiness. In the end, I argue that there is no good evidence that our genes delimit our hedonic potential and that the fact that this misinterpretation of the data has been widely disseminated is potentially harmful to human well-being. (shrink)