This paper argues in defense of theanti-reductionist consensus in the philosophy ofbiology. More specifically, it takes issues with AlexRosenberg's recent challenge of this position. Weargue that the results of modern developmentalgenetics rather than eliminating the need forfunctional kinds in explanations of developmentactually reinforce their importance.

The question of whether the modern evolutionary synthesis requires an extension has recently become a topic of discussion, and a source of controversy. We suggest that this debate is, for the most part, not about the modern synthesis at all. Rather, it is about the extent to which genetic mechanisms can be regarded as the primary determinants of phenotypic characters. The modern synthesis has been associated with the idea that phenotypes are the result of gene products, while supporters of the (...) extended synthesis have suggested that environmental factors, along with processes such as epigenetic inheritance, and niche construction play an important role in character formation. We argue that the methodology of the modern evolutionary synthesis has been enormously successful, but does not provide an accurate characterization of the origin of phenotypes. For its part, the extended synthesis has yet to be transformed into a testable theory, and accordingly, has yielded few results. We conclude by suggesting that the origin of phenotypes can only be understood by integrating findings from all levels of the organismal hierarchy. In most cases, parts and processes from a single level fail to accurately explain the presence of a given phenotypic trait. The debate between the proponents of the modern and extended syntheses is somewhat reminiscent of the duck-rabbit illusion. The two sides are probably talking about the same thing, but from different perspectives. If not, then we argue that the challenge is to do an experiment that rules out the alternative view. (shrink)