From replicators to heritably varying phenotypic traits: The extended phenotype revisited

Research output: Contribution to journalArticlepeer-review


The replicator-based extended-phenotype perspective was important in overcoming the tendency to think about evolution in terms of benefits accrued to individuals or groups. Since individuals and groups are targets of selection, not units of evolution, thinking in terms of individual benefits has sometimes led to absurdities. Dawkins did a great service in pointing to these absurdities, and popularizing and developing the gene's-eye view which avoided a lot of the problems. However, he insisted that the basic Darwinian question with respect to adaptation has to remain "who benefits?", and in order to give a coherent answer to this question he transferred the benefit from the organism to the gene. The question, however, makes it very unclear whether its subject is supposed to be the target of selection or the unit of evolution, and creates the illusion that a single beneficiary can be found. Asking instead, "what is selected and why?" avoids this problem, and remains within the basic Darwinian paradigm, although additional questions such as: "what type of new variations are generated, how, when and where?" have to be added if we are to gain deeper understanding of the evolution of reproducers that have evolved sophisticated variation-generating systems. The replicator view made apparent the many problems inherent in a previous view of evolution, and provided a powerful alternative, which stimulated research. However, this view, including its memetic extension, has many theoretical problems which have become apparent during the last 21 years. The approach suggested here, which is inspired by developmental system theory, focuses on heritably varying traits and reproducers, and proposes alternative units of evolution and alternative targets of selection to those suggested by Dawkins. I have argued that a trait/reproducer-oriented view (i) is more consistent with what we know about the relationship between genes and phenotypic traits, (ii) illuminates the evolution of the traits' development and the diverse sources of a traits' stability and changeability over generations, (iii) gives the organism a more active role in evolution, and (iv) goes some way towards solving the complex problems that arise when we study the relationship between development, heredity and evolution. Nevertheless, it is doubtful that this approach would have been developed without the challenges presented by Dawkins in the EP.

Original languageEnglish
Pages (from-to)353-375
Number of pages23
JournalBiology and Philosophy
Issue number3
StatePublished - Jun 2004


Dive into the research topics of 'From replicators to heritably varying phenotypic traits: The extended phenotype revisited'. Together they form a unique fingerprint.

Cite this