Richard Dawkins introduced the replicator in The Selfish Gene as his contribution to the units of selection debate. The replicator has since gained widespread adoption in biological evolutionary theory, but also in theories of conceptual and cultural evolution. Soon after the publication of The Selfish Gene, discussion arose about what other entities than genes and memes could count as replicators (e.g. Bateson, 1978), and discussion has recently revived, especially in relation to conceptual and cultural evolution. Cause of the confusion over the role and presence of replicators in these domains of evolution is the ambiguity in the definition of the replicator. Dawkins’ loose definition has recently been made more articulate by biologists and philosophers, but consensus remains absent. In the light of the adoption of evolutionary theory in the social sciences it has become increasingly important to have a clear definition of replicator. An analysis is needed of the different definitions that have been suggested. These need to be assessed on their applicability in the various domains where evolutionary theory is used, to advance to a well-defined replicator concept. Furthermore, there is lack of agreement about whether replication is necessary for evolution. The answer to this problem hinges on the question what is regarded as a replicator. A well-defined replicator concept can potentially take away confusion about whether replication is necessary. The need for such a solution is especially pressing in recently emerged discussion between theorists of cultural evolution: in dual inheritance theory replicators are regarded as the limiting case of cultural evolution, whereas meme-theory regards the replicator as essential for cultural evolution to take place. I suggest to research the uses and definitions of the replicator concept, to advance to a definition of the replicator that illuminates its use in the different domains in which evolutionary theory is used.