Doubting Didymus

Mutability is definitely neccessary. If a replicator can not change at all, then evolution simply can never happen in any sense, and the replicator will retain its very first form for all time. That is not evolution by any stretch.

As for DRE, it makes more sense when it is read in context. The authors of the book that I adapted this version from were discussing certain 'replicating' chemical cycles. These cycles could never really 'evolve' in their opinion, though they were technically both replicating and mutable. Thus, they included the criteria of differential replication efficacy. You should keep in mind that the difference does not need to be due to features of the replicators itself, and can involve such factors as drift and environmental change.

If you have a look for where I first posted this, you will see a little conversation I had with Rufus on the topic of DRE, where I agreed that, while it may technically not be neccessary for evolution to occur in theory, it is vital for evolution as we know it.

I do think that the definition of evolution should not be limited to uncontroversially 'living' things. The area is far too grey. No-one doubts that viruses evolve, yet no-one agrees on their 'living' status.

I don't know your stand on the replicating clay crystal theory, but if you imagine for a moment that the clay crystals do satisfy the above criteria, then they certainly must be able to evolve. What are you going to do with a life specific definition when we finally manage a self replicating RNA molecule? Are you going to call it 'alive'? Are you going to call it a 'gene'? What I would do is see if it satisfies my first three criteria and if so, it evolves. If it satisfies the fourth as well, then it could evolve into an organism.

Peez

Mutability is presumably required to generate variability in the population, but I do not see it as necessary to the definition of evolution per se. For example, if a scientist was able to create a population of self-replicators which replicate themselves perfectly, but which are not identical, then evolution can still occur. Of course, there is a definite limit to how far such evolution could go. This was my point. I certainly agree that DRE, when defined so that it does not necessarily imply features of the replicator itself, and so that it refers to the heritable unit, is necessary for evolution as I understand it. I did not mean to state that "evolution" should be restricted in this sense, only "biological evolution". Grey areas remain, just as they do in any area of biology, because "biology" is the study of life and the limits of "life" are fuzzy. I don't see why not, by some definitions of evolution. These are interesting questions, but I am not sure of what you are getting at. I already pointed out that "life" is difficult to define. Why would you expect that a self-replicating strand of RNA is not already alive? From your description I would tend to call such a bit of RNA as a gene.

I believe that the important issue here would be how something evolves. The theories that deal with the evolution of stars, or clay crystals, or a manuscript, or culture, or a population of organisms, are different (as far as I can tell). Just about the only thing that they have in common is change. Certainly some resemble each other more than others, and there may be grey areas, but the study of these phenomena is based on different theoretical considerations. I would suggest that this is the most important consideration in defining "evolution" in any one context.

Peez