[hga]

If you could develop that thesis I'd give it a hearing, but ... well, perhaps start with proposing a mechanism that "interleaves it in some subtle way" or some observations that suggest that's the case.

Me, I hope we're a lot more "modular" and that "genetic seg-faults" will continue to be by definition rare ("by definition" unless they happen after an organism breeds and is otherwise not responsible for its children). What makes you think evolution would go in the non-robust path you think might be the case?

[moens]

Yeah, I would love to have time to develop this thesis as well. Let me think about it a bit, and I'll try to post a five sentence postulate.

As concerns nature, my observation is that (if I can personify nature, and grant it intention for the argument), it does not mind eliminating large groups of children if some (at least one, I guess) viable children remain. As a matter of fact, eliminating a lot of less viable children in favor of few more viable children is kind of the rule for selection.

So we take a group of strongly viable lines, contaminate them, and (in nature's much longer perspective) make them less viable. Marked for elimination.

We usually think this elimination comes from direct competition, and I would agree that in our experience so far this is the observed case. But what is observable in much GM research is that after three or so generations, animals consuming the GM product stop reproducing. The "why" here is still being researched and is at best poorly understood.

So, I argue that nature in this case is taking the robust path. Eliminating contamination.