Kauffman is arguing that biological uses cannot be deduced in advance: lens crystallins from ordinary enzymes, thermoregulatory bumps on dinosaurs into flight feathers. The swim bladder is his standing example, because once a fish has a sac of air and water it is also, unintentionally, somewhere a worm can live. He uses it against the idea that evolution's openness can be specified ahead of time, which is where he says machines get stuck.
is my, they're all over the place.
Of course.
…The same thing. A lung got repurposed. And you can't deduce it. That is why evolution is open-ended and cannot be deduced. It's non-deducible, and it's a construction. There's another step that's really worth pausing over, Daniel. So here's this fish with the swim bladder, and they're all over the place. Once there's a swim bladder, could a worm evolve to live only in swim bladders? Sure. Now, we might think that natural selection crafted a functioning swim bladder because it was a benefit to the fish. But natural selection did not craft a functioning swim bladder so a worm could come to live in it. Wow. That means evolution is crafting structures that enable new niches, but selection did not craft the new niches.
Yeah.
Is the open evolution of the bias? The biosphere keeps creating new pathways of flowering becoming.
Yes.