An experimental science is supposed to perform experiments that find generalities. It's not just supposed to tally up a long list of individual cases and their unique life histories. That's butterfly collecting.

An ecosystem, you can always intervene and change something in it, but there's no way of knowing what all the downstream effects will be or how it might affect the environment. We have such a miserably poor understanding of how the organism develops from its DNA that I would be surprised if we don't get one rude shock after another.

People who have various kinds of politics for whom it is congenial for their apriority politics to say yes, things are getting much worse. They are opposed by people who also have a priori politics saying that this is the best of all possible worlds, capitalism is the greatest thing since sliced bread and you're just alarmists and so on and they can massage the data so it fits them.

We do have tendency, now in biology especially to make up stories, to make theoretical biology a kind of game, in fact we have game theory in biology which is meant to use the theory of games to make predictions or explain things.

When the wrong question is being asked, it usually turns out to be because the right question is too difficult. Scientists ask questions they can answer. That is, it is often the case that the operations of a science are not a consequence of the problematic of that science, but that the problematic is induced by the available means.

In my mind the job of a natural scientist is to bend over backwards to say something which can be demonstrated to be true or at least which is not of such a nature that there's no way to demonstrate that it's false.

If we're going to make a politically responsible and a scientifically sensible claim it should not be stop the world exactly where it is because that's not possible. It has to be to decide what kind of world do we as human beings want to live in.