While the thing presented in the video seems extremely simple (simplistic), after all, it is a 2-point decision with uncertainties, there is a point which hasn't been adressed at all, and that is the integrity of science.
Everybody knows that in decision theory, one needs to have estimates of the probabilities of the outcomes. We have 4 cases in this idealised scenario:
1) There is AGW and we think there is (and act accordingly)
2) There is AGW and we think there isn't (and act accordingly)
3) There is no AGW, but we think there is (...)
4) There is no AGW, and we think there isn't (...)
As explained, you have to weight the risks and benefits in each of these cases. However, risk and benefit go with probability. An event with 1 000 000 000 death, with a probability of 1E-8, is risk of 10 death.
As such, it is extremely important to give as best as one can, estimations of what are the probabilities of the events. Doing so requires adhering to strict scientific principles, especially concerning uncertainties.
The 2 - way decision tree is much too naive, because we have actually a whole scale of possibilities: there can be no AGW, a little bit of AGW, a lot of AGW... and then the consequences can be very varied. Also, the actions and the cost of what we can do is also continuous, and varied.
But let us follow the speaker, and assume that the worst case scenario is 2).
First of all, there can be an ethical and philosophical debate about
how much of our present well-being we are willing to sacrifice for the well-being of future generations - we could all collectively decide that we prefer US having a good time while we still can, and to hell with future generations. Better still 20 years of fun than going down the deep trench right now for the sake of others.
But let us take the stance that we don't decide that. That we think that a catastrophe in 100 years is really a bad thing. So we should then take action now. How drastic should that action be ? Should we, say, decimate population willingly right now in order to avoid it ? We could. We still have enough nuclear weapons to decimate humanity. Is that a measure which makes sense ? Adhering to "avoid 2) at all costs" would dictate that we do that. I think it is clear that that point is silly.
So *how far* are we willing to go to avoid 2) ? Really "at all cost" ? According to the argument in the video, we should launch those rockets immediately!
The final fallacy is that if we have to avoid 2) at all cost, *we should make the probability of it as small as possible*, even if it runs into the face of evidence. So we should by all means avoid "and we think there isn't". That can be done by banning all statements that give a non-near-100%-probability to AGW. "Change people's minds". I'm affraid that that is what is happening. As people are affraid of 2), they prefer artificially over-estimating the probability of AGW, so that decision making always avoids 2) (and 4), hence increasing the probability for 3) ). However, the price to pay for that is that we cheat with science. So one of the extra costs of 3) is that
science will have lost its credibility. I don't know how much "science" costs, but personally, if I have to choose between the well-being of future generations, or the credibility of science, I go for the second option - there is something Faustian to this, I agree
In order to make correct decisions, it is important to make an honest ascessment of the situation, with all certainties and uncertainties included. It makes no sense to make a "private" assessment of the needed decisions, and then to cheat on the uncertainties to enforce that decision. It is why the decision making, the policy making on AGW should be totally disjunct of the science, and it isn't.
Science has one agenda: finding out what happens, and stating what we know, and what we don't know. Policy making should USE the science to make decisions. When scientists play decision makers, we loose both.
The funny thing is that if you hold the kind of argument as in the video about nuclear power, then suddenly you have a lot of people disagreeing. Fill in the cases in the video with and without nuclear power, and think again.
The real point is that the video does touch upon something. However, instead of thinking about action in a monolithic way, we should make this case for every individual action - like picking between going for more nuclear power, or investing into solar energy, or drilling for more oil, or this or that. In other words, making rational decisions. And *each time* it is important to know as well as possible, what are the probabilities for each case.
