I haven't read the book, but I started watching the lecture videos so I can join the discussion. Here the first one:
http://www.justiceharvard.org/2011/03/episode-01/#watch
So I'll start with the
trolley problems:
1) Steering towards the one instead of five: This one seems obvious to everyone, still it's quite disturbing that some people would not steer.
2) Pushing the fat man: The question here is 'what has changed?' The answer: 'Means' -> Direct involvement. Now, intuitions are clearly against pushing the man, but is this a good reason to let four people die unnecessarily? I think no. We
evolved an aversion to (direct) killing innocent people, because of shunning mechanisms in social groups. Being shunned and punished is bad for genes. But the fact that our ancestors evolved this mechanism does not make it 'reasonable' or 'moral'. We also have intuitions towards xenophobia and violence (i.e. in revenge), and we certainly don't think those are moral feelings. Who are we to give abstract 'rules' more weight than the life of four extra people?
Now, if one pins down the exact difference in the situations (the group of students couldn't come up with a precise reason, which is revealing because it shows they're just relying on their gut feelings), it comes down to using someone 'as a means'. But this position would soon run into abusurdities.
Let me introduce trolley problem 2 b): Things are equal to problem 1, two tracks and a choice, steering towards 5 or 1. However, now the track continues in a loop after the 1 person and rejoins(!) the track going towards the 5 people. Assuming the 1 person is fat enough to stop the train, one would be using him as a MEANS. What we have now: An additional piece of track(!) is the cause for four people dying that would otherwise not have died. Deontologists would have to defend this absurdity, arguing for the death and suffering of extra people.
[3) Organ donor example:
This one is phrased very misleadingly. If one looks at the isolated case, clearly the innocent person ought to be killed. But the hospital setting is in no way isolated. Allowing such practices would cause public outrage and fear. People wouldn't want to go to medical check-ups anymore. So those factors have to be considered as well.]
EDIT: Addendum.
I started watching the 2nd video (
http://www.justiceharvard.org/2011/02/e ... two/#watch) where they talk about
cost-benefit analyses for 'human life'. Most people seem to be against even putting a number there, and most would think 'a million' would be way too little. What utter hypocrisy!! Even without the catastrophe in Somalia, we have 27'000 children dying preventable deaths daily(!) in third world countries. Cost-effectiveness analyses have shown that it takes about 400-600$ to save a human life if one donates to the most efficient charities. Any now due to Somalia, the costs are down to less than ten dollars per life! And still not enough people donate. Those people that argue that there cannot be a number, are they spending all but the basic necessities for charity? I don't think so. This is pure hypocricy.
And then the
lion-Christian-colosseum example:
First of all, I think happiness only counts inasmuch as it causes the absence of suffering. Without happiness, there's always longing or boredom which can be counted as suffering. So they weighing is easy, suffering adds up to suffering. Thus, it seems trivial that there's a number of Romans that would justifiy the practice, IF there's no better alternative, if those Romans really have a burning desire to see that Christian tortured, one that cannot be changed in midterm.
Why do people intuitively reject this trivial conclusion? The thing is, we are well capable of empathizing, we can more or less imagine that it is VERY HORRIBLE to be that one Christian. However, we are NOT intuitively good about imagining LARGE NUMBERS. We see this all the time, people make statistical mistakes and have cognitive biases. So it shouldn't surprise us that we can't intuitively grasp the benefits if hundreds of millions of Romans were ecstatic instead of bored. So again, the anti-utilitarians are using misleading intuitions and misleading premises to fool people into reject a perfectly sound, compassionate and consistent ethical framework. Whoever rejects the conclusion is implicitly arguing for extra suffering -- how could that ever be justified?