If you think it would serve utility to ‘withdraw tolerance’ from such-and-such dangerous opinions, you’d better think through all the consequences. Your effort might be an ineffective gesture; in which case, whatever you might accomplish, you will not do away with the danger. Or it might be not so ineffective. To the extent that you succeed in withdrawing toleration from your enemy, to that extent you deprive him of his incentive to tolerate you. If toleration is withdrawn in all directions, are you sure the opinions that enhance utility will be better off? When we no longer renounce the argumentum ad baculum, are you sure it will be you that carries the biggest stick?
David Lewis, ‘Mill and Milquetoast’, Australasian Journal of Philosophy, vol. 67, no. 2 (June, 1989), p. 171
The afterlife is a more heterogeneous affair than people have thought. The point of our earthly lives isn’t to divide us into two groups, one to live forever in unimaginable bliss, the other to suffer unimaginable torment. Instead of being tried, we simply discover who we are. Some, perhaps the most fortunate, find out that they are people for whom the adoration of the deity is the highest form of rapture; they appreciate Christ’s sacrifice and are summoned to the presence of God. Others resist the Christian message and develop different ideals for their lives. They are assigned to places in the afterlife that realize those ideals for them. Atheist philosophers, perhaps, discover themselves in an eternal seminar of astonishing brilliance. Each of us finds an appropriate niche.
David Lewis, ‘Divine Evil’, in Louise Antony (ed.), Philosophers without Gods: Meditations on Atheism and the Secular Life, New York, 2007, p. 237-238
Many […] Christians […] are sincerely compassionate; they genuinely forgive their enemies. Yet they knowingly worship the perpetrator. Perhaps they do not like to think about it, but they firmly believe that, in the hereafter, their God will consign people they know, some of whom they love, to an eternity of unimaginable agony. Moved by this thought, they do whatever they can to urge others to join them in faith. Their deep sympathy with the unbelievers is expressed in efforts to persuade others to play by the rules the perpetrator has set. In worshiping the perpetrator, however, they acquiesce in those rules. They are well aware that many will not fall in line with the rules. They think that, if that happens, the perpetrator will be right to start the eternal torture. They endorse the divine evil. And that’s bad enough.
David Lewis, ‘Divine Evil’, in Louise Antony (ed.), Philosophers without Gods: Meditations on Atheism and the Secular Life, New York, 2007, p. 239
It is hard to see just what has gone wrong. But even if we cannot diagnose the flaw, it is more credible that the argument has a flaw we cannot diagnose than that its most extreme conclusion is true.
David Lewis, ‘Illusory Innocence?’, in Papers in Ethics and Social Philosophy, Cambridge, 2000, p. 155
One way for the utilitarian to deal with the Inquisitor is not to argue with him at all. You don’t argue with the sharks; you just put up nets to keep them away from the beaches. Likewise the Inquisitor, or any other utilitarian with dangerously wrong opinions about how to maximize utility, is simply a danger to be fended off. You organize and fight. You see to it that he cannot succeed in his plan to do harm in order—as he thinks and you do not—to maximize utility.
A second way is to fight first and argue afterward. When you fight, you change the circumstances that afford the premises of a utilitarian argument. First you win the fight, then you win the argument. If you can make sure that the Inquisitor will fail in his effort to suppress heresy, you give him a reason to stop trying. Though he thinks that successful persecution maximizes utility, he will certainly agree that failed attempts are nothing but useless harm.
David Lewis, ‘Mill and Milquetoast’, Australasian Journal of Philosophy, vol. 67, no. 2 (June, 1989), p. 159