Think about free will in terms of a game-theoretic payoff matrix: There either is or isn't really free will, and you can either behave as if there is or behave as if there isn't.
If there is no free will, it doesn't matter. The payouts are the same regardless of what you do. You can't affect anything. The game is rigged, go home, good night.
However, if there is free will, the payouts vary drastically based on how you act, and the only way to get a good payout is to act as if there is free will.
So, acting as if there is free will is, at worst, one of the ways to get the there-is-no-free-will payout, and, at best, the only way to get the best possible there-is-free-will payout.
Note: If you mention Pascal's Wager in your response, you misunderstand the problem with Pascal's Wager. The problem with Pascal's Wager is that it does not give a way to determine which deity to believe in. The argument was constructed with reference to the Christian deity, but it applies equally well to Shiva or me, for example.
You can't abandon a whole methodology just because it can be misused.
http://en.wikipedia.org/wiki/Normal-form_game
http://rationalwiki.org/wiki/Pascal%27s_wager
http://plato.stanford.edu/entries/pascal-wager/
If there is no free will, it doesn't matter. The payouts are the same regardless of what you do. You can't affect anything. The game is rigged, go home, good night.
However, if there is free will, the payouts vary drastically based on how you act, and the only way to get a good payout is to act as if there is free will.
So, acting as if there is free will is, at worst, one of the ways to get the there-is-no-free-will payout, and, at best, the only way to get the best possible there-is-free-will payout.
Note: If you mention Pascal's Wager in your response, you misunderstand the problem with Pascal's Wager. The problem with Pascal's Wager is that it does not give a way to determine which deity to believe in. The argument was constructed with reference to the Christian deity, but it applies equally well to Shiva or me, for example.
You can't abandon a whole methodology just because it can be misused.
http://en.wikipedia.org/wiki/Normal-form_game
http://rationalwiki.org/wiki/Pascal%27s_wager
http://plato.stanford.edu/entries/pascal-wager/