Balancing costs benefits and probabilities

Someone giving the above argument about the NHS might concede that there is some possibility, even if the number of doctors were increased, that the NHS would not improve. Nevertheless, so long as he or she is reasonably certain that it would, it would be simplest to leave the conditional P2 as it is, rather than inserting 'probably' before its consequent, thereby making the argument inductive rather than deduc-

In other cases, however, practical arguments are more plausibly reconstructed as inductive. For we sometimes have to balance costs and benefits in a more complicated way. For example, suppose you are repairing a window. Having removed it, you are invited to a party, which is taking place now. You know the party would be a lot of fun, but although you could very easily repair the window tomorrow, there is no time to replace the window before going to the party; either you leave the window off and go to the party, risking the possibility of rain getting in (assume the

4 In endorsing this style of argument as valid we are assuming that all practical considerations can be subsumed under the categories 'cost' and 'benefit'. We do so for simplicity; but it should be acknowledged that many philosophers hold that rights and duties are practically relevant considerations which cannot be explained in terms of cost and benefit (cannot be explained, that is, from a utilitarian or consequentialist point of view). More generally, the relation of rule-theoretic concepts such as rights and duties to value-theoretic concepts such as cost and benefit is a central concern of the philosophical study of ethics. A way to accommodate this without complicating our argument reconstructions is simply to think of rights and duties as entering into the calculation of cost and benefit in their own right, without any assumption that can be explained in terms of some independent conception of cost and benefit.

window is too high up for there to be a risk of burglary), or you continue working on the window and miss the party. What should you do? Obviously the benefit of going to the party is high, but so would be the cost of rain getting in. If we assume that these are roughly equal, then clearly your decision should rest on the probability of rain: if the probability of rain is less than 1/2, then you should go to the party; if it is higher than 1/2, then you should not. But suppose the cost of rain getting in would be much greater than the benefit of attending the party. In that case you should not risk going to the party even if the chance of rain is fairly low. We can represent this in a table:

0 0

Post a comment