forum 7: week of 27 Feb - pragmatic encroachment
(1) is fairly unproblematic for me. (2) and (3) seem quite related to each other in that both apply in cases where there is a lack of certainty. Indeed, (3) is the assumption doing the most work. (3) is also the most problematic for me, for the following reasons: (a) Supposing that one knows "that O is best" is a huge leap for me because (b) "That O is best" is arbitrary - what does it mean for O to be best? (This question is NOT answered in the Fantl and McGrath article we are reading.) Especially when we assume a lack of uncertainty, the "best" option need not be lopsided in that all other options are "much worse" than the "best" option, whatever that may mean. In other words, the "best" option may not be that much better than the second best option (i.e., the first option beat the second option by a very narrow margin or a close call). This reason is why taking the "maximally" likely option is not always optimal in the probability setting. Hence, I do not buy Fantl and McGrath's argument or reasoning for "if you know that O is best, you are justified in doing O" (page 568) because I almost never know for sure "that O is best"! Establishing the truth of "O is best" is difficult, and the authors (Fantl and McGrath) seem to have swept this important point under the rug.