Method of Refutation by Parallel Reasoning

This conclusion would seem false to the NRA, and this argument might also seem to suggest that the same form of reasoning could lead to opposite conclusions: (3) and (3**). Moreover, (2**) seems true. This premise does not say that it is good for outlaws to shoot their children by mistake. Instead, it says that it would be good if nobody else shot their children by mistake. So far, so good. Notice, however, that this latest argument, (1**)–(3**), does not have the same form as the original argument, (1)–(3), because (2**) and (3**) are about what is good and what ought to be law, whereas (2) and (3) are about what is bad and what ought not to be law.

The next question is whether this disanalogy is important. If not, the latest bumper sticker still might refute the original one. The NRA, however, might argue that this difference is important. The fact that a law has bad effects overall does show that the law should not be passed, whereas the fact that a law would have good effects overall is not enough to show that the law should be passed, since the law still might violate individual rights that can- not be overridden by good effects on others. This claim is controversial, but, if it can be defended, then this parallel argument, (1**)–(3**), fails to show that the original argument, (1)–(3), is invalid. More generally, then, a refuta- tion by parallel reasoning works only if the two arguments really do have relevantly similar structures—that is, only if one argument really is just like the other in relevant respects.

In sum, the method of refutation by parallel reasoning can be used to show that an argument is invalid by presenting another argument with es- sentially the same form in which the inference takes us from obvious truths to an obvious falsehood. In response to such an attack, a defender of the original argument has three main options. The defender might

1. deny that the conclusion of the parallel argument is false,

2. deny that the premises of the parallel argument are true, or

3. deny that the supposedly parallel argument really has essentially the same form.

If any of these responses is justified, then the attempt to refute the original argument by parallel reasoning fails.