Page:The academic questions, treatise de finibus, and Tusculan disputations.djvu/85

 think that I ought to adopt the same course, but merely to give the heads of what he said.

And in the first place, they are blameable in this, that they use a most captious kind of interrogation. And the system of adding or taking away, step by step, minute items from a proposition, is a kind of argument very little to be approved of in philosophy. They call it sorites, when they make up a heap by adding grain after grain; a very vicious and captious style of arguing. For you mount up in this way:—If a vision is brought by God before a man asleep of such a nature as to be probable (probabile), why may not one also be brought of such a nature as to be very like truth (verisimile)? If so, then why may not one be brought which can hardly be distinguished from truth? If so, then why may there not be one which cannot be distinguished at all? If so, then why may there not be such that there is actually no difference between them?—If you come to this point because I have granted you all the previous propositions, it will be my fault; but if you advance thither of your own accord, it will be yours. For who will grant to you either that God can do everything, or that even if He could He would act in that manner? And how do you assume that if one thing may be like another, it follows that it may also be difficult to distinguish between them? And then, that one cannot distinguish between them at all? And lastly, that they are identical? So that if wolves are like dogs, you will come at last to asserting that they are the same animals. And indeed there are some things not honourable, which are like things that are honourable; some things not good, like those that are good; some things proceeding on no system, like others which are regulated by system. Why then do we hesitate to affirm that there is no difference between all these things? Do we not even see that they are inconsistent? For there is nothing that can be transferred from its own genus to another. But if such a conclusion did follow, as that there was no difference between perceptions of different genera, but that some could be found which were both in their own genus and in one which did not belong to them, how could that be possible?

There is then one means of getting rid of all unreal per