{Footnote 1: The truth from which I draw my proof may he either (1) of an objective and universally valid character; in that case my proof is veracious, secundum veritatem; and it is such proof alone that has any genuine validity. Or (2) it may be valid only for the person to whom I wish to prove my proposition, and with whom I am disputing. He has, that is to say, either taken up some position once for all as a prejudice, or hastily admitted it in the course of the dispute; and on this I ground my proof. In that case, it is a proof valid only for this particular man, ad kominem. I compel my opponent to grant my proposition, but I fail to establish it as a truth of universal validity. My proof avails for my opponent alone, but for no one else. For example, if my opponent is a devotee of Kant's, and I ground my proof on some utterance of that philosopher, it is a proof which in itself is only ad hominem. If he is a Mohammedan, I may prove my point by reference to a passage in the Koran, and that is sufficient for him; but here it is only a proof ad hominem,}
XVII.
If your opponent presses you with a counter-proof, you will often be able to save yourself by advancing some subtle distinction, which, it is true, had not previously occurred to you; that is, if the matter admits of a double application, or of being taken in any ambiguous sense.
XVIII.
If you observe that your opponent has taken up a line of argument which will end in your defeat, you must not allow him to carry it to its conclusion, but interrupt the course of the dispute in time, or break it off altogether, or lead him away from the subject, and bring him to others. In short, you must effect the trick which will be noticed later on, the mutatio controversiae. (See § xxix.)
XIX.
Should your opponent expressly challenge you to produce any objection to some definite point in his argument, and you have nothing much to say, you must try to give the matter a general turn, and then talk against that. If you are called upon to say why a particular physical hypothesis cannot be accepted, you may speak of the fallibility of human knowledge, and give various illustrations of it.
XX.
When you have elicited all your premisses, and your opponent has admitted them, you must refrain from asking him for the conclusion, but draw it at once for yourself; nay, even though one or other of the premisses should be lacking, you may take it as though it too had been admitted, and draw the conclusion. This trick is an application of the fallacy non causae ut causae.