Aristotle is funny. In his classes and in his writings (which may have been the same thing, if what we have to read today are really class notes taken by students), he went about pinpointing the truth as precisely as he could; yet in personal encounters, he was not above using a bit of chicanery to win a point.
Artistotle's Topics presents practical advice, systematically arranged, for winning an argument. In service of this goal, it draws in aspects of logic, understanding, and argumentation from his Prior Analytics, Metaphysics, and Rhetoric, and I'm glad I had read at least parts of each of those three books before undertaking this one. Plato taught that rhetoric was to logic as cosmetics was to exercise: the first seeks only appearances while the second aims at an actual good (truth in the case of logic, health in the case of exercise). To fool someone into agreeing with you, Plato said, was a vicious practice. The much more practical Aristotle, on the other hand, knowing that sometimes the good law has to be passed by a less-than-intelligent assembly or the case won before a less-than-perceptive jury, accepted any effective means of winning the moment.
In the Topics, he often speaks of "finding lines of attack," and the book consists mostly of a systematic list of ways to look for chinks in the opponent's armor. For instance, did he make a claim using an ambiguous word? Then disprove it by using the meaning that he did not intend. Never mind that your riposte doesn't address the issue at hand; confusing your opponent helps your case. Likewise, lead him into making points that you can easily refute. And concede only unimportant or tangential points; the uselessness of the conclusions he can draw from them will disconcert him for a moment. As I read through all this advice, it seems to me that Aristotle constantly has an audience in mind, jeering at each stumble and applauding each touch of the rapier.
Readiness to take on any opponent requires proficiency in logic, because of course the fellow might slip up and use a syllogism incorrectly. You, on the other hand, are free to use any specious logic or shortcuts that you can get away with. The venture also requires a knowledge of metaphysics, because every claim your opponent makes involves a subject and a predicate -- terms we learned as kids when we didn't know we were learning Aristotelian metaphysics -- and, because of the nature of existence, the subject and predicate must have an appropriate relationship; in other words, predicates must follow certain rules without which they don't do any good. For instance, the predicate must not use any word used in the subject. (I remember hearing that Aristotelian rule, too, when my grade-school teacher told me I couldn't use a word in its own definition. Of course, credit was not given to the Philosopher.) And a correct predicate must refer to a genus that is ontologically prior to the subject. (For instance, your room has walls and corners, but the walls are prior, because you can have a wall without a corner but not a corner without a wall. So don't use corners to explain walls.) Finally, successful argumentation requires knowledge of rhetoric, because that field of study teaches the art of persuasion through expertise in language and in psychology.
Rather than poking holes in someone else's case, though, all of the book's rules can be turned around to strengthen one's own communication. Aristotle may write almost all of it as a plan of attack on an enemy, but suppose we say, like Pogo, "We have met the enemy, and he is us." The most important rule I see so far in the Topics is that every claim should make the subject more intelligible to the listener. In other words, the predicate of each sentence should be easier to understand than its subject and, as a result, should make the subject better known. Do I always do this? Mostly do this? Ever do this? I should go back and check each sentence of this post to see if it passes the tests, but I'm afraid I'll quickly disconcert myself with the attack and never press "publish."