If, on the other hand, by "die geltenden Meinungen" Zeller means any philosophical or logical theories generally or universally admitted by thinking men as valid, the answer is that there were none such in the fourth and third centuries B.C. Various eminent speculative individuals were labouring to construct such theories, each in his own way, and each with a certain congregation of partisans; but established theory there was none. Nor can any theory (whether accepted or not) be firm or trustworthy, unless it be exposed to the continued thrusts of the negative weapon, searching out its vulnerable points. We know of the Megarics only what they furnished towards that negative testing; without which, however,--as we may learn from Plato and Aristotle themselves,--the true value of the affirmative defences can never be measured.] Whoever undertakes, _bonâ fide_, to frame a complete and defensible logical theory, will desire to have before him a copious collection of such difficulties, and will consider those who propound them as useful auxiliaries.[45] If he finds no one to propound them, he will have to imagine them for himself. "The philosophy of reasoning" (observes Mr. John Stuart Mill) "must comprise the philosophy of bad as well as of good reasoning."[46] The one cannot be complete without the other. To enumerate the different varieties of apparent evidence which is not real evidence (called Fallacies), and of apparent contradictions which are not real contradictions--referred as far as may be to classes, each illustrated by a suitable type--is among the duties of a logician. He will find this duty much facilitated, if there happen to exist around him an active habit of dialectic debate: ingenious men who really study the modes of puzzling and confuting a well-armed adversary, as well as of defending themselves against the like. Such a habit did exist at Athens: and unless it had existed, the Aristotelian theories on logic would probably never have been framed. Contemporary and antecedent dialecticians, the Megarici among them, supplied the stock of particular examples enumerated and criticised by Aristotle in the Topica:[47] which treatise (especially the last book, De Sophisticis Elenchis) is intended both to explain the theory, and to give suggestions on the practice, of logical controversy. A man who takes lessons in fencing must learn not only how to thrust and parry, but also how to impose on his opponent by feints, and to meet the feints employed against himself: a general who learns the art of war must know how to take advantage of the enemy by effective cheating and treachery (to use the language of Xenophon), and how to avoid being cheated himself. The Aristotelian Topica, in like manner, teach the arts both of dialectic attack and of dialectic defence.[48]