I should always take the greatest pains to impress upon my pupils that the irrational condition, which alone justifies the withholding of the truth, must be so obvious that there can be no mistake about it, as in the case of the murderer who, with knife in hand, pursues his victim, or of the insane, or of the sick, in regard to whom the physician positively declares that the shock of bad news would endanger life. But I do think that we are bound to face these exceptional cases, and to discuss them with our pupils. For the latter know as well as we that in certain exceptional situations the best men do not tell the truth, that in such situations no one tells the truth, except he be a moral fanatic. And unless these exceptional cases are clearly marked off and explained and justified, the general authority of truth will be shaken, or at least the obligation of veracity will become very much confused in the pupil's mind. In my opinion, the confusion which does exist on this subject is largely due to a failure to distinguish between inward truthfulness and truthfulness as reflected in speech. The law of inward truthfulness tolerates no exceptions. We should always, and as far as possible, be absolutely truthful, in our thinking, in our estimates, in our judgments. But language is a mere vehicle for the communication of thoughts and facts to others, and in communicating thoughts and facts we _are_ bound to consider in how far others are fit to receive them. Shall we then formulate the rule of veracity thus: Intend to communicate the essential facts to those who are capable of making a rational use of them. I think that some such formula as this might answer. I am not disposed to stickle for this particular phraseology. But the formula as stated illustrates my thought, and also the method by which the formulas, which we shall have to teach in the grammar course are to be reached. It is the inductive method. First a concrete case is presented, and a rule of conduct is hypothetically suggested, which fits this particular case. Then other cases are adduced. It is discovered that the rule as it stands thus far does not fit them. It must therefore be modified, expanded. Then, in succession, other and more complex cases, to which the rule may possibly apply are brought forward, until every case we can think of has been examined; and when the rule is brought into such shape that it fits them all, we have a genuine moral maxim, a safe rule for practical guidance, and the principle involved in the rule is one of those secondary principles in respect to which men of every sect and school can agree. It needs hardly to be pointed out how much a casuistical discussion of this sort tends to stimulate interest in moral problems, and to quicken the moral judgment. I can say, from an experience of over a dozen years, that pupils between twelve and fifteen years of age are immensely interested in such discussions, and are capable of making the subtilest distinctions.