For nothing is more certain than this: all the decisions we have referred to as substantially unjust are supported by undoubtedly acute legal arguments that cannot be gainsaid from the standpoint of formal logic.110 This fact proves that the prevailing method of finding the decision by logical deduction from the legal rule does not work properly. It seems to be an open secret that in the majority of cases we can support both sides of a contention by deductions and constructions drawn from the rule by faultless logic. In courts composed of several justices, where the decisions are notoriously often adopted by a mere majority vote, "the defeated minority could support the opposite decision by drawing from the rule pandectological arguments that would be quite as beautifully logical." If that is admitted, it is quite possible to come to the conclusion that our decisions are "accidental products of juridical construction."

The courts must therefore make it clear to themselves that they are deceived when they take comfort, as they have done in the past, in the fact of having deduced their judgment by formally correct conclusions from legal rules. They should know that the sense of justice of our age, sharpened as it is by social and sociological ideas, demands first of all a just decision, found by the balancing of the substantial interests involved, so as to regulate and govern the legal relations submitted to the judges in a reasonable and comprehensible manner. Such order can be brought out of confusion by no judge who fails to seek first of all what is just, subject to the controlling function of the positive statute. "The certainty of the law is not determined by pandectology, which is little more than a lottery, but by the personal character and substantial experience of the judge"111 To eliminate the full influence of the judge's individuality in favor of a phantom of certainty and equality is con--trary to the nature of law, turns the administration of justice into mechanical machinery and renders it unadapted to real life."112

110 It is even true (see "G" 146, 147) that judgments wrong and unjust from the sociological point of view are supported by greater deductive acumen than those sociologically correct.