That is to say, our perceptions, corrected by our experience and our common sense, must serve as guides for us, and we must seize every opportunity to widen their range and increase their accuracy. For millions of years they have been steadily augmenting our store of knowledge. We know, for instance, that when fire touches us it causes an impression which we call pain and that this impression is invariably the same, and always leads to the same results, in all normal human beings. Therefore, we accept it as an axiom that fire causes pain. There are many other ideas that may be and have been established in the same manner: by the fact that they are universal among sane men. But there is also a multitude of things which produce different impressions upon different men, and here we encounter the problem of determining which of these impressions is right and which is wrong. One man, observing the rising and setting of the sun, concludes that it is a ball of fire revolving about the earth. Another man, in the face of the same phenomena, concludes that the earth revolves around the sun. How, then, are we to determine which of these men has drawn the proper conclusion? As a matter of fact, it is impossible in such a case, to come to any decision which can be accepted as utterly and absolutely true. But all the same the scientific empiric method enables us to push the percentage of error nearer and nearer to the irreducible minimum. We can observe the phenomenon under examination from a multitude of sides and compare the impression it produces with the impressions produced by kindred phenomena regarding which we know more. Again, we can put this examination into the hands of men specially trained and fitted for such work--men whose conclusions we know, by previous experience, to be above the average of accuracy. And so, after a long time, we can formulate some idea of the thing under inspection which violates few or none of the other ideas held by us. When we have accomplished this, we have come as near to the absolute truth as it is possible for human beings to come. I need not point out that this method does not contemplate a mere acceptance of the majority vote. Its actual effect, indeed, is quite the contrary, for it is only a small minority of human beings who may be said, with any truth, to be capable of thought. It is probable, for example, that nine-tenths of the people in Christendom today believe that Friday is an unlucky day, while only the remaining tenth hold that one day is exactly like another. But despite this, it is apparent that the idea of the latter will survive and that, by slow degrees, it will be forced upon the former. We know that it is true, not because it is accepted by all men or by the majority of men--for, as a matter of fact, we have seen that it isn't--but because we realize that the few who hold to it are best capable of distinguishing between actual impressions and mere delusions.