Let us look at a number of similar objects, say a group of five apples. We observe that the relation of the number five to the group of objects in front of us is purely external and accidental. In applying to it the conception 'five' we combine the single objects into a group and give it a name, or numerical label, which has nothing to do with the nature of the items making up the group. This way of thinking, we may observe, is of exactly the kind which the nominalists of the Middle Ages attributed to every conception formed by the human mind. In fact, the process of counting is a process of pure abstraction. The more differentiated are the things which we want to combine into a group through the process of counting, the further this abstraction has to go. We can count apples and pears together under the collective conception of 'fruit'; if turnips are added, we must help ourselves out with the conception 'vegetable products'; until finally we deal only with 'things', without considering any qualitative differentiation. Thus the conception of number is created solely within the human mind, which applies it to things from outside. From the moment when human consciousness was unable to attribute to itself any other than a purely nominalistic mode of comprehension it was inevitable that all explanations of natural phenomena would have two results: (1) the exclusion from observation of everything that could not be conceived in terms of numbers, and (2) an endeavour to find for every numerical relationship capable of empirical proof an explanation which could be interpreted as the result of taking qualitatively identical units and counting them. For this method of forming conceptions is the only one which nominalism can accept with a good conscience. The fact that in so doing it is led ad absurdum has only quite lately occurred to it. For if by the logical following of this path - as in modern theoretical physics - the whole universe is dissolved into units which can no longer be distinguished from each other, then it will become impossible to count these parts, for it cannot be established whether any given one of these hypothetical elemental particles has been counted or not. None the less, Eddington claimed to have found the exact number of particles composing the universe - a number with 80 figures - by using a special calculus, but this number is valid only on the supposition that the particles cannot be counted because they are indistinguishable!9