Now having created these concepts "foxhound," "greyhound," and a lot of other similar ones, I find that they in their turn have a few attributes in common and thus give rise to a new concept "dog." Of course this "dog" is more of an abstraction than ever, the concept of a concept. In fact the peculiarity of this whole process is that, as sometimes stated, the broader the generalisation becomes the less is its depth; or in other words and obviously, that as the number of objects compared increases, the number of attributes common to them all decreases. Ultimately as we saw at the beginning, when a sufficient number of objects are taken in, the concept ("dog" or whatever it may be) fades away and ceases to have any meaning. This therefore is the dilemma of Science and indeed of all human knowledge, that in carrying out the process which is peculiar to it, it necessarily leaves the dry ground of reality for the watery region of abstractions, which abstractions become ever more tenuous and ungraspable the farther it goes, and ultimately fade into mere ghosts. Nevertheless the process is a quite necessary one, for only by it can the mind deal with things. To dwell for a moment over this last point: it is clear that every object has relation to every other object in the world--exists in fact only in virtue of such relation to other objects; it has therefore an infinite number of attributes. The mind consequently is powerless to deal with such object--it cannot by any possibility think it. In order to deal with it, the mind is forced to single out a _few_ of its attributes (the _method of ignorance_ or abstraction already alluded to)--that is a few of its relations to other objects, and to think them first. The others it will think afterwards--all in good time. In thus stripping or abstracting the great mass of its attributes from our object, and leaving only a few, which it combines into a concept, the mind practically abandons the real article and takes up with a shadow; but in return for this it gets something which it can handle, which is light to carry about, and which, like paper-money, _for the time and under certain conditions_ does really represent value. The only danger is lest it--the mind--carried away by the extensive applicability of the partial concept which it has thus formed, should credit it with an actual value--should project it on the background of the external world and ascribe to it that reality which belongs only to objects themselves, _i.e._, to things embodying an _infinite_ range of attributes.