If we now investigate the required _mode_ of such scientific investigation, we are here again face to face with two contradictory modes of handling the subject, each of which appears to exclude the other and to permit us to arrive at no satisfactory result. On the one hand we _observe_ the science of art, merely so to speak, from an external point of view busying itself with actual works of art, cataloguing them in a history of art, drawing up a sort of commentary upon extant works, or propounding theories which are intended to supply the general points of view for artistic criticism no less than artistic production. On the other hand we find science wholly giving itself up in its independence and self-assured to the contemplation of the beautiful, offering generalizations which do not concern the specific characteristics of a work of art, producing in short an abstract philosophy of the beautiful. 1. With regard to the first mentioned method of study, the starting-point of which is the _empirical_ study of definite _facts_, such is the path everyone must tread who means to study art at all. And just as everyone nowadays, even though he does not actually concern himself with physical science, yet deems it indispensable to his intellectual equipment to have some kind of knowledge of the principles of that science[21], so too it is generally considered more or less essential to any man of real cultivation, that he should possess some general knowledge of art; and indeed the pretension to be ranked as dilletante, or even as genuine _connoisseur_, meets with comparatively few exceptions.