208 HOWAED V. KNOX : So naturally and irresistibly, especially in the case of sight, do we carry on the process of thus reducing events to intelligibility, that it requires a great effort to concentrate attention on the succession of perceptions simply as such. Perhaps the easiest way to thoroughly realise that there is no routine in perceptions is to close our eyes in some place where a variety of sounds are to be heard, either in the country or on a busy street ; and to avoid, so far as possible, the ' outward reference ' of our sensations. We shall, I think, have to admit that there is no rule by which the series of sounds that we call a chaffinch's song should suddenly follow on that series which we call a blackbird's song, or why the latter should towards its close be mingled with what we regard as the separate sound of a cow lowing in the distance. So far we have been considering events the supposed antecedents of which lie outside consciousness. But pre- cisely the same thing holds good if we regard the matter from the other side, and consider the forward progress of events. We can always break off our perception of a con- nected series of physical changes by simply turning our back and walking away. This truth applies equally, of course, to cases where there is no voluntary breaking off of perception ; as when a meteor flashes into sight, only to disappear instantly from our gaze. In such cases the last perceived stage in the series is not followed, in percep- tion, by that which is reckoned as its proper effect, or the proper succeeding stage. In other words, so far as our perceptual experience is concerned, it is not true that the same antecedents are always followed by the same conse- quents. There is routine (or what may be regarded as such) and we perceive bits of it ; but there is no routine of perceptions. The fact is that disconnectedness of perceptions is not the exception, but the rule, for anything but very short periods of our mental life. And from what has gone before it follows that even when the perceptions are not disconnected (i.e. even when they do relate to a connected series of physical changes) this does not constitute a " routine of perceptions ". For this thread of perception may be taken up or broken off at any point. The contrast between sequence in sense and the order of physical events is, in truth, even greater than is apparent at first sight ; for where an untrained observer thinks he has actually perceived all the details that go to make up a given complex of phenomena, a rigorous examination will
