WHY DO WE EEMEMBEE FORWAEDS AND NOT BACKWAEDS? 579

that is neither: force has to be brought to bear upon popular usage to impose even so much limitation. The inclusion of the neutral modes, along with pleasure and pain, in the one genus 'Feeling,' must be logically justified on the ground of the importance of the points of community of the three species. Having implicitly handled these in the previous discus- sion, I need not dwell further upon them now. WHY DO WE REMEMBER FORWARDS AND NOT BACKWARDS? By F. H. BRADLEY. To the reader who is new to this question it may wear the appearance of a paradox. He may reply that to go forwards is obvious and natural. But if I ask why should my memory go only one way, why should memory move never from the present to the past he may find that what seemed obvious seems now merely false. Still, if he attends to the subject and confines him- self to bare memory, if he discounts, that is, the cases where we reach the cause from the effect, or in general reconstruct a whole from its interdependent parts, he is likely to admit the existence of the problem. And he may accept the conclusion that the reproduction of a series has but one possible direction, from the earlier to the later. Whether with Prof. Bain he will add this tendency to a long list of " ultimates " (MiND No. 44, p. 469), or will try to find some explanation, I cannot foretell. For myself, though I of course accept the fact of this general tendency, I am not sure that it has no exceptions. I do not believe in the impos- sibility of remembering backwards, and even doubt if sometimes that does not happen in fact. And, so far am I from accepting our habit as an ultimate, that I venture to find no difficulty in seeing how it was acquired, or at all events may have been so. I ought perhaps to begin by attempting to explain how it is possible to reproduce a time- series at all. This would be a far more serious task, and I cannot here undertake it. And so the question must be simply as to the direction of the recall. Yes, a reader may suggest, the problem is, Why, when time itself goes forward, memory is tied also to that direction. But this is not the way to put the question, and we must begin by purging our- selves of such ideas about time. The stream of events does not really run from the past into the future, and it is easy to see that this flow is our own construction. We find, on reflection, that we really do not perceive the future and the events past and present streaming onwards and into it. What we think we see, upon reflection, is a succession of events, in which what we call the present constantly, in part at least, becomes new, and in part slips away backward into what we call the past. And this con-