Page:Philosophical Review Volume 1.djvu/333

Rh  upon controverted points. These latter, however, are almost inevitable in an elementary text-book.

But the faults of the book, of which the above are examples, are not such as seriously to impair its value. It is eminently readable, and is reliable in all essential respects as far as it goes. It is to be hoped that the little book will be instrumental in interesting a class of readers in psychology who would hardly be reached directly by the more advanced works on the subject.

.

This essay, accepted by the University of Leipzig for the degree of Ph.D., begins with a brief consideration of the two classes of hypotheses underlying the considerations of Weber's law, namely, the hypothesis of absolute and that of relative differences. According to the former the absolute differences of sensation are the same as long as the relation of the variations of the stimulus are the same; according to the latter as long as the relation of the variation of a sensation to that sensation is the same the relation of the stimulus to its variation remains constant. The results of the method of successive gradations (mittlere Abstufungen) are regarded as the strongest proof against the relation-hypothesis; when three sensations are so arranged in intensity that the second occupies the middle position between the extremes it is evident that the differences between it and the extremes are considered equal. The relation-hypothesis is of a rather hyper-psychological character. As Dr. Angell has pointed out, according to its supporters the differences between strong sensations are greater than the corresponding (judged as the same?) differences between weaker sensations. It seems that the difference between sensations has been turned into a sensation — a difference-sensation. According to Merkel, a supporter of the relation-hypothesis, Weber's law would be confirmed if by the method of successive gradations the middle stimulus was the arithmetic middle; just how this could be, it is difficult to understand, and if such a result is required to prove that hypothesis, Dr. Angell's experiments have completely disproved it.

In his consideration of the relation-hypothesis Merkel was led to assert that we could determine the half of a given sensation or twice it just as well as the just perceptible difference. Aside from all theoretical considerations the question arises: is it possible to measure intensities