610 SECTIONAL TRANSACTIONS.-^. 



Mr. A. J. D. Lothian. — The Rhyme-structure of ' Paradise Lost,' a Study in 

 Unwitting Habit. 



In his preface to ' Paradise Lost ' Milton abjures the ' jingling sound of like endings ' 

 as neither necessary nor ornamental ; and yet his poem is studded with rhymes or 

 inlaid rhymes, e.g. disoBEDiENce, EDEN, SEED In, or quasi-rhymes, e.g. REGAIN, 

 SEAT, PEAK, SEED, in which the vowel sounds coincide although the consonants 

 vary. 



In poems published three years and four years later, viz. in Diana's Prophecy 

 (History of Britain), and in Samson Agonistes, e.g. from 1. 1669, we find the quasi- 

 rhymes adopted as a conscious system. 



The psychological explanation of Milton's apparent inconsistency is that the 

 system was at first probably unwitting, a survival of normal poetic composition. The 

 rhythm of poetry has something of the dissociating effect whose culmination is found 

 in hypnosis ; it helps the poet to unconscious creation and makes the listener more 

 suggestible, conserving rapport. The listener further comes to accept the recurring 

 sounds referred to above as fit or ' inevitable ' words. They rouse processes ready to 

 discharge and procure assent. Citations were given from the introspection of poets 

 and sound charts provided. 



Tuesday, September 11. 



Dr. R. H. Thouless. — Sensations and Step-experiences. 



The statement that the step between a pair of adjacent stimuli A, B appeared equal 

 to the step between the pair B, C was regarded by Fechner as a judgment that 

 Sa — Sb=Sb — So (if Sa, Sb, Sc are the sensations produced by A, B, C respectively). 

 This assumption of absolute magnitudes of sensation which may be subtracted from 

 one another is not, however, essential to Fechner's Law, which may be stated without 

 it (Ebbinghaus, for example, stated the same fact by referring to the equahty of two 

 sense distances). Fechner's statement is objectionable because he treats the Sh on 

 one side of the equation as equal to that on the other, although, in fact, the sensation 

 received from B when adjacent to A is quantitatively different from that received 

 from B when adjacent to C. Much of the current treatment of contrast seems to be 

 vitiated by the same implicit assumption that there is an absolute magnitude of the 

 sensation produced by a stimulus of particular intensity which under different condi- 

 tions of background may be magnified or diminished by the influence of contrast. A 

 preferable way of treating the matter would seem to be one that discarded both the 

 conceptions of absolute magnitude of sensation and of ' contrast,' and recognised 

 simply that the magnitude of a sensation is not a function only of the intensity of the 

 stimulus producing it, but also of the intensities of surrounding stimuli and of the 

 preceding stimuli (i.e. that it is a function also of the spatio-temporal background). 

 This means that what corresponds to a particular stimulus intensity is not a particular 

 sensation magnitude but a particular class of sensation magnitudes. 



In order to deal with the fundamental statement of Fechner's Law (that the above 

 relationship of equality holds when A/B=B/C) on a hypothesis which has discarded 

 the absolute sensation magnitude, it is necessary to replace the conception of equality 

 of the differences between sensations by that of equality in magnitude of the experi- 

 ences given by the pair of stimuli A, B and the pair B, C. Such experiences given by 

 pairs of stimuli we shall call ' step-experiences.' The statement that two step- 

 experiences are seen as equal is as straightforward a report of immediate experience 

 as that two sensations are experienced as equal. 



We are not justified, however, in attributing to the step-experience the absolute- 

 ness of magnitude which we have rejected as an attribute of the sensation. Experi- 

 ment shows that both are relative in the same sense. Two pairs of stimuli which 

 give equal step-experiences under the same conditions of background may give unequal 

 step-experiences when their backgrounds are difierent (this has been proved only for 

 spatial backgrounds, but we may safely surmise that it is true also for temporal 

 background). Just as in the case of a single stimulus and its sensation, so, corre- 

 sponding to a given pair of stimuli, there is not a single magnitude of step-experience 

 but a class of step-experiences of different magnitudes. The particular magnitude 

 of step-experience which will be produced by a particular pair of stimuli will depend 

 not only on the relative intensities of those stimuli but also on their spatio-temporal 



