July 5, 1901.] 



SCIENCE. 



17 



subjective relations is necessarily manifold 

 and discrete, and so turns scientific study 

 into many separate ways. The peculiar 

 danger of specialism is that it is almost cer- 

 tain to make vision dim, if not to induce 

 complete blindness, or, as virtually the same 

 thing, to create in consciousness curious 

 fancies, strange distortions of reality, seen 

 not with the eye at all, but with the mind, 

 which is always so ingeniously constructive, 

 so original, so imaginative, and one might 

 even say so hypnotic in its power of sugges- 

 tion over the senses. Specialism closes one's 

 eyes and makes one dream. It makes the 

 specialist among physicians see his special 

 ailment in every disorder, and every dis- 

 order in his special ailment, and this so 

 truly that merely to cousult him is to fall 

 his victim. Of course, he can never be 

 wholly wrong, and his unwitting trans- 

 gressions help discovery, but, nevertheless, 

 his situation is full of humor. And in 

 science generally, the specialist dreams, 

 transgressing his own proper bounds with- 

 out clearly knowing that| he has trans- 

 gressed. Why? Because thought, which 

 although often apparently suppressed and 

 abused never actually deserts experiment 

 and observation, is so much greater than 

 vision, than mere sensuous perception. In 

 spite of the specialist being all eyes for his 

 own peculiar interest, the thought that is 

 within him, being bound to conserve an in- 

 divisible universe in every particular thing, 

 leads him, thoughtless devotee that he is, 

 patiently repeating his sacred syllable, into 

 most wonderful visions, projecting his con- 

 sciousness to regions of such logical sub- 

 tlety and marvellous construction as was 

 certainly never known before, unless, per- 

 haps, among those Eastern sages who fed 

 their minds on ' om.' A specialist, he sees 

 the universe, not knowing in his blindness 

 or in his dreaming that it is the universe ; 

 and his danger, the danger of all special- 

 ism, is that he may never awake. 



Thus mathematics and physics and 

 chemistry and biology and psychology, not 

 to say also the social sciences, are depend- 

 ent upon the visions of specialism. Each 

 of them may indeed be special, but thought 

 insists upon making its object conform to 

 reality, which is never special, so that in 

 each there do and must arise abstractions, 

 logical constructions, for the others. When, 

 for example, a physical scientist insists 

 on seeing his world of material phenom- 

 ena only physically, while in reality it is 

 and must be a world of chemical process 

 also, and even of vital and mental char- 

 acter, he is bound to admit to his think- 

 ing what he will call working hypotheses, 

 formally true to his physical standpoint, 

 but what any outsider, in order to explain 

 why they are hypotheses that work, must 

 call compensating conceptions, in short 

 logical constructions that are substitutes 

 for the neglected points of view. A science's 

 working hypotheses are thus as if doors in 

 the paneling by which the other sciences 

 are secretly admitted to a room that seems 

 tightly closed to all comers. Every science, 

 in short, and this the more as the science 

 is special and objective and exact, enter- 

 tains the others unawares. Tennyson's 

 ' flower in the crannied wall ' is nothing 

 in its all-inclusiveness when compared 

 with a well-developed special science. In 

 a sense that is indeed coming to be widely 

 appreciated, no science ever does or ever 

 can live unto itself alone. It may will to, 

 but it does not and cannot. 



But what are these ' working hypotheses ' 

 that work because thej^ are ' compensating 

 conceptions ' or ' doors in the paneling ' ? 

 Some illustration of the foregoing is now 

 imperative. Illustration, however, is diffi- 

 cult, very difficult, for a reason which the 

 scientists will allow me to mention. They 

 know too much about the sciences, while I 

 know too little. Still, as too much knowl- 

 edge is often blinding and so is only a form 



