116 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



be considered as a recording instrument which discloses the 

 appearance of the end-object in the biological medium in 

 much the same way that a camera reveals its manifestation 

 in the physical medium. In other words, the visual image 

 of an end-object is simply that object as it appears in the 

 bodily context. The same is true of all sense-data. Hence 

 the aim of the scientist should be, at this point, as in the 

 case of the physical media, to increase as much as possible 

 the range and variety of appearances. 



That this is in accord with actual procedure, both of 

 common sense and of science, seems clear. One observes 

 the end-object in the widest possible range of contexts. 

 He supplements the information received through vision by 

 that obtained through audition, touch, smell, and taste, and 

 he supplements his private information by calling upon 

 other observers to report their findings. In this way he 

 multiplies the appearances of the end-object by inserting 

 different sense organs and different observers in the medium. 



But in this control of the organism, a new motive has 

 emerged. This motive centers about the desire to establish, 

 so far as is possible, normal conditions of observation. It 

 is commonly recognized that the sense organs are important 

 sources of error. Unfortunately, however, one is not always 

 able to detect the presence of error. This is due mainly to 

 ignorance of the nature of the J5-operators. Nerve physi- 

 ology has not yet reached a stage of development which 

 permits the accurate description of the physical impulse in 

 its course through the organism. Thus one is in the position 

 of being able to read the recording instrument, but unable 

 to interpret its information since he knows neither the 

 nature of that which it was designed to record nor the prin- 

 ciples according to which the instrument operates. All 

 that he can do is to multiply the reports of the instrument, 

 and determine their consistency with one another. For this 

 reason the multiplication of appearances serves not so much 

 to increase information about the object as to test the 

 accuracy of some one bit of information about which the 



