EMPIRICAL TOOLS AND EMPIRICISM 151 



edly equivalent denotations is possible, we are thrown back on our 

 own appraisals of the reliability of different extrapolations. We seek 

 then to readjust the denotations we have assigned to our concepts 

 until once again the data collected do "make sense." Proceeding so 

 we do not wholly give way to human caprice: to "make sense" of ex- 

 perience is, after all, the business of science. Nor do we in any way 

 turn our backs on "hard reality": readjusting our denotations, we 

 simply find that we can make sense of the data collected in one way 

 and not in another. And always, whatever the risks that in future we 

 may have further to reconstruct the denotations of our indicative 

 concepts, we hold firm that indispensable bridge that science ever 

 maintains between the conceptual realm and the perceptual realm— 

 however vertiginous may become the first, and however extended the 

 second. 



Observation and Experiment 



Epistemologically the distinction between observation and experi- 

 ment is a thin one. However complicated the equipment we bring to 

 bear in the laboratory, the ultimate operation in any experiment is 

 always the making of one or more observations. The ancient astrono- 

 mer, making naked-eye observations of the sky, and the modern phys- 

 icist, taking "pointer readings" from his "counter," are engaged in 

 epistemologically equivalent undertakings. However, the important 

 point is not this trivial equivalence but the enormous advancement of 

 our conceptual powers produced when we pass from observation to 

 experiment. In observation we can note only what occurs in the 

 "natural course of events." Passing to experiment, we can for the first 

 time observe what happens in a multitude of other circumstances we 

 ourselves contrive. Though these conditions be "unnatural," the data 

 we obtain may notably advance our understanding of what happens 

 "naturally." Thus we learn much about the normal function of the 

 endocrine glands by studying the abnormal conditions of organisms 

 from which the glands have been removed by surgery. 



Experiment cannot produce, but sometimes practically evokes, the 

 concepts we alone can create. Thus movement observed in nature, 

 rapidly suppressed by frictional effects, suggested the thought that 

 rest is the "natural state" of bodies. Had it been possible to observe 

 movement in the laboratory under conditions that minimize fric- 



