The Social Sciences 73 



approach the use of experiments in certain circumstances. 

 The Common Market can be regarded, for example, as an 

 instructive though incompletely controlled experiment of 

 the theory of free trade. 



Those who remain skeptical about the status of eco- 

 nomics and sociology would do well to reflect that these 

 two disciplines are closer to developing a bona fide experi- 

 mental method than are astronomy, meteorology, or even 

 geology. 



Another criterion that has frequently been used to ap- 

 praise the status of a given field of learning is the degree 

 of precision with which it makes and records its observa- 

 tions. Thus in an older time physics and chemistry were 

 frequently referred to as the "exact sciences" — leaving the 

 reader to infer that all other sciences were somehow inexact 

 or frankly rather messy. It is true that there are some physi- 

 cal and chemical measurements that can be made quite 

 precisely, but, generally speaking, most work in the "exact 

 sciences" is not carried beyond three or four significant 

 figures. This represents an accuracy of 1 part in 1,000 to 

 1 part in 10,000. Many operations in the practical everyday 

 world are carried out with a higher degree of precision than 

 this. Several parts of an automobile engine are honed to 

 a tolerance better than 1 in 10,000 and some may go as 

 high as 1 in a million. Banks balance their books to an 

 accuracy which is presumed to be infinite. 



It is its attitude toward inaccuracy rather than its devo- 

 tion to precision as such that distinguishes science from 

 other activities. Unlike bookkeepers, scientists recognize the 

 probability of error in every observation they make. They 

 not only recognize it, they specifically state it. The methods 

 worked out for calculating the degree of error inherent in 

 various kinds of scientific statements were briefly mentioned 

 in Chapter 1 and will not be discussed here. The important 



