176 WHAT IS SCIENCE? 



The last of the main sources of popular error is 

 connected with a peculiar form of law which brevity has 

 forbidden us to discuss hitherto. We have spoken of 

 laws as asserting invariable associations. Now, a very 

 slight acquaintance with science will suggest that this 

 view is unduly narrow ; it may seem that some laws in 

 almost all sciences (and almost all laws in some sciences) 

 assert that one event is associated with another, not 

 invariably, but usually or nearly always. Thus, if 

 meteorology, the science of weather, has any laws at all, 

 they would seem to be of this type ; nobody pretends 

 that it is possible at present, or likely to be possible in the 

 near future, to predict the weather exactly, especially 

 a long time ahead ; the most we can hope for is to discover 

 rules which will enable us generally to predict rightly. 

 Another instance may be found in the study of heredity. 

 It is an undoubted fact that, whether in plants, animals, 

 or human beings, children of the same parents generally 

 resemble each other and their parents more than they 

 do others not closely related. But even the great 

 progress in our knowledge of the laws of inheritance 

 which has been made in recent years has not brought 

 us near to a position in which it can be predicted (except 

 in a few very simple cases) what exactly will be the 

 property of each child of known parents. We know 

 some rules, but they are not the exact and invariable 

 rules which we have hitherto regarded as constituting 

 laws. 



The pure scientific view of such laws is very interesting. 

 Put briefly it is that in such cases we have a mingling 

 of two opposed agencies. There are laws concerned in 

 such events, laws as strict and as invariable as those 

 which are regarded as typical, but they are acting 

 as it were on events governed not by law but by chance. 

 The result of any law or set of laws depends (cf. p. 167) 

 not only on the laws but on the events to which they are 



