92 THE GRAMMAR OF SCIENCE 



suppose instinct and not reason to be the basis of natural 

 phenomena, reflecting his own source of action into all he 

 observed around him. Indeed, it seems to me more 

 logical to find instinct than reason behind the setting and 

 rising of the sun, for instinct at least does not presuppose 

 consciousness. Perhaps if our dog were a Stoic dog the 

 instinct would seem to him inherent in the universe itself, 

 while had he been reared at the parsonage he would cer- 

 tainly fancy his kennel the product of an instinct super- 

 canine. But both dog and man, in thus arguing beyond 

 the sphere of legitimate inference, are also breaking a 

 fundamental canon of the scientific method. This canon 

 is practically due to Newton, and forbids us to seek super- 

 fluous causes for natural phenomena.^ We ought not to 

 look for new causes to account for any group of pheno- 

 mena until we have shown that no known cause is capable 

 of " explaining " it. In our next chapter we shall see 

 more clearly what is to be understood by the words 

 " cause " and " explanation," but for the present Newton's 

 canon suffices to show us that the Stoics were unscientific 

 in seeking for unknown or unknowable " reasons " inherent 

 in nature, until they had demonstrated ,that the only 

 rational faculty known to them — namely, that of man — 

 was insufficient to account for the rational element they 

 professed to observe in nature. What is reason ? Where 

 may we infer its existence ? Can we proceed from this 

 admissible reason to the rational element in natural law ? 

 — these are the questions the Stoics ought logically to 

 have asked themselves. Our wondefr ought not to be 

 excited by the idea that " so vast a range of phenomena 

 are ruled {sic I) by so simple a law as that of gravitation," 

 but we ought to express our astonishment that the human 

 mind is able to express by so brief a description such 

 wide sequences of sense-impressions. This capacity of 



1 Causas rerwn natiiralmm non phires adniitti dcbere, qtiam qiia Ss' vera 

 sint (Sr= earum Pkcvnomeiiis expUcandis siifficiiait. Natura eiiim simplex est 

 &• reruni causis siiperjlnis non liixitriat. Priiicipia. (Editio Princeps, 1687, 

 p. 402.) This "simplicity of nature" is, of course, pure dogma, but the 

 re^nla philosophandi which forbids us to revel in superfluous causes is funda- 

 mental to our view of science as an economy of thought. 



