Herbe7't Spencer 133 



ledge is the theme of two successive chapters in the early 

 part ^ of his Psychology — as also under the head of ' Trans- 

 figured Realism/ towards the end of that work — and it is 

 fully enunciated in his First Principles. That we can know 

 nothing but phenomena, that everything absolute escapes 

 us — as being for ever unknowable and beyond the ken of the 

 human intellect — is a cardinal principle with Mr. Spencer, 

 who distinctly tells us"^ that all 'objective agencies' pro- 

 ductive of ' subjective affections ' are not only ' unknown ' but 

 also ' unknowable/ 



But every philosophy, every system of knowledge, must 

 start with the assumption (implied or expressed) that some- 

 thing is really 'knowable' — that something is 'absolutely 

 true ' ; and in the present instance it is evidently intended to 

 imply that the doctrine of the 'relativity of all our know- 

 ledge ' is a doctrine which is really and absolutely true. But 

 if nothing that we can know corresponds with reality, if 

 nothing we can assert has a more than relative or pheno- 

 menal value, why does not this character also appertain to 

 the doctrine of the relativity of all our knowledge ? Either 

 this system of philosophy itself is relative and phenomenal 

 only, or it is absolutely and objectively true. But it must be 

 merely phenomenal, if everything known is merely pheno- 

 menal. Its value, then, can be only relative and phenomenal 

 — that is, it has no absolute value, does not correspond with 

 objective reality, and is therefore false. But if it is false that 

 our knowledge is only relative, then some of our knowledge 

 must be absolute ; but this negatives the fundamental posi- 

 tion of the whole philosophy. 



Any philosophy then which starts with the assertion that 

 all our knowledge is merely phenomenal refutes itself, and is 

 necessarily suicidal. Every asserter of such a philosophy 

 must be in the position of a man who saws across the branch 



1 Vol. i. part ii. chaps, iii. and iv. pp. 193-227. ^ Vol. iii. p. 493. 



