MATTER 249 



we distinguish coexisting groups of sense-impressions, and 

 therefore only groups of sense-impressions can be said to 

 "occupy" space. This definition would therefore lead us 

 to identify matter with groups of sense-impressions, and in 

 practical everyday life the things which we term matter are 

 certainly more or less permanent groups of sense-impres- 

 sions, not unknowable "things-in-themselves" beyond sense- 

 impression. Now there can be no scientific objection to 

 our classifying certain more or less permanent groups of 

 sense-impressions together and terming them matter, — to 

 do so indeed leads us very near to John Stuart Mill's 

 definition of matter as a " permanent possibility of sensa- 

 tion " ^ — but this definition of matter then leads us 

 entirely away from matter as the thing which moves. It 

 can hardly be said that weight, hardness, impenetrability 

 move ; these are sense-impressions in the brain telephonic 

 exchange ; their grouping, their variation and succession 

 may lead us to the conception of motion, but a sense- 

 impression in itself cannot be said to move ; it is there at 

 the brain terminal or not there. In order to bring motion 

 into the sphere of sense-impression, we are compelled to 

 associate colour, hardness, weight, etc., with geometrical 

 forms, and in making such constructs (p. 41) we pass 

 from the plane of perception to that of conception. I 

 move my hand ; my power to realise this motion depends 

 on my conceiving my hand bounded by a continuous 

 surface. If the physicist tells me that my hand is an 

 aggregation of discrete molecules, then my idea of the 

 motion of the hand is thrown back on the motion of the 

 swarm of molecules. But the same difficulty arises about 

 the individual molecule. I may surmount it by supposing 

 the molecule to be in itself a corporation of atoms, but I 

 cannot conceive the atom's motion unless it be bounded 

 by a continuous surface or else be a point. The only 

 other way out of the difficulty is to construct the atom of 



1 System of Logic, bk. i. chap. iii. That groups of sense-impressions recur 

 in a more or less permanent form is an experience we have every moment 

 of our lives. There is a "permanent possibility of sense-impressions." We 

 are not forced to assert anything about this possibility residing in a super- 

 sejtsitoiis entity matter. 



