NEW ESSAYS 379 



progressions ; but on the other hand an abstraction is 

 not an error, provided we know that what we ignore 

 is actually there. So mathematicians make use of 

 abstractions when they speak of the perfect lines which 

 they ask us to consider, the uniform motions and other 

 regular effects, although matter (that is to say, the inter- 

 mingling of the effects of the surrounding infinite 19 ) 

 always makes some exception. We proceed thus in 

 order to discriminate conditions [considerations] from one 

 another and in order to reduce effects to their grounds 

 [raisons], as far as possible, and to foresee some of their 

 consequences : for the more we are careful to neglect 

 none of the conditions which we can control, the more 

 does practice correspond to theory 100 . But it belongs 

 only to the supreme reason, which nothing escapes, to 

 comprehend distinctly all the infinite and to see 101 all 

 grounds [raisons] and all consequences. All that we can 

 do as regards infinities is to recognize them confusedly, 

 and to know at least distinctly that they are there ; 

 otherwise 102 we have a very poor idea of the beauty and 

 greatness of the universe, and also we cannot have a 

 sound physics, which explains the nature of bodies IOS 

 in general, and still less a sound pneumatics 104 , which 

 includes the knowledge of God, of souls, and of simple 

 substances in general. 



This knowledge of unconscious [insensible] perceptions 

 serves also to explain why and how no two souls, human 



The chapter deals with maxims for avoiding revolutions. 3 refers 

 to the fallacy of the spendthrift : l Each expense is little, there- 

 fore the whole is little.' Cf. Politics, v. 4, i, I303 b 18 : yivovrai fji(v 

 ovv al aracrcis ov Trepl jjuKpuv d\\' etc piKptav. 



99 E. reads ' the infinite which surrounds us.' 



100 That is to say, the more do actual occurrences correspond to 

 our explanation of them. 



101 E. omits ' and to see.' 



102 * Otherwise ' = ' but if we entirely ignore the infinities in 

 things.' 



103 E. reads ' things.' 



104 See note 86. 



