DISCOURSE ON METHOD. 41 



perhaps more assured, as that we have a body, and 

 that there exist stars and an earth, and such like, 

 are less certain; for, although we have a moral 

 assurance of these things, which is so strong that 

 there is an appearance of extravagance in doubting 

 of v their existence, yet at the same time no one, 

 unless his intellect is impaired, can deny, when the 

 question relates to a metaphysical certitude, that 

 there is sufficient reason to exclude entire assur- 

 ance,^ in the observation that when asleep we can 

 in the same way imagine ourselves possessed of 

 another body and that we see other stars* and 

 another earth, 'when there is nothing of the kind. 

 For how do we know that the thoughts which occur 

 in dreaming are false rather than those other which 

 we experience when_ awake, since the former are 

 often not less vivid and' distinct than the latter^ 

 End though men of the highest genius study this 

 ^question as long as they please, I do not believe that 

 they jwill be able to give any reason which can be <* 

 sufficient to remove this doubt, unless they presup- 

 pose, the existence of God ; For, in the first place, 

 even the principle which I have already taken as a eJL^^ 

 rule, viz. , that all the things which we clearly and ( 

 distinctly conceive are true, is certain only because 

 God is or exists, and because he is a Perfect Being, 

 and because all that we possess is derived from 

 him: whence it follows that our ideas or notions, 

 which to the extent of their clearness and distinct- 

 ness are real, and proceed from God, must to that 

 extent be true. Accordingly, whereas we not 

 unfrequently have ideas or notions in which some 



