CHAP. I.] THE STAETING-POINT. 17 



the writers referred to deny. &quot; Self &quot; and &quot; not-self,&quot; therefore, 

 do not fall alone, but with them every train of thought and 

 every process of reasoning, for no one thought can guarantee 

 the existence of a process of reasoning, still less its validity. 

 Thus absolute scepticism is the logical and inevitable fate of 

 all professing Agnosticism, unless they abandon their un 

 tenable and anti-rational principles. It may, however, be 

 said that certainty is not denied as to the existence of &quot; self,&quot; 

 but only &quot; the highest degree of certainty.&quot; 



Professor Huxley tells us this certainty is not of &quot;the 

 highest degree,&quot; inasmuch as it is not &quot; given &quot; to us by 

 &quot;immediate consciousness.&quot; Something, however, may be 

 said in direct contradiction to this, and in support of the 

 assertion that though both the existence of &quot; a state of con 

 sciousness &quot; and the existence of &quot; self&quot; are known seif-exist- 

 with complete certainty, yet that the existence of pSmamyT 11 

 self is known primarily, and therefore with a higher degree 

 of certainty. Both are, indeed, directly perceived by the 

 mind implicitly in the cognition &quot;thinking self;&quot; both are 

 explicitly recognised only by a reflex act. Nevertheless, the 

 &quot;self&quot; can be known in the order of reflection purely as an 

 existing entity ; but &quot; a state of consciousness &quot; cannot be 

 known in that order but as appertaining to some existing 

 mind, which, in the metaphysical order, is primary to it. 

 The primary act of reflex knowledge reveals &quot;self&quot; to us, 

 whereas the reflex recognition of&quot; mental states &quot; shows them 

 to us as states and modifications of the yet more primarily 

 (in the reflex order) and thoroughly known &quot; self.&quot; I there 

 fore join issue with Professor Huxley, and affirm the direct 

 contradictory to his assertions. I maintain that we know 

 ourselves with supreme certainty, and that we know our 

 several mental modifications (though we know them with 

 certainty) with a certainty which is subordinate and 

 secondary in degree. But waiving this reply, it may be 

 affirmed that if so much certainty be allowed as to eliminate 

 all doubt, and to rationally require unhesitating acquiescence 

 on our part in all we do and say and think, all is conceded 



