Its essential deficiency lies in ignoring the different rank of statements, which expresses itself most clearly in the fact that for the system of science which one takes to be the "right" one, one's own statements in the end play the only decisive role. 

It would be theoretically conceivable that my own observations in no way substantiate the assertions made about the world by other men. It might be that all the books that I read, all the teachers that I hear are in perfect agreement among themselves, that they never contradict one another, but that they are simply incompatible with a large part of my own observation statements. (Certain difficulties would in this case accompany the problem of learning the language and its use in communication, but they can be removed by means of certain assumptions concerning the place in which the contradictions are to appear.) According to the view we have been criticizing I would in such a case simply have to sacrifice my own "protocol statements," for they would be opposed by the overwhelming mass of other statements which would be in mutual agreement themselves, and it would be impossible to expect that these should be corrected in accordance with my own limited fragmentary experience. 

But what would actually happen in such a case? Well, under no circumstances would I abandon my own observation statements. On the contrary, I find that I can accept only a system of knowledge into which they fit unmutilated. And I can always construct such a system. I need only view the others as dreaming fools, in whose madness lies a remarkable method, or -- to express it more objectively -- I would say that the others live in a different world from mine, which has just so much in common with mine as to make it possible to achieve understanding by means of the same language. In any case no matter what world picture I construct, I would test its truth always in terms of my own experience. I would never permit anyone to take this support from me: my own observation statements would always be the ultimate criterion. I should, so to speak, exclaim "What I see, I see!" 

VI 
In the light of these preliminary critical remarks, it is clear where we have to look for the solution of these confusing difficulties: we must use the Cartesian road in so far as it is good and passable, but then be careful to avoid falling into the cogito ergo sum and related nonsense. We effect this by making clear to ourselves the role which really belongs to the statements expressing "the immediately observed." 

What actually lies behind one's