Now the difficulty in Peirce's method, as we have seen, was that this postulate of an external reality unaffected by our opinions would not endure the test for clearness. Every object, says Peirce, reduces to the sum of its effects. The only effect of real things, he says again, is to produce belief. From these two propositions it would seem to follow that reality is a sum of beliefs. But this, of course, eliminates any unchanging reality independent of our opinions about it. We saw further that Peirce defined both belief and meaning as habit and made no distinction between them. Now as belief and meaning are obviously not the same, we are in need of new definitions for these terms. At this point we turned to the interpretation of Dewey. For Dewey the distinction would seem to be that while meaning may well be defined as habit, belief is to be defined as expectation. If we believe in anything, this means that we expect certain results from it. To believe is to suppose that if we were to come into relation with the thing we would find certain effects to come about. From this conception the Deweyan theory of truth would seem to follow immediately. If belief means a sum of expectations, the truth of a belief would mean the verification of these expectations. A true belief simply means one that fulfils expectations. The Deweyan development of the pragmatic method is obviously very much more in harmony with the procedure of science than that of James. James seems to have 'left the track' in his interpretation of the pragmatic method when he related truth to the predominantly valuable. Truth we have found to have no necessary or invariable connection with general value, for many ideas would be acknowledged to be perfectly true while at the same time being either useless or harmful. For Dewey this matter of value has no place in relation to the test of the truth of an idea, for its truth means nothing more than its ability to lead as it promises. We seem, then, it may be said in conclusion, to be confronted with something like the following alternatives: If we believe that Dewey could not have made a correct deduction from the pragmatic method when he developed it into a theory of truth making truth dependent upon fulfilled expectations alone, then very obviously the next step in this investigation is to find the point at which his inference went wrong. This means a re-examination of each step in his reasoning.