Repository | Book | Chapter

The problem of self-evidence, the question whether there are any instances of it and what epistemological value attaches to them, is one that has dogged the footsteps of modern philosophy from its beginning in Descartes up to the present day. The following consideration will show how important the problem is. Our conviction that a proposition is true is often supported by some form of substantiation, in the ideal case by a proof of the proposition. Not all propositions, however, can be substantiated. The substantiation or establishment of any proposition begins with premisses and proceeds by means of inferences to its final conclusions. If we are to accept the conclusion, we must be convinced of the truth of those premisses and the validity of those inferences. Thus no conviction can be a substantiated one unless there are convictions that do not come to us by means of substantiation but are thought to be in no need thereof. Substantiated beliefs depend upon unsubstantiated ones. The same holds for knowledge too. Valid modes of inference ensure that the truth of the premisses is inherited by the conclusions. Thus if our inference is correct and our original premisses true, then the proposition established by the inference is also true. But the truth of the original premisses is not mediated by inference. Mediated and substantiated knowledge can exist only where there is unsubstantiated knowledge. Since an infinite regress of substantiation is out of the question, we are left (so the usual argument runs) with the choice between a dogmatic or conventionalist position that declares certain propositions true without further justification and on the other hand, the assumption that certain propositions possess a self-evidence that guarantees their truth.¹

DOI: 10.1007/978-94-009-5442-7_4

Full citation: