In my previous post, I showed that there is no experiences or appearances of our world that are not also theories of the world. This means that a way is needed to compare these theories. There must be theoretical virtues such as simplicity and explanatory power that mean that one theory is better than another theory.
No theory can be compared to another theory on the basis of any sensory input at all because these are all theories of that very input. Comparing them on that basis would be an instance of circular reasoning. We must either assume that our senses are reliable and form a theory of the world on that basis, or we must compare our theories of the world on a basis that is independent of the appearances.
The problem of skepticism is a problem that cannot be solved by simply assuming that our senses are reliable. We naturally do assume this, but we do not assume it as an answer to skepticism. We assume it because it is impossible to live life while being unsure of the reality of that life. But the moment we attempt to answer the problem of skepticism, we cannot beg the question against the skeptic. If our senses are reliable, then the skeptic is wrong. He suggests that we do not know that; he says that we lack a good reason to believe that our senses are reliable and that skepticism is false. If the skeptic is wrong then there must be a way to determine which theory of the appearances is correct. Simply assuming that our senses are reliable is just another way of agreeing with the skeptic and moving on to a different problem.
The theoretical virtues are simply a way of comparing one theory with another without respect to the appearances themselves. For example, explanatory power is the ability of a theory to explain the appearances. This includes the number and kind of appearances. But explanatory power makes no references to the actual content of the appearances themselves. It is the same with all of the other theoretical virtues.
Theoretical virtues may solve the problem of skepticism, but it is not immediately obvious that they do. There are several problems that must be dealt with first. The first problem is the problem of “the best of a bad lot”. Suppose that we have two theories of electrical appearances, and the theoretical virtues say that one is better than the other. One theory says that electricity is caused by miniature lightning spirits. The other says that it is caused by tiny, green gremlins. The problem is that both theories have nothing to do with the true cause of electricity! The second problem is the theoretical virtues themselves. Are they real? If they are real, and not merely imaginary, how might individual people – who are all different – come to have knowledge of them as they really are? Finally, even if we can avoid skepticism by use of the theoretical virtues, this may still mean that the average person has no defense against skepticism. Is that true?
Next, I will discuss the nature of theoretical virtues.