Hume and Peirce on Belief, Or, Why Belief Should Not be Considered an Epistemic Category

Joseph C. Pitt Hume and Peirce on Belief, Or, Why Belief Should Not be Considered an Epistemic Category Most of the history of epistemology in western philosophy has been involved in one way or another in an attempt to define "knowledge" or "to know" or "knowing".1 The first 2000 years or so were characterized primarily by seeing knowledge as some form of certainty. Hume's attack on the certainty of knowledge in his Treatise of Human Nature more or less put an end to that effort. The twentieth century saw a lot of effort expended on attempts to characterize knowledge as justified true belief even after Gettier's famous little paper. Whether or not there is a possible connection between knowledge and certainty, I want to argue here that the move to characterizing knowledge as justified true belief, or as belief in any form, is a mistake. In support of what is admittedly controversial, I will enlist Peirce and Hume. What I am after is a way out of the now traditional disconnect between knowledge and action, or to put it more dramatically, man and the world. The main problem facing epistemology since Descartes has been how to put man back in the world. Another way to put this is to characterize the main Cartesian move not so much as dualistic but as solipisitic. Man the knower is isolated within the confines of his own thoughts. Knowledge is something an individual has primarily because of certain inputs, depending on the agenda it can be limited to sensory inputs or it can include divine input or social. But whatever the source of the input, the problem has been to show how the knower can be certain with regard to content of knowledge, and if not certain, then at least confident in his or her knowledge. But, even if confident with respect to the knowledge, as the product of individual cogitation, it is by definition solipistic, and, more to the point, nothing follows about what the knower should do. So, in the long run the problem is not solipistic epistemic justification, it is how to turn the knower into a doer. It matters not if you are an empiricist or a rationalist, on most contemporary accounts, knowledge remains something locked in your head, having little obvious connection with the world. But, it will be objected, empirical knowledge is about the world, so how can you say it is not connected to the world? So long as knowledge is Transactions of the Charles S. Peirce Society Spring, 2005, Vol. XLI, No. 2 344 Joseph C. Pitt conceived of as the product of individual cogitation, it is not clear that it is about the world, since it is completely subjective. And when you try to determine if what you know is the same as what I know, enter the problems of translation and under determination and on and on. The grab-bag of 20th century philosophical problems is a direct function of the Cartesian solipistic move into which almost everyone subsequently bought. Hume wrestles with it in the Treatise as he shows that knowledge is impossible on these grounds and belief is reduced to a feeling. Kant produces the magical schematism to get him from the mental to the physical. 20th century epistemologists waste their time over Gettier problems to no avail since justified true belief is still an internalist issue. If the problem is to explain the connection between what we believe to be the case and our actions, hence the world, then no account of knowledge that makes it the product of individual internal processes can solve that problem since the world has been rendered irrelevant. But there is a way out. It is the pragmatist's way. If you accept as a premise that the mark of knowledge is successful action, the connection between knowledge and action is established from the beginning. It is no longer a question of certifying the input. The output is the key player here. However, I would argue, it is not quite that simple — since the connection between knowledge and action is assumed here, not proven, nor can it be. That is, the connection between...