1. The Fundamental Problems of Epistemology



            Traditionally, epistemology or “the theory of knowledge” has been concerned with problems like the nature of knowledge, the distinction between knowledge and belief, or the general problem of justification for claims to knowledge. Our approach here will be different. We will first try to determine the general nature of cognitive behavior as produced by knowledge based or “epistemic” systems and then analyze what differences exist in these performances and the significance of these variations.  We attempt to develop an epistemology of process starting with the realization that any manifestation of knowledge by an intelligent system necessarily involves a process which is amenable to a fundamental analysis concerning the information that generates it.


            We assume to start that all cognitive systems are information determinant that is, that their behavior in all forms is caused by information in the system. This behavior need not be externalized, thoughts for example, or internal states representing dispositions to perform we will also regard as behavior. Behavior here then is more akin to certain types of operational states of a system than to what is normally thought of as action. Cognitive systems for our purposes here are “smart” or “intelligent” systems; they are entities that are capable of internally generated, information based, syntactic behavior. Syntactic behavior involves at a minimum a certain type of temporal structure, that is, it involves the sequencing and timing of behavioral elements. An open question here is whether or not all cognitive systems are epistemic, that is, do they necessarily involve something we might regard as knowledge. Our suspicion is that they are not, that some systems, e.g., robots can be cognitive without really knowing anything.


            Human beings are both cognitive and epistemic although not always both at the same time. This is one of the themes that we intend to pursue here. They can for example talk a great deal – cognitive behavior – without really knowing anything about what they are talking about. We can talk about ghosts for example but unless there really are ghosts then we can really know nothing about them. But the most important issue with human cognition will probably center on understanding the difference between talking about something, and talking because of something, i.e., the information basis of linguistic behavior. This is where the problems of epistemology are really to be found if we are concerned with something more than chasing words.


            The issues of traditional epistemology generally concern considerations regarding propositional knowledge, i.e., statements like “x knows that y is the case”, where “Y” is some sort of proposition. Now while propositions need not involve universal terms, they generally do.  Thus most of the problems of epistemology involve universal terms either directly or indirectly. A large number of questions, we will argue, are then dependent on the nature of universal terms.  The issues do not merely involve questions about what it  means to say  that “X knows Y” where we would be concerned essentially about the meaning of the word “knows”, but more generally as in the case of scientific laws, of a deeper issue, subsumtion, i.e., how do we decide that “this case is like the other (s)?” The first problem of traditional epistemology is then in our view not that of knowledge per se, it is the problem of inclusion: If we want to decide if “X knows Y” is a valid claim, we must first understand the nature of what is claimed in “Y”. That is, we must understand the nature of the universal terms in “Y”.  The assumption that “knows” is the same in all cases and does not depend on “Y” has to be proven. The nature of universal terms, we shall see, in many cases is largely determined by how the nervous system works as an information processing system, and chiefly by the fact that the nervous system, on the input side, generally reduces the amount of information involved in its processing activities.


            We will also be concerned with what might be called a proof of the doctrine of empiricism, the idea that all knowledge of the external world comes ultimately through experience. The development of this proof leads to many other important considerations concerning human cognition and epistemic behavior in general including a information based critique of metaphysics.