Internalism/Externalism and Foundationalism/Coherentism

Philosophy 524

Fall 2008

 Internalism vs. Externalism

The first question to ask about the trees is, "How do we construct them?" In other words, what should the Ps look like and what should we say about the links between the Ps? There are two general approaches to take here: internalism and externalism.

Internalism: this is grounded in the assumption that merely by "reflecting upon [one's] own conscious state", one can identify and arrange a set of epistemic principles that could then be used to determine whether any given belief is justified or not. On this view, the Ps identify propositions that a believing subject accepts, and the links between them are justificatory links that are internalistically specifiable.

1. That is, one need only use introspection to get at the epistemological nature of one's own belief system.
   
   
2. The Problem: there is no logical connection between justification and truth. We are able to use internalism to arrange beliefs from most justified to least justified, but we are unable to collapse justification into truth; that is, a proposition may be as justified as anything we believe and still be false. We can get our internal house in order, but there is no way for us to know if the principles we use correspond to the principles that govern the world.

Externalism: this is developed in reaction to internalism; in particular, the externalist is unwilling to accept the gap that remains between justification and truth if you adopt internalism. Here the Ps can include propositions that are not accepted, and the links need not be available to the subject via introspection.

1. The Externalist Solution: to tie justification to the way in which beliefs are produced. An externalist epistemology is grounded in principles that explain how it is possible for truth to give rise to justification.
   
   
2. Two Forms of Externalism: Reliabilism and Causalism.
   
   
   1. Reliabilism focuses on the processes that give rise to beliefs, and lays down the condition that one is justified in believing P if the subject's belief in P was produced by a process that is reliable. The problem with this is how can we make sense out of what it is for a process to be reliable? Chisholm argues that the only way that one can make sense out of this is if you introduce internalistic elements into the picture.
      
      
   2. Causalism focuses on the causes of the belief, arguing that justified beliefs are beliefs that are caused by the proposition's being true. Similar problems confront this approach: how can we make sense out of the notion of causality here?
      
      
3. Chisholm's reaction to both of these approaches is that there is no way they can succeed unless they involve internalistic elements, which would leave the externalist open to the same problems that motivated them in the first place.

Foundationalism vs. Coherentism

Return to the trees on the handout and the second question: "How do we justify these trees?" This can be brought into high relief by considering the Regress Problem.

1. This problem is created by the observation that the trees above might never reach bottom -- where will the demand for justification stop?
   
   
2. There are four responses available here:
   
   
   1. A belief can be justified by a belief that is unjustified. (Wittgenstein?)
      
      
   2. A belief is justified by another, which is justified by another, ad infinitum. (See Sosa.)
      
      
   3. A belief is justified by another, which is justified by another, which is justified by the one we started with. (Coherentism)
      
      
   4. A belief is justified by another, which is based on a belief that justifies itself. (Foundationalism)

First, we will consider Foundationalism. Chisholm develops what we will call a Modest Foundationalism. Such a view has the following features:

1. An asymmetrical relationship exists between the foundations and the superstructure.
   
   
2. Doubts about any psychological beliefs being indubitable or incorrigible are allowed.
   
   
3. There are restrictions on which beliefs can lie at the foundation. (Call these "basic beliefs.")
   
   
4. The foundational relationship is justification of belief rather than knowledge, although knowledge is the goal.
   
   
5. Superstructure beliefs may be only inductively supported by basic beliefs. (This is the point of Chs. 5-7 -- there we see that justification can transfer from proposition to proposition without there being any logical implication between those propositions.)
   
   
6. Coherence is allowed some scope in the justification.
   
   
7. There is a distinction made between having a justification for a belief and being able to show that one has a justification.
   
   
   1. This is modest because of features 2, 4, and 5, and to a lesser extent 6 and 7.
      
      
   2. Problems: establishing that there are any self-justifying beliefs; finding enough so that we can justify all the beliefs that are intuitively justified.

Coherentism: the tree is justified because of the way in which these beliefs cohere with the rest of our beliefs, none of which are foundational.

1. But how can we get around the a world/the world problem here?
   
   
2. The epistemic relations are not transitive, and this would appear to undermine the view