Theory of Knowledge
Fall 2006
Lecture Notes: The Causal Theory

I. The Causal Theory

1) The main idea: see (C), p. 82. Quote from Goldman, p. 461. Review some simple examples - a perceptual case, memory. Non-knowledge cases, e.g., you guess that p and there is no causal connection.

2) How this deals with the Gettier cases - there is no causal connection in those cases.

3) Why it gets complicated: blows to the head and deviant causal chains.

4) Goldman’s proposal: see , p. 459. Appropriate causal connections: perception, memory, correctly reconstructed causal chains.

5) Alleged advantage of the causal theory over the TAK, p. 460: doesn’t require ability to state justification. Thus, it better explains memory knowledge. Example about Lincoln. But maybe TAK can be defended in a different way than he suggests: Apparent memory experience is the justifier.

6) Goldman’s responses to objections, pp. 455-6: all of this is ok. p. 460: Introspection can be added to the list of approved processes. Note especially the point on the last page about finding out that a person has knowledge.

7) Note also that this view does not have the implication that in order to have knowledge one must know that a causal connection of the right sort obtains. You can be ignorant of all that.

8) Problems:

a) Knowledge of facts that aren’t causes, e.g., generalizations. See p. 84-5, text. Goldman’s solution to this - see p. 458 of anthology. He states this principle:

If x is logically related to y and if y is a cause of z, then x is a cause of z.

On p. 459 he explains how this works with the generalization. But this seems problematic. The generalization: The Cubs never win is “logically related” to its instances in the same way. So it seems to follow that we know it. (Even if it is true, we don’t know it.)

            Notice that Goldman slips in the idea of “warranted inferences” when he discusses this. But that seems to be the traditional idea creeping back.

            Other examples in the same category: moral knowledge, arithmetic knowledge.

b) Knowledge w/out correct reconstruction of the causal chain - see Example 5.3. As noted, Goldman could reply that if you have a justified belief about the causal chain, then you have knowledge. But, again, the theory is similar to the traditional one.

c) Causation but no knowledge - roughly, the idea is that you can get the causal chain right without having knowledge. See Example 5.4. You just guess at the causal chain and get it right. Goldman can say that you have to be justified in your beliefs about the causal chain. But then his theory is indistinguishable from the traditional one.

Conclusion: to get anything plausible, you have to move much closer to an evidentialist view.