Theory of Knowledge
Fall 2006
Lecture Notes: Truth Tracking

1) Go over the general idea: use thermometer model. Go over p. 475 from Nozick’s article. All these points should be familiar.

2) Explain TT, including the use of subjunctives and the difference between the subjunctive conditional and causal connections. Note: this is a theory about knowledge, not justification. The tracking condition is supposed to do the work of the combination of the justification condition and the “no essential falsehood” condition.

3) Does this do better than the causal theory on the cases that caused problems? It does seem better for Examples 5.3 (not discussed in class) and 5.4 from the text. In each case, if the proposition in question were false, the person would still believe it. Also, the generalizations issue for the causal theory, discussed last time seems to work out better on this view.

4) Why Nozick complicates the story - Examples 5.5 and 5.6.

5) The solution is developed in (TT*). The key thing here is the appeal to methods. Examples, as developed in class, show how difficult it is to get a satisfactory account of what the methods are.

6) The problems for this are serious. There is a more general way to make the point. The idea of the tracking theory (and, to a lesser extent, the causal theory) is that knowledge is a matter of one’s beliefs being correlated with what’s true. But that does not seem to be either necessary or sufficient for knowledge. One way to see the point of the Granny case - Example 5.6 - is as showing that this just isn’t necessary for knowledge - you can know without being a truth tracker. The details added to get around this are distractions from the central idea. Being correlated with the truth isn’t nec. for knowledge. And the fake barns case - Ex. 5.7 - shows that it does not draw the right distinctions.

            Part of the problem is that evidence does not play a large enough role in this theory. In the granny case, she does have excellent reasons in the actual case. It’s just that in situations in which the proposition is false, she would have good but misleading reasons. That does not make her actually not know.