What does Goldman mean by “naturalizing epistemology”?
Linking epistemology with the sciences…
Epistemology thought of as a branch of psychology
Epistemology is a study of how the human mind works.
What does Kornblith mean by “the Replacement Thesis”?
It is a proposal to subsume epistemology under psychology.
Epistemology would be replaced by a branch of science.
Goldman rejects the Replacement thesis. But, he thinks science has a role to play in epistemology. How can science shed light on what is and is not possible in epistemology?
It could tell us the limits of human cognitive powers.
It would tell us what our (human) minds can and cannot do.
We might think that the rules of rationality (what is and is not rational to believe) are established by the rules of logic. The rules of logic say that our beliefs should not be inconsistent. Why should we think that this is not really a rule of rationality?
There are some inconsistent beliefs that seem to be perfectly rational.
Even if I believe that it should not be cloudy today and so will not be cloudy, it might make sense for me to believe it is cloudy if it looks cloudy.
One might think that a belief is rational if and only if it is produced by a rational mechanism. How can science help us then with whether human beliefs can be rational?
It can tell us whether humans have a mechanism for complying systematically with probabilistic rules.
If we do not, then, the standard for rationality cannot be probabilistic rules.
If we do, then, it can.
Psychologists say that we do not have a mechanism for complying systematically with probabilistic rules. Why is this fact relevant for doing epistemology?
If we do not, then the standard for human rationality cannot be reasoning according to the rules of probabilities.
People have inadequate mechanisms for probabilistic reasoning. So, instead, they use representative heuristics. How does the example of 31 year-old Linda (on p. 149) illustrate this point?
Proposition f is more likely than proposition h. If we were using probabilistic reasoning, we would say that f is more likely than h.
But, instead people say that h is more likely than f. That’s because instead of probabilistic reasoning, people use a representative heuristic.
How does the example of Dick, the 30 year-old man, illustrate the same point? (p. 150)
The description of Dick is neutral as to whether he is an engineer or a lawyer.
So, if we are told that the base rate of lawyers to engineers is 70 to 30, then the probabilities that he is a lawyer should be 70%. If the base rate is 30 to 70, then the probabilities that he is a lawyer should be 30%.
But that is not the answer people give. So, they are not using probability reasoning in giving their answer. They are going on the representative heuristic.
Here is a principle. My belief in a hypothesis h is not justified if the evidence for h is compatible with some alternative hypothesis h*. This principle threatens skepticism about our knowledge of grammar. How does it do that? Given that the evidence for the correct rules of grammar are underdetermined, we need to appeal to science. How can cognitive science help in establishing justification for our beliefs about the correct rules of grammar? (p. 156)
There are different hypotheses as to the rules of grammar. And the linguistic data is consistent with more than one hypothesis. So there is no way to know which is the correct hypothesis. So then, we have skepticism about the rules of grammar.
The sciences can tell us which rules or hypothesis we are innately going to choose.
Whatever hypothesis we are innately going to choose is the correct one.
Studying what we in fact choose will tell us what the correct rules will be.
So, science can help with trying to identify the universal rules of grammar.