Philosophy, Film, Politics, Etc.

Monday, July 26, 2010

Gettier and the De Dicto / De Re Distinction

I want to discuss the difference between de dicto and de re beliefs and its implications for Gettier cases. This will be an elaboration on my earlier treatment of Gettier cases. I am mainly restating my argument in terms of the de dicto / de re distinction. Unfortunately, I have read very little of the literature on this distinction, so I cannot situate my arguments in a scholarly fashion. (Any pointers to relevant papers would be greatly appreciated.)

The difference between de dicto and de re beliefs is sometimes illustrated with an example like this one:

(1) Ralph believes that someone is a spy.

There is an ambiguity in (1): Does Ralph believe of a specific individual that he is a spy? Or does Ralph believe that at least one person is a spy, without having a belief about any particular individual being a spy?

If Ralph's belief is about a specific individual, then it is de re. If Ralph's belief is not about a specific individual, then it is de dicto. My contention is that for a person to have a justified de dicto belief, as opposed to (or in addition to) a justified de re belief, then that person must have reason to believe that some condition is satisfied above any reason for believing that some particular individual satisfies that condition. De re justification is not sufficient for de dicto belief.

Here's an example from a Gettier case:

(2) Smith believes that the man who will get the job has ten coins in his pocket.

Does "the man who will get the job" in (2) refer to a particular individual? If so, then Smith's belief is de re. However, if Smith believes that somebody is such that they will get the job and they have ten coins in their pocket, without having this belief about a particular individual, then Smith's belief is de dicto.

In the Gettier case, Smith has a justification for a de re belief. He is informed by the president of the company that Jones will get the job, and Smith has a justified true belief that Jones has ten coins in his pocket. Thus, (2) is true as a statement of Smith's de re belief about Jones.

As it turns out, Smith got bad information. Jones will not get the job, but Smith will. It also turns out that Smith has ten coins in his pocket. From this, we can draw several conclusions. First, Smith's de re belief about Jones is justified but false. However, if we take (2) to denote a de dicto belief, then Smith has a true belief. It is true that somebody is such that they will get the job and they have ten coins in their pocket. However, I do not think Smith has this de dicto belief. While Smith is clearly justified in his de re belief, he has no justification for the relevant de dicto belief. If there is no justification for a de dicto belief--if we have no reason to think that Smith even has such a de dicto belief--then there is no Gettier problem. (For there to be a Gettier problem, Smith must have a justified true belief that is not propositional knowledge; if all Smith has is a justified false belief, then there is no problem.)

It is evident that there can be justification for a de dicto belief which does not justify a de re belief. For example, if the president of the company told Smith that whoever gets the job has ten coins in his pocket, then (2) is true under a de dicto reading. However, Smith is not given this information. If we want to consider a de dicto reading of (2), we have nothing to go on but the justification for Smith's de re belief. The question, then, is whether or not a justified de re belief automatically justifies a corresponding de dicto belief.

It is possible to have corresponding de dicto and de re beliefs. For example, suppose that Smith was told by the president that the person who gets the job has ten coins in his pocket. Smith then has a justified de dicto belief, but he is not sure if it is Jones or himself that will get the job. Smith might also believe that Jones is the better candidate, and so he believes that the person who will get the job is Jones. That is a de re belief. It is clear that these are two distinct beliefs from the fact that they can have different strengths. Smith may be very confident that the president was telling the truth, so his de dicto belief may be quite strong. However, he may not be confident in his estimation of Jones' ability; his de re belief may be very weak.

While it is possible in some cases to have corresponding de dicto and de re beliefs, we have no reason to think that this is happening in the Gettier case. Smith believes of Jones that he is the man who will get the job and that he has ten coins in his pocket. Smith has no belief that somebody else might get the job who also has ten coins in their pocket. At least, we have no reason to attribute such a belief to Smith, which means that there is no apparent justification for Smith to have such a belief. The de re belief gives no cause to attribute to Smith a belief about whether somebody other than Jones might get the job while also having ten coins in their pocket. Therefore, the de re belief does not justify the de dicto belief. If Smith does have the relevant de dicto belief, it is not justified. Hence, no Gettier problem.

To make my critique of Gettier complete, I make a similar analysis of Gettier's second example: the case in which Smith is justified in believing that Jones owns a Ford, and so forms a variety of disjunctive beliefs, including this one:

(3) Jones owns a Ford or Brown is in Barcelona.

As I wrote in my previous treatment of the topic, (3) can be restated as

(4) One of the following is true: Jones owns a Ford; Brown is in Barcelona.

Smith's justification for believing that Jones owns a Ford is enough to justify his belief that (4) is true. However, is this a de dicto or a de re belief? More specifically, does "one of the following" denote either of the two propositions in the disjunction, or does it denote a particular proposition?

Smith has no justification for believing that either one of the propositions may be true. He only has justification for believing that one of them is true. Therefore, Smith believes (4) in so far as "one of the following" is given a de re reading: it denotes "Jones owns a Ford." This belief is false, for it is not the case that Jones owns a Ford. However, it happens to be the case that Brown is in Barcelona, even though Smith believed that Brown was nowhere near Barcelona. If somebody believed (4) under a de dicto reading, they would have a true belief. However, Smith has no reason to believe (4) under a de dicto reading. He only believes that Jones owns a Ford. Thus, again, no Gettier problem.