Tuesday, August 10, 2010

Oppy on Disagreement (Part 2)

This post is part of a brief series on Graham Oppy's article "Disagreement". The article provides an overview of the epistemology of disagreement and considers its application to cases of religious disagreement.

In part one, I introduced the basic concepts and motivating questions behind the epistemology of disagreement. As we saw, the key question is "What is the rational response to cases of disagreement between doxastic peers?". A doxastic peer is someone who has the same cognitive capabilities, and access to the same information, as you do.

In this part, we consider some examples of disagreement. The examples provide some instruction as to how to answer the key question. However, one of Oppy's key points is that while clear cut answers might be possible in uncomplicated hypothetical examples, once we get closer to real-life cases of disagreement things become much less clear-cut.

1. Cases in which you should adjust your beliefs
We begin with four cases of disagreement in which it would seem rational to change your original belief. They are:
  • Restaurant: A group of diners agree to split the bill after a meal in a restaurant. Two of them add-up the items on the bill in order to determine what each individual's contribution should be. The first says the contribution is $63; the second says it is $61. Both are confident that they added-up the figures correctly and both have been shown, in the past, to be equally good at this kind of arithmetic. 
  • Wrist Watch: Twin brothers are given identical watches for their birthdays. Before going to sleep they synchronise the time on their watches. When they wake up, the first twin reports the time as being 6.45; the second twin reports the time as being 6.51. There is no reason to suspect that either is lying or cognitively disabled in any way.
  • Horse Race: A group of punters are watching a horse race. They have each backed different horses. The finish to the race is very tight, with several horses appearing to cross the line at roughly the same time. Nonetheless, one of the punters jumps up, elated that his horse has won. Several other punters do the same thing, even though they backed different horses.
  • Perfect Pitch: Two expert musicians, with equal training and experience, are listening to an operatic performance. Their enjoyment of the performance is ruined when they disagree about whether a singer has hit a C or a C-sharp.
Why would it be reasonable to alter the degree of confidence you have in your belief in these types of cases? What do these cases have in common?

Well, they involve people with equal experience and epistemic access to the domain in which the disagreement takes place and, further, that domain is the kind of domain in which even experts may occasionally get things wrong. So it would be strange to assume that one's original judgment must be correct.

2. Cases in which you should not adjust your beliefs
We can contrast the examples just given with the following cases. In these cases, adjustments seems less rational:
  • Elementary Math: Two departmental colleagues, who have known and worked with each other for over a decade, are trying to work how many people will be attending an upcoming conference. The first one, reasoning out loud, says that two people will be attending on Wednesday and two on Thursday. That means four in total. The second colleague disagrees, saying "But 2+2 does not equal 4".
  • Perception: Three roommates, who have been living together for several years, are silently enjoying lunch. The first roommate asks the second to pass the salt to the third, who is gesturing at the other end of the table. The second roommate refuses on the grounds that the third roommate is not present.
  • Directions: Two neighbours, who have lived in the same area for several years, meet each other while out walking around the neighbourhood. The first tells the second that he is going to his favourite restaurant. The second -- who is also a fan of the restaurant -- tells him that he is heading in the wrong direction. The first disagrees.
  • Bird: You and your best friend are in the university library. You are both staring out the window. You notice a bird flying past. You turn to your friend and ask "What type of bird was that?". Your friend claims there was no bird.
Why would adjustment be irrational in these cases? What do they have in common? 

Well, they would seem to involve cognitively basic judgments. That is, judgments about immediate perceptual experience, personal memory or elementary math and logic. These are not the kinds of judgments we expect to be mistaken about.

3. More Problematic Cases
Those cases were relatively straightforward. The following cases are more tricky because they involve complex judgments about complex topics. It is not clear what should be done in these cases.

  • Meteorologists: Two highly experienced and respected meteorologists are asked to predict the likelihood of rain tomorrow. They both avail of the most up-to-date techniques for computer modelling of the weather and they feed the same data into their models. After doing so, the first meteorologist says there is a .55 probability of rain tomorrow, the second says there is a .35 probability of rain.
  • Detectives: Two experienced and successful detectives are both investigating the same case. The evidence is very complicated. Some of it points toward one suspect; some of points toward another suspect. After analysing all the evidence, they both decide upon a different culprit.
  • Doctors: Two impeccably well-educted doctors, with oodles of clinical experience are called-in to consult on a particularly difficult case. They both run the same tests and consider the same case history. After doing so, the first physician thinks that the patient has chronic fatigue syndrome. The second thinks the patient has lupus.

We might worry about whether the experts in these examples are really peers since it would be difficult to determine if they have the exact same experience, ability and evidence. Still, they are close to serious, real-life cases of disagreement. How we respond to these types of disagreement can be significant.

As can be seen in the following case:
  • Two Expert Judges: Suppose there are two expert judges who agree in 90% of cases and disagree in 10%. Suppose further that they are each correct in 95% of cases. So of the 10% of cases involving disagreement, each judge will be right in 5% of the cases. It just happens that there is no way of knowing which 5%.
If you are following their judgments, what should you do when they disagree? That depends on whether you want to get at the truth or avoid error. If you want to get at the truth you would be better off following one of the judges (since you will then be right 95% of the time). But if you want to avoid error, you should suspend judgment.

In the criminal law, the goal is usually thought to be to avoid error. The famous English jurist William Blackstone once observed (I'm paraphrasing) "'tis better for 10 guilty men to go free than for one innocent man to be punished." The problem is that this could be crippling. For imagine if we had several experts, who are usually right, but disagree across a broad spectrum of cases. In that event, we would have to suspend judgment in nearly every case.

In the next part, we will move on to consider religious disagreement.

No comments:

Post a Comment