Our next set of topics will be about a set of questions concerning scientific knowledge. Distinguish skeptical questions from “Socratic” questions. There will be questions of both sorts. Some background ideas about knowledge, not in the book.

 

I. Kinds of Knowledge

1. Knowing individual things: He knows President Seligman.
2. Knowing how to do things: George knows how to read.
3. Knowing who: No one knows who the Academy Award winner will be.

4. Knowing whether, when, why, what, etc.

5. Knowing that: Jones knows that ducks quack. 

 

The last is propositional (factual) knowledge. Maybe the others can be defined in terms of it. Eg., possibly (3) can be defined in terms of propositional knowledge: No one knows any proposition of the form “X wins the Academy Award”.

 

Propositional knowledge is the kind of interest here. Science is supposed to yield factual knowledge of the world.

 

II. Propositional Knowledge

 

A. Knowledge and True Belief

 

Consider the proposal that knowledge = true belief. [Write out as analysis.]

 

1) True belief is necessary for knowledge.

Possible objection: I knew all along that the butler did it, and then it turned out that he didn’t. You might have felt sure, but you didn’t know.

2) True belief is not sufficient for knowledge. Lucky guesses. Even informed guesses are not knowledge. Suppose I believe that the guy from “Capote” will win the Academy Award. I might even feel confident of this. But I don’t know that he will win. Note: that this is about the future is not crucial. Suppose the awards were already given out but I didn’t watch or hear reports. I now believe that he won (past tense) and I’m right. But still no knowledge.

 

B. The Traditional Analysis of Knowledge

1) What’s needed in addition to true belief is traditionally said to be “good reasons” or “evidence” or “justification”.

2) So, the traditional analysis of knowledge holds that knowledge is justified true belief. That is:

S knows that p iff i) S believes that p; ii) it is true that p; iii) S is justified in believing that p.

3) These three conditions are independent: any two can be satisfied while the third is not. Explain.

 

C. Justification

1) Justification of the sort needed for knowledge requires very strong reasons. There can be reasons that make belief more reasonable than denial or suspension of judgment, but still come up short of reasons good enough for knowledge. E.g., weather forecasts, perhaps.

2) But knowledge does not require absolute certainty. Perhaps we can have something like evidence that puts the proposition beyond all reasonable doubt, though not beyond all possible doubt. Eg., the lights are on, water freezes at 32 degrees (under normal conditions), etc.

3) Some skeptics insist that knowledge does require evidence that excludes all possible doubt. But we still need to distinguish the relation we have in cases of what we ordinarily regard as knowledge from other cases. Why not continue to use the word “knowledge” for that?

4) A related idea is that justification, even knowledge level justification, does not guarantee truth. Victims of perfect hallucinations, or extremely well-done deceptions, may have justified but false beliefs.

5) Evidence and practical reasons: There is a big difference between having evidence for the truth of a proposition and a motivation for believing it. Examples: athlete, patient.

6) There’s difference between reasons for belief and reasons for acting. Fork in the road example.

7) Sources of justified belief (and knowledge): perception, introspection, memory, reason. These are psychological processes in us.

8) What about intuition, faith, and mystical experience? Note a difference between intuition as something weird and spooky and as merely acute perception (which no doubt exists and is, I think, a source of justification). Whether the others are a source of knowledge and justification is a more contentious issue. Notice that we have considerable evidence about the nature and merits of perception and memory and we lack comparable evidence about the merits of faith or mystical experience. So there does seem to be a basis for treating them differently.

9) William K. Clifford’s Principle: “It is wrong always, everywhere and for anyone to believe anything upon insufficient evidence.” It’s not clear what he meant by “insufficient”. On one reading, his principle is completely trivial. But there is a way to interpret it that makes it not trivial (though perhaps not all that controversial). When you consider a proposition, your choices are belief, disbelief, suspension of judgment. The principle says that it is wrong to believe when the evidence is not supportive. Note: this means overall evidence. We can generalize: wrong to disbelieve when the evidence is supportive, wrong to suspend judgment when the evidence is not counterbalanced. In general: overall evidence determines which is the right attitude.

            The significance of Clifford’s principle is that it says that evidence is the determining factor. How important something is, how nice it would be for it to be true, etc. make no difference.

            On this reading, Clifford’s principle is not unduly cautious. Belief can be the right attitude even if the evidence is not strong enough for knowledge. Weather forecast example - right to believe that it will rain, but not strong enough justification for knowledge. But it is wrong to suspend judgment when the evidence does point one way or another.

10) A popular slogan is: everyone has a right to believe what they want. As a legal principle, that seems admirable. We don’t want thought police. But as an epistemological principle, this slogan is a mistake and it conflicts with the suggested interpretation of Clifford’s principle. Ambiguity of “wrong”: legally wrong, morally wrong, epistemically wrong. Clifford’s principle is best taken to be about what is epistemically wrong.

 

III. Unjustified Beliefs

 

There is a large literature detailing situations in which people have unjustified beliefs - beliefs that are not supported by their evidence. Thinking about some of these cases can help to clarify the concept of justification. Since some of the errors are potentially relevant to thinking about the justification of beliefs about scientific matters, it will be useful to review them here. These notes will provide only brief descriptions of the examples reviewed in class.

 

1) Refusal to revise one’s view in the light of conflicting evidence. Examples: a) The Millerites are (were?) a religious group who predicted that the end of the world was coming soon. As the dates passed, each time they pushed back the date a little rather than acknowledge the overall weakness of their general view. b) Suicide notes example. c) Psychotherapists who stick to their diagnoses in the face of undermining evidence. It is possible to make up explanations for why we are like this: pride, value of consistency.

Note: in example (a) it is far from clear that the people had any good basis for their belief in the first place. It differs from (b) and (c) on this.

Another note: it seems that the order in which you get the evidence does not matter. Detective example: evidence that suspect A is guilty, and you believe it. Then you get comparable evidence that it’s B. You don’t get to stick with A because you already believed it. There’s no good principle favoring inertia here.

 

2) Confirmation bias: a tendency to look for evidence that confirms what we already believe rather than evidence that goes against it. Example: reaction to conflicting studies, arguments - both sides became more convinced of their initial views. Along these lines is another fact: we tend to hang out with like-minded people.

But notice the difference between a) getting biased information, and b) treating competing evidence in an inequitable way. The latter is an error of rationality. The former is a different sort of error - perhaps we can say that there are errors in dealing with the evidence you have and errors in collecting evidence. But if you know that your sources are not representative, then you have reason to treat them differently than if you don’t know that. And if you don’t have reason to think that your sources are in some way biased, then it’s not clear that you’ve made any error of rationality at all.

Yet another point to note is this: If you have good reason to think that one group is more reliable about a topic than another, and you hang out with the people you have good reason to believe to be reliable, and then your beliefs are shaped by them, it’s hard to see that you are making any kind of error.

 

3) Availability error: basing judgments on evidence that is vivid or memorable, rather than evidence that is reliable. Example: Using personal recommendations rather than statistics (as in forming beliefs about which car is most reliable.) We need to be careful about this: using the evidence we have available is the reasonable thing to do.

This makes for puzzling examples. Example about lunar effect: people believe the position of the moon can affect behavior, making people crazy (lunatics). Careful study undermines this. Belief persists. One explanation is that the cases of crazed behavior in a full moon are more memorable. Media reporting is another. But on the latter option, what is our evidence about this? It’s not so clear that there is an error in reasoning.

 

4) Ignoring the evidence: from time to time we engage in such things as wishful thinking or fearful thinking. Example: optimistic and pessimistic students. In such cases our emotions rather than our evidence guide our beliefs. When the resulting beliefs are unsupported, they are epistemically unjustified even if they have some important personal value.