Evidence and Explanation
Suppose that there is a set of facts that is evidence of a conclusion.
We speak of weak evidence, good evidence, conclusive evidence. We qualify so that the strength of the evidence is stated.
If A, B, C, and D are taken to be evidence of E, we say that A, B, C, and D are evidence for E. Good evidence, strong evidence, conclusive evidence.
There was a trend for a while in philosophy to flip things around, so that E was said to explain A, B, C, and D, instead of saying that A, B, C, and D are evidence of E. The language of explanation was used instead of the language of evidence.
What I noticed is that there was no corresponding qualification on how well the explanation explains the facts, or if there was, it was not as easy for people to evaluate. There was simply the statement that E explains A, B, C, and D, but there was no adverbial qualification on "explains".
There was the "best fit" model of explanation, but I believe that many people, even philosophers, do not reason well when thinking in terms of how well an explanation fits a set of facts. People reason much better when thinking in terms of evidence and conclusion, not explanans and explanandum. They have a better sense of proportion, so that they can appropriately qualify the strength of the evidence.
We speak of weak evidence, good evidence, conclusive evidence. We qualify so that the strength of the evidence is stated.
If A, B, C, and D are taken to be evidence of E, we say that A, B, C, and D are evidence for E. Good evidence, strong evidence, conclusive evidence.
There was a trend for a while in philosophy to flip things around, so that E was said to explain A, B, C, and D, instead of saying that A, B, C, and D are evidence of E. The language of explanation was used instead of the language of evidence.
What I noticed is that there was no corresponding qualification on how well the explanation explains the facts, or if there was, it was not as easy for people to evaluate. There was simply the statement that E explains A, B, C, and D, but there was no adverbial qualification on "explains".
There was the "best fit" model of explanation, but I believe that many people, even philosophers, do not reason well when thinking in terms of how well an explanation fits a set of facts. People reason much better when thinking in terms of evidence and conclusion, not explanans and explanandum. They have a better sense of proportion, so that they can appropriately qualify the strength of the evidence.
Comments (1)
This is cool.
Stats people talk about "how much of A is explained by B" but you're absolutely right that the closest we come in everyday language is something like "A is (or isn't) a good explanation for B", where "good" seems to mean "convincing", which leaves open the possibility that it's not the "real" explanation.
I was recently thinking about the reverse (which might be relevant here): if it takes {A,B,C,D} to bring about E, then having any 3 of the 4 in-place makes the last one behave as if it's the whole set, as if it's the explanation, the other 3 are just context. (Or we do statistics.)
Good topic. Have to think about the psychological point at the end more.