I'm sure this is a simple question. What I am referring to is disbelieving someone on Day 20 because they have lied every day previous to Day 20.
Another example is the boy who cried wolf. The 50th time he cries "Wolf!" nobody believes him.
As a matter of practicality it's perfectly reasonable to disbelieve someone on Day 20 or take the cry of "Wolf!" to be false, as we've established through experience that a person lies, and so we judge probabilities (unconsciously maybe) and choose not to believe them.
However I'm asking from a purely logical perspective. If a witness lies many times, their credibility is severely hurt, but it would technically be a fallacy to say their last testimony was a lie because:
1.The witness has lied ten times to 13 questions.
2.The witness has a tendency to lie.
3.The last statement from the witness is a lie. (invalid conclusion)
What comes to my mind is "hasty generalization". However "hasty generalization" is defined by Wikipedia as:
... a conclusion about all or many instances of a phenomenon that has been reached on the basis of just one or just a few instances of that phenomenon. It is an example of jumping to conclusions.
Faulty generalization (hasty)
Notice it says based on just one or just a few. I think this excludes my examples because the conclusion is not based on just one or few instances, but in some cases many, for example 30 instances.
I agree that it is "jumping to conclusions", but I feel there's a more accurate term for it. I feel there's a term for something like:
- Just because it's happened in the past (even every time) doesn't mean it'll happen now (even though the odds are in favor it happening).
Also, I know this is related to the problem of induction, (You don't know that the sun will rise tomorrow), but that's not a name of a fallacy.