who was this 70s researcher who questioned evolution?

quote:

---

Proving X to be false does not mean that Y is any more likely to be true.

---

This thread doesn't look likely to go anywhere interesting, so instead I'll challenge the above assertion. All other things being equal, proving X false does mean that Y is more likely, assuming X and Y are alternative possibilities. More specifically, if new data show that X is impossible, but do not distinguish between Y and any other possibilities, then the posterior probability of Y is p(Y|data) = 1/(1-p(X)), where p(X) is the prior probability of X being true (i.e. the probability before the new data arrived).

quote:

---

Really? So proving that earth is not a flat disk mounted on the backs of turtles makes it more likely that the earth is actually being pushed around its orbit by butterflies?

---

Yes, it does. Of course, increasing a probability that is already vanishingly small by an infinitesimally small amount isn't going to matter very much in practical terms, but why do you think it has no effect?

quote:

---

I don't believe you've thought this principle through.

---

In the original formulation, let the prior probabilities of X and Y be p(X) and p(Y), and the probability of all other possibilities be p(Z). Given these definitions, p(X) + p(Y) + p(Z) = 1. After encountering the new data (call it D), p(X|D) = 0 and by Bayes' theorem
p(Y|D) = p(D|Y)p(Y) / (p(D|Y)p(Y) + p(D|Z)p(Z)).
I stipulated that the new data does not distinguish between Y and Z, which I take to mean that Y and Z and equally likely to yield the observed data if they are true, i.e. p(D|Y) = p(D|Z). Given that, the above expression for p(Y|D) reduces to p(Y) / (p(Y) + p(Z)). Our relative certainty about Y (that is, our estimate of the probability that Y is true after seeing the new data, relative to our starting estimate) is given by
p(Y|D) / p(Y) = 1 / (p(Y) + p(Z)) = 1 / (1 - p(X)),
which was what I claimed.

I don't know what thread you've been reading, but in this thread I was replying to the general statement, "Proving X to be false does not mean that Y is any more likely to be true." As far as I know, you can only prove something false if its probability isn't already zero. Therefore, your argument about the effect of new data on statements with a prior probability of zero is irrelevant. Yes, of course you can't reduce the probability if it's already zero; presumably, if the poster I was responding to had meant that, he or she would have said so, rather than making the statement in question.

quote:

---

I understand how probability works

---

So far the data in your posts suggest otherwise.

quote:

---

There was never any probability or any evidence for the earth being mounted to the backs of turtles. The fact that people might nonetheless have believed the proposition means absolutely nothing.

---

In the complete absence of evidence, you would be able to assign a probability of zero to the idea that the earth was mounted on the back of turtles? On what basis? it's true that some people's belief in its truth would not mean anything, but neither would your disbelief.

quote:

---

I'm sure you know better than that. Something with zero probability of being true must be false, and can often (but not always) be proven to be false.

---

Then you're using a completely different meaning of "probability" than I am, since I'm using probability to describe our knowledge of truth or falsehood. If something has not been proven to be false (either because of evidence or because it is a logical impossibility), we do not know it to be false, and therefore it cannot have a zero probability of being true. That's certainly the meaning that's appropriate to this context, since we're talking about how the probability of truth changes with new evidence.

quote:

---

Can I prove that the number of degrees in a triangle in planar geometry is never greater than 2 right angles? Yes I can. Euclid in fact showed that the number of degrees in a triangle is always exactly equal to two right angles.

---

I don't see your point here. What was the probability that the theorem was correct before Euclid proved it? Seriously, I don't know whether you think it was always equal to one or not. And what does this have to do with proving something false that's already known to be false?Getting back to the original point, do you agree that disproving X does make Y more likely, assuming Y is not already known to be false (and making the other assumptions I've already described)?

quote:

---

If you two want to have an argument about whether frequentist or Bayesian statistical interpretations are more correct then that is probably the topic for a different thread surely?

---

I had come to the same conclusion (although it's not clear to me that a frequentist approach is actually being offered). I will propose a new topic when I have a chance.