Timeline for Is it ever reasonable to infer impossibility from high improbability?
Current License: CC BY-SA 4.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 6 at 15:52 | comment | added | Michael Hall | @KristianBerry, but isn't it also intellectually unethical, and circular logic, to believe in a highly improbable sequence of natural processes that haven't been identified, (much less replicated) and proclaim "well, here we are, so they must have happened."? | |
| Oct 19 at 17:41 | comment | added | Kristian Berry | @yters I have tried to avoid evaluating ID theory on non-technical levels but I must confess that I know that ID theory has been unethically constructed. It is intellectually unethical to believe in such theories, because of the equivocation, abuse of quotes, etc. that has gone into them. It is not surprising at all, since the major ID theorists are motivated by even more unethical worldviews (their preferred distortions of Christianity or Islam). But there's little chance of you tricking me into believing in the pseudomathematics/pseudoscience at play (I got over Christianity eventually). | |
| Oct 19 at 17:34 | comment | added | yters | @KristianBerry ever heard of p-values? Used all over the place in science. Those are a (shoddy) prespecification. | |
| Oct 19 at 16:48 | comment | added | Kristian Berry | @yters rewording the formulation of the concept doesn't make its justification noncircular. I don't accept that the concept of prespecifiable targets is realistic in the first place. I don't share all the abstract presuppositions of those who think it realistic. The concept is not forced upon me by science or mathematics in general, either. So, it might privately justify things for those who accept it, but it is not sufficient for public justification. | |
| Oct 19 at 16:11 | comment | added | yters | @KristianBerry all that is necessary for prespecification is the target definition has to be independent of the event. Let's say the sharpshooter used a machine gun to spell out the Gettysburg address. In that case, we don't have any predrawn targets. However, the Gettysburg address is independent of random machine gun fire, so it counts as a prespecified target. Same with DNA. All the descriptions of the marvelous biological mechanisms are independent of evolution (since they are drawn from fields outside of biology, such as engineering), so the mechanisms count as prespecifications. | |
| Oct 19 at 14:24 | comment | added | Kristian Berry | @yters that is presupposing that there is a prespecified target in this context, which is equivalent to presupposing a prior purpose. So the argument is effectively circular. That's not necessarily a bad thing, as all systems of concepts end up with at least a small circle of mutually defined, otherwise irreducible terms, but it is then again a matter where we have to presuppose what the argument is meant to show, to show it, so it is not dialectically effective. Anyway, I have seen plenty of ID-theory statements about small probability events being impossible to recognize equivocation also. | |
| Oct 19 at 14:16 | comment | added | yters | This answer is a misunderstanding of the ID argument. It is not saying small probability events do not happen. Every event that happens is small probability. The ID argument is actually talking about something called a rejection region, which must be prespecified before the event. Think of this as the opposite of the sharpshooter fallacy, where a single, very small target is drawn before the sharpshooter shoots. If he manages to hit the target, then odds are he is a very skilled shooter, ruling out chance. DNA has hit an extremely small prespecified target, ruling out chance + necessity. | |
| Oct 18 at 20:22 | comment | added | Barmar | For the past, it can be useful if you have multiple hypotheses and compare the probabilities of each. But the ID people only claim that abiogenesis is highly improbable, they can't calculate the probability of a designer. | |
| Oct 18 at 20:20 | comment | added | Barmar | @PeterRankin That's the problem with applying probability in arguments like the OP's in the first place. "What are the odds?" is a useful question when making predictions about the future, not when asking questions about the past. | |
| Oct 17 at 19:57 | comment | added | Olivier5 | If detectives think they can solve the mystery of life's origin, let them try. Though I would think biochemists and biologists are better equipped in terms of methods and truth standards. | |
| Oct 17 at 16:54 | comment | added | JonathanZ | Detectives are a horrible standard. | |
| Oct 17 at 16:28 | comment | added | Peter Rankin | But detectives make reasonable inferences on FAR lower probabilities. If someone claims to have solved a scrambled Rubik's cube blindfolded, without seeing the scrambled cube first, wouldn't you infer that they must have looked? I would! :) And that's only about 1 in 10^20. I think the die analogy, coupled with how the "why would we infer" clause is worded, is overlooking the sheer magnitude of exponential numbers like 10^200. | |
| Oct 17 at 13:43 | comment | added | Kristian Berry | @PeterRankin if the die is fair, then the probability that it will land on a "useful" side is still just 1/(10^200) (if only one side is "useful"). This doesn't mean that, if it does land on that side, it was too improbable to be possible. I would have a higher probability of winning some lottery if I bought 10,000 out of 20,000 available tickets, but if I won that lottery, it would not follow that I had bought that many tickets. People win lotteries by buying a few tickets (or even just one) all the time. | |
| Oct 17 at 13:38 | comment | added | Scott Rowe | @PeterRankin but, if it did land on something meaningful, someone would be talking about it. And, here we are. | |
| Oct 17 at 12:55 | comment | added | Peter Rankin | I think it would be better to phrase the analogy as, "What are the odds you would guess the side it landed on beforehand?" Because the probability it will land somewhere is 100%, but the chance it will land on anything meaningful (e.g., functionally useful, or specifically predicted) is a different matter. | |
| Oct 17 at 12:45 | history | answered | Kristian Berry | CC BY-SA 4.0 |