Short Lecture About Science For Marc Colten Yesterday I tried to answer a posting in the '

Master Index Current Directory Index Go to SkepticTank Go to Human Rights activist Keith Henson Go to Scientology cult

Skeptic Tank!

Short Lecture About Science For Marc Colten Organization: FB Physik, Universitaet Kaiserslautern, Germany From: (Thomas Kettenring) Message-ID: <> Newsgroups: sci.skeptic Yesterday I tried to answer a posting in the "CSICOP; origins and problems with" thread, but the news server refused to do anything. Now the discussion has gone further, in a sense, and my posting is sort of obsolete. I'll cut it down to what I find necessary and post it after this. But it looks like Marc still has not read about null hypotheses, so here goes: ----------------------- A null hypothesis is a hypothesis that a certain quantity has the value zero. If we know nothing of the quantity, we assume the null hypothesis until it is proven that the value is different from zero. It is easy to see that a null hypothesis can be disproven but not proven. We can't prove the value is zero, we can only say "it's very close to zero, it's less than so-and-so." But we can say "it's more than so-and-so, it can't be zero." Several cases were discussed here: - Flatness of the earth: The quantity is the curvature of the Earth (the inverse of the Earth radius). We assume it is zero (that is, "the Earth is flat") until it is proven to be different from zero (that is, we observe something that can't be observed when the Earth is flat). - Continental Drift: The quantity is the velocity of two continents in respect to each other. We assume it is zero (that is, "the continents don't move") until it is proven to be different from zero (that is, we observe something that can't be observed when the continents don't move). - Telepathy and Co: The quantity is the size of the effect of telepathy. We assume it is zero (that is, "there is no such thing") until it is proven to be different from zero (that is, we observe a telepathic effect). - Flying Reindeer: The quantity is the number of reindeer on Earth capable of flying. We assume it is zero (that is, "there is no such thing") until it is proven to be different from zero (that is, we observe flying reindeer). ("Observe" includes scientific rigor in all cases above, of course. Stories of flying reindeer or jigsaw with continents don't count.) You may ask, "why do we need a default? Can't we wait until we have enough proof either way?" But as I said above, we can never have enough proof for a null hypothesis, so if the null hypothesis is true, you will for ever stay in the I-don't-know state. Even if it isn't you'll have no answer for a while. So this method, the method of the donkey that can't decide which of two haystacks to eat and therefore starves to death, is utterly useless. But if we haven't been able yet to disprove the null hypothesis, that means that for all practical purposes the quantity in question is zero, or so close to it that the difference doesn't matter yet. So the null hypothesis is a good working hypothesis. As soon as it doesn't work anymore, you can toss it. You may ask, "why don't we use a hypothesis close to the NH, instead of the NH itself? For example, we have very few telepathic persons around, they can do it only part of the time?" Occam's razor is the answer. "Don't unnecessarily multiply entities." If you don't need the telepathy hypothesis to explain the world, don't use it. The NH is always simpler and therefore better, if compatible with the data of course. ----------------------- And another one about openmindedness with respect to new theories: ----------------------- You seem to think (correct me if I am wrong) that there is one correct way to behave for a scientist when hearing of a new thought; a sort of middle-way skeptical openmindedness. But if you do indeed think that way, you are demandingo much from scientists. The scientific community consists of humans. Don't you understand that after I have examined the thirtieth perpetuum mobile and found that it doesn't work, as the others didn't before, I get slightly annoyed and don't want to have anything to do with it? There are other things to do that are more promising. Now that doesn't mean that nobody will listen to people who think they have found something new. A small percentage of scientists will spend time for things others (or even they themselves) find rather hopeless, either for the outside chance that there may be some truth there, or out of a sense of duty to the search for truth. *One scientist alone can't do science.* It's the whole of them who do it. You need others to examine your work, to test it and look for mistakes. But one specific scientist is not obliged to test any specific thing; they can do what they want. If you have built a perpetuum mobile and find nobody who wants to test it, you can't blame them for not thinking it worthwhile. Give them a real reason to do it; general statements about narrow-minded dogmatists won't help. Returning to "the correct way a scientist has to think": It is needed that there is a diversity of opinions in the scientific community. The community as a whole has to be open to new ideas, not single scientists. Bringing everybody on line won't help; it will make science monolithic. There has to be "memetic diversity", so to speak; lots of different ideas. But there also has to be selection. Bad ideas aren't popular. If you want to decrease the size of that selection by demanding that everybody gets real open-minded, you are hindering science. Everybody will suddenly test perpetua mobilia, and the people who build them will be encouraged and make more of them. (Of course they will claim conspiracies too, as their true genius still isn't recognized.) We have to decide between the Skeptic Skylla who now and then devours someone who has found something real but has not enough evidence yet (i.e. Wegener) and the Charlatan-Charitable Charybdis which sinks the whole ship of science. Some wheat will always land in the chaff, unless you don't separate the two. ----------------------- thomas kettenring, 3 dan, kaiserslautern, germany I am the most humble person in the world.


E-Mail Fredric L. Rice / The Skeptic Tank