re: #882 freetoken

Well, one hopes that would be the case. But as in the recent McClean/Carter paper in JGR has highlighted, the peer review process is not above being gamed by the crafty with a plan.

Over time though, and this is what many of the detractors of science ignore, the peer review process tends to correct errors because of the high premium placed on correcting errors. This is quite different from punditry, where yesterday’s blog essay is quickly swept under the rug.

I half agree with you. The fact is that there are preciesly two real questions asked in any real scientific debate.

They are:

1. Did you really see what you think you saw?

and

2. Does that really mean what you think it means?

If someone says something interesting in either question, like say finding a surprising new result, eveyone asks question one first.

If the result claimed is big enough to garner wide attention but just wrong, like say cold fusion, almost instantly, enough other scientists will do enough digging into question one to kill it if it is false. If someone actually cooked their data, they are pretty quickly tossed out of the community. This is one of the reasons why I never understood how anyone could do it. They always, sooner or later get caught. About the only way to not get caught is to write something in a journal so obscure that no one sees it untill after you die. At which point, what’s the point?

As to question two, there is more room for debate because there is more of an issue of theory vs. experiment. Those debates all get settled in one of three ways. 1. more evidence comes in from a different experiment that verifies/disproves the initial idea. 2. Someone points out a mathematical error. 3. A better understanding of the mechanism makes the old model this is based on obsolete.

So that said, the scientific community is very self correcting and very self regulating. It needs to be not because we have a thing for error correction per se, but rather because getting our own work right, and our own understanding of things as individual scientists means understanding it and getting it right.

Now this is peer reviewd science done legit. A great example of the other type would be tabbacco doctors who were paid to tell you that smoking was not a health risk. That went on into the 80’s and later. The peer reviewd science was settled on smoking back in the 50’s