This is a huge issue and is the reason meetings are so invaluable. It sucks that so many people end up doing the same experiments someone’s already done, but it’s hard to get, “nope, doesn’t work,” results published no matter how rigorous they are, unless it’s unexpected data that causes a lot of controversy.
**Do scientists often admit they were wrong? **
Often, yes. Always, no.
Neither case involved “scientists” admitting error. The first example involved medical practitioners during a relatively primitive era in medicine. The second cites the popular press.
While individual scientists may in some instances resist giving up on favorite theories, science itself is self-correcting and good at acknowledging error once good evidence is presented. This is viewed as a weakness by purveyors of pseudoscience. “Mainstream medicine is always finding out it was wrong about something, but homeopathy hasn’t changed in hundreds of years!”
True, but there are so many ways of being wrong that I don’t think there are enough journal pages to hold them all.
To second your major point, there is a big difference between thinking something and going into the lab to discover that you were wrong (very few will have an issue with this) and admitting something you’ve published a dozen papers on is wrong.
I’ve seen - admittedly in engineering that admission of being wrong is reflected in no more publications. We have a problem of scaling, in that a method that works well on tiny circuits fails big time on real ones. It is very common to see a paper showing results on tiny circuits that would be revolutionary if it scaled, and then total silence.
Not that I’ve ever seen, and I’ve read thousands of reviews as a program chair and an editor. What you do see is that relevant work by the reviewer needs to be cited, and the author needs to explain why the new work goes beyond the old work. In my experience reviewers tend to be biased for work in their specialty, even work that contradicts or extends their work, since it makes their specialty seem more important.
It is common practice to make sure you reference work by those most likely to review your paper.
Science can be thought of as a gigantic process for (eventually) winnowing out the wrong, whether individuals calling themselves scientists are willing to admit they were wrong or not.
This is the way it differs from other forms of obtaining knowledge.
Shhhh! Some of the readers don’t belong to the guild. You don’t want just anybody to to get published do you?
I’m not sure if you are joking or not. Those funded by the tobacco companies got money but not much reputation. How much did they publish after the link became clear? Who has the higher reputation?
In private industry, sure. You have to get all publications cleared, but they are far more worried in general about revealing something about a future product prematurely than offending the company. I’ve done a lot of grant reviewing for NSF, and spoke quite a bit to the director of the division I reviewed for, and I highly doubt anyone in government cared what the papers say - and they don’t get screened. Nothing I work on is politically sensitive, but that is true for 99% of science.
During the Bush administration it was the Republican commissars editing and distorting research findings, not the scientists themselves.
[/QUOTE]
I’m on the editorial board of a journal, and we need more papers! Anyhow, my academic friends like lots of submissions to places where they publish, because it seems the lower the accept rate the more prestigious the conference or journal.
Extends certainly, but I doubt that you can site any studies that show reviewers are more likely to recommend favorably papers whose results contradict their own in a major way rather than results that confirm their results.
Most scientific papers you’ll read proposing some new idea will conclude with the authors of the new idea themselves pre-emptively admitting that the new idea is probably wrong.
I’ve never seen a study - it would be fairly hard to do. The papers I see present methods of doing things, and it is very common for the previous work section to present weaknesses in previously published work. I’ve never seen anyone object to this. In general, not doing it is more likely to get a paper rejected. To be sure, cases where some previous result is contradicted are rare. There is a book about the dinosaur killing asteroid controversy, and it seems that both sides got their stuff published, though perhaps in different places.
BTW, and competent editor needs to take this kind of issue into account when making a decision. It might be good to send a paper to a foe to get a different view, but a reject from someone deadset against a theory might not be enough to cause a rejection, especially if more neutral parties like it.
If you think about it, most of the work people do ends up being wrong, and if they never admitted it, they’d be stuck doing the “wrong” stuff all their lives. Science is a constantly self-correcting process. Some of the most stubborn scientists out there, though, tend to be paleoanthropologists. Everyone wants their pet fossil to be a human ancestor.
Of course, that idea is probably wrong…
It’s certainly something that comes up in the history of science a lot. People became quite afraid of contradicting Lord Kelvin’s estimate of the Earth’s age; you just did not want to offend the grand old man by asserting that his estimate couldn’t possibly be right.
It’s been pointd out in many books on the subject that the scientific community had a lot of difficulty accepting the theory of continental drift (you might be shocked at how very recently it was finally generally accepted) and for years were building ridiculous counter-theories on top of counter-theories in an effort to deny something that the evidence was making very clearly true.
Scientists are human, like anyone else, and it’s going to be a rare case where a guy who’s worked ten years on something will easily give it up, even in the face of the scientific method.
But things move forward all the same.
In the high-energy physics community, papers get retracted on occasion. I ran across one such phantom paper the other day. The authors had originally made the pre-print publically available on the ArXiv, but then during the peer review process it was discovered to be in error, so they removed it (but not before other people had cited it, leading to some confusion.)
More to the point, high-energy physicists have spent the past few decades speculating as to what the LHC (Large Hadron Collider) will find. Many predicted it would discover evidence of supersymmetry. If it fails to find supersymmetry (which is seeming increasingly likely), the proponents of the theory will either find some way to wiggle around the facts (“maybe the superpartners are just really heavy…”) or they’ll decide they were wrong. I can’t speak for all of them, but the ones I know are currently leaning towards option 2. (“I’ve got to publish this book before supersymmetry is ruled out,” one of them recently told me.)
That conclusion (published in the scientific literature) will probably be retracted as well. 
Oh, I think it probable that he’d be negotiating a new job with the Japan Whaling Association. 
Except that it doesn’t. As I noted above, there are maybe a dozen or so well-known instances of a majority of scientists in a particular field being seriously reactionary and obstructionist toward a revolutionary new theory in that field.
And everybody brings up those same well-known instances to illustrate the closed-mindedness of the scientific process, while ignoring the fact that the vast majority of change and correction in prevailing scientific theories has taken place without encountering any serious obstructionism at all. Yes, these famous instances do come up a lot, but the behavior patterns they illustrate are not at all typical.
And yes, as I also noted above, many scientists can indeed be stubborn and prejudiced when it comes to relinquishing their own pet research hypotheses that they’ve invested a lot of time and effort in.
But most of the new research that the average scientist sees doesn’t seriously impact his/her own pet hypotheses, if any. So s/he has little motivation to reject it out of hand.
So as isaiahrobinson and Jackmannii pointed out, the overall effect is that while every new hypothesis may meet resistance or denial from the few individual scientists whose toes it steps on the hardest, that resistance is generally just an insignificant eddy in the broader current of more-or-less indifferent acceptance on the part of the research community as a whole.
My understanding is that a lot of the resistance was because the advocates of continental drift didn’t have any plausible mechanism for it to happen.
So the average geologist said “Sure, South America and Africa fit into each other (not just physically but fossil and other evidence). But how did a whole friggen’ continent get thousands of miles away?”, and when the Drift enthusiast said “Um, a wizard did it?” the average geologist said “Hey, come back when you’ve got a theory that makes sense all the way through”.
Then, when basic geological research discovered the molten magma and floating solid crust, most geologists accepted drift very quickly.
I don’t think that its that individual scientists tend to admit they’re wrong (which makes sense. You spend 30 years learning, teaching, and researching one theory and then decide “well, that was all garbage”. It’s that eventually those scientists die off and those schools of thought don’t get so many new people in to replace them.