Biggest recent mistakes in physics or any science.

[Whalstib]

Any ideas? I was prompted as I found a very nice 1970 50th Ed. of the CRC Chemistry and Physics Handbook for $1 today! At fist I was thinking how much of this can be regarded as accurate today? After a moments thought I figured the vast majority of it is probably quite accurate.

I mean how often to physics formulas and the like change nowadays? Perhaps in chemistry they are measuring down to 100 significant figures on atomic weight or something equally so precise it doesn't matter much in the general scheme of things.

Which led me to my post. Have there been any taken for granted serious science premises that have been refuted or refined to a great deal? I was reading the history of the age of the Earth and it kept getting older as radioactivity was discovered but the likes of Lord Kelvin albeit based on current knowledge "accurate" was WAY off! But how about recently? has the computer age refined any ideas so drastically to take note? Will I have information in my 1970 handbook which is archaic and useless?

Anyway I think I like my $1 deal and leafing through it has plenty I don't understand and probably won't but know where to look!

Thanks,

Warren

 

[Ryan_m_b]

I understand what you're asking and I'm not to sure of the answer. One thing to remember is that many measurements of the past were accurate in the sense of being the most accurate possible. That's all our science is today as well but often is vastly more accurate which has led to uncovering many incorrect but accurate conclusions of the past.

My main study isn't actually physics but biology. Biology has arguably (preparing to get shot down...) advanced more than any other science over the past 50 years. There's nothing that has overturned well established previous beliefs but every few years we seem to discover how little we know. When we come close to thinking we've nailed a subject a whole other field opens up beautifully intertwined with the previous revealing we didnt know as much as we thought!

Classic example, we thought that DNA gave rise to RNA which gave rise to Protein. Proteins could then influence DNA, we basically thought that RNA was just a messenger in the system. In the past decade we've discovered a whole world of RNA regulation of DNA and Proteins. That's fairly standard in biology, once you have something nailed down it slips from your grasp at the last

 

[nismaratwork]

I think the Catholic Church's stand against Galileo was a pretty big one.

 

[Whalstib]

I think the Catholic Church's stand against Galileo was a pretty big one.

Not quite Simplicio,

A more accurate example might be found with Kepler who doubted Galileo. In fact many prominent scientists of the day were skeptical.

The lesson, don't insult the man who holds the keys...

W

 

[Whalstib]

I understand what you're asking and I'm not to sure of the answer. One thing to remember is that many measurements of the past were accurate in the sense of being the most accurate possible. That's all our science is today as well but often is vastly more accurate which has led to uncovering many incorrect but accurate conclusions of the past.

yes this is my point. My example of the age of the world was accurate BEFORE and understanding of radioactivity. It was based on cooling rates and of course if something is radio active it won't cool as predicted. Very simple explanation but you get the picture.

So has science gotten to such a point of accuracy that little changes? Is there little difference between the 1970 CRC Handbook and the 2011 ed.?

W

 

[Jimmy Snyder]

That's how I was able to determine my age. When I was younger, they said the Earth was 2.5 billion years old and now they say it's 4.5 billion years old. That makes me 2 billion years old.

 

[nismaratwork]

Not quite Simplicio,

A more accurate example might be found with Kepler who doubted Galileo. In fact many prominent scientists of the day were skeptical.

The lesson, don't insult the man who holds the keys...

W

True, but I was struck by the fact that science is constantly being hounded by people enamored of old beliefs. Hell, there's a flat-earth society... I'm not sure that there IS a biggest mistake.


Wait... I take that back. Castle Bravo is probably the biggest mistake, in terms of sheer miscalculation of yield.

 

[Whalstib]

True, but I was struck by the fact that science is constantly being hounded by people enamored of old beliefs.

Right!

We tend to think it's cranks outside of or on the fringes of science but as Kepler proved science makes mistakes.

I also found an old copy of "The World We Live In" by Time/Life back in ~1957. Most of it for the depth it goes to is quite accurate considering the audience, but the geology is humorously flawed. of course pre-plate tectonics so still working within a framework of acceptable science.

Perhaps mistakes was too strong a phrases, misconceptions as most were based on accepted science for the most part.

When I get through with the cover to cover review of both the 2011 CRC and 1970 CRC I will be better able to comment<G!>

W

 

[nismaratwork]

Right!

We tend to think it's cranks outside of or on the fringes of science but as Kepler proved science makes mistakes.

I also found an old copy of "The World We Live In" by Time/Life back in ~1957. Most of it for the depth it goes to is quite accurate considering the audience, but the geology is humorously flawed. of course pre-plate tectonics so still working within a framework of acceptable science.

Perhaps mistakes was too strong a phrases, misconceptions as most were based on accepted science for the most part.

When I get through with the cover to cover review of both the 2011 CRC and 1970 CRC I will be better able to comment<G!>

W

The beauty of science is that at its best, it admits its mistakes and seeks to find all possible answers as to how they were made, and what the truth might be.

Conditional knowledge may not be as calming as certainty, but it's been quite the boon to humanity. Bit of a kick in the pants for other animals though...

 

[Dembadon]

[STRIKE]I'm not particularly proud of bloodletting.

http://en.wikipedia.org/wiki/Bloodletting[/STRIKE]

Edit: I misread and misunderstood your post. Consider the above irrelevant.

 

[Cleonis]

Have there been any taken for granted serious science premises that have been refuted or refined to a great deal?

I was reading the history of the age of the Earth and it kept getting older as radioactivity was discovered [...]

I can't think of an example of a mistake.

But there have been surprises.

In cosmology, there is the discovery that at some point in the Universe's history the rate of expansion has started to increase.
(However, I don't think the estimates of the age of the Universe were much affected by that finding)

In particle physics it was long taken as established that neutrinos are massless. If neutrinos would have a rest mass then that rest mass would be far smaller than the mass of any other particle. It was deemed by far more likely that neutrinos just don't have a rest mass.
However, the phenomenon of neutrino oscillation puts a lower limit on neutrino mass. So the textbooks have to be rewritten on that one. (Incidentally, it's my understanding that neutrino mass does not involve the Higgs mechanism.)As ryan_m_b mentions, molecular genetics has been particularly rich in surprises.
Before the human genome project estimates of the number of genes in the human genome ranged from 50,000 to 100,000 genes. Findings from the human genome project indicate there are somewhere in between 20, 000 and 25,000 genes.

A very large percentage of human DNA (and DNA of related creatures) consists of long repetitive sequences, that have long been regarded as superfluous DNA. It was thought of as a result of overdiligent copying of DNA,or whatever. It was often referred to as 'junk DNA'. However, the genome projects show that many parts of this DNA material are highly conserved over the course of evolution. Similar stretches of DNA are found in the mouse genome.
And as far as we know: structures that are highly conserved over the course of evolution are very important structures. So while the function of the long repetitive sequences is still unknown, the evidence suggest it shouldn't be viewed as 'junk DNA'.

 

[nismaratwork]

[STRIKE]I'm not particularly proud of bloodletting.

http://en.wikipedia.org/wiki/Bloodletting[/STRIKE]

Edit: I misread and misunderstood your post. Consider the above irrelevant.

A good idea though... maybe Transorbital Lobotomy would trump that however, if medicine were to be included in the OP.

 

[nismaratwork]

I can't think of an example of a mistake.

But there have been surprises.

In cosmology, there is the discovery that at some point in the Universe's history the rate of expansion has started to increase.
(However, I don't think the estimates of the age of the Universe were much affected by that finding)

In particle physics it was long taken as established that neutrinos are massless. If neutrinos would have a rest mass then that rest mass would be far smaller than the mass of any other particle. It was deemed by far more likely that neutrinos just don't have a rest mass.
However, the phenomenon of neutrino oscillation puts a lower limit on neutrino mass. So the textbooks have to be rewritten on that one. (Incidentally, it's my understanding that neutrino mass does not involve the Higgs mechanism.)


As ryan_m_b mentions, molecular genetics has been particularly rich in surprises.
Before the human genome project estimates of the number of genes in the human genome ranged from 50,000 to 100,000 genes. Findings from the human genome project indicate there are somewhere in between 20, 000 and 25,000 genes.

A very large percentage of human DNA (and DNA of related creatures) consists of long repetitive sequences, that have long been regarded as superfluous DNA. It was thought of as a result of overdiligent copying of DNA,or whatever. It was often referred to as 'junk DNA'. However, the genome projects show that many parts of this DNA material are highly conserved over the course of evolution. Similar stretches of DNA are found in the mouse genome.
And as far as we know: structures that are highly conserved over the course of evolution are very important structures. So while the function of the long repetitive sequences is still unknown, the evidence suggest it shouldn't be viewed as 'junk DNA'.

How about particle "swarms" after high energy collision... that was a bit of a shock (to me at least).

 

[Dembadon]

A good idea though... maybe Transorbital Lobotomy would trump that however, if medicine were to be included in the OP.

Technically, medicine would qualify, as the OP's subject reads "... or any science." However, the practice of bloodletting was mostly used before scientific medicine. Not to mention it was scientific medicine itself that helped to disprove the effects of the practice as it pertained to various ailments in the 19th century.

 

[gendou2]

When I think big mistakes in science for the past year, I remember the NASA announcement of the Arsenic-tolerant bacteria.
I felt like the paper made the big mistake of inflating the importance scope of its findings.
http://www.ironlisa.com/WolfeSimon_etal_Science2010.pdf
http://science.nasa.gov/science-news/science-at-nasa/2010/02dec_monolake/

I also remember a recent paper by Roger Penrose citing evidence in the WMAP for cyclic cosmology.
My take is that the statistics in this paper tasted like fudge, a mistake atypical for a guy like Penrose.
http://arxiv.org/abs/1011.3706
http://telescoper.wordpress.com/2010/11/28/doubts-about-the-evidence-for-penroses-cyclic-universe/

 

[nismaratwork]

Good call on that first one (didn't read the second yet)... man was that overhyped! I was waiting for an alien snail or something, but noooooOOOooooo...


@Dembadon: Good point, although the fact that leeches have found a valuable place in microsurgery now is ironic beyond belief; talk about the old and the new coming together!

 

[Gokul43201]

Not quite a mistake, and not really after the 70's, but this one's close: Parity. Until 1957, virtually everyone in the particle physics (and probably the broader physics) community, essentially took it for granted that parity conservation was universal. Sometime in the mid-50's, Yang and Lee did a survey of the literature, and found that this had yet to be experimentally verified in the case of the weak interaction, and that such an experiment would be worth doing. Most of the community probably didn't want to bother setting up a very difficult experiment only to show the obvious. Finally they managed to find a sympathetic ear, and Wu's Co-60 experiment shattered a few decades worth of complacence.

http://en.wikipedia.org/wiki/Parity_(physics)#Parity_violation

 

[gendou2]

Another great blunder in 2010 was committed by Walter Wagner.
He filed a lawsuit against CERN to stop the LHC from starting up.
He claimed the particle accelerator constituted a "credible threat of harm".
This is a joke to anyone who has ever seen cosmic rays in a bubble chamber.
http://www.physorg.com/news203573289.html

 

[Vanadium 50]

Gokul, if you read the Wu article in PRL, you will discover that immediately after that paper there is a paper by Garwin, Lederman and someone I am blanking on that also discovered parity violation. So while I can't tell for sure if this was a hot topic in 1957, the fact that two groups published simultaneously using very different techniques suggests that it was maybe not so simple as "nobody wanted to do the experiment".

 

[Whalstib]

Not a mistake but I have heard there were errors in Feynman's Lectures or at least I heard the latest edition either updated or corrected some content. This isn'tr science per se and but writing and of course that would open the flood gates for inaccuracies just based on typoes etc.. which is not what I'm referring to in this thread.

Would the Lectures still have solid valuable information for a physics student or are they considered archaic?

Thanks,
W

 

[jhae2.718]

The Lectures are classic IMO; I have the New Millennium Edition.
(My physics professor also recommended it when I asked her.)

 

[lisab]

I don't know if falling for fraud would count as the kind of mistake the OP is thinking of, but that autism-vaccine hoax was a terrible black eye for the medical research community, IMO.

 

[Proton Soup]

the idea that we were going to cure obesity and heart disease by making everything "lo-fat"

 

[Loren Booda]

The failure of our science educational system. The attitude of the Scopes monkey trial, for instance, continues with the chosen ignorance of over half of our country. It isn't belief in an alternative to science that is the problem, it is intolerance to objectivity from fear.

Over half of all college students in the U.S. fail to graduate.

 

[jhae2.718]

Over half of all college students in the U.S. fail to graduate.

Is it really that bad? (I'm assuming the statistic is for all majors, and not just STEM...)

 

[Gokul43201]

Gokul, if you read the Wu article in PRL, you will discover that immediately after that paper there is a paper by Garwin, Lederman and someone I am blanking on that also discovered parity violation. So while I can't tell for sure if this was a hot topic in 1957, the fact that two groups published simultaneously using very different techniques suggests that it was maybe not so simple as "nobody wanted to do the experiment".

Yup, it's the very next article.

(some of this may be a little bit off, but the general order of events, I think, is correct) Lederman et al, were Wu's colleagues at Columbia (though she performed her experiment at NIST). She told them of her preliminary results as soon as they came out, and Lederman et al immediately performed an independent verification. While they were doing that, Wu was repeating\testing to make sure she hadn't been looking at instrumental artifacts. I think Lederman et al even had a paper ready before Wu, but waited for her to submit first. This is purely a guess, but I'm not sure they'd have been as easily convinced to give it a shot if someone else hadn't tried it first.

Edit: Here's a more accurate recounting - http://ccreweb.org/documents/parity/parity.html#Madame%20Wu

 

[Gokul43201]

Not a mistake but I have heard there were errors in Feynman's Lectures or at least I heard the latest edition either updated or corrected some content. This isn'tr science per se and but writing and of course that would open the flood gates for inaccuracies just based on typoes etc.. which is not what I'm referring to in this thread.

Would the Lectures still have solid valuable information for a physics student or are they considered archaic?

Thanks,
W

I've used them almost as a textbook. Do you have more details on what errors were corrected?

Edit: Nevermind. I think I just found them. Will take a look. http://www.feynmanlectures.info/flp_errata.html

 

[Phrak]

Any ideas? I was prompted as I found a very nice 1970 50th Ed. of the CRC Chemistry and Physics Handbook for $1 today!

I have a 1934 edition. What to compare figures?

 

[Ryan_m_b]

As ryan_m_b mentions, molecular genetics has been particularly rich in surprises.
Before the human genome project estimates of the number of genes in the human genome ranged from 50,000 to 100,000 genes. Findings from the human genome project indicate there are somewhere in between 20, 000 and 25,000 genes.

A very large percentage of human DNA (and DNA of related creatures) consists of long repetitive sequences, that have long been regarded as superfluous DNA. It was thought of as a result of overdiligent copying of DNA,or whatever. It was often referred to as 'junk DNA'. However, the genome projects show that many parts of this DNA material are highly conserved over the course of evolution. Similar stretches of DNA are found in the mouse genome.
And as far as we know: structures that are highly conserved over the course of evolution are very important structures. So while the function of the long repetitive sequences is still unknown, the evidence suggest it shouldn't be viewed as 'junk DNA'.

Yet another surprise was that there are three times as many proteins as there are genes! Thanks to post transcriptional and post translational modification.
One small correction though 'Junk DNA' isn't a proper term in genetics, it might sometimes be used colloquially but it's not a proper term. Even long stretches of conserved DNA don't necessarily point to function, as far as we know the large majority of the genome serves 'no' function.

NB: by no function i mean no function of contributing to gene expression, it may still have important structural functions etc

 

[nismaratwork]

Yet another surprise was that there are three times as many proteins as there are genes! Thanks to post transcriptional and post translational modification.
One small correction though 'Junk DNA' isn't a proper term in genetics, it might sometimes be used colloquially but it's not a proper term. Even long stretches of conserved DNA don't necessarily point to function, as far as we know the large majority of the genome serves 'no' function.

NB: by no function i mean no function of contributing to gene expression, it may still have important structural functions etc

Well, while you're correcting the fellow/lass, maybe mention "intron" and "exon"?