Quite a while back, Clifford Johnson at Cosmic Variance had a post seeking nominations for "The Greatest Physics Paper Ever." Back after a long hiatus, he's now holding a vote among five finalists: Isaac Newton's Principia Mathematica, Albert Einstein's General Relativity, Emmy Noether's paper on symmetry and conservation laws, Dirac's theory of the electron, and the Einstein, Podolsky, and Rosen paper on quantum non-locality.
(Newton's Principia Mathematica had a comfortable lead when I last checked, so if you're a partisan of one of the other candidates, go over there and vote...)
Of course, as some have noted, these are all theory papers. This is a natural result of the overabundance of theory types in blogdom, but it still just doesn't seem right. For most of the history of science, after all, the development of theory has been driven by experimental results, not the other way around.
So here's an attempt to restore the balance. I'm soliciting nominations in comments for the best experimental work in physics. What experiment do you think was the greatest, that is, what is the one experiment that you think did the most to change the field of physics (hopefully for the better)?
Nominate your favorite experiment in comments. If I get enough comments (which is not by any means a sure thing, as this ScienceBlogs thing may be too new to have the necessary audience), I'll compile the results, and we'll put this question to a vote as well.
I'll put a few suggestions below the fold, but here's your chance, experimental types. Why should the theorists get to have all the fun?
An obvious starting point for this sort of discussion would be the "Most Beautiful Experiments" list generated a few years back. I posted about this at the time, but it's recently resurfaced with spiffy animated graphics (a site so nice, Cosmic Variance linked it twice.) There are some problems with the list (like the fact that the Leaning Tower of Pisa story is probably apocryphal), but it's not a bad place to start.
Of the experiments on the list, the best are probably Young's double slit and Rutherford's discovery of the nucleus. Young's experiment provides clear and unambiguous evidence of wave-like behavior in light, with nothing more than a couple of slits cut in a card. It scores big points for elegance, as well as for directly contradicting Newton, a bold move in 1800.
Rutherford's experiment is more complicated, but like Young's, it completely smashed the prior theory. My favorite part of the story is that the discovery supposedly came about because Rutherford had hired a new student, and set him a silly make-work task to get him used to the apparatus. What he found required a complete re-thinking of existing models of the atom, and produced one of the great quotes in experimental history:
It was quite the most incredible event that ever happened to me in my life. It was almost as incredible as if you had fired a 15-inch shell at a piece of tissue paper and it came back and hit you.
(That's Rutherford describing the discovery.)
Among experiments not on the "Most Beautiful" list, I'd nominate the Michelson-Morley experiment, and not just because Morley was a Williams grad. The real role of the experiment in launching relativity is somewhat unclear, but it's certainly a fabulous example of the ingenuity of experimentalists. Michelson set out to make a ridiculously difficult measurement, and in the process invented a technique that's still in use (and that I'll be teaching in lab tomorrow morning) for precision measurements.
Another good candidate (and I'll stop here) would be a Bell Inequality experiment, but here we hit the major problem with sorting out great experiments. Many of the best experiments in physics only put limits on particular quantities, limits which are constantly being questioned and refined. In the case of Bell's Inequality, all the experiments to date have left loopholes that allow some possibility for local-variable theories (albeit increasingly baroque local-variable theories) to survive. There hasn't really been a definitive Bell Inequality experiment.
Still, the experiments by Alain Aspect and co-workers are awfully impressive, and most physicists regard them as fairly convincing proof that quantum mechanics is a non-local theory, and that Einstein, Podolsky, and Rosen were wrong (in an inspired manner, but still wrong). Given that quantum non-locality is one of the absolute strangest things in the entire world, I'd say that an experiment proving it deserves a place among the all-time greats.
So, what's your nomination for the greatest experiment in physics?
- Log in to post comments
Do "observations" count as well as experiments? I'd nominate both Galileo's discovery of the moons of Jupiter, and Hubble's discovery of the expanding universe. And I'd vote for Galileo.
The Micheldon-Morley thing was the first thing that leapt to my mind, so I guess I second that nomination.
MKK
Well, I'm a theorist, but I'll put my 2 cents in. How about Galileo's inclined plane experiments? Or Cavendish's work on gravity? Faraday's discovery of induction?
Ok, that was 3 cents.
Another vote for Rutherford here. That was truly groundbreaking.
The discovery of P violation by Wu & co. Utterly unexpected, quite upsetting to intuition, and led very directly to ideas at the foundation of the Standard Model.
How about Aspect's experiment conforming Bell's inequalities...
Great idea. How about Penzius and Wilson? True they were trying to wash away their signal, so the clever design aspect is lacking, but the CMBR is still the strongest piece of evidence for the Big Bang.
Wu's parity violation was a striking experimental result. But it simply wasn't utterly unexpected: Lee and Yang had suggested the experiment because they saw the possibility that parity could be violated. Doesn't make the results less striking, though.
How about the joint discovery of the J/psi boson? That was unexpected and played a crucial role (the "November revolution") in launching the standard model.
Second the nomination for Galileo's inclined planes. People were still being burned for alleged witchcraft in what we laughingly call the "civilized world"; respectable scholars still regarded alchemy as a legitimate method of inquiry; meanwhile, Galileo's doing some of the earliest experiments in the western world that deserve the name "scientific".
I'll also nominate Newton's experiments on circulating fluids, which allowed him to pretty decisively rebut his main contemporary rival theory of the solar system, Descartes's vortex theory, by showing that rotational speeds in fluids do not exhibit the behavior shown by the planets.
I'd definitely vote for Michelson & Morley, and not just because I'm a relativist. I have to admit I'm a sucker for two things:
- Null results, when everyone expects to measure an effect, and
- The kind of arrogance that Michelson (justifiably) had in believing he could measure the effect of motion through aether.
As I recall, everyone expected aether effects on the order of (v/c)^2, where v is Earth's speed through the ether. For obvious reasons, these effects were considered too small to be observable. Michelson knew he could design an interferometer precise enough to measure the effect. When it failed, he built better ones -- all with the same outcome.
It was precisely Michelson's care in devising an unassailable experiment that let theorists like Fizgerald to look for alternative explanations.
Here's two more of my favorites (again from a speed of light perspective):
- Galileo's failed attempt to measure the speed of light with lanterns. The brilliance was in recognizing that a null result in the presence of reaction time only puts a *lower bound* on the speed of light.
- Roemer's rather ingenious explanation (okay, not necessarily an experiment per se) that discrepancies in eclipse times of Jupiter's moons can be explained by a finite speed, and his estimate of that speed.
Personally, I love all of the 20th century QM experiments, but I'd be hard-pressed to find a favorite. What about Stern-Gerlach or Davisson-Germer?
Busy day at work today, so I'm a little slow getting to all these good suggestions...
Sean: Do "observations" count as well as experiments?
In a certain sense, I suppose all experimental work counts as "observation" of nature. It's just a little trickier to observe some of the finer points...
dr. dave: Well, I'm a theorist, but I'll put my 2 cents in. How about Galileo's inclined plane experiments? Or Cavendish's work on gravity? Faraday's discovery of induction?
Cavendish is another good one, especially as his basic technique is still the state of the art for measuring gravity. Adelberger's group at Washington has really pushed it to the limit, but the central element is still a torsion pendulum.
Mr Upright: Personally, I love all of the 20th century QM experiments, but I'd be hard-pressed to find a favorite. What about Stern-Gerlach or Davisson-Germer?
Davisson-Germer is possibly my favorite story of the lot, because the only reason their experiment worked is that they broke their apparatus. I'll be lecturing about that in a week or two, and I always enjoy that story.
Well, there are so many great experiments in physics ! In addition to the above, I would like to add
H.K.Onnes - Superconductivity
R.L.Mossbauer - Mossbauer effect
Bednorz & Muller - High Temperature superconductivity
And to do justice to the centuries before-
Tycho Brahe - The enormous amount of astronomical observations that lie at the beginning of physics as we know it !
Michael Faraday- Electrolysis/Magnetic Induction
Heinrich Hertz - Production and Reception of Electromagnetic waves
H.K.Onnes - Superconductivity
R.L.Mossbauer - Mossbauer effect
Bednorz & Muller - High Temperature superconductivity
I'm a little ashamed to admit that I don't know as much about these as I should. We recently had a student resurrect our Mossbauer experiment as a senior thesis (it'll debut in a few weeks in the upper-level lab class), so I've learned a bit about that, though I still don't quite understand what it's about.
Heinrich Hertz - Production and Reception of Electromagnetic waves
I heard a really interesting talk when I was in grad school by a guy questioning whether Hertz really could've seen what he reported. It involved a bunch of reconstructions of the apparatus, and the hall in which the experiments were performed.
He seemed to be saying, as I recall, that Hertz only got the results he did because of some odd quirks of the set-up, and that the success of the theory was partly just luck. I could be badly misrepresenting the talk, though-- it was several years ago, now.
Millikan's oil drop
The whole field of Mossbauer spectroscopy is beautiful--such an ingenious exploitation of a counterintuitive effect to get extraordinarily clean and precise results. One of my personal favorites is the Pound-Rebka experiment, which used Mossbauer spectroscopy to measure the gravitational redshift/blueshift between the bottom and top of a building. Not least because the effect was so tiny that they were even pushing the ability of that technique to measure it, and had to go to heroic efforts to account for things like relativistic time dilation due to thermal motions.
(Also, when I was in grad school, my office was a few feet away from the historic shaft. It's often described as an elevator shaft, but it isn't; it was built into the building in the 19th century to do free-fall experiments. In a sense, this was another one.)
Well, I am not very well-versed with these experiments either. But, every one of them earned a Nobel prize, so the presentation speech/nobel-lectures might be of some help.
H.K.Onnes - Superconductivity
R.L.Mossbauer - Mossbauer effect
Bednorz & Muller - High Temperature superconductivity
The great thing about the experiments is of course the surprises they bring. All three of the the above were mostly unanticipated by theory. In fact, it took almost thirty years to understand the theory behind superconductivity - and this in the golden age of theory when QM/QFT were discovered ! And even after BCS theory, High Tc Superconductivity was definitely a surprise - and if I understand right, its theory is not yet in place..
And as far as Chad's comments on Hertz go, I should admit that my knowledge of experiments that underlie electromagnetism is quite rusty. You might be onto something though as a glimpse through the first few pages of this book reveals. Unfortunately, our institute libary doesn't seem to have it. So, we can do nothing but hope that somebody out there who has read the book The Creation of Scientific Effects : Heinrich Hertz and Electric Waves (Paperback) by Jed Z. Buchwald would help us out!
The whole field of Mossbauer spectroscopy is beautiful--such an ingenious exploitation of a counterintuitive effect to get extraordinarily clean and precise results. One of my personal favorites is the Pound-Rebka experiment, which used Mossbauer spectroscopy to measure the gravitational redshift/blueshift between the bottom and top of a building. Not least because the effect was so tiny that they were even pushing the ability of that technique to measure it, and had to go to heroic efforts to account for things like relativistic time dilation due to thermal motions.
I think my problem is that I come from a community where using a spectroscopc reference to stabilize a laser frequency to the level of a part per billion is an undergrad summer student project. I have to really stop to think about the fact that there's really no other way to do spectroscopy on nuclei before I realize how impressive the Mossbauer effect is.
By the way, Loganayagam R., I apologize for the delay in publishing your comment. The system flags anything with more than three URL's, as an anti-spam measure, and holds it for approval.
I'm going to tinker with the settings a bit, and see if I can avoid that...
I vote for Michelson-Morley. It immediately got people thinking outside the Young-elastic-ether box, and led directly to relativity, whether Einstein did or did not have it in mind in 1905.
Newton's great optics experiment showing that light was composed of many colors. He virtually had to invent experimental rigour in order to make the demonstration work.
second the nomination of Roemer circa 1675 timing of Io eclipses
(Mr. Upright | January 12, 2006 07:56 PM)
If I can second more than one nomination, then my second second would be Cavendish gravity
(dr. dave | January 12, 2006 07:37 AM)
Both involved measuring a fundamental constant for the first time in a novel and elegant way----the speed of light and Newton G.
I am trying to reduce it to 3ish so i am really REALLY skipping some beautiful work.
i nominate:
Tycho Brahe. Without him we wouldn't have had Kepler and then Newton's 1/r^2.
Hertz' observation of E/M waves (confirmed Maxwell's equations).
Plank. Wether he liked it or not he helped to creat the quantum world.
Hubble and the expansion of the universe...
this is impossible. I keep coming up with them.
Two of my favourites:
Faraday's experiments connecting electrocity and magnetism.
Hubble's use of Leavitt's standard candles (Cepheid variables) to determine once and for all that our galaxy was one of many....the universe was way bigger than was supposed......by a long way.
Cheers,
-cvj
an old link: PhysicsWebv '02 (amazing is that Galileo never perfomed his two experiments!)
I suggest Cavendish one.
I vote for Michelson and Morley's experiment (in fact it's two votes as my wife votes for it too). But I want to add another experiment which should be held as important: the measurement of the Lamb shift.
Let me put in a plug for Friedrich and Knipping. What, you don't remember them? How about their boss, Max von Laue?
Von Laue's great insight was the realization that the expected wavelengths of X-rays and the expected spacings between atoms in crystals were similar enough that it should be possible to use a crystal as a 3D diffraction grating. He assigned Friedrich and Knipping to investigate; their first try used copper sulfate (bad choice - triclinic!), but then used a sphalerite crystal (isometric) and found symmetrical diffraction spots on their films.
Quoting Kittel: "The work proved decisively that crystals are composed of a periodic array of atoms and marked the beginning of the field of solid state physics."
The 100 year anniversary of the paper is coming up in June of 1912.
Not especially groundbreaking as an experiment, I suppose, but I still appreciate the utter simplicity of using a pendulum to determine the mass of the earth. My students can usually get within 10% of the accepted value, which may not impress them too much (it seems), but I find it fascinating.
Faraday!
I'm surprised nobody has offered up the Wohler synthesis of Urea from "inorganic" reagents. At the time it was certainly an experiment. And it set the foundation for modern chemistry.
Subsumed in above, but emphasised: for minimalist elegance, Newton's variously coloured pieces of cloth laid on snow in sunlight. (Bronowski also rhapsodised)
1.Hertz's observations of radio waves confirmed Maxwell's unification of electromagnetism and gave us modern technology
2.In 1919 Eddington's observation of the bending of starlight by the sun confirmed Einstein's theory of General Relativity and the curvature of space-time. This overthrew Newton.
3.Chadwick's table-top experiment in 1932 which discovered the neutron and gave us nuclear physics.
4.J J Thomson used a cathode ray tube to discover the elctron - was this the discovery of the cathode ray tube? If so it gave us TV.
5.What about experimental ingenuity? The late Val Telegdi was quite good at that.
NB Wilson & Penzias' experiment was not the first observation of the blackbody radiation - E McKellar first detected it in 1941 and T A Shmaonov observed it again in 1957.