Wednesday, August 08, 2012 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Dirac, Lawrence, Mott, Penrose: anniversaries

Today, on August 8th, we celebrate anniversaries related to three British and one American scientist. (Thanks to Gene for his reminder about the calendar.)

Ernest Orlando Lawrence was born in South Dakota in 1901, i.e. 111 years ago, to a family with Norwegian roots. He studied in South Dakota, Minnesota, Chicago, and Yale. That experience made him one of the first important scientists who were fully educated in the country of Vinetou (I guess that the Americans won't know what country it is!).

He would work on the photoelectric effect in the mid 1920s but in the late 1920s, this member of the Bohemian Club (no relation with my nation!) would already be known as the Atom Smasher.

He received the 1939 Nobel prize in physics, aside from other prizes, for his invention of the cyclotron he made at the University of California. He had realized that a LEP-like :-) linear accelerator he saw in Rolf Widerøe's article could need to be very long so he invented the circular, LHC-like shape instead. The diameter of his first prototype was about 10 centimeters, approximately 100,000 times smaller than the LHC and the price was about $25, about 500,000,000 times smaller than the LHC. ;-)

But there are ongoing Summer Olympics in London so it's fair to focus on British physicists. The calendar happens to kindly collaborate. ;-)

Just hours ago, I honestly didn't know that the second stanza of the U.K. anthem sketches a sensible attitude of the U.K. to the EU. "Confound their politics, frustrate their knavish tricks."

First, Paul Dirac was born in 1902, exactly 110 years ago. I wrote about his life in 2009 so unfortunately, this true giant of the 20th century theoretical physics won't get extra space here.

Instead, let's spend more time with my distant misspelled English cousin Nevill Francis Mott who received the 1977 Nobel prize in physics and who died on August 8th, 1996. You surely expect a detailed description of his Nobel-winning discovery. But we have a problem here, Houston. ;-) Let's first say some things that are not problematic.

Mott was born in September 1905 to parents who were both physicists in Cavendish. He spent much of his childhood in a village and was educated by his mother. He would study or work in Cambridge, Manchester, and Bristol.

The Nobel prize shared with Phil Warren Anderson and John Hasbrouck van Vleck was "for their fundamental theoretical investigations of the electronic structure of magnetic and disordered systems". However, Mott's own work, one focused on amorphous semiconductors, belongs to the narrow list of physics Nobel prize discoveries that are known to be wrong. He "derived" a lower bound on the zero-temperature metallic conductivity. However, this lower bound was invalidated by four guys from the Bell Laboratories just three years after his visit to Stockholm.

This is the reason why I won't discuss the detailed semiconductor research by Mott – although we could learn something from the mistakes, too. Instead, let me mention that Mott's name appears once on this blog, in the review of Sidney Coleman's excellent lecture Quantum Mechanics In Your Face. Already around 1930, Mott managed to understand some things about the foundations of quantum mechanics that are still perplexing to many confused physicists today. In particular, he explained why a particle leaves a straight track in a cloud chamber even if the initial state of the particle is an s-wave (the same probability amplitudes in all directions).

He showed that the directions of the individual bubbles are correlated, essentially equal, so while we can't predict the exact direction, we can use quantum mechanics to prove that there is a single direction. Various anti-quantum zealots still misunderstand those basic things even now, in 2012 (in more diplomatic words, I am referring to the "little disagreement" I have with Brian Greene about quantum mechanics: thanks to Shannon for the report haha), when they incorrectly claim that quantum mechanics predicts macroscopic properties of macroscopic objects to be fuzzy classical mixtures.

It doesn't do anything of the sort. Due to the correlations and entanglement that those sloppy folks routinely overlook, quantum mechanics predicts that tracks as well as our perceptions are totally sharp – one of the possible discrete outcomes – and properly correlated with other macroscopic events they should be correlated with. Quantum mechanics just says that we don't know which outcome – e.g. which direction of the track – will be taken. But quantum mechanics does imply that the probability is 100 percent the final answer will be sharp. Not knowing where a bone is located doesn't transform the bone to a jellyfish – a seemingly simple observation that physicists like Brian Greene demonstrably have problems to comprehend.

Just to be sure, Mott's contributions to physics don't boil down to a wrong paper about semiconductors and remarks about rudimentary quantum mechanics. He's known for many valid insights. He explained the latent image on a photographic emulsion; the Mott transition between metallic and non-metallic states (when the temperature gets too low, electrons suddenly don't have enough energy to escape the vicinity of the host atoms and become unable to conduct electricity); Mott insulators (materials predicted by the band theory to be conductors whose conductivity is lost because of the electron-electron interactions); and – from quantum-mechanics-inspired mathematics, Mott polynomials. It's somewhat remarkable that from his numerous achievements, the Nobel committee just happened to pick a wrong paper. ;-)

That completed my text about Nevill Francis Mott.

Finally, someone who is alive and kicking. Sir Roger Penrose OM, FRS (whatever it means), one of the most creative physics-ambitious mathematician of the 20th century, celebrates his 81st birthday today. Congratulations!

He was born in Essex in 1931. His brothers include mathematician Oliver Penrose and chess grandmaster Jonathan Penrose. He studied mathematics in London and focused on tensor analysis under John Todd. Penrose would pioneer twistors, new beetle notation depicting tensors and spinors and spintensors (I still remember those exciting times at the high school when I borrowed his and Rindler's books from the scientific library), and he would prove himself to be a remarkably talented source of cute mathematical curiosities such as the Penrose triangle, Penrose stairs, and Penrose tilings, a rigorous mathematical invention predicting and "explaining" the existence of quasicrystals with a five-fold symmetry and others (he should have probably shared the recent Chemistry Nobel Prize for quasicrystals).

He influenced as basic fields of maths such as linear algebra with the Moore-Penrose pseudoinverse matrix and other concepts.

Of course, in physics, before twistors became a real part of physics, Penrose has been especially important as the source of insights about Einstein's classical general relativity and one of its most important applied subdisciplines, cosmology. In 1965, he demonstrated that the collapse of dense stars would produce singularities. This conclusion was made more inevitable and rigorous by Stephen Hawking who would prove the Hawking-Penrose singularity theorems.

Penrose is also the father of the Cosmic Censorship Conjecture, a claim that the Universe avoids processes that would produce naked singularities i.e. singularities unshielded by an event horizon and therefore "visible" to the horny observers in the rest of the Universe. His intuition was that singularities are ugly and seem to lead to the loss of predictivity (because quantities diverge) so general relativity has to have some hidden powers to avoid such objects or, at least, to make them unable to affect most of the Universe.

However, later research showed that the original strong claims about the cosmic censorship (which were, by its character, mathematical hypotheses about the properties of Einstein's equations) were invalid. I think it's fair to say that experts agree today that the most general justification of the conjecture is based on faulty logic. General relativity may become unpredictive in the presence of extreme features such as naked singularities but general relativity isn't the whole story and has no reason to be the whole story. Quantum gravity may remain and probably does remain predictive even in the presence of previously "scary" features such as the naked singularities: quantum gravity (i.e. string theory) just offers us more robust, complete, and precise rules what should happen in difficult situations instead of simple bans saying that difficult situations shouldn't arise at all. So general relativity is indeed unpredictive and kind of inconsistent when similar situations occur but physics – the exact theory of gravity which has to be quantum – is not inconsistent and that's the only truly needed condition in physics.

(Just to keep you updated, the no-hair theorem is another "big conjecture" in general relativity and it's surely considered invalid above four dimensions today, too.)

You see that the Cosmic Censorship Conjecture is a bridge in my discussion from Penrose's discoveries that are valid to those that aren't. The CCC had to be weakened many times as its previous strong versions have been invalidated; one had to add increasingly strong assumptions about the genericity of the initial state but even those conjectures usually found their counterexamples. It's hard to say what is the "current viable" proposal. As far as I know, it's fair to say that the whole program of CCC is dead.

The wrong thinking behind the CCC – the assumption that classical general relativity has to be the full story – has affected many opinions of Penrose in which he is much more demonstrably wrong. For example, he would vigorously attack string theory's insights about conifolds, claiming that the presence of conifolds in the moduli spaces of Calabi-Yau manifolds makes string theory inconsistent or unpredictive. This is of course pure rubbish. Conifolds don't present any problem. In fact, the smooth and totally consistent description of physics on conifold-shaped spacetimes in string theory is one of the most obvious and striking victories in the Duality Revolution. New light degrees of freedom, light objects, and amplified non-perturbative corrections to dynamics (such as wrapped D-branes and wrapped world sheet instantons) whose existence follows – and whose detailed interactions follow – from string theory save the day. Moreover, mirror symmetry allows us to prove equivalences between physics on singular manifolds (from the GR viewpoint) and physics on totally smooth manifolds, and so on. We know for sure that Nature doesn't fanatically abhor geometries that look singular in the approximation of GR; in fact, we know that Nature loves many of them exactly as much as She loves some totally smooth spacetimes because many such pairs are actually exactly equivalent.

Everything that Roger Penrose has said about quantum gravity – and maybe even about quantum mechanics – in his life is probably trash, of course (which is not too shocking because those are conceptually and technically difficult disciplines and Penrose has received no formal education in those modern physics topics). Sorry if I have overlooked anything but it looks rather unlikely. I've been exposed to this stuff for about 20 years when I heard about his (and Hameroff's) crackpot theories on gravitationally collapsing wave functions that induce consciousness (gravity plays absolutely no role in the foundation of normal quantum mechanics and is too weak to matter in ordinary physics situations; the well-definedness of the observations has nothing to do with gravity and has been understood by folks like Mott since 1930 or so, see above; consciousness has nothing to do with gravity and nothing to do with non-existing collapses of the wave functions; the wave function doesn't really "collapse" at all, at least not in any material sense, because it is not an observable).

To enumerate all the pieces of rubbish that Roger Penrose has said about quantum gravity (e.g. holography), quantum mechanics, and physics in general as well as about cosmology – especially about inflation (see also here and here) – and even his far-reaching claims that physics is wrong would as hard as to enumerate his valuable contributions.

But of course, even when I am drowning in hectoliters of garbage of this kind, I still consider Penrose to be an important and creative thinker (it would be silly to deny this fact for someone whose blog has 83 entries that contain the word Penrose) and I wish him a happy 81st birthday! ;-)

Add to Digg this Add to reddit

snail feedback (9) :

reader Mobius Trip said...

So, then, what do think consciousness is correlated to?

reader Tom Weidig said...

FRS = Fellow Royal Society

I met Hameroff once at the Tucson conference on human consciousness. Yes he is jumping on every little idea in quantum physics without much scientific insight.

reader Luboš Motl said...

Hi Mobius Trip, consciousness is a concept with various dimensions. The spiritual ones are very intriguing - do other men; women; chimps; dolphins; dogs; bacteria; DNA molecules; hydrogen atoms realize their existence? Do other people have the same subjective feeling when they see a blue spot as when I see a green spot? - but one should understand that they can't really be addressed by science and many ideas about consciousness in this category are unscientific.

For example, consciousness isn't any fluid that would resemble water. Even if you imagined it to be fluid, its number wouldn't be conserved. Consciousness is nothing that can be "measured" as easily as the volume of water. It's a vague, ill-defined concept.

Then you have aspects of consciousness that can and should be scientifically addressed. Neuroscience is the part of science that does so. What the consciousness is tightly linked to from the scientific viewpoint is clear: it is linked to physical systems that are able to collect, process, and send information (such as the brain, the only truly good example we know). The existence of consciousness is impossible without the brain-like material carrier. Physical systems that function similarly should be admitted to have qualitatively similar kinds of consciousness.

But just to be sure, science only applies to things that can be measured, so what we are finally talking about are the objectively detectable, and not the subjective, phenomena accompanying consciousness, things that we may formally associate with consciousness but they're completely material and "dull" in this sense, such as thinking. ;-)

reader Dilaton said...


I want to get such a cute 10cm accelerator to add it to my buch of keys and impress my colleagues :-P

I know like Penrose diagrams for black holes but I did absolutely not get it from his "Road to Reality" ... I was probably to dumb, it was Lenny Susskind who made me understand them in his cosmology lectures :-D

Thanks for this nice article :-)

reader Luboš Motl said...

Thanks, Dilaton! I completely forgot about Penrose diagrams, an important thing. I will try to explain them sometime...

Pocket accelerator could be fun. ;-)

reader Gene said...

At UC Berkeley during the summer of 1958 I designed a rather complicated microwave test cavity with many probes for the study of the high frequency properties of a YIG (yttrium-iron garnet) sphere. I procured an old electromagnet that had been sitting around the lab for the measurements. I found out later that E.O. Lawrence had built that very magnet and had used it to demonstrate the world's first cyclotron.

About eight years later, at Varian Associates in Palo Alto, my group used an identical YIG sphere (who's properties were now understood) to build the first tunable Gunn oscillator and this became important in making a frequency-agile radar system at very low cost. Compared to the previous method using a backward-wave oscillator (BWO), the cost was reduced by two orders of magnitude and the signal-to-noise ratio improved by three orders of magnitude.

Defensive radar systems use frequency hopping to make it impossible for an incoming missile or aircraft to lock onto their target. This frequency hopping is essentially the same method invented by the movie actress, Hedy Lamarr, in the 1930's. It is often referred to as frequency-domain multiplexing.

reader Seele said...

Penrose wasn't trying to explain the subjective experience of consciousness that you have alluded to above (and that some people, such as Daniel Dennet even deny). He doesn't really deal with this at all. Instead, he was looking for a physical, but non-computational explanation for how the mind apparently is capable of leaps of insight above and beyond what could be arrived at by purely computational processes. If the brain is simply a neural network it can still be modelled without residue by a universal Turing machine and is effectively a computer, operating within the theoretical limits of a computer. Penrose is arguing that there exist thought processes that go beyond these limits, and he tries to offer a physical explanation of how this could be possible.

reader Luboš Motl said...

Dear Seele, thanks. I am really confused by these speculations. The essence of brain processes is surely analogous to the essence of processes in a computer - some electrical signals are running and using a grid of neurons/transistors that may perhaps reconnect as well. However, the detailed architecture of the brain is clearly not isomorphic to a particular microprocessor we have developed and not even to a general class of computers as discussed in computer science. For example, some operations in the brain may be non-deterministic and fuzzy in the same sense as they would be in an unreliable computer. The brain may also rearrange its internal geometry of the connections, and so on. There are lots of other ingredients, and probably more important ones than what I mention, that make human brains different, more "heuristic" etc., than existing electronic computers.

But why would one postulate some completely different processes that don't really work and are supposed to be linked with cutting-edge fundamental physics such as combinations of quantum mechanics and gravity (I don't want to use "quantum gravity" because Penrose's musings have nothing to do with what we normally include in the term)? As far as I can say, there's no evidence that the human brain is "really this different" from a normal computer. Even if a miracle occurred and one could show that the human brain's inner workings were totally different, there would still be many miracles left needed to show that they had something to do with gravity, or "collapses", and so on. The whole scheme just doesn't make any sense. As far as I can say, it fits squarely to the musings by totally average crackpots who just want to impress others by their having learned how to pronounce some sexy words in difficult science such as neuroscience, quantum mechanics, and gravity, and they're combining these sexy tools in random ways to charm others. There's nothing that makes sense in those mechanisms and explanations from a scientific viewpoint. Am I wrong?

reader Peter F. said...

"Am I wrong?".
No, Lubos, at least I don't see that you are! I very much like to see how you appear (to me) to think about this topic. The only weakness (or possible blindspot) in your worldview that I think and feel I have observed (and that is somewhat related to consciousness and its various intensities and "contents") is to do with 'taking into account which factors play a part in how we are motivated to behave the ways we do'. However, I am happy to know you as I know you anyway! :-)