Monday, June 12, 2006 ... Deutsch/Español/Related posts from blogosphere

Science vs. democracy

The scientific blogosphere is getting weird because of many reasons.

A fellow blogger and beloved colleague has proposed that the state-of-the-art physical picture of reality should be determined by making a compromise or consensus among all the people who have earned a PhD degree in physics, and later justified this idea in

It does not seem that she was joking. I can't believe that our colleagues might believe that this is how science can work. It just looks incredible to me. Stunning. Nothing against her, of course, except that she obviously misunderstands the difference between having an "opinion" and having a scientific argument.

Among the physics PhD degrees, you can find thousands of crackpots who believe that relativity is wrong, we are surrounded by extraterrestrial aliens, the Hydrogen atom has a state beneath the ground state called the hydrino, and dozens of other unusual theories. Because I know most of them, with a little bit of exaggeration :-), I could also enumerate most of them ;-), but I choose not to. Clearly, these people have no impact on the actual physical picture of reality because they can't offer any scientifically verifiable, convincing, or reliable arguments that could have any impact.

Of course that if you ask one random senior person at the IAS, for example Juan Maldacena ;-), you will get a far better picture than if you average all physics PhDs in the world. The more inclusive your ensemble is, the more distorted picture you obtain by averaging. Is not it obvious? It just can't be otherwise.

When I say that the certain large groups have "no impact", I literally mean zero, not a very small number. It is a very important finer point. If one allows his or her opinion to be distorted by external political pressures, even if it is just by a little bit, he or she is abandoning scientific integrity and consistency of his or her understanding of the real world. It is really impossible in science to be influenced by other pressures than scientific arguments. This principle is extremely important and must be followed very rigorously.

Imagine that you violate this principle a little bit. You take a crackpot, for example one of the hundreds of people who like to contribute to "Not Even Wrong", and you decide that her or usually his opinion about a physical quantity such as the total or maximum number of spacetime dimensions in quantum gravity should influence what we think - just because the crackpot's voice should be heard in a democratic community, blah blah blah. Imagine that you allow him to change the maximum dimension by a small amount such as "10^{-6}".

It should be obvious that using three millions of coherent crackpots, you could reduce the spacetime dimension of M-theory from 11 to 8. ;-) Surely you can easily find three million crackpots and crackpots can often behave as a coherent state.

I think that every reader who has at least a vague idea what physics is must know that such a method of determining the critical dimension as well as the result are completely absurd. M-theory as we know it just can't work in the total number of 8 dimensions. If you choose a different quantity than the critical dimension, the method will still be absurd and the situation will be analogous.

If you ask a qualitative question such as the question whether it is possible to find regimes in a consistent theory of quantum gravity in which strings or branes are created, it is again obvious that the estimated probability that the statement is right just can't be influenced by external pressures, not even by one part per million, otherwise you end up with an absurd, inconsistent picture.

When well-defined ideas are being discussed, one either has an argument or he has no argument. There is no way to shift the scientific picture a little bit by pseudo-arguments.

Real science is of course not determined by random holders of PhD degrees. PhD degrees mean nothing in science. They can mean something for sociology but they can never be used as scientific arguments for anything. Not even higher scientific ranks can directly influence the scientific picture of reality. When you become a postdoc, a professor, or a university professor - or when you are awarded the Nobel prize, it has no direct consequences for science as long as it is science.

Of course, in reality, there exists a high correlation between the ability of people to influence how science looks like and their degrees, jobs, or other achievements. But this correlation can never be used as a scientific argument for anything as long as science remains science. It can only be used as a proxy by those who are not participating in the expert scientific process themselves and for whom the authorities are the easiest way to get some idea about difficult questions. But these people, by definition, cannot have any power to influence the direction of science.

In the real world of science, the major developments are typically caused by an extremely small group of people who are sufficiently hard working, smart, and lucky. It is enough to have 1 person who finds the right result, and science can be revolutionized. Whether or not others will understand the new insights is a different question, and it is clear that the more complicated insight one makes, the smaller number of people will really understand it.

In most cases the authors of important breakthroughs are complete insiders; sometimes they are outsiders. But neither of these things matters. The only things that matter are arguments, their agreement with the observations, and their theoretical rigidity.

So if thousands of vitriolic journalists and crackpots are trying to eliminate or "undo" insights we have made about any questions, for example evolutionary biology or quantum gravity, by threatening scientists, by spreading huge amounts of lies on various websites, my message to them is that their effort can never influence what science thinks until they find a scientific argument. You don't like that life was created by natural processes? You don't like that quantum gravity involves extra dimensions, strings, or branes? You would like to make the creation of life at least slightly more compatible with the Bible, or the theory of quantum gravity to be slightly more similar to crackpots' discrete fantasies? Too bad: you can't do it, not even infinitesimally, unless you actually have some new insight.

Feel free to intensify your screaming, to write 20 times nastier things, to invent 10 new pseudonyms for your disgusting anonymous personal attacks on the web, to invite 100 new crackpots, to write silly books and virulent tirades in the newspapers, and to establish new political parties. Still, your impact on the actual science will be zero. Nothing. I mean exactly zero, not approximately zero.

Science will never be affected by these things unless you talk about corrupt science. Of course, what I described above is an idealized picture. In the real world, the people who establish the new political parties can actually have some impact once they start to execute inconvenient scientists - and some scientists will be influenced much earlier than 1 day prior to their execution. ;-)

But the history teaches us that in the long run, the political manipulators always lose at the end. What you effectively do is simply to move your good scientists to a different place - much like the Nazi Germany that made a huge gift to the U.S. science in the 1930s - and to make your country scientifically inferior, controlled by people similar to you. You will lose the war and the result in the long run is clear, too.

It is just the intermediate timescale that is really threatened, and I happen to care about this timescale, too.

Committees and political constraints of science

The blogger mentioned at the beginning has even proposed to establish "advisory committees" that would be deciding what research directions are promising and what conclusions about general questions that scientists are allowed to make and what conclusions they are not allowed to make. I just can't believe she's serious because what she effectively proposes is to transform the scientific community to a new edition of the 16th century Catholic Church.

Science critically depends on the freedom to obtain any results - freedom that allows to abandon any dogma whenever any single person finds a sufficient body of evidence that the system works differently than thought previously. In fact, the moments when this thing occurs are the most important moments in the history of science. I just can't believe that someone deliberately wants to make these things harder or impossible.

A strong advocate of the "advisory committees" would surely say that the modern committees would be smarter and they would not prevent people from discovering the heliocentric systems and other things. But such a hope of the advocate is obviously absurd. The geocentrist Catholics were also considered to be "smart scholars" during their era.

It is also completely clear what kind of people would try to become members of these committees. Radical feminists would try to get in and attempted to ban any research that disagrees with their crazy belief that the male and female brains are physiologically indistinguishable. Quantum gravity crackpots would get in and tried to suppress all research that disagrees with their childish, discrete idea about spacetime. And so on, and so on. We could eventually dream about the 16th century as a relatively liberal era in which science had the opportunity to thrive.

Once again, it is absolutely critical that scientists have the freedom to reveal the truth whatever it is and they are unconstrained by the pre-determined truth defined by someone else - and it does not really matter whether "someone else" is called the Inquisition or an "advisory committee". It's just about the choice of the words but the principle is the same in all cases.

Bureaucracies can neither determine the truth nor the relative importance of classes of ideas in the future. The more specialized field we look at, the more devastating effects such "committees" would have. Additional analyses of these problems appear in the fast comments.

Add to Digg this Add to reddit

snail feedback (2) :

reader Leucipo said...

So, you are against science evaluation procedures using correlations? Interesting. Can I infer you are also against citation counts?

reader Luboš Motl said...

Dear Leucipo,

you have not written what is exactly "interesting", so I can't judge your non-existent "interesting" idea. I think that what I wrote is not interesting but rudimentary description what science is and what science is not that every kid interested in science should understand from the late high school years.

It should be clear from my text that the weighted average as determined by citation counts is a much better way for an outsider to get an idea about the importance of results in a scientific field that he or she does not directly belong to. It is a method that gives Witten about 1,000 more weight than Peter Woit, and it has other realistic advantages.

But on the other hand, it should be totally clear that not even citation counts can ever be used as scientific arguments for anything. If we accepted the idea that citation counts themselves can be used by actual scientists to declare an idea true, the results would be obvious. Very bad ones.

Randomly and spontaneously created fads and waves of fashion would be gaining exponentially growing influence, because old citations would attract new citations, and so forth, and the percentage of actual content in all of people's beliefs would be diminishing and it would be replaced by irrational fads and cults.

Obviously, in reality, things work in this way to a certain extent (well, especially in certain less prestigious fields such as the climate science), but it would be a catastrophe if this mechanism became primary in any sense, especially in serious fields such as high energy physics.


(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//','ga'); ga('create', 'UA-1828728-1', 'auto'); ga('send', 'pageview');