Sunday, August 27, 2006

Intelligent design and eternal inflation: it would be very foolish...

One of the most provoking recent thoughts of Leonard Susskind about the philosophy of science can be summarized by a quote that appeared in a January 2006 issue of Nature:
  • It would be very foolish to throw away the right answer on the basis that it doesn’t conform to some criteria for what is or isn’t science.

This statement has many aspects - sociological, philosophical, and scientific ones. Let me sketch the basic points that will be discussed more carefully:

  1. I completely understand where Susskind is coming from - he primarily criticizes those who have naive and narrow-minded ideas about the methods how evidence can be accumulated in science. Most of the article will be dedicated to this point.
  2. It was absolutely understandable that Susskind's idea was going to be misused by creationists, including some co-authors of "Traipsing into Evolution", an ID book, who have mentioned the quote on CSPAN2 today (thanks for TheGraduate for the information).
  3. Susskind probably couldn't predict the twists and turns how his ideas were going to be misused.
  4. You humble correspondent agrees with the creationists that if there were two isomorphic situations, we should be making correspondingly isomorphic decisions.
  5. The Reference Frame, however, thinks that the situations are very far from being isomorphic.

Okay, let's start with a more detailed discussion.

First of all, Lenny Susskind is convinced that the picture of eternal inflation involving a gigantic set of string-theoretical vacua is a well-established consequence of our theories that have been supported by sufficient, albeit indirect, evidence.

Doubts about the landscape and eternal inflation

I happen to disagree with him in this point. When we use Occam's razor properly, the very process of quantum tunneling in between two vacua is something that breaks the usual notions of time, space, and causality, and it is very questionable whether science can ever talk about the "moments before our bubble was created" and their mutual causal relationships. More concretely, we don't have too many convincing papers that would study string theory during these unusual transitions.

Also, I view the evidence for the large number of metastable non-SUSY vacua to be insufficient. More importantly, I think that there is no evidence that we won't find a more complete selection mechanism.

On the other hand, I don't think that the previous sentences can be treated as unbreakable dogmas. A scientist should never fall into such a trap. It is always possible that a sufficient body of evidence is found that will force us to change our preconceptions like my preconceptions from the previous paragraph, or Susskind's preconceptions from the paragraph above it.

In fact, I can quite easily imagine that I just don't understand certain crucial things about the landscape dynamics or eternal inflation well enough, and if someone explained them to me, it could hypothetically end up with my admitting that Susskind's conclusions are inevitable according to the best theoretical analysis we can make. Scientifically speaking, it seems clear to me that it is now our turn - the anti-anthropic community's turn - because recent results about the large number of solutions have made their picture (including a quasi-solution of the cosmological constant problem) more concrete and our picture with a more well-defined selection mechanism more speculative.

We discuss these questions in the framework of string theory because we think it is the only consistent theory of quantum gravity and the best framework to ask all these questions in a controllable fashion. But in a more general sense, this anthropic discussion transcends most technicalities of string theory. Non-string theorists usually don't say much about these issues because without string theory, they can't offer any rational arguments in one way or another.

Readiness to revise opinions

Just like I am ready to be convinced about Lenny's picture by a more careful analysis or - less likely - by an experiment, I have no doubts that Lenny is ready to abandon his picture if a more predictive scenario is proposed or a proof of instability of the non-supersymmetric landscape, inconsistency of eternal inflation, or any other problem of his picture is demonstrated by scientific arguments.

People disagree about the probability that the interim anthropic picture of the vacuum selection problem will be completely revamped. But they certainly do agree that the recent results have strengthened the case for the anthropic principle and that either answer can only be accepted if non-trivial technical arguments are given, as opposed to philosophical dogmas.

When I say "people", I of course mean real scientists in the field. Besides the scientists, we have a lot of zealots who are organized at various blogs and who write various books that are critical about science. These people don't think that such "important" questions such as the existence of a gargantuan number of other Universes should depend on "shaky" things such as mathematics. These critics of physics believe that the answers to these questions are pre-determined by philosophy, by religion, or - as they arrogantly call their dogmas - by their "definition of science".

I couldn't disagree more. The very point of science is that it has no dogmas and all of its assumptions can be questioned. If we think scientifically, all of our beliefs and pre-conceptions - including the most viscerally believed ones - are subjects to revision if a sufficient body of evidence shows that these beliefs can't agree with the most recent observations or considerations.

All the answers to the questions like

  • how many classical solutions the fundamental theory has?
  • how uniquely can we predict this quantity or another when we know the full theory?
  • how directly can we observe a particle predicted by a theory?
  • should we accept the existence of other Universes?
  • does our Universe have a finite life-time?
  • does it make sense to ask what happens after the vacuum decays, or what happened before it was created?

are questions that should be studied by scientific methods and that should not depend on any dogmas. These particular questions look esoteric and philosophical in nature. Indeed, they are at the cutting edge of the research of quantum gravity and the researchers can't agree about generally acceptable answers at this moment.

Nevertheless, we have seen many remotely similar questions in the past, such as

  • can there be any classical system with hidden variables behind the probabilities in quantum mechanics?
  • does black hole evaporation preserve the information?
  • what are the individual microstates of black holes?
  • can space change its topology?
  • can a theory in "D" dimensions be equivalent to another theory in "D+1" dimensions?

and many others that could have also seem as philosophical questions but they have eventually been solved. Doomsayers who have argued that science would break down in its attempts to answer certain questions have been proved wrong thousands of times. Even in the 19th century, we would find a lot of examples. Auguste Comte, a philosopher, argued that the chemical composition of the stars would remain a mystery forever. It took less than 10 years until spectroscopy was discovered and Auguste Comte was shown to be a very naive man, despite the fact that he had founded a science-friendly direction in philosophy called positivism.

So the message is clear: science has the capacity to answer questions, including the questions that look very difficult and divine. On the other hand, the answers of science are never known in advance. It's the whole point of science - and the main lesson of all of its revolutions - that we can trust no prophets who have been told the truth by divine forces.

Just like Google's servers must keep on calculating what is the best physics blog at a particular moment ;-), we must still look at all theoretical and experimental evidence and check and re-check our assumptions. And the particle physics community, string theory community, and all other serious science communities are surely doing so. (As you can see, the climate science community does not quite fit the bill.)

In many cases, the new data make us less certain about certain assumptions and sometimes the assumptions can simply be shown incorrect. On the other hand, it often happens that the developments show us that some uncertainty is not really justified. For example, we think that the basic postulates of quantum mechanics are much more inevitable and universal than people would have believed a few decades ago. It's because all predictions of the theories that have assumed limitations of quantum mechanics have been falsified, and all situations that were believed to be inconsistent with quantum mechanics - such as black hole evaporation - turned out to be consistent when the calculations are made more carefully or when a more complete description of the physical system is used. Note that the conclusion of the previous sentence depends on many interesting papers written in the last decades that couldn't have been predicted in the 1920s.

It is extremely important for theorists to be open-minded about all of their assumptions. But the scientific open-mindedness is a very special kind of open-mindedness: it is open-mindedness that commands us to change our opinion when someone (which can include ourselves) gives us new rational evidence, experiment, or a calculation - not when someone smiles at us or, on the contrary, sends his hungry dogs or crackpots against us or when he blackmails us by writing grumpy books that contain no rational arguments.

The critics of science simply never understand the previous paragraph. The concept of not knowing the answer at the beginning and finding the answers by rational arguments is well beyond their grasp. They either believe that their favorite dogmas are unbreakable and should be spread as universal truths without any evidence - or even against available evidence - or they believe that the best evidence for an assumption in science is a sociological argument or an ad hominem attack.

You can't be surprised that the authors of the books critical about science can't be taken seriously by the scientific community because they are missing a very essential idea that makes sciences scientific.

Indirect and abstract arguments

The next related point I want to comment on was labeled as "1" in the list at the beginning. Some naive people interested in science - which often includes the science critics from the previous paragraph - often believe that certain ideas can only be proved in one particular way.

We have already mentioned Auguste Comte who was surely imagining that it was necessary to pay a spaceship that would travel towards the Sun and that would build a chemical lab in the middle. Such a project could be hard even in 2006 but a more important problem with Comte's idea is that the chemical lab simply wasn't necessary. Spectroscopy was enough. Because of his lack of imagination, he was simply unable to imagine that a new kind of evidence could have become available that would instantly make all of the "untestable" data completely transparent.

I could add dozens of similar examples from the century that followed. Instead, let me jump directly to the 1960s when the quarks were proposed for the first time. Once again, many people were just too naive and they thought (and still think) that a theory of quarks would remain "unscientific" forever because the quarks clearly don't exist in isolation - they "can't be seen", in this sense. If they can't be seen, QCD is dead, QED. ;-) Fortunately, there was no blogosphere back in the 1960s, otherwise the dumb counterparts of Peter Woit would have made the life of the proponents of quarks unbearable.

Needless to say, QCD, the theory of quarks, is perfectly scientific and supported by overwhelming evidence. Even long before the evidence was overwhelming, all good particle physicists knew that there was no way how the theory could have been wrong. Yang-Mills theories were the only theories that allowed asymptotic freedom (and various scaling laws and sum rules) and the bound states of the quarks and antiquarks had the right quantum numbers to match the hadrons.

The evidence to support QCD was less direct than the most naive fans of science could have expected - or could have demanded. This trend - the fact that the required steps to find evidence or proof are increasingly more complex, subtle, and abstract - has been a characteristic feature of science in general and theoretical physics in particular at least for 300 years. There is no way how this trend could be suddenly reverted. Theoretical physics is bound to demand increasingly deep and ever more complex mathematical and abstract reasoning, especially if it continues to be increasingly difficult and expensive to obtain new experimental data.

You don't like it? Are you irritated that theoretical physics requires mathematics? I don't know how to tell you my recommendation clearly enough, but let me try: f*ck off. You're just far too stupid. Physics is not for you, okay? Naivity is deadly in this field and the ability to think about sufficiently abstract links between mathematical ideas and observed phenomena is absolutely paramount. One can't do meaningful theoretical physics without this ability and in fact, she can't even understand theoretical physics without this ability.

If you think that the only way to prove string theory is to see a pair of shoes with a string, then you're a moron and you should better avoid any debates about theoretical physics of the last 100 years because noise, laughter, or anger are the only things you can contribute.


Let me summarize the previous section in another way and present some additional examples; I will borrow some specific comments from Lenny Susskind and thank him. Thank you, Lenny. ;-)

All serious scientists know that a theory that makes quantitative statements about the real world just can't be unfalsifiable - only vague theories such as "the Universe is based on the female principle" can be unfalsifiable. Unfortunately, the general public doesn't understand that the universal "unfalsifiability" criticism of scientific theories never has any scientific value.

Moreover, the scientists and critics of science who have used this criticism have always been wrong. A large portion of the important true ideas has been criticized as untestable or unfalsifiable - this criticism is an indirect sign of depth of the theories - and the critics have always been wrong. They always incorrectly and arrogantly overgeneralized their own inability to falsify a statement into the inability of science to make such a decision. For example:

  • Auguste Comte criticized any science about the interior of stars as untestable because he lacked the imagination how the new technologies - such as spectroscopy - could suddenly make the data easily accessible.
  • The Lamarckian evolutionists criticized the Darwinian evolutionists because Darwin's picture talked about the history hundreds of millions years ago which the Lamarckians found "untestable". They had other theories - that the individual lives are imprinted into their carriers' genes and i our children - that were, indeed, falsifiable. In fact, very easy to be falsified.
  • The creationists criticize Darwinism as unfalsifiable even in 2006: while it's incredible, I really mean the 21st century.
  • Most psychologists were criticized by behaviorists who argued that there exists no internal emotional life inside human beings - only things that can be seen by naked eyes exist. The behaviorists were wrong, too.
  • Quarks were said to be untestable because they can't be seen in isolation. They still have not been seen and they will never be seen. But we are certain that the theory is correct. The critics were completely wrong again.
  • Inflation was criticized because one can't look before the surface of the last scattering - it is too far in time - which makes it unfalsifiable. Nevertheless, the case for inflation is getting really strong recently and the new CMB data can already exclude classes of inflationary models.
  • String theory is criticized for having too many stationary points and for not being "falsifable", too, and all serious scientists realize that this criticism is rubbish, too, because virtually all of the scientific work is a counterexample.

This kind of criticism is wrong and vacuous and shows the critics' lack of imagination and their lack of real arguments. Working science never relies on these general philosophical arguments that are randomly used against theories that someone finds inconvenient. It is true that many of the critics' own theories were indeed falsifiable, too. They were often easy to be falsified which didn't make them more valuable: quite on the contrary.

Nature does not provide simple-minded people with affirmative action that would guarantee that Her secrets are easy to be found and easy to be confirmed by them. She really doesn't care. Are you dissatisfied with this lack of policy? No problem, you can always try to move into a different multiverse.

Isolation of the science community

The last paragraphs are meant to address the points "2,3,4,5" and will be rather brief. Lenny Susskind wrote his book - and made all his comments - assuming that all the readers would resemble the particle physics community. In other words, he assumed that all of his readers would be convinced in advance that evolution is right and creationism is wrong. With this assumption, most of them - the radical left-wing ones - should be even happier when Lenny offers them such a large number of Universes that will make the picture of the world even more materialistic, chaotic, and Godless. ;-)

Surprisingly for Lenny, some creationists and some of the most intelligent designers can also read and send e-mails to each other. It was completely inevitable - and, at a superficial level, fair - that they would use his words as sociological evidence for their religious agenda, namely their attempts to introduce Intelligent Design to schools. It was equally obvious that this trend would also be used by the critics of physics.

All of these events were inevitable, predicted at Harvard, unfortunate, but not lethal. The main explanation of the different outcomes is, of course, that Lenny's favorite theories involving eternal inflation are tightly connected with the state-of-the-art theories that seem to describe all phenomena we have ever seen. They are fully integrated into a system of falsifiable - but not falsified - ideas that people study in order to have the most accurate description of the real world that is possible. Sometimes this reasoning is more abstract than the naive observers of science could like but it is still the best science that we can do with the available experiments and observations and with our brains.

Once again, the theory that Lenny has in mind contains not only eternal inflation and controversial ideas about the anthropic principle but also the picture of string theory that logically implies all of the tested ideas of quantum field theory and general relativity. It is a very non-trivial combination of features of Susskind's picture - it is a combination whose importance is severely underestimated by the simple-minded proponents of "alternatives of string theory" and a combination whose breathtaking explanatory power cannot be compared to anything in Intelligent Design, not even remotely.

The creationist friends should try to see that if two people are doing a seemingly identical thing, it's not necessarily an identical thing.

And that's the memo.

No comments:

Post a Comment