Tuesday, November 23, 2004 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Aliens' answer to the anthropic principle

Cumrun Vafa has just pointed out a new article in the Time magazine to me. The address is

http://www.time.com/ ...

The article starts by explaining who are cranks, and why many scientists turn on their e-mail filters to avoid all contacts with the crackpots. However, the article itself is then becoming increasingly entertaining, frustrating, or irritating, depending on your mood.

It explains that many parameters of the world seem to be adjusted in such a way that life is possible, and roughly speaking, the text proposes four main different explanations why it's so - and the first three explanations are related:

  • God. The most obvious explanation of these coincidences is that there exists God who created the world that allowed Him or Her to start to produce the humans to His or Her image at the end of the week. Instead of God, the creator could have also been Zeus or Odin, as some experts propose. Unlike the previous cases in which science superseded God by another explanation, we may have reached the point where God becomes the only savior, the text explains.

  • Aliens. Although James Gardner, an attorney from Portland, Oregon may seem as another crackpot, the text argues that there is a widespread consensus that Gardner is a genius instead: people like Martin Rees and researchers from SETI and the Santa Fe institute support him. Gardner's theory says that our Universe (whose technical name is Biocosm TM) has been manufactured by a superior race of superintelligent extraterrestrial beings, the so-called U.F.O. Übermenschen.
  • Multiverse. The author uses inflation theory, Darwin's theory, and especially string theory to argue that there are many Universes, and roughly one of them has the right properties for us to be born. Superstring theory was "renamed" to M-theory, and no one except for the specialists should be interested in the reasons why it was renamed. Strange as it sounds, most physicists agree that it is the most likely candidate to unify GR and QM. Shamit Kachru and his friends calculate the number of Universes according to M-theory - because string theory has been renamed - and the result is 1 followed by something like one hundred zeroes (recall that one hundred is 1 followed by two zeroes).
  • Rational arguments. At the very end, the article also offers several reasonable voices - like Brian Greene, Steven Weinberg, Alan Guth - who are not sure whether there are very many Universes around, and who still find it plausible that there will be a rational explanation for the currently unknown features of the Universe. In fact, Lenny Susskind at the very end remains a bit open-minded, too - he almost sounds as a voice of reason. The article does not mention Weinberg's anthropic prediction of the cosmological constant, and it also makes other errors in attributing various ideas to the people.
Well, this popular article shows the direction in which theoretical physics is going to be developed if we eventually accept the anthropic meta-arguments as a part of science, instead of saying "we don't know yet the answer". Just like our Universe will be viewed merely as one representative among "1 followed by 100 zeroes" universes, the physical description itself will be understood as one possible explanation among many others that include different races of aliens and gods. It's just obvious that if we can't make hard and testable (and tested) predictions, the corresponding part of physics will be reduced to "yet another religion".

Add to del.icio.us Digg this Add to reddit

snail feedback (12) :

reader Anonymous said...


If one day in the future somebody ever finds an exact general solution to all the equations governing quantum string/M-theory, do you think it will eliminate the anthropic principle from our vocabulary?

reader Lumo said...

It's just guessing. My guess is that 80% probability is that this full knowledge will move the term "anthropic principle" to textbooks of history, 20% that it will survive. I don't know for sure. Cheers, LM

reader Anonymous said...

Even if string theory works out, we'll need the anthropic principle as an explanation for the existence of Lenny Susskind.

reader Lumo said...

No no no, the existence of Lenny Susskind is explained by the holographic principle: the information is just on the surface, not in the bulk. ;-) I hope that it the previous contributor was LS himself, he will understand that this was just a joke! :-)

reader mark said...

The Deity and Alien explanations are really one in the same - creation by deus ex machina.

It doesn't really matter whether you call it " God", " Yahweh", "Vishnu" or "Darth Vader". A superpowerful extraterrestrial able to populate life on a planetary scale would have technology indistinguishable from magic or miracles from our perspective.

reader Plato said...

Thanks again Arun

He who has Art and Science also has religion
But those who do not have them better have religion.
Professor Chandrasekhar

reader Anonymous said...


The quote actually belongs to Goethe, as the article clearly states.

reader Anonymous said...

I don't see why #3 should be lumped in with #1 and #2. Both #1 and #2 require additional explanations as to where the gods or aliens came from, what their universe looks like, and who finely tuned it so they could arise.

The multiverse, on the other hand, requires no additional explanation. It may or may not be right, but it solves the problem of "how did things get so finely tuned?" without complicating things any further than they need to be.

As for the hope that string theory is going to uniquely predict from first principles several parameters which happen to be exactly what's required for life of any kind to exist... well, I'll have to trust the experts here, but it doesn't seem to me a likely prospect for any theory. Evolution is how we explain the miraculous chain of events within our corner of the universe which conspired to create life, so it seems quite natural to expect that's how it would work on a larger scale.

reader Plato said...

thanks for correction.......

Professor Chandrasekhar used to call himself an atheist when pressed for his belief system. A perusal of this lecture transcript, however, reveals him to be a deeply religious man, in the sense best described by Goethe:I think the point and quote is attributed correctly as you point out, but there is no difference, if Professor Chandrasekhar is considered in this same context?

Lubos is very tricky with his words to Peter Woit in regards to autism, I thought I would try it too:)

reader Lumo said...

Why is 1,2,3 related?

A contributor says:

"I don't see why #3 should be lumped in with #1 and #2. Both #1 and #2 require additional explanations as to where the gods or aliens came from, what their universe looks like, and who finely tuned it so they could arise."

Sorry, but I think that #3 also requires additional explanations, unless we find it OK to freely work with 10^1500 universes with more or less arbitrary properties - and our Universe being one of them.

It's exactly your opinion that "we don't have to require any deeper explanation or mechanism for (3) - the anthropic principle" that clumps it with 1,2. It's exactly what 1,2 are saying. God created all the animals as they are - all of their properties are what they are, and you should not ask "why".

It does not matter whether you call the mechanism "God" or "chance", the scientific consequences of the mechanism are absolutely analogous: there is no controllable framework neither to compute the "right" vacuum (or properties of animals) nor to compute the probabilities of different vacua.

Take 100 pounds of trash, shake it, and create a dog. It's unlikely, right? But you can say that there are 10^{10^{23}} Universes, and it's exactly ours where it was possible for God to create Dog in this way. From the physical viewpoint, the God will look equally supernatural as if you avoided the explanation with "chance" and other Universes.

Whether you call it God or chance, it's unlikely, and it's exactly this unlikely combination of facts that requires a deeper explanation, and 1,2,3 reject this point.

reader Travis Garrett said...

Hey Lubos, this is Travis, I'm a grad student doing numerical relativity (LIGO etc...). I was curious about your opinion on the chances that the mathematical ensemble exists - i.e. Tegmark's level 4 multiverse (e.g. astro-ph/0302131), the collection of all mathematical objects. I think it seems reasonable that the ensemble would exist: first since it appears likely that mathematical structures could give a complete description of reality (strings...) so that reality actually is a mathematical structure, and then furthermore there should be no reason why only one mathematical structure should exist (ours) and not all the others as well (certainly we are just particular permutations of atoms, and can thus not endow the mathematical structure we are embedded in with any special properties - like existence). Note that this would in fact be favored by Occam's razor, which picks out the simplest theory, not the one with the smallest design space (but then Occam's razor is nothing more than a statistical trend...). Crucially I think the possibility of the ensemble makes testable predictions - although sort of weird meta-predictions. That is, in the ensemble there will only be a couple structures whose most basic elements correspond to the deepest physics we have discovered so far, but there will be many more where our current physical laws are just a particular limit of some yet more complex theory. Just as Newtonian physics is the slow speed, weak field limit of GR, and the h->0 limit of QFT, which are in turn the large scale limits of M-theory, so in turn we would in time discover new physical phenomenon that would necessitate reinterpreting M-theory as a particular approximation of some more complex mathematical structure. That is, if the ensemble does exist, then this statistically would be the case, since there are so many more complex theories than simple ones, and thus if the decades, centuries, and millenia role on and we keep on finding ever more complex physics then we can grow increasingly more confident that the ensemble exists. What do you think? It's perfectly reasonable I think, although empirically it'll take an annoyingly long time to start to decide (i.e. do we keep on making progress and finding new things or not).

* It probably seems at first that the ensemble would be a completely unworkable mess where you could never predict, say f, because the inverse f^-1 also exists (so to speak), but I don't necessarily think that is the case since mathematics has structure. You couldn't, say, change number theory so that there are many more primes than composites (at least when ordered in increasing size). Indeed, I suspect that things become much more tractable (although still very hard!) if you consider theories of terms of increasing order of complexity, and also perhaps only consider theories that are algorithmically compressible, as our universe certainly is...

reader Lumo said...

Hi Travis, welcome! I am not sure whether it is a level 4 multiverse ;-), but I believe in the independent existence of mathematical structures - but it's still a different question from one whether these ideas are relevant for the world around us. Sorry, I can't tell you more. All the best, Lubos