[hep-th/0407213] Scientific alternatives to the anthropic principle

Authors:  Lee Smolin
Abstract:  It is explained in detail why the Anthropic Principle (AP) cannot yield any falsifiable predictions, and therefore cannot be a part of science. Cases which have been claimed as successful predictions from the AP are shown to be not that. Either they are uncontroversial applications of selection principles in one universe (as in Dicke's argument), or the predictions made do not actually logically depend on any assumption about life or intelligence, but instead depend only on arguments from observed facts (as in the case of arguments by Hoyle and Weinberg). The Principle of Mediocrity is also examined and shown to be unreliable, as arguments for factually true conclusions can easily be modified to lead to false conclusions by reasonable changes in the specification of the ensemble in which we are assumed to be typical.
We show however that it is still possible to make falsifiable predictions from theories of multiverses, if the ensemble predicted has certain properties specified here. An example of such a falsifiable multiverse theory is cosmological natural selection. It is reviewed here and it is argued that the theory remains unfalsified. But it is very vulnerable to falsification by current observations, which shows that it is a scientific theory.
The consequences for recent discussions of the AP in the context of string theory are discussed.
[PDF]  [PS]  [BibTex]  [Bookmark]

Discussion related to specific recent arXiv papers
Post Reply
Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

[hep-th/0407213] Scientific alternatives to the anthropic pr

Post by Antony Lewis » May 17 2006

This is an interesting paper, and well worth reading. He attacks various rather silly definitions of the anthropic principle, and describes his black hole evolution model, which, at the very least, is a very nice logically possible example of a multiverse with interesting predictive properties.

However he also attacks some sensible versions of the anthropic principle. In fact he seems to be even using one in his 'scientific alternative'! This sensible version is basically the mediocrity principle, Copernican Principle, self-sampling assumption - whatever you want to call it: the idea that if there are many observers we should expect to be typical. A very simple and I'd have thought rather uncontroversial example is Bostrom's explanation of why cars go faster in the other lane:

http://plus.maths.org/issue17/features/ ... index.html

The author does not seem to have any valid complaints against this use of the principle. A feature about this argument is that you can only make statistical refutations - but then anyone dealing with data is used to living with that anyway.

In particular I find his argument against the Doomsday argument rather odd. His argument seems to go like this: there were observers 10 000 years ago; these observers could have made the Doomsday argument; these observers would have said the world was likely to end soon; these observers would have been wrong (we exist!); hence the Doomsday arguement is wrong. What seems to be wrong with this is that of course the 10 000-year old observers are untypical: if you make a statistical argument about typicality, of course a small number of atypical observers will be wrong! So I don't understand his criticism of this use of the anthropic principle (the Doomsday argument may well be wrong, but I don't think it is wrong for this reason).

Finally, the evolving black holes. The idea here is that if universes bud off from black holes, and parameters mutate between universes, the ensemble will come to be dominated by universes that are fine-tuned to produce maximal numbers of black holes. But to then make the (scientific) prediction that parameters we observe should be tuned to maximise the number of black holes, he needs to argue that most observers will be in the most abundant type of universe. i.e. precisely a sensible version of the anthropic principle - we should be a typical observer. If you don't make this argument there is no reason why we shouldn't be living in one of the universes which is very rare (atypically makes few black holes) but happens to contain life.

In conclusion, I think his title should be: "Scientific versions of the anthropic principle". Then I agree with almost everything he says (save above comments). Similar comments apply to e.g. astro-ph/0605173 where the authors are also using the mediocrity priciple to locate ourselves in the most long-lived vacuum, while at the same time appearing to be claiming a non-anthropic argument.

Garth Antony Barber
Posts: 59
Joined: July 19 2005
Affiliation: Published independent

[hep-th/0407213] Scientific alternatives to the anthropic pr

Post by Garth Antony Barber » May 17 2006

You have to be clear about which definition of the AP is being used.

The basic principle, such as that of Stephen Hawking's: "The universe is as it is because we are," simply recognises the fact that, given our existence, the properties of the universe have to be consistent with the properties necessary for life somewhere within it.

The next stage of the AP is to develop reasons why this should be so; the Weak AP, the Strong AP, the Participatory AP and so on.

Smolin's objections to the AP seem to focus on the coincidence they imply and he seeks to explain this apparently propitious nature by physical principles.

In his CNS theory this is explained by universes evolving properties that maximise the number of black holes, which he claims are also properties propitious for life.

Even if he is correct in this assertion, is not this propitious nature of BH production rather a coincidence?

Garth

Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: [hep-th/0407213] Scientific alternatives to the anthropi

Post by Antony Lewis » May 18 2006

I'm not too worried about the 'coincidence' of maximal black hole abundance and life-friendlyness. However the theory would be most convincing if there were some parameter that could be tweaked to increase the likelihood of life but only at the expense of massively decreasing the number of black holes.

Incidentally, on the question of 'am I Chinese?': Let's estimate a probability: assume that the number of people in a population who have time to spend wondering about the anthropic principle is proportional to the number of new PhDs. Given I'm one of these people, am I Chinese? About 11% of recent global PhDs are Chinese[1], so in fact with these assumptions it is about as likely I'm Chinese and asking the question as that I'm British and asking the question. In any case 4/5 of alive humans are non-Chinese, so most people should expect to be non-Chinese. (Asking 'am I British?' is an a postiori question, so I shouldn't be surprised if it has an a priori low likelihood - I have to be something).

Of course Smolin is right that the correct reference class is often not obvious, and to have a convincing anthropic result you need sharply peaked probabilities.

Garth Antony Barber
Posts: 59
Joined: July 19 2005
Affiliation: Published independent

[hep-th/0407213] Scientific alternatives to the anthropic pr

Post by Garth Antony Barber » May 19 2006

Incidentally, on the question of 'am I Chinese?': Let's estimate a probability: assume that the number of people in a population who have time to spend wondering about the anthropic principle is proportional to the number of new PhDs. Given I'm one of these people, am I Chinese? About 11% of recent global PhDs are Chinese[1], so in fact with these assumptions it is about as likely I'm Chinese and asking the question as that I'm British and asking the question
Is that an example of the "Completely Redundant Anthropic Principle"? !!

Garth

Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: [hep-th/0407213] Scientific alternatives to the anthropi

Post by Antony Lewis » August 25 2006

In case people missed it there is a very nice paper on anthropic problems from Radford Neal, now at math.ST/0608592. He makes significant progress, though unfortunately doesn't yet have any convincing treatment of the inifinite case.

Post Reply