Recently Ken Olum wrote an article about a related paradox involving the
anthropic principle, see here:
''Conflict between anthropic reasoning and observation
Anthropic reasoning often begins with the premise that we should expect to
find ourselves typical among all intelligent observers. However, in the
infinite universe predicted by inflation, there are some civilizations which
have spread across their galaxies and contain huge numbers of individuals.
Unless the proportion of such large civilizations is unreasonably tiny, most
observers belong to them. Thus anthropic reasoning predicts that we should
find ourselves in such a large civilization, while in fact we do not. There
must be an important flaw in our understanding of the structure of the
universe and the range of development of civilizations, or in the process of
This is similar to the doomsday argument. Unlike this paradox, the doomsday
ineffective in the case of an infinite universe or an ensemble of all
possible universes (actually, I believe it is
always flawed). Maybe one has to take into account that at a certain point
biological life forms will be replaced by machines. It is likely that a
future civilization consisting of machines will consist of only one
superintelligent being, instead of many independent individuals.
Assuming that the AP is valid and given that we haven't colonized our solar
system yet, we should asume that this is the typical situation intelligent
observers find themselves in. They are just about to leave their planet and
start colonizing nearby planets. This suggests that the replacement of
biological life forms will typically start at this very moment. The reason
is that space is a very hostile environment for biological life forms. Going
to nearby planets and colonizing them is so expensive that it is cheaper to
build intelligent machines to do the job.
----- Original Message -----
From: Hal Finney <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Saturday, January 10, 2004 12:24 AM
Subject: Peculiarities of our universe
> There are a couple of peculiarities of our universe which it would be
> nice if the All-Universe Hypothesis (AUH) could explain, or at least
> shed light on them.
> One is the apparent paucity of life and intelligence in our universe.
> This was first expressed as the Fermi Paradox, i.e., where are the aliens?
> As our understanding of technological possibility has grown the problem
> has become even more acute. It seems likely that our descendants
> will engage in tremendous cosmic engineering projects in order to take
> control of the very wasteful natural processes occuring throughout space.
> We don't see any evidence of that. Similarly, proposals for von Neumann
> self reproducing machines that could spread throughout the cosmos at a
> large fraction of the speed of light appear to be almost within reach
> via nanotechnology. Again, we don't see anything like that.
> So why is it that we live in a universe that has almost no observers?
> Wouldn't it be more likely on anthropic grounds to live in a universe
> that had a vast number of observers?
> The second peculiarity is the seemingly narrow range of physical laws
> which could allow for our form of life to exist. Tegmark writes about
> this at http://www.hep.upenn.edu/~max/toe.html. He shows a chart of
> two physical constants and how if they had departed from their observed
> values by even a tiny percentage, life would be impossible. In the
> full paper linked from there he offers many more examples of physical
> paramters which are fine-tuned for life.
> So why is this? Why does it turn out that our form of life (or perhaps,
> any form of life) can exist for only a tiny range of variation?
> Why didn't it turn out that you could change many parameters a great
> deal and still have life form?
> I don't see anything a priori in the AUH that would have led to this
> prediction. Now, it may just be one of those things that "happens to
> happen", a fundamental mathematical property like the distribution of
> primes or the absence of odd perfect numbers. Self-aware subsystems
> just mathematically turn out to only be possible in a very tiny region
> of parameter space.
> Now, you might be able to make the argument that "tiny" is not well
> defined, that there is no natural length scale for judging parameter
> ranges. Tegmark could as easily have zoomed in on the appropriate region
> of his graph and shown a huge, enormous area where parameters could be
> moved around and life would still work.
> However I think there is a more natural way to put the question, which is,
> what fraction of computer programs would lead to simulated universes that
> include observers? And here, if we follow Tegmark's ideas, the answer
> appears to be that it is a very small fraction. (Of course, you still
> need to use your own judgement to decide whether that is "tiny" or not.)
> In a way, then, these two questions are both related, and perhaps the
> same. They both ask, why so few observers? One question looks around the
> interior of our universe, and the other looks at the set of all universes.
> In each case, it seems that intelligent life is terribly uncommon.
> Hal Finney