Tipler book review Craig R. Helfgott pointed out the following book Tipler, Frank J.: The physics of immortality: modern cosmology, God, and the resurrection of the dead, Doubleday 1994, 527 pages. QB981.T57 firestone library. and golly, that was fun to read! This has to be one of the weirdest books I've come across... Tipler knows about a lot of physics, quite a bit of it rather related to our course, and he has a lot of ideas of various kinds, all intermixed with some really weird religious wackiness. Tipler's fanaticism makes it a lot more fun to read than these 2 pop science books: Penrose, Roger: The emperor's new mind concerning computers, minds, and the laws of physics, Oxford University Press, 1989. [Totally incoherent.] & Hawking, Stephen W.: A brief history of time Bantam Books, 1988. [Written quickly in an effort to make money, I think. Marred by a completely idiotic claim that after the universe stops expanding and starts to contract, entropy will realize it should decrease not increase, bouncing balls will gain energy not lose it, etc. Was Hawking on drugs? This is all the more amazing since Hawking was the major player in showing that entropy is non-decreasing even in the presence of black holes and general relativity, assuming the laws of Hawking radiation and ordinary thermodynamics. When a bouncing ball's energy is converted to heat, that represents something like 10^20 bits worth of entropy increase, which means the universe is then about 2^(10^20) times more probable a state-ensemble than it was before. Reversal of this is a heck of an improbable event.] For another dose of wierd science you could check out Dyson, F.W.: The search for extraterrestial biology, reprinted in Selected Papers pp. 555-571. [Concludes the universe "must" be a lot different that observations indicate that it is. Dyson can't understand what went wrong. I suspect both his physical and his sociological arguments are flawed...] The basic thesis of Tipler's book is... "Life" (= computing devices) will spread out & engulf the universe, and then change the basic structure of the universe in an effort to prolong the subjective-time existence of life despite the fact the universe is doomed to undergo a "big crunch." The whole geometry of the universe will be converted into, and maintained in, a peculiar unstable solution of general relativity called the "Taub universe". In this universe, the big crunch happens in only 1 direction, the other directions stay the same length. And there is a temperature gradient life (heat engines) can feed on. Assuming it requires energy of order kT to perform an "irreversible bit operation" (the subject of our lecture 3!), Tipler does some kind of time-space integration and concludes that the total number of irreversible bit ops that will be performable before the big crunch is infinite, hence life will "live forever." (The idea that life might find it difficult to exist at unboundedly high temperatures presumably does not bother Tipler.) Within these infinite computing resources, a simulation of all possible human beings and universes will be performed, since - why not, what the heck else to do with all those spare cycles - and thus in virtual reality we will all be "resurrected", presumably with appropriate Eternal rewards for the Good and punishments for the Evil! And... more amazingly... in appendix G, I see a sketch of something remarkably like my paper on Church's thesis, including a cite to (an earlier) Gerver paper!! Also Tipler claims his theory predicts the mass of the top quark (185+-20 GeV) and Higgs boson (220+-20 GeV)... and the top quark has now been discovered and its mass is 199+-42 GeV. But... really this prediction was made by Cabibbo et al based on desiderata much less radical than Tipler's future universe... Tipler's desiderata just happened to agree with theirs. Also Tipler predicts a Hubble constant value <= 45 km/sec per megaparsec; the current experimental values seem to be 50-100. But here are 4 objections in increasing order of severity: 1. Current measurements of both luminous and dark mass indicate the universe has factor 10 times too little mass density to be closed... (hence big fizzle, not big crunch - too bad for Tipler theory) 2. Tipler takes a maximally optimistic view of future technological progress, e.g. Superhuman level AI by the year 2002 running on 1000-teraflop computers... nanotechnological 100-gram interstellar self-reproducing spacecraft powered by 8-Km wide light sails weighing only 1 Kg, and fresnel lenses 10^9 Km wide to beam light at them (don't worry - the whole galaxy-colonization project would cost only about $260 billion, i.e. 5 times the Apollo mission's cost). 3. If each bit operation converts kT into heat... and infinite computing is done... the universe will acquire infinite entropy. However, a closed universe (that is, of bounded diameter) with bounded mass-energy (hopefully mass-energy is still conserved!) as postulated by Tipler page 138, necessarily has BOUNDED entropy as we'll see later in the course. So I think something is wrong! 4. most seriously... Tipler seems to have a major misunderstanding of "chaos." A hamiltonian system is "chaotic" if it is exponentially sensitive to initial conditions. That means if you perturb the initial conditions INFINITESIMALLY, the final conditions will also be perturbed INFINITESIMALLY, but an exponentially larger infinitesimal. It is NOT correct to get the idea that chaos implies that FINITE perturbations in the initial conditions will be exponentially amplified at the same ratio. For example, imagine kneading a ball of putty 100 times. Each knead folds the putty over and squishes it back. Result: the putty is folded into 2^100 "layers" (ignoring the fact this is smaller than an atom...). An infinitesimal volume element of putty is flattened by a factor of 2^100 in 1 direction and stretched 2^50 in the other 2 directions. This is analogous to a chaotic Hamiltonian flow in phase space (which preserves volume also). Tiny creatures inside the putty which can do infinitely precise measuring and prediction, thus could figure out how to swim 10^(-100) meters at the right moments and thus arrange to end up wherever they wanted in their putty ball universe. BUT, even if they could swim 1 mm, they would NOT be able to end up 10^97 meters away, or even 1 meter away. No, they would still be stuck in the same 3 cm ball of putty. Tipler thinks life could alter the large scale structure of the universe by making tiny changes and taking advantage of exponential amplification caused by chaos acting over cosmological time. In fact, he sees this as obvious and trivial. However, I suspect that really, no matter what life does, the large scale structure of the universe will remain qualitatively unaltered, and also, the measurement and computing requirements are too much, even if Tipler's vision were possible. (For one thing, Life would seem to have to take its own future actions into account - requiring impossible self-knowledge as in the Halting Problem?)