November 27, 2006
Professor Lee Smolin
31 Caroline St.
Waterloo, Ontario
Canada, N2L 2Y5
LSmolin@perimeterinstitute.ca
Dear Professor Smolin:
The purpose of this letter is principally to enable its author to sort out his ideas about The Life of the Cosmos (“TLOC”) . I found TLOC to be one of the best and most thought-provoking books I have read. Particularly good was your discussion of the philosophical underpinnings of modern cosmology and physics, a problem generally glossed over by popular physics books many of which seem more calculated to stimulate an “oh wow” reaction than further thought.
I am, however, sending this letter to you for two reasons. First, I would naturally love to receive some response to my thoughts presented below. Perhaps your publisher is paying some starving graduate student to respond to letters written to you by adoring dilettanti. Second and more realistically, I hope to inspire you to address some of the matters I discuss here in a future book or in a new edition of TLOC.
I proceed with assurance that you are still interested in explicating views contained in TLOC. Although TLOC was written over a decade ago, my reading of your more recent papers on the Internet indicates that you have not abandoned the issues you discussed in TLOC. Your recent debates with the proponents of the Anthropic Principle (“AP”) and the more doctrinaire advocates of String Theory (“ST”) seem to flow directly from principles presented in TLOC. I am not, however, primarily interested in those debates although they probably are playing a large role in your life now.
While certainly not in a position to advise you on anything to do with physics or the philosophical issues relating to it, I am in an excellent position to present you with raw data on ignorance and errors typical of the prime audience for popular physics books. I have been reading popular physics books, which I refer to as “physics for phools” books, for over 40 years since I tried to understand relativity from George Gamov. I continue to read about two such books a year. While I am probably better read in philosophy than most readers of phools books, I am innocent of any real knowledge of math beyond the sixth grade level and professionally work as a lawyer representing environmental organizations in matters relating to water pollution.
Given this introduction, I trust you will not read any suggestion, question or argument presented below as an attempt to cast doubt on anything you have written or to challenge any of your conclusions. Rather these thoughts, even when presented as arguments, are all to be taken as indications of places where the ignorant may need further assistance.
This letter is divided into four sections. After raising a number of matters that I believe might be handled in a future work, I present thoughts (and misconceptions) on philosophical issues raised in TLOC.
I. Problems for Better Explication
There are a few things that I think need to done to help readers better understand the topics addressed in TLOC.
A. Terminology for Discussing Everything
Authors of popular physics books should adopt a convention as to how to refer to everything that there is. As Leonard Susskind mentions in his 2004 debate with you on the Anthropic Principle, the word for everything there is used to be “universe.” Now that we are talking about multiple universes, a new term needs to be invented for everything. Maybe that is what you meant by “cosmos,” but that’s not clear and there are presentations in TLOC that are less clear for this reason.
In the balance of this letter, I will adopt the convention that the word for everything (including all universes and anything that could be outside any universe) is “cosmos.” Accordingly, I may ask how many universes there are in the cosmos but will not ask what lies outside the cosmos.
B. Where does all the stuff come from?
Somewhere there should be some answer to the prejudice of the ignorant that something does not come from nothing. I gather that black holes have only a finite amount of mass and that it is the amount of this mass that determines some of the properties of black holes. How then is it possible that a black hole can give birth to another universe that has multiple black holes each of which in turn can give birth to another universe that will contain multiple black holes? Where does the stuff come from for all this new creation?
I have heard of the concept of vacuum energy and false vacuums but I do not see how this is the supply of stuff for an infinite number of universes emerging from black holes. Probably I am missing something simple here, but one assumes certain duties when one writes books for the simple minded.
C. Time in the cosmos or just in the universe (or universes)?
Time and causation in TLOC need further explication. Your Cosmological Natural Selection (“CNS”) theory relies on causation between universes at least to the extent that changes between universes must be “small.” (TLOC p. 117) Further, it is stated that “[t]ime does not end in the centers of black holes” (TLOC p. 116) However, “there is no meaning to time beyond change.” (TLOC p. 355) and it has commonly been thought that causation and time only make sense as applied within the universe. Indeed, if space and time all have to do with the contours of a particular universe and relations within a particular universe, it is unclear what meaning time or causation could have as applied to the cosmos or outside of any particular universe.
Somewhere it must be explained how time is to be handled in relation to the cosmos as a whole. It appears from your treatment of “time” as relational that the concept of time simply has no bearing as to the cosmos but it would appear that some concept of cosmic time emerges from your theory that black holes give rise to new universes that resemble the universe of the parent black hole.
D. The Relational Theory of Space and Time
TLOC handles the debate between Newton and Leibniz and the difference between special and general relativity particularly well. Although I have read about a dozen presentations of this material, I think it finally sort of clicked.
I have one quibbling issue, though. I suspect that the dizziness one feels from spinning and the sensations one gets from accelerating have more to do with human physiological processes than with the properties of general relativity. At least I believe that the Darwinian reasons for us perceiving acceleration and rotation differently from inertial movement have more to do with the needs of dealing with matters on earth than with orienting ourselves relative to the universe as a whole.
We are not typically in ship cabins without windows and do have ways of identifying our movements relative to the surface of the earth. That such relative movement is normally perceived with our eyes while there are additional senses that perceive acceleration and rotation would seem to be a matter of biology.
II. Principles of Scientific Explanation
The nature of science and what counts as a scientific explanation really seems to be the big fish to fry here. It is at the center of the Anthropic Principle debate and, from what I see from Science News and your article for the New York Academy of Science, your recent disagreements with some practitioners of string theory.
It is a notorious fact of philosophy that whenever one starts trying to set forth the principles for ethics, knowledge or anything else some joker asks what the basis is for those principles. The effort to supply that basis leads to setting forth a set of meta-principles for which justification can also be demanded. It was realized, at least as early as the time of Aristotle, that at some point one has to tell the joker to accept the principle as obviously true or, at least, shut up. The principle of sufficient reason cannot require that everything be justified.
In TLOC, you discuss or allude to a number of criteria that should be used in determining whether a theory is potentially true or scientific. Behind some of these criteria there are further arguments stated either in TLOC or in more recent articles. These criteria are that scientific theories:
1. Agree with observations
Everyone agrees with this criterion in principle although there are often none so blind as those that will not see. Also, what we are looking for may affect what we observe both as a matter of psychology and quantum physics.
2. Rest on few unexplained facts and principles
It is generally assumed that a theory with fewer laws, ultimate entities and forces is better than one with more. Everyone seems to assume this is true and probably this is a suitable place to just tell the joker to accept the criterion and shut up. The holy grail for this principle would be to prove, as Descartes attempted, that all the observable facts can be explained as a matter of logic from the mere fact of existence. ST is a bit less ambitious and hopes to explain everything using a theory that everything consists of one type of string that is vibrated in various ways in various dimensions to make up the cosmos. The inability of ST to reach a unique set of basic parameters is seen by some as further support for the AP
In any event, the criterion that a theory should use few laws and entities is taken in this letter as self-justifying.
3. Are self- evident or, at least, not ad hoc
In addition to having few entities, laws and initial conditions, we want the given to be self-evident and not to appear ad hoc. Just saying it all depends on a god is simple enough but does not explain anything. We must start somewhere, but we do not want to start somewhere that ends the game before it begins.
This principle of science is harder for much of the world to accept than the first two stated. The religious person might ask what is wrong with accepting a dues ex machina explanation if it leads to a happy ending to the play. Science, however, is right in rejecting this argument as being inconsistent with the game it wants to play.
But the joker still has questions to pose here that the scientist cannot simply reject as out of bounds. When is something self-evident or acceptable as an initial condition? The idea that universes can spring out of nothing, even if there is a lot of energy contained in the modern concept of “nothing”, strikes a lot of people as far from self-evident. At some point do we just accept things as true because we observe them, even if we have no basis for explaining them, because “that is the way things are?”
If the string theorists could do away with the landscape, get the number of entities and laws down to only a half dozen, fix on a particular number of dimensions and (using some means of observation yet unknown) observe the ultimate entities and dimensions, one could still ask why they stopped there. What is the reason there are these entities, dimensions and laws? Indeed, this question might be asked of anything short of Descartes’ proposal that everything be shown to be necessary from the bear fact that there is some form of existence.
4. Verifiable
This is the main weapon you launch against AP and ST. In a recent article, you state that, “[o]ne of the most fundamental principles of science has been that we only consider as possibly true those theories that are vulnerable to being shown false by doable experiments.” But it does not seem that everyone accepts this “fundamental principle.” There is a very active debate as to the extent in which the principle that a theory should be able to be falsified is a requirement of science.
In TLOC and other of your writings this principle is not really taken as given. It is argued for using appeals to other practical and philosophical principles including the suggestion that, due to failure to adhere to this principle sufficiently, many people pursuing ST have wasted a lot of time and research money.
Still, to some degree, this seems to be a matter of trying to agree on what game to play. One could play a game like soccer that allowed players to carry the ball in their hands and the game might be fun, but it would not be soccer. Although Susskind and other ST defenders say that they think that their theories will eventually be able to be tested, they are less concerned about verifiability than you are. Whether this is “unscientific” is not clear because the boundaries of science are not set forth as clearly as the rules of soccer.
Let’s assume that someday the ST people work out their math and physics to the point that they can offer an interpretation of the universe that uses just a few rules, entities and dimensions that involve only aesthetically pleasing numbers, but that these entities, dimensions and laws could not be tested in any way given the equipment we are likely to have in any of our lifetimes. I gather Smolin would say that the ultimate theory has not been achieved but the proponents of ST would say that it had. You might argue that accepting an unfalsifiable theory is just not science and that if mankind were willing to accept aesthetically pleasing theories that could not be tested, it would have stopped doing science after being presented with any one of the “beautiful” religious or philosophical systems of the ancient world. The hypothetical string theorists could answer that their theory is not arbitrary, it is aesthetically pleasing and mankind should just accept it and work on curing the common cold. I do not see a way to resolve the debate at that point except by letting you keep working on a verifiable theory and them retire.
To put the matter another way, is insisting that science must be verifiable to insist that the ultimate reason for a phenomena accepted by science must be an observation? Must one keep asking for a sufficient reason until one can just point to what one sees and say “see, it’s there.” What I will call the aesthetic theory of science is willing to stop with principles that are aesthetically pleasing, simple and consistent with observations.
Finally, the requirements for when something is verified are not clear as to cosmological issues. We will never be able to receive any sort of direct evidence of another universe. The most that will happen is that positing that there is a universe creating mechanism that creates many universes is the most elegant way to explain facts about our universe that would otherwise have to be attributed to chance. This is unlikely to be considered verification by people who generally verify things by seeing or touching them.
5. Gratifying to the human spirit
Everyone, whether they admit it or not, also seems to want a theory of the cosmos that allows progress, avoids “heat death” and otherwise gives some sort of meaning to human life. Probably no one thinks that this is a scientific principle but direct and indirect appeals are made based on this principle all the time. (See e.g. TLOC pp. 179, 218, 373)
This does not look so bad. Why do we do science anyway? Assuming it is not contrary to observable facts, why should we not prefer a theory that makes us happy over one that makes us miserable?
Indeed, I have toyed in the past with developing an “anti-scientific” theory that is based entirely on what would be true of the cosmos if things were as we would find them most satisfying. Most of our current religions would fail badly on this point because, having been designed when life was short and brutish, they make human life on earth a far too trivial affair to satisfy people committed to being here long. Given the expectation of a fairly long life, but the knowledge of eventual death of ourselves as individuals and of all things in time, one is looking for a cosmos that allows lives something like an organized sport in which it is important to play each game to win but there is another game tomorrow and a new season after this season’s winner has had a period to celebrate. We would like an eternal return with infinite variety, memory of past episodes and the possibility of continued growth.
It will be seen below that measured using the “gratification” criterion, the CNS theory does not really seem to do much better than the weak AP.
III. Theories for Explanation of the Remarkably Felicitous Shape of the Universe
A particularly strong element of TLOC is the way the governing parameters for the universe are set forth and the presentation of the menu of possible theories for addressing the reason for this felicitous and unlikely collection of governing parameters. The menu consists of:
- Strong AP (god)
- Weak AP (very large number of universes chosen at random)
- There is only a single unique mathematically consistent theory (the rationalist’s dream)
- Fundamental “parameters may actually change in time” (TLOC p. 55-56)
TLOC rejects Strong AP as unscientific and, although I think this would have pained your hero Leibniz, the Rationalist theory that everything is as it must be as an a priori matter, because it “strains credulity.”(TLOC 55-6) No argument here as to either of these points although it is possible that, without explaining everything, the need for logical and mathematical consistency might have considerably cut down the number of possible universes that have to be created for the AP to work.
After setting forth the menu, TLOC then goes on to explain the CNS theory.
IV. How does CNS fair against the principles of science and specter of the Eternal Return
It seems from the AP debate and other papers that you still fully support the CNS theory at least as to charges that it contradicts what is now known about black holes.
As I understand CNS theory, it holds:
1. At least one universe simply came into existence (proven by observation),
2. It is very unlikely that any universe that simply came into existence would have the parameters we observe in our universe (proven by observation of the parameters and the fact that they are so scattered and unrelated in scale),
3. The cosmos has at least two universe-creating mechanisms that work after the initial universe(s) come into being: bounce from the collapse of a universe and creation from black holes, (postulated),
4. The black hole mechanism is capable of creating many more universes than the other existing mechanism(s), (postulated),
5. Universes that create black holes have highly unlikely parameters (proven through physics I do not understand)
6. At least the mechanism that creates universes from black holes creates universes that have similar parameters as its parent universe, (postulated by Smolin, but claimed to be impossible by Susskind and others)
7. Because universes with parameters that favor creation of black holes create more progeny, such universes are favored by natural selection and are more common, (Darwinian logic)
8. We are probably on a ordinary universe (mediocrity principle) -
Therefore, the universe in which we find ourselves is one that favors creation of black holes.
Further, this theory is testable, unlike the AP, because we can observe whether the parameters in our universe are those that facilitate creation of black holes. Thus, it does better on criterion #4 above and, assuming you are right about black holes, does as well on #1 as AP and the theory that new universes are created only by eternal inflation.
Arguably, CNS has less simplicity (criterion #2) than eternal inflation because eternal inflation may only need one universe-creating mechanism while CNS needs two, but that is not clear. One should also score AP and CNS as a draw on #3 as neither theory seems ad hoc although unfortunately neither would probably have impressed Descartes as self-evident either.
You seem to think that CNS does better at gratifying our desire for evolution and progress and avoids the Eternal Return that Nietzsche and others found to be so dreary. (TLOC pp. 179, 371) It is not apparent why you would feel this way.
Under CNS, “[t]here must be enormous numbers of other universes” (TLOC p. 118). Why there would be any less under CNS than under the assumption that there must be a sufficient number of universes for observers to become probable under the AP? Indeed, there does not seem any reason why there is anything short of an infinite number of universes in the cosmos under either theory. Both theories postulate universe-making mechanisms that create huge number of universes without suggesting any limit to the operation of the mechanism(s). To argue that the processes run out of gas at some point seems completely arbitrary. The only difference between CNS and AP in generating an infinite number of universes is that, being less random, a higher proportion of the infinite number of universes contains black holes and, as a result, a higher proportion of the universes contains observers.
It is the infinite number of universes combined with the finite number of possible universes that generates the Eternal Return. CNS does not suggest a way around that.
Also, the dreariness is not lessened by any real evolutionary progress. The evolution of universes, if it exists, is quite limited. Once the parameters for peak production of black holes are found, the black hole universe-creating mechanism will hover around those parameters eternally. Further, I did not learn anything from the infinite number of other [Kantliebes] in other universes. Nothing I learn here will be of any benefit to any of them. The only progress is within a universe and each such universe may face a heat death or a big crunch. Disallowing the possibility of an observer outside of the universe(s) (TLOC p. 334) only makes things more dreary by eliminating any hope of a memory and of anyone getting any lasting benefit of the infinite number of dice games.
If the number of universes in the cosmos is infinite and the number of possible universes is finite, it really seems like we can almost go back to Plato and see the “form” as primary and the “matter” as secondary. Everything possible is realized an infinite number of times. Each new universe is just another boring affair and the point of science would be to determine all the possible universes. We might be back to Leibniz, as well as the other rationalist philosophers, more than you would like.
Thank you for your consideration.
Respectfully submitted,
original was signed
Doctor Smolin sent no response but then I did not really expect him too. It would be best if he dealth with some of these issues in his next book.
![[info]](http://l-stat.livejournal.com/img/userinfo.gif){kind=small}