The creek of reason
Was only ever a narrow ribbon of hope in the desert
But now it is a chain of stagnant waterholes
Full of shopping trolleys and broken bottles…
I am here in another inland city with a minor university, a minor university that has decided not to offer a chemistry degree anymore and so has handed over the chemistry it is contractually obligated to teach its third year students to our minor university: specifically, to me. I am here to supervise four going-on-five days of experimental work, where from 8:30 to 5:30 with a break for lunch they will grind away doing experiments. These experiments involve the exercise of real, unglamourous, skills- distillation, recrystallisation, not poisoning oneself or settling oneself on fire. The lab we are in is splendid. Such a high ceiling. Taps everywhere. It has the look and the smell of the organic chemistry labs of my own undergraduate days, but it is much too big for us. There is plenty of equipment, plenty of apparatus, plenty of room to rattle around in. All my life this sector of science, my sector of science, has, in this country at least been shrinking.
I never liked the term ‘central science’, but it is. It is the real key to solving our problems in health, our problems with the environment, in developing new materials that will allow
Of course, these things are still rolling along, I tell myself. Science is not really a dried-up creek, it is still an unstoppable tsunami. Only not here. Or, not much.
Sometimes I google something to find out more about it and find something I have written about it, almost immediately.
Sometimes what I have written is embarrassingly wrong.
Watched the documentary about murdered Russian journalist Anna Politkovskaya on SBS the other night. I was impressed by how flash
As we enter the age of stupid…
My aversion to many-universes models was excessive and irrational. After thinking the matter over carefully, I no longer think there are any logical or moral objections to a many-universes model. I will continue to reject these models on two bases, but I will seek to moderate my language and will not automatically reject them. These two bases are:
(1) Parsimony. We can explain things adequately without them.
(2) Utility. They do not shed any useful light on the most interesting and crucial aspects of quantum mechanics, and hence are not very exciting.
I will now outline my initial objections to many-worlds models, and the answers to those objections I have come up with on further reflection.
The first objection is the aesthetic-moral one made so effectively by Larry Niven in the story ‘All the
[I need to find the story again to find the quote I want to put here, so in the meantime you can link to this discussion about the Buffyverse.]
But, in agreeing with this objection, I am not thinking quantitatively. And I am not thinking correctly about *how* microscopic splits of universes lead to macroscopic variations.
It is not equally likely that I will keep driving merrily along the road, and that I will veer suddenly to the left and run over a nun. The bundle of lives that it is ‘me’, if it has a 1 in 10 trillion chance of veering aside and hitting the nun, has been shaped by the choices it has made to be ‘better’ than some other bundle of lives that has a 1 in 10 million chance of veering aside and hitting the nun. Furthermore, I do not know that a choice I can think about is actually possible: perhaps my freedom is more restricted than I believe, and I actually have no finite probability of veering aside and running over the nun.
Perhaps this is best expressed in terms of another one of those Socratic dialogue thingies.
Assertion: It seems that the possible number of universes is infinite, such that any imaginable universe is not only an image of truth, but a fraction of truth. Therefore the many-universes model is immoral and aesthetically repellent.
Response: However, the evidence suggests strongly that our universe has not existed forever, which means it had a beginning. This beginning surely imposes some set of initial conditions on the bundle of possible universes, which implies that the number of multiple universes is not infinite.
Assertion: Although the possible number of universes may be finite, it seems that we cannot quantify, to any extent whatsoever, the probability of one universe existing rather than another. Therefore the many-universes model is still immoral, yadda yadda.
Response: However, statistical thermodynamics provides a very good model for how we could do this. Imagine, if you will, a simple universe of a large number of red balls that can be placed at any of a large number of energy levels separated by a finite amount of energy. Let us imagine that this universe has a certain total energy, that it begins at some point in time with the balls assigned to some unique configuration of energy levels, and that balls are free to move from one energy level to another- not completely freely, but with a high probability to adjacent levels and a decreasing probability to more distant levels.
This is a reasonable model for a classical universe. We can’t know exactly what the configuration of balls is, but whenever any change happens in this universe, we can say it is in the direction of the configuration of balls that can be made in more different ways. This is just a statistical law, which arises from having lots of balls, but it is so solid that it is the basis for the second law of thermodynamics, and hence for everything.
Figure 1. Two sets of red ball universes with total energy 42. There is only one way of arranging the balls that gives rise to the universe on the left. There are many ways of arranging the balls that give rise to the set of universes on the right.
Is there anything to stop us from saying that the physical meaning of probability is that everything happens, and every time a ball moves from one place to another the universe splits?
No, there is nothing to stop us, we can certainly say this. This might be a useful way to look at things in a classical universe. It gives a very straightforward physical interpretation of probability. There are no real aesthetico-moral objections, since the choices (or deterministic non-choices) we make are the same ones we would make in the classical universe, according to the laws (or guidelines) or thermodynamics. We just interpret them as probabilities rather than unique events. I think, given how very large large numbers are, if we say that every particle in the universe can choose (or be impelled) to change state once a chronon, our macroscopic choices (or non choices) take place at such a high level of emergent phenomena that most of the unnerving possibilities that make us turn away from the many-universes model, like Larry Niven, with fear and loathing, are in any meaningful sense of the word, impossible.
Assertion: Okay, okay, maybe you’re right. But what about the two-slit experiment?
Response: The two-slit experiment?
Assertion: Let me explain. Actually, you can probably get a much better idea if you go out and get hold of a copy of Feynman’s ‘Lectures in Physics’. It is in Volume III.
If you have two slits in an electron-proof thingy such that a single beam of electrons can go through both of them at once, then ping onto an electron detector, you can get two different outcomes, depending on whether you put something at the slits to detect whether the electrons go through them or not.
Figure 2. Things that can happen in the two-slit experiment with a beam of particles.
You get this result- electrons acting like particles would act if you detect them, and like waves if you don’t detect them- even if you shoot them through one at a time, so you can detect each electron striking the other side individually, ping, ping, ping.
The two-slit experiment is relevant because it seems to imply communication between bundles of universes. An electron chooses to go through one slit or another: the universe splits. But the overall features of the observed universe depend on the choices of many electrons. How can this make sense, in a many universe model? How can the many-universes model cast any light on this? Should we not see, in a universe of red balls, the particle-like distribution of particles in all cases.
Response: Yes, this is one of the more wacky things about quantum mechanics. Actually, I fail to see how the many-universes model can cast any light upon it. Perhaps Marco (pbuh) can explain.
However, the two-slit experiment can still make sense in a many-universes model. It does *not* imply communication between universes, because wave-particle duality can mean something like the De Broglie pilot wave model, which is perfectly consistent with electrons going through independently to create an interference pattern. This does not rely on any spooky ‘now I’m a wave, now I’m a particle, ooga-booga’ weirdness, which Dr Clam has decided he finds more irritating than the many-universes model.
Assertion: What about the different behaviour of fermions and bosons? A few chapters later on in Feynman’s ‘Lectures on Physics’, there is this really nifty discussion of scattering. Particles that are different from each other scatter in one way, which is the same sort of way, more or less, as macroscopic lumps of matter. Particles that are identical scatter in one of two completely different ways, depending on whether they have integer spin (bosons) or half integer spin (fermions).
Feynman writes somewhere that he tried to put together an explanation for this difference between fermions and bosons into an undergraduate lecture, but found he couldn’t do it, which he says means that we don’t really understand it. Whatever it is, doesn’t it just knock the stuffing out of the many universes theory? Here we have macroscopic consequences arising from probabilities that don’t seem to behave anything like probabilities behave in your universe full of balls. And that’s the wacky way all probabilities behave in quantum mechanics. Sure, you can save your many-universes theory by adding lots of wacky ad hoc rules about how ‘balls’ of different kinds should behave, but what good is that? It hardly makes it a useful predictive theory, huh, huh?
Response: Again, this example is one of the more spectacularly weird things about quantum mechanics. And maybe shoehorning it into a many-universes model would just be papering-over a dodgy theory. But maybe it could actually shed some light on the problem. For instance, our model of the universe of balls sort of implies that we can tell the balls apart: but if the balls are indistinguishable, then there is only one way to get to this configuration, just like there is only one way to get to this configuration, and they are equally ‘probable’. Or am I dreaming?
Figure 3. Two sets of red ball universes with total energy 42. If all the balls are indistinguishable, isn’t it true that there is only one way of arranging the balls that gives rise to the universe on the right, just like the universe on the left?
White King: Tell him he’s dreaming.
9.10.8
I have been spending my time here in this inland city, doing wet chemistry during the day- which I am not so good at, I would not be the one doing this in a more knowledgeable time- and at night I come back and watch the Sky business news- because I have a very low resistance to television- and read ‘Accelerando’. Only a chapter at a time, because then my brain is too stretched and I have to go and take a nap. I thought tonight that maybe I am wasting my time, messing about with dumb matter, when the real game is elsewhere. But someone has to figure out these things.
I am reminded, in these exciting times, of a story I once read by Leo Szilard, where capitalism is compared to the manic depressive cycles of the insane.
Stross has something that is superficially plausible in ‘Accelerando’, on making command economies as effective as market economies, using expert systems that can evolve optimal allocations of resources without the need for competition in the real world. This sounds good, but what it leaves out is the ‘Garbage In, Garbage Out’ principle. We have all these sophisticated computer models for all sorts of things, and so many of them are just rubbish, because they need more and better inputs than we can provide. A pseudo-market in silico cannot function like a real market until we are massively, massively, massively documented in real time.
When that day comes, we will have to get rid of so many of our stupid laws it isn’t funny.
Usury is a sin. Muahahahaha!
The world is flat - with gigantic frigging walls built everywhere.
Like, our dog runs off. He doesn’t go very far, and he hasn’t gotten into a great deal of trouble (yet), but we live in the sort of neighbourhood where one day, someone will shoot him. We went to a good deal of trouble putting up one of those electronic dog fences, which worked real well for a while. However, he is a dem clever dog, and when the battery in his zappy collar was low he worked it out, and just barreled out through it from then onwards. We thought we would get a more uber zappy collar, and the ones available locally were real expensive, but there were much cheaper ones on the net from overseas- like half the price- so we ordered one of them. Customs stopped it. It is apparently legal to sell them in
Like, I get all these emails from Indian undergraduate project students looking for a place to come and do a three month or so final year project. Unfortunately, I always have to write back and tell them what I have found out after making inquiries, that there is a 0.0001% chance of their getting a visa to come and do this.
Like, our university demands one particular English test. It is the TOEFL instead of the IELTS, or vice versa. I can’t quite remember. I ended up waiting nine months for a student from the Middle East I had offered a scholarship- who had perfectly decent English skills- to get in to do the proper test at the heavily oversubscribed testing centre in his country, and what with one thing or another, I had to hand the scholarship money back before he could take it.
They are absolutely batshit insane – to quote Lexifab, in another context – about occupational health and safety at the place I am visiting. A fellow came around this morning to move my pushbike – because he got an email from someone else in the Geschutzapparat, telling him to come around and move my pushbike – because it was a safety hazard, leaning up against a wall inside the lab in a way that impinged too much on a thoroughfare that none of us actually fared through during the course of a day.
It is the only place I have ever been where prescription glasses do not count as safety glasses in the teaching lab. I have just ignored this directive.
10.10.8
You should try and complete this statement my son made to me the other day: ‘My mind gets confused when I…’
I am about 43% of the way through Accelerando. My brain is feeling stretched, as it is meant to. At the same time, I am living through what seems to be an exciting economic phase transition.
It is probably right and good that the tertiary sector as presently constituted should begin to wither away. It is configured for the old age of stupid, not the coming age of stupid. We need to educate a leaven of people who actually know things to get us through the coming age of stupid, using the tools of this/that age.
A symptom of the new age of stupid is the idea that simulated ‘experiments’ can replace experiments. We need to find a way to escape from the tyranny of safe.
I predict that in another month, after the ‘worse than expected’ RMP growth numbers come out (worse than expected by who?) everyone will be looking critically for the first time at all the RMP economic data we’ve seen for the past decade and telling us that it is dodgy as. Roll on Economics 2.0!
‘…think about knight’s moves in four-dimensional chess.’
11.10.8
Finished Accelerando. Once the pace accelerates past what the characters can cope with so their story is left behind, it is relatively easy for a simple human like me to keep up.
For what its worth, I don’t think it is possible- within the universe- for there to be a conscious entity which stands in relation to a human being as a human being stands to a tapeworm. This analogy crops up a number of times in Accelerando.
Datum: The complexity of possible problems scales in a dizzying way. For example: Two body problem, easy. Three body problem, impossible. (NB: Number of bodies in the universe, 10180 ono. )
Datum: There is a limit to how fast information can pass from one place to another. It isn’t all that fast, compared to speeds we can imagine.
Datum: There is a limit to how much stuff can be packed into a particular volume. It isn’t all that much, compared to smallnesses we can imagine.
I think if you put these three things together, it will work out the horizon of possibly tractable problems in our universe will turn out to be not all that far removed from the horizon of problems solvable by human intelligence in our universe.
I believe there could certainly be conscious entities which stand in relation to a human being as a human civilization of several quadrillion humans stand to a human being. But I don’t think that degree of ability to hold information, to come up with new ideas, to link existing ideas, would be qualitatively different from human intelligence, to the same extent that we are different from tapeworms.
17 comments:
Re: the two slit experiment.
A- Fire a beam of electrons at the slits and get the familiar wave pattern. This represents how all the new universes arising from that point are the same. They all have that same wave pattern. When all the universes have the same result - that is science.
B- Fire one electron at the slits with an electron detector at a point somewhere along there (attach a thermonuclear device to the detector if it detects the electron if you like). There is considerable diversion between one set of universes (armageddon) and the others. This is clearly outside the cup of reason = not science. The act of observing a chancy event puts an observer (the detector in this case) in a very privileged position. If nothing observes it, all the universes emanating from an event end up being the same.
I define humans as "special observers" as there is complex processing of observations, thus with our "creativity" or "free will" actively selecting the universe we end up in.
"If nothing observes it, all the universes emanating from an event end up being the same."
Nup. There is a distribution of probabilities of different events happening, and hence of universes, whether you combine the wavefunctions (no observation) or the probability amplitudes (observation). The way probabilities combine does not behave the same way in the two different systems, and my point was that I don't see any way the many-universes model throws light on this.
Perhaps what I *really* need to do is to mount a spirited analytical defence of what may be our real grounds for disagreement, your assertion that models that attribute a privileged role to the observer are not utter rubbish. ;)
Nup. There is a distribution of probabilities of different events happening, and hence of universes, whether you combine the wavefunctions (no observation) or ...
I'm not quite sure of what you mean by this. By "observation" in the quantum sense, I am talking about a differential dependency on the outcome of an event (whether it is physically observed at that time, or the results of which leave evidence such that a future physical observation can depend on it)
The distribution of probabilities make no difference if there is no differential dependencies (observations) on whether things go one way or another in these universes. Yes there is a spectrum of universes generated either way, but they'll be identical until something gets "observed".
There is one distribution of universes if "something gets observed" and another distribution of universes if something doesn't. All the universes won't be 'identical' in either case. Do the maths!
I can scarcely follow this discussion, as is the case with all topics centred on any kind of mathematical equation more complex than basic algebra. However, I make two comments:
1) I like the idea of multiple universes because I secretly harbour the hope that one day the world will be forced to unite in order to repel an invasion by dinosaur-riding laser-wielding Hussars of the Greater Prussian Ultra-Empire. Your mathematical estimation of the likelihood of this (near-certain) eventuality is of no relevance!
2) Clam, now that you have defeated 'Accelerando', you may be interested in Stross' Laundry novels ('The Atrocity Archives', 'The Jennifer Morgue') which are a pleasing mashup of Len Deightonesque cold war spy thrillers and the Cthulhu mythos. Have I mentioned these before. I don't know. Come to think of it, it may have been you who pointed me towards his short story online about Oliver North triggering an Azathoth (or something) vistation that destroys the world. In which case, ahem.
Have you read Fork! by Andrew Shellshear. It is reallly nifty science fiction based on parrallell universes.
I for one welcome our mustachioed dinosaur-riding overlords...
Hmm, I think it may well have been me, dave. But I have not yet read the novels. This shall be remedied!
I fear I shall have to detail my thoughts on two slits/multiple universes thoughts more thoroughly and meticulously at some stage. For me, once I understand what is going on, the maths doesn't really make a difference. Of course, there is a very small chance that the maths contradicts my thought-experiment model inside my head, but that doesn't really sound like me. It is possible that the difficulty lies at the meta-physical or theological level - ie does God play dice/ does the dice only stop rolling when it gets observed?
Having re-read your two slit experiment, I remembered what the point was. To detect which slit the electron passed through, one must in some way disrupt its natural flow. To simplify the problem even further to demonstrate where multiple universes comes in, lets remove the backstop such that we can't tell where the electrons end up. Then the only thing we are detecting or not detecting is which slit the electron(s) passes through. I assume its a 50-50 thing in this case. If it is detected there is a fork in the universes (universe A = electron passes through slit A, ditto B). If you don't detect them (and ensure conditions like a vaccuum etc. such that there is no way of telling), there is no fork in the universes at that point.
This fork (or lack thereof) accounts for the different set of universes when there is a backstop enabling you to see where the electrons end up.
By concentrating on the slits you are missing the point. I think.
An electron hits the detector, somewhere. According to your model which ascribes a special role to the detector/observer, the universe then 'forks'.
But, it does not fork into two universes: instead, it 'forks' into one of a very large number of universes, *the probability of which* is given by one of those two curves in Figure 2. How can the many universes theory give any insight into the shape of those curves? *Why* does splitting the set of possible universes into the set where the electron went through A and the set where the electron went through B give a different probability distribution than the case where we left the set of possible universes unsplit?
You have convinced me that the many universes hypothesis is not a priori stupid, nor morally repellent. What you have to do now is convince me that it is in any way useful. :P
Think about this. How can one electron change the universe if nobody or nothing ever gets to "observe" it. It's as if that electron never happened, as far as the universe is concerned. The question then becomes theological/metaphysical. What are those electrons doing when nobody is looking at them? To me the universe doesn't fork until somebody or something (even a backstop/ oxygen molecule or whatever) observes where it is. I assume the experiment will work differently again if the experiment is done in a medium in which you can see the precise path of the electrons. Then there will be a spectrum of universes the way I see it.
Pish tosh. How can you talk in a room where there is nobody listening? Easy, you can listen to yourself. (Or your invisible lepreachaun friends can listen to you. If you are like me, they do this all the time).
What makes an oxygen atom a better observer of an electron than the electron itself? After all, the wavefunction of the electron has a finite value at every location in the universe. I bet if you could talk to an electron, it would say:
"How can this abstraction called a Marcoparigi possibly influence the universe in any way? It is just a convenient mathemtical fiction for correlating the actions of a whole bunch of fundamental particles like me and Eric. How can it be self aware? You funny guy, pretending to be Marcoparigi. Who are you, really?"
Is it not, however, still a metaphysical question? Science is about the observable universe. Whatever the electron knows about itself is not accessible by science, but only by unverifiable thought experiments. Quantum effects are very, very strange if looked at from a point of view that science can find out *everything*. All we know is that the universe doesn't follow "impossible" paths. However there is an indescribable number of possible paths.
There is a limit to our self-awareness. An intelligent enough "other" can know more about us than we can know about ourselves.
Of course it is a metaphysical question.
How does the Many Universes model help us answer it?
In my Cosmology post I touched on why parallel universes model was satisfactory. The motivation in a multiverse model is to have the same laws in the Universe as we have in the universe. I guess my axioms would essentially be Spinoza's + my marconomic ones.
One other detail is that My multiverse theory invokes an axiom that assumes that the time and space available to hold the multiverse is infinite, and that all possible universes exist as matter/antimatter pairs moving away from each other at the speed of light. Whether the other universes *actually* exist is a moot point. They can at least be consistent alternative history/futures. The singularity of the big bang is the point in space time where the universes originate.
Another axiom is that it is at a point that something is observed where the universe forks based on the actual observation. An electron hitting an oxygen atom, for instance, is when the oxygen atom can *observe* the electron and the electron can observe the Oxygen atom and work out where it is. This is the point at which the universe it ends up in is "chosen" whether "God" chooses it to guide history the only way it can with natural proximal causes, or we end up in the universe that will last the longest, who knows.
Sure, from our previous experiments we have a probability graph of where the electron "could" go, but this probability graph is the same for all universes. The only thing different about the universes (due to this event) is where the electron ends up.
If you observe which slit the electron goes through, the universe forks into two single slit experiment universes.
The motivation in a multiverse model is to have the same laws in the Universe as we have in the universe.
This is a presumptuous motivation, and on our previous experience as a species, has a negligible chance of reflecting reality.
I still am absolutely at sea as to exactly what you think a many-universes model explains better than a common or garden single-universe model. And I still don't seem to have conveyed my point about different sorts of probability not reflected at all in a dichotomous 'forking universe' model. So perhaps it is not a point, but a blunt mood. I may try once more. Or I may give up!
You seem to be seduced by the joy of making up cosmologies out of whole cloth- which is attractive, I admit- but why matter/antimatter universe pairs emitting from a singularity that begs explanation? Why not giant turtles?
I remain,
Yrs Faithfully,
D. Clam, Esq.
You seem to be seduced by the joy of making up cosmologies out of whole cloth-
I guess I thought that was the whole point of metaphysics. Quantum effects bring up things we cannot observe and have trouble relating to. For those things we tend to make up things to account for those observation blind spots. One needs to fit that into a grand scheme, or cosmology, which is also made up. I just hope that my made up schemes are consistent with each other and with observations we can make.
which is attractive, I admit- but why matter/antimatter universe pairs emitting from a singularity that begs explanation?
Of course to answer one of my questions that begs explanation - If the universe could have just quantum tunnelled itself into existence, why can you not get a (quantum tunneled) universe within a universe that would make a supernova seem tame. Well, I know that matter/antimatter pairs can spontaneously appear within our universe, but would quickly be neutralised by the overwhelming amount of matter.
Why not giant turtles?
Well, because I don't believe in giant turtles.
I, for one believe that science has got about as far as it will as far as cosmology (or a grand theory of everything) goes. I even think science has overstepped the mark in things like evolution - twisting it into answering what ought to be theological or philosophical questions. I guess what it comes down to, is that quantum theory challenges the belief in a deterministic universe.
Post a Comment