Quantum Russian Roulette...in France

There's something freeing about ridiculously speculative ideas. In sifting through the shards of crackpottery you get to let your mind wander to the furthest recesses of Meinong's Jungle. So let's have some fun.

A while back I mentioned that quantum mechanics is beset by questions of interpretation: the mathematical relations in it correspond well to experiments. Better than any other area of physics, in fact, but that doesn't mean we know what they're telling us. They seem to suggest that systems are in multiple states ("superposition") until a mysterious event ("measurement") forces them to choose just one. As Wikipedia can attest, there's no shortage of suggestions for what's really going on there. The so-called Copenhagen interpretation has historically had the most weight, so much so that its philosophical precepts are often seamlessly integrated into the introductory QM curriculum to the extent that undergrads don't even know their "objective" physics is coming with a helping of philosophy (hint, kids: it always does). The squares of coefficients in expansions of a wave function (the piece of math that's taken to be a complete description of a physical system) are to be taken as the probability of the system collapsing into a given possible state during measurement. Thus the notion of a collapse is built right into that interpretation.

The interpretation I want to focus on here is called the many worlds interpretation (MWI). This one rejects the notion of a collapse from many states to one. In the MWI, there's nothing magical about measuring something, no (meta)physical difference between that state before and after. Instead of things being in a superposition of every possible state and then dropping out of it, states always remain in superpositions. Sometimes this is stated as the universe "splitting" whenever a measurement occurs but this is sloppy language that implies the existence of multiple universes. The split is merely the smooth evolution of wave functions simultaneously down different streams of possibility. MWI only involves one universe but there are many stories playing out concurrently (many superposed worlds) in that universe. Schrödinger's cat is dead and alive, both before and after we peek at it. In peeking, our state (I see it alive vs. I see it dead) gets mixed up with the state of the cat and we ourselves are in a superposition. We exist in many different worlds but since these worlds or branches of our existence can't interact we're only aware of one of the possibilities.

Now that we've got that background down, we're ready for the good stuff. There's a thought experiment, perhaps most closely associated with physicist Max Tegmark, called quantum suicide. It posits that if the MWI is correct suicide is pretty much impossible. Tegmark envisions a (pretty sick) experimental apparatus in which a machine gun is aimed at an experimenter and hooked up to a tool that measures the spin of a certain particle along a certain axis when the gun's trigger is pulled. If the spin of the particle is found to be up, the gun fires a single bullet. If the spin is down, nothing happens. The experimenter lives in one possible outcome, he dies in the other. So what happens when we try it out? Tegmark argues that we have a superposition of the two possibilities: 1) spin-up, experimenter dead and 2) spin-down, experimenter alive. However, since presumably conscious experience ceases in the first possibility, one's conscious experience can't travel down that road. That doesn't mean that this possibility doesn't occur (both possibilities play out), it means that you can't experience it. This is intimately related to that continuity of consciousness notion I've mentioned before.

In practice, this would imply that you--the experimenter--can pull the trigger as many times as you want, the gun will never seem to fire and kill you. Your lack of dying will get exceedingly improbable after a while, effectively proving the MWI to be the correct interpretation of quantum mechanics. However, someone watching you do the experiment can see you die (their conscious experience is free to flow down that branch). So while you can't experience a world in which you're dead, others can. Which intuitively makes sense. I should note that not everyone agrees with Tegmark's conclusions but let's ignore those buzzkills for a bit.

For a moment, let's accept the (admittedly absurd) conceit that the new Large Hadron Collider in France could destroy the world. What might we expect to see if MWI is correct? I assume this would be a bit like a massive quantum suicide experiment--or rather, more like a worldwide cult suicide attempt. If operating the LHC inevitably leads to the destruction of the planet and the death of everyone on earth, none of our conscious experiences could ever travel down a branch of existence in which the machine works. In other words, the only observation we could ever have is the machine breaking. Every single time we try to use it. Is this what we've observed every time we try to use it?

Well, it's hard to be sure because the first time they tried taking it out for a real spin last year they broke it.

On 19 September, just nine days after protons were circulated in both directions of the €3bn LHC, an electrical connection between a dipole magnet (one of 1232 that bend the protons around the ring) and a neighbouring quadrupole magnet (one of 392 that focus the proton beam) failed during circuit tests in the last of the LHC’s eight sectors. At the time, a current of 8.7 kA was being pushed through superconducting cables the width of a stick of chewing gum to generate the enormous magnetic fields required to bend protons at high energies.

But they're going to start it up again next month. So we'll see. If it breaks again and again maybe we'll have some slim evidence that 1) the LHC is destroying us all in another branch of reality and 2) MWI is correct, in the same way that an endless series of clicks from Tegmark's quantum gun would indicate that MWI is correct and he can never lose a game of quantum Russian roulette. Or it could mean that the LHC is a multi-billion dollar piece of shit.

Of course, most of this is utter nonsense. But everyone's allowed to be a crackpot once in a while.

Bounded in an Inflatable Nutshell

Thirty years ago, big bang cosmology faced a problem. Actually, it faced a trinity of problems. It couldn't account for the uniformity of the observable universe, the near-perfect flatness of the universe, and the absence of certain particles that theoretically ought to exist. These three birds were taken down with one big stone when Alan Guth realized that a brief period of superluminal expansion early in the universe's history could account for all of these observations. The notebook pages on which he scribbled his historic calculations--with "Spectacular Realization!" written triumphantly at the top--are on display at Chicago's Adler Planetarium.

But inflation--now a cornerstone of modern cosmology and supported by additional observation evidence--comes with some heavy ontological baggage. It implies that what we know as the universe, a vast expanse bordered by distant quasars and gamma-ray bursts, is not the totality of existence. Instead, this observable universe is a merely a pocket of a much larger Universe, a bubble in an endlessly larger cosmic sea. And there are other bubbles in that sea, pockets of the Universe that have inflated or are inflating. Each one is a universe in its own right, some subject to different physical laws than our own observable universe. The bigger Universe (with a capital U) containing all the bubbles is likely infinite but one question worth exploring is how many types of other bubble universes exist beyond the horizons of our own. The arXiv blog has a post up detailing a new paper from one of the pioneers of inflationary cosmology, Andrei Linde, where he explores the question and comes up with an answer: 10¹⁰¹⁰⁷:

How many of these could we actually see? What's interesting here is that the properties of the observer become an important factor because of a limit to the amount of information that can be contained within any given volume of space, a number known as the Bekenstein limit, and by the limits of the human brain.

Linde and Vanchurin say that total amount of information that can be absorbed by one individual during a lifetime is about 10¹⁶ bits. So a typical human brain can have 10¹⁰¹⁶ configurations and so could never disintguish more than that number of different universes.

10¹⁰¹⁶ is a big number but it is dwarfed by the "humungous" 10¹⁰¹⁰⁷.

"We have found that the strongest limit on the number of different locally distinguishable geometries is determined mostly by our abilities to distinguish between different universes and to remember our results," say Linde and Vanchurin

This is a bit like the human eye distinguishing shades of a given color--how many shades are "really" there is less relevant than the number of shades the human eye/brain can distinguish. Two bubble universes might be slightly different in some physical sense but if you plop a human in first one then the other and he can't find a way to distinguish between them, then it doesn't really matter how different they are. That's why the number of possible types of universe they calculate is reduced to the vastly smaller (but still enormous) 10¹⁰¹⁶ number.

This reminder of the vastness of the inflationary Universe makes me think of an argument I once read that inserts anthropic reasoning into an inflationary universe and finds that something's got to give:

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 anthropic reasoning.

The argument is pretty simple. Start with the assumption that of all the sentient observers in all the multiverse (the collection of bubble universes in the sea-Universe), you're not particularly special. The majority of civilizations in the multiverse will be relatively small but imagine that a handful of "supercivilizations" exist that spread across a galaxy in their universe (and perhaps on to other galaxies) and pack individual star systems a billion times more densely than we do. Then there's a very good chance that the majority of sentient beings exist in such a supercivilization. To take a real-world example that might make this a little clearer: in the United States, only 0.4% of firms have more than 100 employees, yet employees of these big firms account for 63% of paid employees in U.S. firms. The reason, of course, is that even if there are lots and lots of little firms, the big firms are the ones that have lots of people. So if I picked a worker at random from the working population of the U.S., it would be a smart guess that he works for a big firm and not a small one (even though big firms are rarer than small ones).

The same logic applies here, except the numbers are significantly more exaggerated. Whereas all but about 2 in 5 U.S. private employees work for a large firm in our example, the authors in that paper calculate that "all but one individual in a hundred million belongs to a large civilization." So if you pick a sentient observer at random, it would be a very good bet that he belongs to one of those supercivilizations because that's where most of the sentient beings are. But we Earthicans are in the curious position of not finding ourselves in such a supercivilization. So is our starting assumption that we're not special (i.e. that if we pick someone at random out of the multiverse we're free to pick ourselves) wrong? Is inflation--with its untold numbers of other bubble universes--wrong? Is there some general principle that forbids or nearly rules out the development of supercivilizations?

Ultimately we don't know but the author does offer some suggestions. In addition to the ones I just mentioned, he suggests maybe we are part of a supercivilization but just don't know it (paging Agent Mulder!), maybe some sort of selection bias is going on here, or perhaps even "the idea of 'individual' will be different in the future. That last one is particularly interesting:

Perhaps civilizations more advanced than ours consist of only a single individual, or only a single individual per planet, in whatever sense of individual is necessary for anthropic reasoning. In that case, even though the civilization is very widespread, the number of individuals is small. A similar idea is that individuals of those advanced civilizations are so different from us that they cannot be considered part of the same reference class, and we should not reason as though we could have been one of them.

Does this remind anybody else of the Kurzweil predictions we talked about in What Lies Ahead? Particularly the bits like:

● Individual beings merge and separate constantly, making it impossible to determine how many “people” there are on Earth.
● This new plasticity of consciousness and ability for beings to join minds seriously alters the nature of self-identity.

Something to mull over.