Interrogating Many Worlds

The "many worlds" interpretation of quantum mechanics is an idea that gets thrown around a lot in physics. It also has made it into models of magick and how magick works, such as Lon DuQuette's famous statement "it's all in your head, you just have no idea how big your head is." As I pointed out a while back, DuQuette is not putting forth a purely psychological model of magick with that statement. Instead, he is proposing that practical magick works by means of a sort of "reality selection."

The many worlds interpretation of quantum physics is a way of modeling quantum interactions without employing the "wavefunction collapse" of the Copenhagen interpretation. In Copenhagen, until we measure a quantum event it exists in a superposition that includes all possible values. This superposition is shown to be wavelike by simple experiments such as the slit experiment. When the superposition is measured, this wavelike superposition "collapses" into a single value.

In many worlds, on the other hand, every possible value in the superposition happens in its own "world," or really "universe." By measuring a quantum event, we select which of those worlds we wind up inhabiting. But they all are out there - a practically limitless number of universes, each of which "cleaved off" from the others based on single quantum interactions. Many worlds resolves the issue of what wavefunction collapse "really is" by asserting that it doesn't exist. It only looks like it does because we only perceive the universe we end up in.

DuQuette's idea exploits this concept by viewing magick as a way to "direct" which universe you wind up in, or maybe "navigate" through the various possibilities. He explained in one of his talks that in order to make something happen, you transform yourself into the sort of person that those things happen to. This frames his model of practical magick as a sort of magnetizing influence that results in you experiencing the world that you want to experience - or in Thelemic terms, the world that is in harmony with the change you are casting for.

My model of magick uses a single-world interpretation simply because it is easier to represent "near misses" in single-world where something more like a wavefunction collapse happens. This is more conceptual than mathematical, since many worlds returns the same mathematical results as a single-world interpretation. From a magical perspective, though, it's problematic. Let's say that you cast for a thousand dollars and get a hundred dollars. How does the universe in which you would have gotten the full thousand relate to the universe in which you only got the hundred?

In single-world it's easy - your spell produced a probability shift in the direction that you wanted, in that it got you some money. The shift you created was only strong enough to overcome the odds of you getting a hundred, not a thousand, so that's what you got. In many worlds, though, you effectively have to work out a coordinate system onto which you map the "hundred-universe," the "thousand-universe," and everything in between. Then you have to work out "how close" the "hundred-universe" is to the "thousand-universe" to work out the probability variance between the two.

It does resolve mathematically, but conceptually it's a mess. So I lay out my models in "single-world" because they make more intuitive sense that way. With two interpretations that explain the same results, it's always best to go with the simpler of the two, especially if you are trying to explain something in a coherent fashion.

So that's a long introduction to this article that I came across recently, which is a couple of years old but is also one of the best expositions I've seen of the problems with the many worlds interpretation.

After the Danish physicist Niels Bohr articulated and refined what became known as the Copenhagen interpretation — widely regarded as the orthodox view of quantum mechanics — in the 1930s and ’40s, it seemed that the central problem for quantum mechanics was the mysterious rupture created by observation or measurement, which was packaged up into the rubric of “collapse of the wave function.”

The wave function is a mathematical expression that defines all possible observable states of a quantum system, such as the various possible locations of a particle. Up until a measurement is made and the wave function collapses (whatever that means), there is no reason to attribute any greater a degree of reality to any of the possible states than to any other. It’s not that the quantum system is actually in one or other of these states but we don’t know which; we can confidently say that it is not in any one of these states, but is properly described by the wave function itself, which in some sense “permits” them all as observational outcomes. Where, then, do they all go, bar one, when the wave function collapses?

At first glance, the many-worlds interpretation looks like a delightfully simple answer to that mysterious vanishing act. It says that none of the states vanishes at all, except to our perception. It says, in essence, let’s just do away with wave function collapse altogether. This solution was proposed by the young physicist Hugh Everett III in his 1957 doctoral thesis at Princeton, where he was supervised by John Wheeler. It purported to solve the “measurement problem” using only what we know already: that quantum mechanics works.

But Bohr and colleagues didn’t bring wave function collapse into the picture just to make things difficult. They did it because that’s what seems to happen. When we make a measurement, we really do get just one result out of the many that quantum mechanics offers. Wave function collapse seemed to be demanded in order to connect quantum theory to reality. So what Everett was saying was that it’s our concept of reality that’s at fault. We only think that there’s a single outcome of a measurement. But in fact all of them occur. We only see one of those realities, but the others have a separate physical existence too.

In effect, this implies that the entire universe is described by a gigantic wave function that contains within it all possible realities. This “universal wave function,” as Everett called it in his thesis, begins as a combination, or superposition, of all possible states of its constituent particles. As it evolves, some of these superpositions break down, making certain realities distinct and isolated from one another. In this sense, worlds are not exactly “created” by measurements; they are just separated. This is why we shouldn’t, strictly speaking, talk of the “splitting” of worlds (even though Everett did), as though two have been produced from one. Rather, we should speak of the unraveling of two realities that were previously just possible futures of a single reality.

When Everett presented his thesis, and at the same time published the idea in a respected physics journal, it was largely ignored. It wasn’t until 1970 that people began to take notice, after an exposition on the idea was presented in the widely read magazine Physics Today by the American physicist Bryce DeWitt. This scrutiny forced the question that Everett’s thesis had somewhat skated over. If all the possible outcomes of a quantum measurement have a real existence, where are they, and why do we see (or think we see) only one? This is where the many worlds come in. DeWitt argued that the alternative outcomes of the measurement must exist in a parallel reality: another world. You measure the path of an electron, and in this world it seems to go this way, but in another world it went that way.

That requires a parallel, identical apparatus for the electron to traverse. More, it requires a parallel you to observe it — for only through the act of measurement does the superposition of states seem to “collapse.” Once begun, this process of duplication seems to have no end: you have to erect an entire parallel universe around that one electron, identical in all respects except where the electron went. You avoid the complication of wave function collapse, but at the expense of making another universe. The theory doesn’t exactly predict the other universe in the way that scientific theories usually make predictions. It’s just a deduction from the hypothesis that the other electron path is real too.

This picture gets really extravagant when you appreciate what a measurement is. In one view, any interaction between one quantum entity and another — a photon of light bouncing off an atom — can produce alternative outcomes, and so demands parallel universes. As DeWitt put it, “Every quantum transition taking place on every star, in every galaxy, in every remote corner of the universe is splitting our local world on earth into myriads of copies.” In this “multiverse,” says the physicist and many-worlds proponent Max Tegmark, “all possible states exist at every instant” — meaning, at least in the popular view, that everything that is physically possible is (or will be) realized in one of the parallel universes.

That long excerpt is just a selection. The article is long, but it explains the issues in plain language and manages to cover a lot of bases. As I mentioned, it is a couple of years old and doesn't address this recent finding, which at the very least poses a possible experimental challenge to many worlds.

In the simulation, a piece of information is simulated to be sent backwards in time. That information is then damaged. However, when the information returns to the “present” it is largely unaltered and, counter-intuitively, with travels further into the past the final piece of information returns with less damage. Such an effect only works in quantum mechanics, in simulations conducted via quantum computers, because time travel is not yet possible.

“On a quantum computer, there is no problem simulating opposite-in-time evolution, or simulating running a process backwards into the past,” said Nikolai Sinitsyn, a theoretical physicist at Los Alamos National Laboratory, in a statement.“We can actually see what happens with a complex quantum world if we travel back in time, add small damage, and return. We found that our world survives, which means there's no butterfly effect in quantum mechanics.”

To test the butterfly effect, the researchers used an IBM-Q quantum processor with quantum gates, which simulate forwards and backwards cause and effect. Standard computers and processors use ‘bits’ in their chips, which exist in two positions - either ‘on’ or ‘off’ – which is the makeup of binary.

Quantum computers use ‘qubits’ rather than bits, which can be both be on and off simultaneously, as well as somewhere in between. In the simulation, a person sends a qubit back in time. An intruder in the past measures the qubit, which disturbs it and changes its quantum correlations. This is because even slight contact between an atom exhibiting quantum behaviour and another atom will immediately move the atom out of its quantum state.

The simulation is then run forward, to bring the qubit to the present day. It was found that, rather than the information being unrecoverable due to extrapolation of the small inciting incident – the act of stepping on a butterfly in the common metaphor – it was protected from minor tampering. “We found that the notion of chaos in classical physics and in quantum mechanics must be understood differently,” Sinitsyn said.

This result is a serious challenge to many worlds, or at the very least a serious challenge to the idea that every quantum event creates a new universe in which it occurs. In other words, there may be "many worlds" that bifurcate at particular points, but there must also be some mechanism that allows the "split worlds" to come back together or not bifurcate in the first place for quantum interactions which as "small" - at least for some definitions of "small." How that would work conceptually or heuristically is anybody's guess at this point.

So basically, this implies to me that it's not "all in your head" no matter how big your head is. Magick, then, would be best interpreted as shifting probability in a single world rather than selecting one of many possible parallel worlds. Or at least that's what the data suggests up to this point.