Chad Orzel, Contributor
Quantum physics is weird. That’s probably the most famous thing about the theory— the one thing everybody knows is that “quantum” means “radically different from everyday physics.” This has both good and bad consequences: on the bad side, unscrupulous individuals can exploit this to pass off “quantum” scams; on the good side, it means there’s a market for books explaining how quantum physics manifests in everyday life. (OK, that’s mostly a benefit to me personally…)
As I’ve argued before, though, when you look at things the right way, the weirdest thing about quantum physics is that we think quantum physics is weird. That is, the quantum rules governing the behavior and interactions of fundamental particles are the only rules in physics, applying to everyday-size objects as well as microscopic ones. It’s not possible to make a sharp division between scales with quantum rules for small things and classical ones for big things— that’s the real point of the Schrödinger cat thought experiment. The world is quantum, all the way up.
But, of course, our everyday experience doesn’t look quantum, but instead seems to be well described by classical laws of physics. This is a fabulous example of the “more is different” phenomenon where a simple set of fundamental rules applied to an enormously complicated arrangement of objects give rise to a difficult-to-predict set of “emergent” rules governing collective behavior of larger objects. The classical rules aren’t a different set of physical laws, they’re the consequence of the quantum rules acting on enormous numbers of objects.
Of course, the space between these two sets of rules is a fertile area for research exploring the details of how this process happens. Most of this is essentially philosophizing— thinking up plausible stories about how classical reality emerges from quantum rules, and evaluating them on the basis of logical and mathematical consistency. As experimental technology improves, though, every now and then pieces of these big fuzzy philosophical stories become testable in a concrete way, and that’s tremendously exciting.
The proximate cause of me writing about this is a set of recent experiments described in an excellent piece by Philip Ball that get at one of the central predictions of “Quantum Darwinism.” This approach is largely promoted by Wojciech Zurek of Los Alamos National Lab, who has long been associated with the idea of decoherence (picking up on work by Heinz-Dieter Zeh), in which environmental interactions are responsible for destroying quantum superpositions leading to the single observed outcomes of classical measurements. Quantum Darwinism is an outgrowth of this, which springs from the recognition that the most essential characteristic of classical reality is its objectivity: everybody agrees about the state of a classical object.
Zurek notes that in order for multiple observers to agree about the state of a quantum object, they each must be getting some information about its state from its larger environment, which is built up of a lot of other quantum objects. Interactions between the quantum object and each of the individual components of the quantum environment lead to their states becoming entangled: the state of the quantum object of interest and the state of a particular component of the environment are correlated in such a way that measuring one gives you information about the other. That entanglement is the source of the information that individual observers are using to determine the state of the quantum object of interest.
In order for an object to take on the sort of objective state that we associate with classical reality, all of those observers must agree as to what that state is. But each of those observers must be getting their information from a different subset of the environment, more or less by definition. Which means that the various components of the quantum environment must all contain copies of the same information about the quantum state of the quantum object of interest. And that’s what makes “Quantum Darwinism” Darwinian: the only states for which there can be a consensus are ones that can survive having multiple copies of their state information made through interactions with the quantum environment. When you go through this process, you can show mathematically that the sorts of states that survive are very classical ones: the outcome of the measurement is either 1 or 0, not a superposition of 1 and 0 at the same time.
It’s a lovely story, and lends itself to a catchy name, but it probably also seems like something that would be impossible to test. But there is, in fact, a clear prediction from this that can be teased out: in the quantum-Darwinian scenario, the environment stores many copies of the state information that can be read out by many different observers, so there’s little marginal benefit to any observer from measuring more of those copies. That is, measuring one copy tells you something you didn’t already know and measuring two is good because it confirms your first measurement, but the twelfth measurement of the same thing isn’t really increasing your knowledge very much.
That may not seem terribly useful, given that any macroscopic experiment is necessarily embedded in an environment containing uncountably vast numbers of quantum particles, but recent developments in quantum technology have made this experimentally accessible. That is, physicists have learned clever tricks for controlling the interactions of quantum systems in ways that allow the isolation and investigation of manageable numbers of quantum particles. That’s the key to the three experiments described in Ball’s article: two of them (one, two) use photons that interact with each other in ways that entangle their states, while the third uses atoms of a rare isotope of carbon interacting with a nitrogen atom that’s taken the place of one of the carbon atoms in a diamond.
In all three cases, they can arrange for the system that’s designated as the quantum object of interest to be in a particular indeterminate quantum state, allow it to interact with those designated as components of the environment, and then measure the states of the environment to extract information about the state of the object of interest. When they do that, they find exactly the pattern quantum Darwinism predicts: measuring one component of the environment leads to a big increase in your knowledge about the object of interest, but measuring a second one doesn’t get you that much more. And the final state of the object of interest is a classical one, as it must be.
These are, of course, preliminary and very small experiments— for the nitrogen-in-diamond one, the “environment” consisted of only three atoms of the right carbon isotope. Really nailing down what’s going on will require lots more tests, involving much bigger environments. When you think about it, though, it’s absolutely mind-blowing that we’re able to do this at all— the level of control needed to make these tiny artificial “environments” would’ve been unthinkable back in the 1970’s and 1980’s when Zeh and Zurek started talking about this stuff. Today, we’ve got impressively clean tests, and good prospects for scaling up the experiments to really look at how our classical world emerges from its quantum environment.