New paper by Deutsch on quantum teleportation
Wed, 2 Jun 1999 10:46:54 -0700

David Deutsch is one of the pioneers of quantum computing and a staunch advocate of the no-collapse "many worlds" interpretation of QM. He has a new paper at (PDF format) which re-interprets some famous quantum experiments. He argues that quantum information is essentially local, contradicting the accepted interpretation:

> All information in quantum systems is, notwithstanding Bell's theorem,
> localised. Measuring or otherwise interacting with a quantum system S
> has no effect on distant systems from which S is dynamically isolated,
> even if they are entangled with S. Using the Heisenberg picture to analyse
> quantum information processing makes this locality explicit, and reveals
> that under some circumstances (in particular, in Einstein-Podolski-Rosen
> experiments and in quantum teleportation) quantum information is
> transmitted through "classical" (i.e. decoherent) information channels.

In the early days of QM, two separate mathematical formulations were developed, by Heisenberg and Schrodinger. They were soon shown to be mathematically equivalent, and Schrodinger's was simpler to work with, so that is the one which is usually used today. However Deutsch goes back to the Heisenberg formulation in this paper and he shows that this method allows us to localize the information flows in quantum experiments in a way which is very difficult to do in the conventional Schrodinger picture. The result is a completely different understanding of the EPR paradox and of quantum teleportation.

Quantum teleportation is a widely misunderstood (and IMO misnamed) phenomenon which we have discussed here in the past. It is conventionally interpreted as providing instantaneous transmission of quantum information. What is sometimes overlooked is that for the QT to work, classical information must be sent about the state of the system to be "teleported". This classical information is then combined with the supposedly "magically" transported quantum information at the destination, to allow an exact re-creation of the quantum state of the original system.

In Deutsch's paper, he rejects the idea that any sort of instantaneous transmission of quantum information occurs. His careful analysis shows exactly where and how the quantum information transmission is done. Amazingly, it occurs in a sort of "hidden channel" associated with the *classical* information transfer.

As part of the quantum teleportation process, ordinary measurements have to be made on the quantum system and the results sent via a normal, classical channel to the destination. These classical results of the quantum measurements, when they are transmitted from the source to the destination system, have associated with them what Deutsch calls "locally inaccessible information". This is a new type of quantum information which can't be learned by local measurements on the particles involved (Deutsch draws an analogy to a message encrypted with a one-time-pad, where the information content cannot be revealed by any local measurement on the message by itself).

This type of information is furthermore invulnerable to quantum decoherence and so can be carried by classical means. Apparently, even if you wrote the results of the measurements on paper and sent them by postal mail to the destination, the letter would carry along with it this hidden quantum information which would then arrive at the destination along with the classical message. (This is my interpretation.)

Deutsch summarizes,

> Given that quantum theory is entirely local when expressed in the
> Heisenberg picture, but appears nonlocal in the Schrodinger picture,
> and given that the two pictures are mathematically equivalent, are we
> therefore still free to believe that quantum theory (and the physical
> reality it describes) is nonlocal?
> We are not - just as we should not be free to describe a theory as
> "complex" if it had both a simple version and a mathematically equivalent
> complex version. The point is that a "local" theory is defined as one
> for which there exists a formulation satisfying the locality conditions
> that we stated at the end of Section 1 (and a local reality is defined
> as one that is fully described by such a theory).

And since the Heisenberg interpretation provides such an interpretation, quantum mechanics should properly be viewed as local.

This paper is part of what I see as an increasing trend to demystify quantum theory. There are several approaches, including the transactional interpretation, ideas based on time symmetry, no-collapse approaches, which are moving us away from the spooky action at a distance which was so upsetting to early physicists like Einstein. It may well be that in another few decades we will have a quantum theory which would be much more philosophically acceptable to the early pioneers of QM.