Decoherence between chains of questions (or histories, as they are called) is a mathematical condition that ensures that the answers to questions will obey the classical logic of probability theory, rather than quantum logic. The general idea is illustrated by the ERP effect referred to earlier. Here the positions of the emitted particles are affected by minute interactions with the rest of the universe, so that they become essentially classical observables with no holistic connection between them: they decohere. Their spins, however, because of their essentially non-classical nature, are very little affected by external interactions and so exhibit a holism: they cohere.
On the quantum logic approach, the questions (or observables) come first and the state (which ultimately is the state of the universe) then becomes a global determinant of the probabilities of various answers to these questions.
The consequences of this is that physics becomes fundamentally non-local. We do not start off assuming that the universe is composed of independent atoms. So global effects do not require special mechanisms to make them happen; rather, special mechanisms are required to break things down to the point where physics becomes local.
There is, alas, one basic problem with this approach: we can't do it! Current decoherence schemes of describing quantum theory have clarified a great deal, but they are still ultimately unsatisfactory in that they prescribe a god-given collection of decohering histories. As Hartle and Gellmann explicitly stress, however, the universe is not observed from the outside. Rather, the questions that consitute the physically realised decohering histories after any point in time are determined by the past history up to that point. The universe, in other words, observes itself, and we don't know how to handle the fact. (Is it an accident that this is the characteristic feature of mind?) What is needed is a method for formulating the way in which the collections of meaningful quantum questions in each situation are engendered not from the outside, but from within.
In order to explain how this non-Newtonian view can shed light on the natrue of mind it will be useful to refer to the ``Hard problem'' of Chalmers (1995): the problem of explaining why, apparently, brain processes give rise to an experience; why there is such a thing as the view from inside such processes (what it is like to experience them) as well as the scientific observation from outside. For the dualist there is no problem at this point, because the dualist can postulate a separate soul that observes the brain processes from the inside and thereby generates the experience. The problem for the dualist comes later, in making sense of a dual world.
For the non-dualist, things are not so simple. It might be thought that one could resort to supposing that some special part of the brain, the me-brain, observes other parts of the brain and thereby experiences what is going on there. But if one postulates this, then one must ask the same question as before of the me-brain and thereby fall into an infinite regress. This sort of explanation is still rooted in dualistic thinking, in that it separates the experiencing entity from the information processing entity, even though these are now at the same level of brain function.
With Chalmers, I would hold that the basic experiential aspects of consciousness cannot be explained in terms of existing physical categories, but require the addition of a fundamentally new area of science, associated with but not reducible to existing physics, corresponding to experience. Chalmers uses the analogy of electric charge, which is associated with mass in being located on massive particles, but which is none the less a quite distinct quality. If one has a cloud of electrons, the total mass and the total charge are (almost) exactly proportional to each other quantitatively, but they remain qualitatively distinct. I am claiming that consciousness is a quality that is carried by brain processes, as charge is carried by massive particles, and which therefore inherits some of the structure of the brain processes, but which is distinct from them. Because this experiencing is carried by some brain processes we can deduce logical, structural things about experience from observations of brain processes; but the fundamental nature of experience remains separate from the brain processes.
It is here that non-locality enters. Since the structure of experience has to reflect the structure of the processes that carry it, and since mind is non-local, the structure that carries mind has to be non-local. That means, straight away, that it is futile to look for mind in generalisations of charge at the particle level, as has been attempted in many speculative writings that try to identify consciousness with fundamental particle properties such as muonness. The carriers have to be global entities, and at present the only candidates for these are the quantum states that can be defined in the quantum logic (that is, the non-Newtonian) approach to physics. Note that if we restrict ourselves to a narrow wave-mechanics view of quantum theory then the quantum states are still so closely tied to space and time that it is hard to see their non-locality. The quantum logic approach, on the other hand, takes this non-locality as its starting point.
The first step in the direction of making a global quantum state the carrier of consciousness was taken by Ian Marshall in proposing that global Bose condensations might underly consciousness. But such a move assumes that the only quantum states that are of relevance to consciousness are those characterised by large-scale physical order. These are the only ones that might manifest interesting physical properties analogous to superconductivity. On the other hand, there is no reason why we need be bound by these physical considerations. Since we do not yet know what the new science of consciousness is going to be like, in our present state of ignorance there is no obvious limit to the sort of quantum states that might be significant to experience.
A key aspect of the sort of non-Newtonian approach that I have been refering to is that there is a whole range of possible quantum states that could potentially have a role in the universe but which in our present theory have no interpretation. The total quantum state of the human brain, in all its complexity, is potentially the ground within which mind can operate, is potentially the carrier of mind. On the crypto-Newtonian view of current physics, however, most such states would simply be regarded as arrays of apparently meaningless patterns of phases distributed across the particles of the brain, and any possible information that they contain would be dismissed as meaningless. Yet any discrimination of such states that can be formulated mathematically (provided that it is reasonably stable: technically, provided that it is represented by an global operator that almost commutes with the Hamiltonian) can define a possible quantum question (observable).
I suggest that things would be different if we were to turn round our whole pattern of description. First we need to turn round physics, so that we could see the local Newtonian picture as a specially disintegrated case of the fundamentally global reality. Within this more general non-Newtonian view there would be a natural place for mind at the point where the current effective state of the person, and the universe, as it is given by our past history, determines which chains of future questions are to make up reality. This will involve both those states that are now regarded as meaningful, including those considered by Marshall and Penrose, and also those others, dismissed as meaningless by Newtonian physics but having an internal meaning to consciousness. Second, we need to turn round our whole approach by putting mind first. We would be a position to understand how it was that mind could actually do something in the cosmos; not by acting as a separate force in addition to the Hamiltonian, nor by determining directly the answers that are obtained to quantum questions (thought that may happen indirectly) but by determining precisely which decohering histories of questions are realised in the process of self-observation that is embodied in consciousness.
Our much prized capacity for free will is not exhibited at all when we make a choice among totally defined alternatives with all the available information clearly specified. This is something that a well-constructed adaptive learning programme on a computer could do. Real human creativity is rather exercised in our ability upset the board; to redefine the alternatives; to set up new and unforseen connections. It is this that is the essence of quantum logic, where the range of possibilities is not fixed in advance. What I call `choosing' whether to eat quiche or nut roast for dinner is in reality only my post hoc emotional reactions to a largely non-conscious decision process. A true exercise of free will would be to respond by cooking a nut quiche.
In this picture the action of mind is not restricted to the brain, but extends to the whole system: a ``system'' meaning everything that is in interaction at a given moment. Mind breaks out of the skull.