From Quantum Reality by Nick Herbert
"The ontology of materialism rested upon the illusion that the kind of existence, the direct 'actuality' of the world around us, can be extrapolated into the atomic range. This extrapolation, however, is impossible . . . Atoms are not things." Werner Heisenberg
Quantum Reality #4: The many-worlds interpretation. This quantum reality, first dreamed up by Hugh Everett in 1957 while a Ph.D. candidate under John Wheeler, takes the quantum measurement problem seriously and solves it in a bold and flamboyant manner.
The measurement problem can be stated in many ways. Everett saw it like this: the orthodox ontology treats measurement as a special kind of interaction, yet we know that measurement interactions cannot really be special since M devices are no different from anything else in the world. How, then, asks Everett, can we strip the measurement act of its privileged status and achieve within physics that democracy of interactions which certainly prevails in nature?
Bohr, for instance, assigns special status to measuring devices, conferring on them a classical-style actuality not possessed by the atomic entities under their scrutiny. Von Neumann, on the other hand, does not consider M devices special: he describes them in terms of possibility waves just like atoms. However, the price von Neumann has to pay to purchase this equality of being is the necessary elevation of the measurement act to special status. Unlike any other interaction in nature, measurement has the power to collapse the wave function from many parallel possibilities (the premeasurement superposition of possibilities) to just one (the actual measurement result).
Following von Neumann's picture of quantum theory, Everett represents everything by proxy waves, but he leaves out the wave function collapse. When a quantum system encounters an M device set to measure a particular attribute, it splits as usual into many waveforms, each corresponding to a possible value of that attribute. What is new in Everett's model is that correlated to every one of these system wave functions is a different M-device waveform which records one of these attribute values. Thus if the measured attribute has five possible values, the quantum-entity-plus-measuring-device develops into five quantum systems, each with a different attribute value paired with five measuring devices each registering that value. Instead of collapsing from five possibilities to one actual outcome, the quantum system in Everett's interpretation realizes all five outcomes.
To account for the stubborn fact that no one has ever seen one M device turn into five, Everett makes a not-so-modest proposal. The apparatus actually does split into five different parts, says Everett, but each part occupies its own parallel universe. A human being--one of Everett's critics, for instance--dwells in just one of these universes (at a time) and cannot perceive the other four. Likewise the inhabitants of the other four universes are not aware of their parallel partners.
The "ordinariness" of quantum facts in spite of the real existence of multiple universes is accounted for in Everett's model by the fact that each human observer perceives only a single universe. We do not know why human perception is limited to such a small sector of the real world, but it seems to be an unavoidable fact. We are not directly aware of these alternate worlds, but our own universe would not be the same without them.
Everett's quantum theory without collapse describes the world as a continually proliferating jungle of conflicting possibilities, each isolated inside its own universe. In that world (which we might call super reality) one M device splits into five. However, humans do not happen to live in super reality but in the world of mere reality, where only one thing happens at a time. We can picture Everett's super reality as a continually branching tree of possibilities in which everything that can happen actually does happen. Each individual's experience (lived out in mere reality, not super reality) is a tiny portion of a single branch on that lush and perpetually flowering tree.
All interactions in Everett's super-real world are of the same kind: two systems come together, get correlated, then start to realize all their mutual possibilities. A measuring device is just like any other quantum entity except that its macroscopic attributes happen to be especially sensitive to some attribute (usually position) of an atomic entity with which it may become correlated. Lots of entities become correlated with photons, but few qualify as photon detectors because their visible attributes are not significantly changed by this photonic association. Our phosphor/screen combination is different: it prints a mark on a tape whenever it correlates with a photon's position attribute. In Everett's model, M devices are not essentially different from anything else except in certain unimportant details.
Everett's many-worlds interpretation of quantum theory, despite its extravagant assumption of numerous unobservable parallel worlds, is a favorite model of physicists because of all quantum realities it alone seems to solve the measurement problem with no arbitrary canonization of the process of measurement. In Everett's picture all measurement devices and measurement acts are fundamentally of the same nature as all other devices and acts. Strictly speaking, there are no "measurements" in the world, only correlations.
Einstein objected to suggestions of observer-created reality in quantum theory by saying that he could not imagine that a mouse could change the universe drastically simply by looking at it. Everett answers Einstein's objection by saying that the actual situation is quite the other way around. "It is not so much the system," Everett says, "which is affected by an observation, as the observer who becomes correlated to the system." The moral of Everett's tale is plain: if you don't want to spilt, stop looking at attribute-laden systems.
At a recent conference on the nature of quantum reality, Berkeley physicist Henry Stapp suggested an advantage that Everett's quantum reality confers on biological evolution and similar improbable but not impossible processes. Suppose, says Stapp, you could calculate the odds for life to begin on Earth and found them to be infintesimally small but not actually zero. In the conventional single-universe model of things, something with a very small probability is effectively impossible: it will never happen. However, in the Everett picture everything that can happen does happen. If life on Earth is possible at all, then it is inevitable--in some corner of super reality. In Everett's bountiful multiverse, every little "could be," no matter how improbable, gets its time to shine.