During the 1920s and 30s a strange picture of reality emerged from quantum mechanics. A physical object acted like a wave until it was observed and then it suddenly "collapsed" into a particle. There were two horribly unacceptable things about this idea. The first was that the wave existed only as a wave of possibilities, encoding many possible states of reality all superimposed on each other: there was no particle in existence until the wave collapsed.
The second was a little more subtle. Von Neumann showed that the quantum state of the wave was, mathematically, inherited by the particle detector and not collapsed to a definite measurement until it was observed by a scientist. This "Von Neumann regress" meant that for a unique reality to emerge, a conscious observer was needed.
This "Copenhagen interpretation" of quantum mechanics was deeply rooted in the mathematics and so has remained the mainstream model ever since.
In 1935 the physicist Erwin Schrödinger famously highlighted its horrors in his thought-experiment involving a cat that was both alive and dead at the same time, until it was observed. But quantum mechanics worked and so his quibbles were not taken too seriously.
It was not until 1961 that Eugene Wigner highlighted another paradoxical consequence of all this, in a scenario that has come to be called Wigner's Friend. He pointed out that if his friend locked himself in his laboratory to perform Schrödinger's experiment, at the end he would know whether the cat was alive or dead but Wigner would not. To Wigner the friend would be described by a quantum superposition of two states, knowing-the-cat-is-alive and knowing-the-cat-is-dead. Only when he observed his friend telling him the outcome would his friend's state collapse to one or the other.
But of course, as far as the friend was concerned, he had been in that collapsed state ever since he opened the box. The two physicists disagreed flatly in between times, giving quite incompatible descriptions of the friend's quantum state in between his opening the box and telling Wigner.
But Von Neumann's regress doesn't stop there. One can take the dilemma a step further: to yet another colleague, walking along the corridor to come and see Wigner and find out about his other friend. To this colleague, both Wigner and the friend were now in a superposition of knowing-the-cat-lived and knowing-the-cat-died states until he reached Wigner's office and asked him the result. This colleague held yet a third view of the friend's reality, each scientist having observed, directly or indirectly, the friend's knowing-the-cat's-fate quantum state come into existence on different occasions.
And so on, in an infinite regress of observers and conflicting quantum realities. Once one accepts Schrödinger's logic as valid, the infinite regress becomes unavoidable. As quibbles go, Wigner's caused barely a ripple and nobody for a moment questioned the science which gave rise to it.
However, another such infinite regress of observers had previously received much shorter shrift from the physics community and the theory which embodied it had been resoundingly rejected.
J.W. Dunne was a sometime soldier, Fellow of the Royal Aeronautical Society and dreamer of apparently prophetic visions. A sober rationalist, he was deeply disturbed by these visions and sought to understand what was happening to him. His starting point was the nature of Time, pointing out the distinction between his fixed timeline in Einstein's four-dimensional spacetime and his experience of travelling along that line. His logic led him in his 1927 book An Experiment with Time to propose another time dimension which only the observer's mind inhabits, thus setting the observer outside the physical life being lived just as a scientist exists outside the experiment being observed.
But Dunne's logic also applied to his new observer in their own time dimension. The apparent steady pace of one's journey along this fifth dimension now also needed to be explained. The answer was as before, another new time dimension and a higher observer still to inhabit it. And so on. Thus a serial regress of time dimensions and observers emerged, continuing on to infinity. He called his dimensions Serial Time and his philosophy Serialism.
Most Philosophers and physicists were unimpressed with Serialism. A few accepted the first new level of time for the conscious observer to inhabit but the logic which led to the regress was universally rejected.
Bickering with the Society for Psychical Research over experimental support for dream precognition, along with the religious nature of his philosophical conclusions about the Infinite Mind at the end of the regress, led to further scepticism in a world where atheistic materialism had become the intellectual fashion.
Meanwhile the endless regress of so-called phase spaces, offering a private little world for each and every quantum wave in the universe, became clear and was happily accepted into the mainstream.
Dunne argued his case as best he could, following up in 1934 with a study of the implications of Serialism for modern physics, The Serial Universe. But he did not understand the mathematics of the quantum equations and Schrödinger had not yet produced his cat, it was a lost opportunity. Despite his pleas to be treated with the same seriousness as the relentless "phase" spaces, his infinite regress with its aroma of theology was not to be tolerated. A generation of materialist physicists put his books down and moved on.
By the time Wigner came along, Dunne had passed away and Serialism had long been forgotten by the physics community. Nobody noticed the close logical parallel between Wigner's serial regress of observers and Dunne's.
Meanwhile, in 1957 the paradoxes inherent in the Copenhagen model had led Everett to adopt (what DeWitt later called) his "many-worlds" interpretation: every possibility inherent in the wave actually happens in a universe somewhere, and its collapse under measurement is merely our discovery as to which of these universes we inhabit. When the box is opened, the universe splits in two, one with a live cat in it and the other with a dead one. Every new quantum wave continues to create new possibilities and its parent universe spawns a cascading multitude of new universes, one for each possibility, in a vastly-extending exponential regress. Everett was a better mathematician than Dunne and his presentation followed the quantum equations as closely as the Copenhagen interpretation did. Moreover, it was free of overt metaphysics. Perhaps too the neat sidestepping around Wigner's friend offered the lesser of two percieved evils. There can be little other explanation as to why this uttely bizarre regress came to be taken more seriously than Dunne's much gentler one. It was decades before the flaws in Everett's logic were widely recognised by the mainstream and his untenable concoction seen for what it was. (There may be other kinds of parallel universe, but that is not the point here).
Everett's folly demonstrates graphically how it is not necessary to accept infinite regressions in the mathematics as physical realities in themselves. Nevertheless it is hypocritical to accept Everett's premise despite their being inherent in it, while also rejecting Dunne's precisely because they are inherent in it. Let us at least be consistent in our approach. One might add the sum-over-histories, renormalisation problem, parallel inflationary universes and the sequence of projective spaces in certain string theories to the infinite regresses readily accepted by the modern physicist. Besides its nose-curling metaphysical glaze, Serialism had also suffered from being the first and hence unfamiliar. Although its mathematics is of schoolkid level it is not wrong. And it has one big advantage over Wigner's, in that all the regresses go on inside a single mind, one does not need to invoke an an endless queue of observers cluttering the corridors and Schrödinger's Cat sleeps on in its box, dreaming its own dreams while it may.
Nobody has yet gone back and said, "Well, if we accept the Copenhagen interpretation as presented by Schrödinger then, as Wigner and others has shown, infinite regressions of consciousness come with it, so let's take another look at Dunne's." Maybe it's time somebody did.
Updated 19 Oct 2017