QUANTUM LOOPINESS

Dick Pountain/14 December 2006/10:42/Idealog 149

The great physicist Richard Feynman once opened a seminar with this profound but amusing stand-up routine: "Einstein was a giant." <long pause> "His head was in the clouds but his feet were on the ground." <long pause> "But those of us who are not that tall have to choose!" <universal laughter>. Among his audience was another of my heroes, Carver A. Mead, who has most definitely chosen to keep his feet on the ground, while never quite losing sight of the clouds.

Mead, a former student of Feynman's, is now the Gordon and Betty Moore Professor of Engineering and Applied Science at the California Institute of Technology (and yes, that is indeed the same Gordon Moore of Moore's Law). But Mead himself has a better claim than anyone to be father of the electronics revolution: in 1968 he delivered a highly original lecture on the scaling properties of semiconductor transistors, answering a theoretical question raised by the same Gordon Moore, who was then working at Fairchild Semiconductor. Moore went on to found Intel Corporation and turn those scaling properties into the processors that drive our modern PCs.

It's almost impossible to imagine Great Britain ever breeding a Carver Mead. His father was an electrical engineer who worked on the great hydroelectric projects of the Roosevelt era: he grew up in a 'camp' by a dam in California's Sierra mountains, occupied entirely by electrical engineers and their families. By the time he was 12 electrons were as real to him as marbles, and he went straight from highschool to Caltech where he's worked now for 50-odd years. His "Introduction to VLSI systems" (written with Lynn Conway) is without any doubt The Most Important Book Of The Twentieth Century That No-one's Heard Of. In it he lays out the principles for making all the chips we currently use, plus a couple of generations that we haven't even started to make yet. 

In recent years Mead has turned his attention to the way biological systems process data, and how to emulate them in silicon. A friend of mine used to work next-door to him at Caltech, and I had the pleasure of visiting his lab and watching an analog artificial retina chip he was developing in action. It uncannily reproduced the behaviour of the human eye by saturating if it looked at the same spot for more than a second, so it had to be randomly twitched to keep on seeing.

More recently still Mead has focussed his own eyes back on the clouds. In his 2002 book "Collective Electrodynamics" he set out on an apparently quixotic quest to overthrow the Copenhagen Interpretation of quantum mechanics. That's the version put forward by Niels Bohr, the version everyone knows a little bit about, where particles can be in two places at once (but you can't know exactly where) and where the whole universe becomes rather a fuzzy and indeterminate place. Bohr's interpretation triumphed in the late 1920s against the alternative views of Einstein and Schrodinger who were appalled at the introduction of statistical uncertainty at such a fundamental level ("I cannot believe that God plays dice..."). 

Mead agrees with Einstein that statistics have no place at the very bottom level of physics: they don't reflect physical reality but merely our problems about knowing that reality. He bases this belief on the fact that several hugely important phenomena have been discovered since those great Einstein-Bohr debates (he helped discover some of them) which if known then would have altered the outcome. These new phenomena include superconductivity, lasers, the Quantum Hall effect and the Bose-Einstein condensate, and they all share one property - they involve the behaviour of *coherent* wave systems. Mead proposes that to understand the fundamental properties of matter you must start from the properties of coherent waves, the exact opposite direction from the Copenhagen approach which needs its statistical and probabilistic methods because it starts from incoherent phenomena.

Without Mead's towering reputation you might suspect him to be one of that Green Ink brigade who write to editors refuting relativity or proposing perpetual motion schemes. I'm predisposed to believe him not just for his reputation but because what he proposes is what I learned in chemistry back in the 1960s: I'm at home with the Schrodinger equation, and the notion that everything's made of waves frightens me no more than that everything's a wave and a particle at the same time.

 There's also a philosophical reason why I want to believe Mead, having to do with a subject I often bang on about here, namely excessive rationalism. The Copenhagen Interpretation introduces the observer's mind into physics, which is of course why a lot of people like it, because it allows any number of loopy sci-fi plots to masquerade as popular physics: the many-worlds hypothesis, teleportation, telepathy, you name it. That truth is that the universe would behave exactly as it does were we not here to see it (which looks increasingly likely by the way), and Bohr's introduction of the observer into quantum mechanics will some day be seen as a theoretical disaster. The ground isn't such a bad place to keep your feet on: it does after all yield the silicon from which the Pentium inside your PC is made.