Quantum Evolution, by Johnjoe McFadden

Johnjoe McFadden has constructed a theory that combines quantum mechanics and evolution, sort of an end-run connecting the two areas of science I would have thought least likely to be directly connected. His basic premise is that quantum mechanics has a direct, significant effect on living organisms by providing a mechanism for “directed evolution”. Normally, when you think of evolution, you think of completely random, blind mutations causing all kinds of diverse changes in the organism. A select few will happen to help the organism survive a bit more, while the vast majority will be detrimental and cause the organism to become dinner for someone else. Those mutations that happen to cause advantages will naturally stick around, since the organisms carrying them will outlive all other variants of the species. At the heart of evolution is the completely random, unavoidable, defects that appear in the genome of the organism — there’s no way for the organism to intentionally cause mutations that introduce beneficial changes. Or is there?

He cites some experimental work here and there that seems to suggest that bacteria can somehow trigger a mutation that works in their favor. For example, bacteria that can’t digest lactose, but are nevertheless fed only lactose and deprived of their natural food, will very quickly (suspiciously quickly) cough up the right mutation to change their digestive system to be able to digest lactose. Of course it could happen by chance, but the bacteria are somehow able to spur just the right mutation much more quickly than you’d expect by pure chance. Now I don’t really know how true this is — perhaps the evidence for “directed mutation” is not as strong as he suggests. At best it seems to be a raging debate within biology as to whether certain experiments indicate “directed mutations”, or are just being performed improperly.

Perhaps directed mutations aren’t possible, but lactose-eating bacteria aren’t the only suspicious evolutionary juicers he brings up. He also refers to perhaps the biggest gaping hole in evolutionary theory remaining — the origin of life. Just exactly what the chain of evolutionary steps between inert organic chemicals and the simplest unicellular organisms isn’t understood, and may never be understood well (given that we can’t go back in time to watch the process). Dr. McFadden argues that the likelihood of DNA, proteins, and most importantly molecules that can reproduce themselves evolving purely through random chance is quite low, and needs a kick to make it plausible. That kick also happens to be from quantum mechanics.

Johnjoe suggests that quantum mechanics gives the extra bit of push to get unlikely events in evolution over the wall separating completely improbable and “just probable enough to have happened on Earth to cause life”. His argument hinges on everyone’s favorite magic wand of quantum mechanics, the nature of an “observation”. Imagine a tide pool on ancient earth, where a little bit of brown glop over in the corner contains some molecules that are on their way to becoming self-replicating. (The general feeling is that the very first step towards a living organism must have been a single molecule that is capable of making copies of itself, of “self-replicating”.) If you picture small bits of organic molecules coming together randomly to happen to make just the right molecule that can copy itself, you’d probably wonder just how unlikely that is to happen. Nobody’s managed to make a self-replicating molecule in the lab, in the several decades since this research area bloomed. What chance is there that it happened completely by chance?

McFadden argues that the odds were helped along by quantum mechanics — he suggests that the initial fumbling around in ancient tidepools to generate the first self-replicating molecule was quantum-mechanical, so the candidate molecules existed as superpositions of many different configurations simultaneously. It’s as if the first molecule could try out all the possible building blocks (a carbon over here, a methyl group or two down there) at once, most of which didn’t impart the power of self-replication to the molecule. But at some point, one of those quantum states did consist of a self-replicator, and by definition, once it popped into existence it started to make copies of itself. The copying took off, feeding back on itself to generate lots more copies of this molecule, as each new copy can start making copies itself. At some point the copying process somehow causes an “observation” of the quantum system, and so the molecule pops out of its quantum superposition and collapses into the self-replicator state. In a nutshell, the original molecule uses quantum mechanical superposition to try on a gigantic number of possible designs, one of which will happen to be self-replicating. And that self-replicating version of the molecule serves as the observer to collapse the quantum wavefunction.

It’s a clever idea — if you picture traditional classical physics, it would take forever for just the right atoms and molecules to come together by pure chance to become self-copying. It’s just too rare to come across a molecule that can copy itself, that even the billions of years the earth has been around is too short a time for this to happen by chance. But evoking quantum mechanics, the molecule can try out all kinds of different arrangements simultaneously, and the one variety that can self-replicate will pop the molecule out of the quantum superposition automatically. This is reminiscent of how quantum computers will (eventually) be used — they are designed to farm out parallel computation to various quantum states of the computer, all of which can exist simultaneously in a quantum superposition.

Hell, why didn’t I think of that? Because I admit I don’t understand quantum mechanics, and I’m not the only one. No one ever completely explained just exactly what a “quantum observation” is to me, and who gets to do the observing. There was a lot of talk of electrons hanging out in the bottom of quantum wells, or of hydrogen atoms hanging out in otherwise empty space, or other very simple, idealized situations. In the real world, though, you’ve got all kinds of things interacting all the time — exactly who gets to be in a quantum state, and who gets to “observe” it to collapse it out of that quantum state? From what I can tell, it’s still an open question in physics research, and so I’d say it’s a bit premature to insert quantum observation into your theory. If you can’t exactly explain what the proto-molecule was doing in a quantum superposition, and how the self-replication observed its way out of the superposition, then this is all pretty much science fiction.

To be fair, McFadden does try to explain this, and maybe I’m just not getting it. It seems to have something to do with the sheer volume of copies the self-replicator will make — if it makes enough copies, the big ol’ wad of copies will get so big as to qualify for existence in the classical domain. And when things get classical, that’s when wavefunctions collapse. It seems to me, though, that my lack of understanding is as much from him abusing the ideas of quantum mechanics as it is from me not doing my quantum mechanics homework back in school. Nobody really understands well the interface between quantum and classical physics in the sort of “mesoscopic” world inhabited by biochemistry, so he’s probably a good 50 years to early to be traipsing around in this field of research.

Things get reeeallllyy wacky at the end of the book, when McFadden turns his attention to the brain, specifically to consciousness. Brain cells are electrical — all those little ions shuffling back and forth across the cell membrane cause little electromagnetic fields in the space around the cells. You can even pick up these EM fields from the outside, to give a rough idea of what’s going on inside the brain — that’s what brain scanners like EEG and MEG are doing. While most cell-to-cell communication uses action potentials and neurotransmitters, the overall EM field generated by neighboring cells probably has some effect on the brain. Any given neuron is mainly affected by the handful of other cells that make direct connections with it, and so what that neuron does (either fire an action potential or not fire) at any given moment is basically a result of what the cells connecting to it are telling it. But the cell might be nudged a little bit by the overall EM field caused by the collective action of thousands of neurons surrounding it — maybe it’s pushed a little closer towards an action potential because of the net field outside the cell wall.

McFadden argues that this larger-scale electromagnetic field in the brain might be a second-level, parallel information processing mechanism. Eh? To put it simply, he argues that the EM field might be the part of the brain that’s conscious — the part of you that is aware of the world, and is thinking about these sentences as you read them, is the electromagnetic field inside your brain. This is completely against the grain of modern neuroscience — pretty much everyone thinks that consciousness (and any other information-processing your brain does) is done by the action potentials, by the individual neurons sending messages to each other. Just like your computer processes information by sending electricity around in little wires, your brain processes information by sending little pulses of electricity around among nerve cells. All the little wires and gizmos in your computer emit EM fields into the open air around them, but they’re not used for anything inside the computer and are generally regarded as a nuisance to be damped out. Nerve cells will similarly emit EM fields (because after all, they are shuttling electric charges around), but the main form of communication between cells consists of big “action potentials”, not these EM fields. For a typical nerve cell, EM fields from neighboring cells are just the faint sounds of dogs barking down the street, while action potentials transmitted from neighboring cells are the SWAT team kicking down the bedroom door. (Nerve cells tend to do a lot of crystal meth). Why would you bother to stick some of the processing power of the brain in the puny EM fields, when you’ve got all these robust action potentials scooting about? Besides, what could the EM fields do independently of the action potentials? The EM fields ought to be faint echoes of the states of the neurons in the local area, sort of an average of the voltage of all the cells nearby. How could there be anything going on in the EM field that isn’t already going on inside the cells themselves?

Obviously, I’m fairly critical of his theories. After all, what does he know — he’s just a world-reknowned expert in microbiology with a PhD and an active research lab. I’ve got a website. Despite my skepticism about his theories, his book is quite entertaining and admirable for its gusto. Hell, it’s worth the read just for the first few chapters that give a basic overview of quantum mechanics, evolution, and biochemistry — wild theories aside, he’s a much better and clearer science writer than some esteemed luminaries of science I could name (I’m lookin’ at you, Dawkins). Give it a read, at the very least it’ll make you think.