Increasing Diversity By Killing It

Veritasium recently highlighted what he calls "The Longest-Running Evolution Experiment" on YouTube. The experiment uses E. Coli bacteria and it's been running for 33 years.  This means that there have been 74,500 generations of bacteria.

To put that in perspective, assuming a generation in humans takes about 20 years, it would take humans 1,490,000 years to have this many generations.  For the record, if you ask a Darwinist, they will say that modern humans have only been around for 300,000 to 800,000 years.  Indeed, going back 1.5 million years, our ancestors would be Homo erectus.  The point is, there are huge differences between H. erectus and H. sapiens that supposedly came about in those roughly 75,000 generations.

On the other hand, if you look bacteria after the same 75,000 generations, they are basically unchanged to this day.  Not only that, but E. Coli can be found back well before this 33-year-old experiment began too. And in all that time, no mutant bacteria formed which would be classified as anything other than E. Coli.

But this is a bit aside the point I wanted to make in talking about this bacteria now.  The point raised by the video is that the E. Coli that exists today ought to have evolved to better fit into the environment of the laboratory, and comparing older strains with modern strains show that modern strains of bacteria are, in deed, "more fit."

This is, in fact, how evolution is typically presented. Organisms become "more fit" in their environment.  The problem is that this overlooks one extremely obvious point: becoming more fit for a particular niche environment does not mean that you are more fit as an organism, as a whole.  What I mean can be seen if we hypothesize a bacteria that has 50% capability of survival in a lab and 50% capability of surviving in a kitchen and 50% capability of surviving in a bathroom.  After thousands of generations, we measure that the bacteria now has a 95% capability of surviving in a lab, and that's all we measure. We then declare that the organism is "more fit", despite the fact that for all we know the new organism has a 0% capability of surviving in a kitchen and in a bathroom now.

The point can be even more readily made by considering what happens when a human feeds wildlife.  Birds, for example, may learn that to get food they just eat the seeds from a feeder all winter long. But what happens when the old woman who used to feed them dies and there's no more seeds?  The birds die too, because they have lost the ability to get food on their own.

So the question is, can birds that learn to eat seeds from a feeder be considered "more fit" than birds that know how to search for food on their own?  Only in the extremely specialized context of that specific environment and only assuming that environment never changes could such a bird be considered "more fit."  In all other points of view, it's actually harmful to the bird to make it dependent upon humans.  

E. coli naturally lives in the intestines of a human being.  Would we still consider the E. coli to be "more fit" if we discovered that all these lab grown bacteria would die if placed back into an intestine?  Does the fact that they are the best at living in the lab really mean the organism is "the best" itself, given that without being able to survive in humans, all of these bacteria have no ability to survive the instant there is no more funding for this experiment?

The reason that becomes important is more than just semantics. Evolution is supposed to explain why organisms become more complex over time, yet all these experiments actually show is organisms adapting to a single variable that we have artificially decided is the only thing we should measure for.  In fact, it ought to be predicted that they would become simpler as a result.  After all, if E. coli doesn't need to survive in the stomach because it's environment is now restricted to a laboratory, then the ability to survive in any other environments is wasted effort on the part of the organism.  It's better to streamline the organism and remove that ability.  But, clearly, this is reducing the available functionality, not increasing it.  And in fact, natural selection is a winnowing process by definition.  Death is not a creative function.  You do not increase diversity by killing off something.

So can you really call this an experiment in evolution?  Only in the sense that the bare-bones definition of "evolution" is change through time, and certainly these E. coli have changed through time. But to try to extrapolate from that some grand scheme of Darwinian progression is simply pushing the data way too far from what it actually provides.