The wonder of evolution

This isn’t really a question but if there are evolutionary biologists out there feel free to fight my ignorance.

Evolution can achieve amazing things over millions of years. I can understand how evolution led to things like eyes, ears, digestive tracts, bipeds, five fingers, venom, wings, and so forth.

There are two things, however, that blow my mind:

  1. Evolving the ability of an organism to heal itself. I can understand how certain structures and chemical processes may have arisen but the idea that if you get damaged, it will heal (within certain constraints) seems like an impossibly complex feature.

  2. Evolving the ability to learn. Similar to healing, this isn’t just structures but the continued development of an organism within its life span. Humans developed the ability to continue to learn new facts, behaviors, and skills throughout their lifetimes, and nearly all other animals seem to have this ability to some degree or another. The idea of evolving a brain that can learn is stunning, made even more so by our meager ability to replicate it artificially.

It’s enough to make me believe in intelligent design…nah.

Remember, the evolution on multicellular organisms started in unicellular organisms. “Healing” was closing a hole in a membrane, and that can happen spontaneously in the first place. Adaptations that can speed up a process that happens in some cases anyway is something that would be preserved. And multicultural organisms? They replace old cells anyway. In fact, the trick is to stop the spread of new cells.

Far from being “an impossibly complex feature”, cells are in a constant state of homeostasis, both building and breaking down proteins to regulate cellular activity, remove toxins and antigens, et cetera. Multicellular organisms actually have a process of apoptosis or programmed cell death, where cells that have outlived their function or acquired too much damage are deliberately broken down and the components reused to build new cells. These abilities also translate into repair of trauma-damaged cells. If you are referring specifically to the clotting of blood, regeneration of body parts, and formation of scar tissue, these are processes that evolved very early in multicellularism, likely to facilitate organism growth as it passed through stages of development rather than specifically to protect against traumatic injury.

The functionality of even simple living neural networks to learn and adapt is stunning and it takes no small amount of imagination to conceive of how it came about and developed in complexity to the point of the sapience that humans (and to a degree, many other animals) enjoy. But I wouldn’t describe it as surprising; the evolutionary advantage of being able to to learn and spontaneously adapt by making predictive behavioral modifications in response to changing environmental conditions and hazards is so self-evident that the real surprise is that it took as long as it did. The complexity of the mammalian (and avian) brain is certainly a marvel of biology and an area of science that is perhaps even more mysterious than the foundations of physics, particularly in how brains form again and again from relatively simple genetic cod and epigenetic factors to produce highly complex architectures that are nonetheless quite similar from individual to individual belies our ability to describe it in detail but that is what makes neuroscience so interesting.

That we have not been able to simulate consciousness on digital hardware is both a reflection of our very limited understanding of the actual processes of consciousness, and the immutability of said hardware. In popular science and fiction literature the brain is often compared to or described as a computer, but in fact the brain is fundamentally different from a digital computer in many ways but in the particular aspect that it is not a state machine but rather a constantly changing matrix of highly complex connections. Although “neural networks” in computing have been around for a long time and have found great utility in machine learning, their complexity pales to the sophistication of an actual brain in terms of how adaptive it is to stimuli. The complexity of the brain is so great that we cannot actually simulate a single neuron to a really high degree of biophysical fidelity, and are nowhere near being able to simulate something even remotely as complex as a relatively simple mammalian brain in software. When we do get to the point of making digital computers that are legitimately self-aware, my wager is that how they work and their conception of the world will be vastly different than our own.


I talk to a fair number of strangers on the phone, often for extended periods, and they sometimes get comfortable enough to chat about their personal politics. I had a “vendor” (we occasionally buy some $15 cables from him) casually go on a rant about how evolution was a bunch of BS and even had a scenario to back up his claim. It was something like this:

“You know Rolex watches? They’re full of gears and parts and springs and everything else. Well, suppose you take one all apart and put it in a bag and shake it. Yeah, well, no mater how many times you shake it, you’ll never get a completed watch out of it and that’s just a watch. Something like a brain would never happen.”

I certainly wasn’t going engage this $150/year vendor to tell him that, yes, you would get a watch given enough time. Or that, in parallel, every time you shake the bag, a (nearly) duplicate bag or bags are created so the first bag doesn’t need to make a functioning watch. So there are millions and billions of bags and so shaking parts might make a complete watch relatively more quickly than you might think.

I might have gone on to refer to something almost everyone knows at least a little about: dog breeds. With people doing the selecting (instead of nature), people have bred dogs for assorted uses and traits and it only took, at most, some thousands of years. And they’re so different that’s it’s a great illustration of how evolution can change things in timescales we can better comprehend.

I wasn’t aware that this was a formal ‘analogy,’ thanks.

Thanks for the link. While I have little patience for the silliness of “intelligent design” debates, some of the historical background made that an interesting read. I particularly like Richard Dawkins’ argument that “design is top-down, someone or something more complex designs something less complex.” The watchmaker is necessarily more complex than the watch, and by extension, whatever being designed the watchmaker must be correspondingly more complex than the watchmaker. Who, then, designed the designer? Dawkins concludes “that (a) this line continues ad infinitum, and (b) it does not explain anything. Evolution, on the other hand, takes a bottom-up approach; it explains how more complexity can arise gradually by building on or combining lesser complexity.”

Another interesting comment is that although William Paley is among the most famous originators of the watchmaker analogy, he did not in fact oppose the concept of evolution: “A closer look at Paley’s own thinking reveals, however, a God who works through the laws of nature, not beyond them like the modern ID theorists’ designer. Paley had no objection to species changing over time. It’s only in today’s highly polarized culture-war climate that we don’t bother to notice that one of the forefathers of intelligent design theory might have been perfectly comfortable with evolution.”

I had been unaware of the Kitzmiller v. Dover Area School District trial in 2005, a sort of modern version of the Scopes Monkey Trial. It appears that a bunch of lunatics in a Pennsylvania school district had mandated the teaching of intelligent design in public schools. The judge, after five weeks of argument, reached the solid conclusion that “intelligent design” was not science, and was a thinly veiled pretext for religiously motivated creationism, and therefore mandating its teaching violated the Establishment Clause of the First Amendment.

There is no shortage of Christians who tout their atheist-turned-Christian-because-of-the-wonder-of-life testimony and IMHO it is the reason that scientific debates on this topic generally go nowhere and persuade almost no one. Both evolutionists and creationists agree that life is extremely complex and amazing. It all comes down to something more subjective - whether they feel this amazing healing, or ability to learn, - comes from a deity higher up or “just” evolution. And for some it crosses that threshold and they embrace intelligent design and others it does not and they do not.

I have a feeling that if you looked into their backgrounds most of them were never atheists, but say this to “prove” how strong their religion is.

Assuming this person is honestly looking for a discussion, that would not be my argument. You’re buying into a flawed model of evolution, which is a bad first step.

And bringing in many-worlds is a distraction from the issue. Instead, I would point out that this analogy has two fundamental flaws.

  1. It assumes there’s a target for evolution.
  2. It has no equivalence for natural selection, which is far more important than randomness to evolution.

However, chances are this person is just putting up a strawman argument, which means it’s probably not worth arguing about it with them.

The problem with all the “watch” or “jumbo jet in a junkyard” analogies is that they fundamentally misrepresent evolution as being driven solely by random chance. It isn’t.

Take the watch. Imagine a bag of watch parts shaken in a bag, though really we should imagine a million bags of watch parts being shaken.
We then inspect the bags and whenever a couple of elements interact in a way that is useful for the purpose of performing a function we set those aside and populate the millions of bags with those structures and repeat the process. And we repeat it millions of times.
There are other complications and things to take into consideration but here is a little video that covers the same concept in an illuminating way. It is far from perfect (and admittedly so) but it is a far better analogy of what is going on in evolution than the “watchmaker” analogy as traditionally presented.

if they cant stop the new cells doesn’t it evolve into cancer at some point?

Not necessarily. There are a small number of cells in all tissues that do not have limits on how many times they can reproduce. They are stem cells and they are essential to healing wounds. When there’s an injury, nearby stem cells migrate to the injury site to do repair work.

Now stem cells are thought to often be the originators of cancer, but more has to change than just no limits on cell division. Something else that tells the cells to divide as fast as it can is required. One type that I’m familiar with is chronic myeloid leukemia (CML).[*] In CML, parts of chromosomes 22 and 9 swap places. That produces an unnatural combined protein called BCR-ABL1 that causes white blood cells to rapidly reproduce. White blood cell progenitors have a switch that, when signaled by a specific other chemical, causes it to produce another white blood cell. BCR-ABL1 is a modified form of this switch that is stuck on “on”, so way too many cells are produced.

There are drugs that treat this form of cancer by blocking that mutated switch. Gleevec was one of the first targeted therapies for cancer and there are now several others for CML as well as similar drugs for other cancers that are caused by similar mechanisms.

[*] No I do not have CML. I just got interested in it a number of years back and did a lot of research.

Evolving a clock

Yeah, that’s one of the main features of cancers. Cells that don’t know when to stop replicating.

Here is an example of evolution in action. I read about this many years ago. Some computer scientists started with a random program. That is a list of commands. Nearly all such programs would do nothing or nothing interesting. But they were all run along with unsorted lists. Maybe there were a million such programs. Any that did anything at all useful were duplicated many times with slight random changes. They were looking for programs that began to sort the lists. Rinse and repeat. Many, many times. Eventually a sort program emerged that turned out to be faster than any known sort program. Interestingly, when they looked at the program, it was incomprehensible. Evolution produces solutions, but not necessarily comprehensible ones. Biology is hard.

Many, many years ago. There was a paper in the 1959 Fall Joint Computer Conference (or maybe Spring Joint) by someone from IBM who did this. And genetic algorithms have been used to generate tests and new circuits. The difference from natural evolution is that selection is done based on some kind of goal, not just reproductive fitness.

Just to mention that the only people who use the term “evolutionist” in my experience are creationists, who try to say that evolution is a faith not a conclusion from the evidence.
Those who reject the evolutionary explanation often seem to not really understand how long a billion years is and how many generations there are to let new features evolve.

I’ve never yet heard a good explanation of why one of these people was an atheist. Half of them seem to think atheism is equivalent to not going to church. Now atheism does not require a logical justification, it is just the lack of belief in any god, but this kind of testimony would be a lot more convincing if someone had a rational justification for their atheism.

I am not saying that they have no justification for their previous atheism. I am saying they (or at least a lot of them) are outright lying about being former atheists, much the same way some claim to be former Satanists.

As Dr Manhattan said,

Reassembling myself was the first trick I learned.

Organisms, by necessity, know how to assemble themselves. Healing is simply assembling oneself again might be the article you read

“It didn’t kill Osterman. Did you think it would kill ME?”