Open Questions Meet Settled Science

This is a direct follow-up to my last post.  In that post, I explicitly questioned Darwinian Evolution - to the extent that the first comment asked me about other articles that do the same.

I want to defend Darwin here and everything he got right, because I read a lot of articles from people who read what experts say and piggy-back on that to proclaim that Darwin's work is being "overturned" or similar war-of-ideas-and-Darwin-is-losing kind of language.  This is the wrong way to understand the scientific process of expanding beyond the clear limits of Darwin's theory.  But I suppose part of the reason people turn to it is because they don't understand the right way to understand this process.  I'm going to spend much of my time here working through that, partly by comparing Darwin's contributions to those of Newton's discovery of universal gravitation.

The Universe Before Newton

When I was young, I always had a problem understanding why Newton got so much credit.  What exactly was his great discovery?  Everyone always said he 'discovered' gravity and this changed the world.  Really?  As in, nobody knew about gravity before Newton?  This is a basic observation babies learn at the dinner table while dropping their utensils on the floor.  How did Newton get credit for discovering that?

The problem is that Newton's actual discovery has been so thoroughly integrated into general knowledge that it's difficult to even understand what his discovery actually was.  I want to take you back to pre-Newtonian times to get a handle on this.  Specifically let's look at the discussion among Pumba, Simba, and Timon from The Lion King.  They're all staring out into the sky wondering what the stars are.  Simba gives this weird explanation about the "Kings of the Past" and Timon explains they're just fireflies.  Pumba maintains they're giant balls of fire and gas burning billions of miles away.  Simba leaves when his friends make fun of his heartfelt explanation.  He has some mystical experience talking to his dead father and a baboon.  Whatever.  Let's keep the camera on the meerkat and the warthog.  Pumba claims the Earth revolves around the sun, and that the planets do to.

Timon: Look, Pumba, you gotta give up on these crazy ideas.  There's just no way they're real, and any idiot can prove it.  First off, if we were spinning around so much you'd think we'd all fly off - why aren't we all flying off?
Pumba: Well, Timon, that's because there's a force pulling at us to go straight at the Earth.  That force is stronger than the flying-off force, so we stay here.
T: That doesn't make any sense.  If that were true there'd be a giant coriolis effect.
P: (Struggling to remember what the big word means) There is one, but the world is so big we don't notice it.
T: Plus there'd be parallax with the stars, halfway through the year they'd be in one position, then six months later they'd be different, like seeing out one eye then the other.
P: Maybe there is a parallax, but the stars are so far out there we can't see it.  That's why the stars have to be billions of miles away.
T: Do you have any idea how far out there they'd have to be for that?  Are you even thinking this through?  If they're that far away they'd be too small to see.  All your explanations require me to believe these crazy ideas that are just excuses for you not knowing what you're talking about.  You're building up this big complicated story, when a much easier story works just fine.  Your problem is that you're ignoring all the physics.  Your idea stinks ... just like you do.  Did you fart again, Pumba?
P: Sorry.

Try to explain how the Earth goes around the Sun, and how the stars and planets work, based solely on the observations you can make with your traditional five senses.  (I'll even throw in proprioception and any of the other senses, too, but it won't do you much good.)  Without modern telescopes capable of measuring the tiny parallax, without being able to send things to the moon and other places in the solar system, you have to convince people of a lot of difficult ideas.

  • The fixed stars actually move
  • The 'planets' that do move are going around the sun
  • All this stuff is really big and really far away
  • Things way up in the sky operate using the same physics we operate under down here on Earth
This last point is probably your most difficult and it's the least obvious.  What makes most sense is that these tiny things way up in the sky should just fall to the Earth like anything you toss up in the air.  But there's something about being so high up there that makes stuff up in the sky immune to falling.  In fact, the farther up there it gets, the less it's affected by forces here on Earth.  This was Aristotle's contention, and it fit the observed phenomena perfectly.  Birds could go up in the sky somehow, but they still always came down.  Meanwhile, clouds seemed to float around effortlessly.  They didn't do anything to stay up in the sky, being held up in defiance of gravity.

Let's go higher still to the moon.  The thing is clearly solid, yet it sits there, unmoving, in the night sky.  If the same force of gravity that pulls you and me down when we jump into the air pulls on the moon, where's the evidence for it?  It makes no sense.

The problem is that you can't say the moon and the planets and everything else moves in any way that's similar to how we move down here on the Earth.  For one thing, if they were moving through all that distance, why haven't they slowed down?  They're moving through the air, which means there's got to be some kind of air resistance (even if this idea wasn't formalized yet) so how are they moving so far and so fast through the sky for so long on their own?  In order to maintain this model, you still have to make up a separate set of mechanics for the sky than the one you use for what happens down here on Earth.

Okay, now fast forward thousands of years to the Copernican Revolution.  The problem Copernicus, Galileo, and all their followers had were all the arguments laid out above.  Those are good arguments.  They're obvious to anyone thoroughly familiar with mechanics as people understood them at the time, and they're virtually insurmountable using unaided observation from the ground.  This was the real reason people doubted the movements started by Galileo and Copernicus.  They had some nice ideas, but they also had major holes in their theories - holes we wouldn't figure out how to address for decades to centuries later.  And when serious natural philosophers brought up those problems they were waved away as though they weren't important.  Or they got outlandish explanations with no physical evidence to back them up.  This gave detractors the impression that Copernicans and Galileians weren't serious.  They were unwilling to answer huge problems with their theories - because they had no answers.

Enter Kepler and Newton

Kepler figured out a way to describe the motion of planets, using mathematical concepts and some basic laws of motion.  He laid the groundwork for treating objects way out in the sky (although not stars - yet) as though they followed similar physics to what we observe on Earth.  It wasn't exactly the same, because here on Earth gravity overpowers so much of everything, and out in the Solar System Earth's gravity is only a minor player, but it was a major step in taking the movement of these celestial bodies seriously.  Also the idea that there's a (near) vacuum in space makes inertia for large bodies easier, without having to hand wave away the problem of "but why doesn't all this stuff slow down over time?"  That's not an obvious assumption to make.  Usually when an engineer says, "let's start by assuming ideal conditions" they're not talking about anything that could exist in real life.


Newton's great discovery wasn't that gravity exists.  As a phenomenon it clearly did.  Newton's genius was his ability to craft a set of mathematical equations that could describe both the gravity we experience on the surface of the Earth and the movement of celestial objects.  This was huge!  For the first time, the sky could be described as part of the universe we live in, not something that operates by completely different rules.

Again, if this simple transformation in thinking isn't clear for you go back to what it felt like to live in a pre-Newtonian world.  You could go out into the night and observe stars, planets, the sun, the moon, and clouds all moving about way up in the sky beyond anyone's reach or experience.  Their movements were mostly regular, but sometimes random.  And importantly, there was no obvious force acting on them to cause their movement.  All this activity up in the sky could be cataloged and sometimes even predicted, but it couldn't be explained in any satisfying way.  Sure, there were still questions remaining after Newton's discovery, but the major mechanism was solved.  We finally had a model of the universe we could build on; a frame onto which we could fill in details.

Years later, while trying to fit this theory into the much finer observations available in the late 19th/early 20th century, Einstein figured out that Newton's model needed to be revised.  He reworked a lot of the math, and modified many of the concepts.  I'm not going to minimize what Einstein did for the model of the universe physicists now hold in their heads.  Einstein's ideas really did supplant the model Newton was operating under.  But without taking anything away from General Relativity, I think we can clearly conclude that we wouldn't have gotten here without Newton's model.  We might want to say that Newton's model was wrong, in that it doesn't describe things as they really are, but that feels off.

Sure, Newton's model needed to be revised, but the core idea - that we can describe universal phenomena with the same physics we use to describe colloquial phenomena - is still at the heart of physics today.  Even quantum physicists, who will start by pointing out that "when things get really small we have to rely on a different set of physical phenomena" will point out that these same phenomena happen at larger scales, they're just absorbed into the noise of large numbers of interactions.  The whole point of unifying all the fundamental forces of nature is an outgrowth of the new dogma that there shouldn't be a separate set of universal laws for different locations or circumstances.

So yes, people go around writing articles that introduce Relativity by saying, "We now know that Newton was all wrong about gravity", but that unfairly dismisses how Newton single-handedly brought us from a place none of us would recognize today into a rational universe of physics.

Darwin did this same thing for biology

Biology before Darwin

One of the biggest problems Darwin faced, in his day, was the paleontology community.  That's because the fossil record didn't help his case.  Prior to Darwin, we had the story told by the fossil record.  This story had layer upon layer of fossils, but those fossils didn't tell a story of evolution.  Instead, it looked like there would be long periods of time where rich sets of species would thrive, then something would happen to wipe out a large percentage of those species, and a while later a whole new set would spring up out of nowhere.  Where did these species come from?  From the evidence in the fossil record, the dominant theory of the day was that the emergence of all these species came from successive 'creation' events.  Really!  This was the dominant theory of the time, based on the evidence available.

Again, we need to get into the mindset of the people back then.  Today, we think this is absurd because it requires us to theorize something is coming from nothing.  When does that ever happen?  But to the people of the time life came from nothing all the time.  Take maggots, for example.  If you leave meat out for awhile, you get maggots.  This is the spontaneous generation of life from nothing - other than that you have to start with the right conditions for life to occur.  One biologist came along and said, "no if you cover it up you don't get maggots", but that was easily explained by saying the meat needed to have access to air.  You don't get life if you smother a person, and you don't get life spontaneously sprouting out of meat if you smother it.  Simple, right?

It wasn't until a biologist covered the meat with a screen - thereby preventing flies from laying their eggs in the rotting meat - that we had solid evidence against spontaneous generation.  So this idea that life can't come from nothing is - surprisingly - a recent concept.  If we go back to before that time, it seems logical to explain these explosions of life using the same spontaneous generation seen at smaller scales elsewhere.

Sure, this fit with the Biblical story of Noah and the flood - disaster leads to elimination of speciation, followed thereafter by a new outgrowth of large numbers of species.  That made it harder to be an atheist and a biologist, when foundational to the theory of speciation was literally crationism, but this wasn't because the philosophers let their beliefs blind them to the evidence.  I'm sure that happens to a certain extent any time your personal closely-held beliefs track with your scientific work.  But it's unfair and incorrect to simply chalk this whole period up to "bad science was driven by religious bias".  That's not what happened.  Instead, the best evidence available led in the direction of de novo creation - even down to the species level in the fossil record - so that's where the field went.

Along came Darwin, and he started pointing out ways creatures could be seen to morph over time to adapt to their environment.  He explained how this could be carried down from generation to generation, where only those adaptations that were beneficial would survive.  His theory had two important concepts that - while not original to Darwin - were combined together in his theory in a coherent way that gave an alternative to the explanation that everything just came about fully-formed.  This still didn't explain the fossil record, but it explained a lot about what people saw in front of them every day.

One thing most people don't realize about Darwin is that he thought you could carry traits with you that you acquired during your lifetime.  So, for example, if you were more physically active, built up a lot of muscle, lifted weights, and generally spent a lot of time in the gym; in Darwin's view you could then pass those traits along to your children.  This is because even though Mendel did his work around the same time as Darwin, Mendel was an obscure monk whose work wasn't discovered (and integrated into Darwinism) until much later.

In fact, without Mendel's ideas, Darwin started to fall out of favor until around the 1920's.  There were too many problems with the theory that couldn't be explained.  Around this time Mendel's work with the pea pods was re-discovered and the idea of genes was incorporated into Darwin's model.  This is called the neo-Darwinian synthesis, and it still predates the discovery of DNA as the genetic material by three decades.  Mendel's ideas added a constraint Darwin couldn't have known about: mutations happen in the germ line, such that acquired traits can't be passed down.

That means in Darwin's explanation of the evolution of the eye he couldn't use the step that a piece of clear material got lodged in the light-detecting hole, which happened to help focus the light better (describing the development of the lens).

Post-Darwinian Molecular Challenge

All this brings us a little context to the last post's challenge of Darwinian evolution.  It's not that Darwin should be completely discarded.  It's that there's so much it can't explain we know we need whole new theoretical constructs to build off of that simply don't fit on the frame Darwin built.  As with the neo-Darwinian synthesis we need new ideas to make the whole thing work again.  Some things we absolutely need to keep, like natural selection and common descent.  Incidentally, these also happen to be the things many hard-core anti-evolution people target the most.

Take common descent.  There's no problem in molecular biology with common descent.  There's this concept of a phyogenetic tree, where we can trace genetic similarities back based on how close you are to a speciation branch point.

Let's try an analogy here.  Say there's a copyright lawsuit, where someone made a photocopy of a textbook page for personal use.  He liked this page and since he didn't have any qualms about piracy (and Fair Use is ridiculously vague in IP Law) he made a few copies to share with his friends.  Those friends shared with their friends, and on and on.  Now, nobody's photocopier is perfect, and each time someone made a copy they kept any extra spots introduced from things like specks of dust on the copier bed.  So although the content is mostly the same from person to person, we could take a stack of pages from different people and, by meticulously comparing the tiny imperfections in different copies, figure out whose copies are from which branch of copies along the piracy tree.
This is useful in describing how recently a trait was developed, and in what order certain aspects of it developed.  It also helps us understand aspects of a system.  For example, there's a gene that's really important in humans for ensuring eyes develop.  You can take that same gene - the human gene - and put it into fruit flies.  Except you don't express it in a controlled way, so the flies start developing eyes in really creepy places like in their legs and at the ends of their antennae.  It's pretty gross, but it's also fascinating that a human gene works in a fruit fly.

This is like finding out that the word for "mountain" in French sounds very similar to the English word.  That's because the two languages share a lot of history.  Meanwhile, if you try to do the same thing for Japanese most of the similar-sounding words (cognates) are for more recent ideas like "furaidopoteto" (which sounds similar to 'fried potato') for French fries.

We see this same thing with genes, where other mammals share more genes with us than something that branched off earlier, like reptiles.  And we share more with reptiles than something that branched a lot earlier, like mushrooms.  When I say we have issues applying Darwinian theory to understanding molecular biology, I'm not saying we have a problem with this idea of common descent.  Indeed, common descent is the part of evolutionary theory that makes molecular biology easier to understand.  We can quantify it mathematically, and pull out patterns that speed up research by allowing us to do research in fruit flies.  That helps us understand what's happening in humans.

Allow me one powerful example among so many I might choose.  A researcher in Germany was playing around with genes in fruit flies.  She was randomly disrupting different genes to figure out what might happen to the flies if they didn't have certain genes.  It was a wild fishing expedition, but these things can be quite useful sometimes.

One of these mutations caused her fruit flies to grow a hairy fungus all over their bodies.  She looked at them and said, "Cool!"  Except she was German, so instead she said, "Toll!"  The gene she disrupted turned out to be a cell membrane receptor.  It recognized certain patterns that are common on things like fungi but not on fruit flies.  That way, it knew when the receptor recognized one of these substances it needed to attack that fungal invader.  She called the gene a Toll receptor.  Cool.

Once we knew these things existed, researchers in mice and humans started looking for similar genes.  Since both the mouse and human genomes have been sequenced, we used some sophisticated search tools that allow us to find genes that share lots of similarities, even if they're not exact.  In mice, humans, and other mammals we found whole families of genes similar to the Toll receptors found in fruit flies.  We called them Toll-like receptors, and they can recognize all kinds of foreign-looking invaders in the same way the fruit-fly protein works.  Thus, the idea that organisms have common motifs - and that those motifs are more abundant the closer an organism is related - is literally an active element of research to this day.

The Clash of Open Questions and Settled Science

I really want to defend Darwin here, as much as I can, which is why I compared him to Newton.  Isaac Newton built the foundation from which others, like Einstein, were able to rise higher.  Without Newton's discoveries, Einstein wouldn't have been able to figure out relativity.  This same idea applies to Darwin.  Charles Darwin brought us into a biological world where organisms are interrelated, and mechanisms have to flow from one to another.  His work shifted the whole world of thinking in biology, and it paved the way for us to make the discoveries that make his theories inadequate.  The fact that they don't explain everything very well anymore is because we used his theory to rise as high as we have, like climbing a ladder to the top rung.  It wasn't that he was wrong given the information he had available to him.  All his discoveries were significant, given the state of the evidence at the time - similar to Newton's discoveries.

So when I see articles cheering the "downfall of Darwin" or saying we need to jettison his ideas, it always feels wrong.  For one thing, we're never going back to a pre-Darwinian world view, just as we're never going back to a pre-Newtonian one.  It took me a lot of explaining just to bring those strange worlds back into perspective, and I think we can agree nobody who seriously questions either Darwin or Newton wants to return to either of those ways of thinking.

We want to avoid certain failure modes of thinking.  We don't want to treat Darwin's theory as sacrosanct in a way that says, "these ideas are inviolable".  But we do want to appreciate the contributions Darwin made, while still allowing ourselves to figure out where the holes in the theory are.  It's in those holes that so many major discoveries are found.

There's no reason to outright reject Darwinian evolution, or say the man was a charlatan just because his ideas aren't sufficient to explain speciation or the origin of life.  It's enough to admit, "we don't really know how this applies at the molecular level".  We will someday, but it may not be any time soon.

Comments

  1. I don't want to be one of those people who keeps saying, "Well, okay, that's fine, but what about this?" I'm more hoping to lean on your vastly superior knowledge of biology to ask some questions I've always been curious about. This time around my question is about speciation. I've heard that despite all the selective breeding we've done we've never really created an entirely new species, and that attempts to do so artificially have all ended up with fatal mutations. First is that really the case? Or have I been mislead?

    Second, unless and until we can do that, it would appear to be a big challenge to both natural selection and common descent, right? Or not?

    ReplyDelete
    Replies
    1. On natural selection: This is a real phenomenon that actually happens. We've observed it directly. In that sense, we wouldn't say that downstream projected effects of natural selection are a "big challenge" to it. Going back to the Newtonian analogy, we might say that universal gravitation is a real phenomenon we've directly observed. Einstein might come along later and say those observations are just an outgrowth of space-time warping, but that's not the same as fundamentally challenging Newton's idea. Newton said, "I think the same force that's operating on apples in trees is operating on the sun, moon, and planets." That's still true, even if the ultimate answer about how the phenomenon we call gravity really works is "it's more complicated than that."

      That's basically what I'm trying to say here. Yes, natural selection is real, and we shouldn't discount that. Yes, it's likely playing an important role in evolution. But as with Einstein and gravity, we're at the point in molecular biology where natural selection doesn't do a good job of explaining the specifics. There has to be more to the story here, and we should be really interested in figuring out what that is. Instead, there's this inane 'war' between what people think of as science vs. religion (but is really just militant atheists vs. biblical literalist Christians) that's biasing everyone away from honest inquiry. Too often I see religious people shy away from deeper inquiry for fear they'll have to either reject their faith or devote themselves to 'debunking' established theory.

      On the other hand, too often I see atheists reject any attempt to question whether the neo-Darwinian synthesis is adequate to tell the story without major revision. They're okay with, "we haven't figured out this transition point between these two species yet, but we will!" But they're not as open to, "This explanation doesn't really fit current theory. We need to accept that we don't understand it so we can go looking for new mechanistic answers." This is why I put this in the Open Questions series, because NOT allowing it to be an open question really inhibits good science from both directions.

      Too often I see bad explanations (or most charitably, 'wildly inadequate explanations') thrown together to dismiss a legitimate disconnect between what could plausibly be accomplished through neo-Darwinian mechanisms and what we actually observe. But then on the other side I also see a misguided attempts to discard or discredit real mechanisms because they fail to tell the whole story. This can go so far that people are willing to replace credible scientific theory and observations with wild hokey junk science.

      Another analogy would be like defending whether ion transport across the cell membrane is real. One side says, "All the functions of a cell can't be explained by invoking ion transport, so therefore it doesn't exist!", and the other side argues, "NO! We've clearly observed ion transport, so it must drive all cellular functions!", when the obvious and answer is, "Yes, it's clearly part of the story, but there has to be more to it. Can't we accept that and move on to discover more about how life works?"

      Delete
    2. On common descent: I'd mostly repeat what I've said above, except that we can't directly observe common descent because it happens over such long time scales. However, I'd personally be surprised if some other mechanism eventually wins out over common descent to explain the evidence. It really looks exactly like what we'd expect it to look like if all organisms descended from a common ancestor. The language diversity analogy is apt here, because we really do see lots of similarities between more closely 'related' species and we can track those changes over time. I know the theological hand-waving answer to this is often, "of course God isn't going to re-invent his own methods multiple times; he'd use similar mechanisms for all creation" but that does a much worse job of explaining the specific, mathematically quantifiable drift of inconsequential genetic variation from species to species.

      Thus, common descent strongly implies speciation is happening - that one species is branching off from another on down the line from single-celled organisms to the currently observed mega-variety - and that the fundamental idea of evolution is sound. Given the amount of time it generally takes for speciation to occur, it's not surprising that we've never observed it. This makes sense, since it would presumably be advantageous for a species to be able to mate with and acquire the beneficial mutations of wildly disparate members of the species. Branching off into a different species with which the parent organism can't interbreed is an unfortunate trade-off that comes with any specialization that makes new innovations by the new branches incompatible with each other.

      So yes, common descent is the best explanation of the consistent evidence from molecular biology for speciation. This is where neo-Darwinian theory is at its strongest. But a solid explanation for how that happens in practice is still lacking. And the traditional story of natural selection driving molecular innovation is so unhelpful that you're probably better off ignoring it most of the time when you're talking about the structure of NF-kappaB, or JAK/STAT signaling. It's a great shorthand for larger mechanisms, like escape from predators or survival strategies, though.

      Is evolution happening? Yes, almost certainly. How is it driven on a molecular level? We don't know; it's an open question. What can it tell us about the origins of life? Not much, unfortunately. Should we throw it out and start over? That would be folly. How about we build on it, being honest with ourselves about when its insights are essential, and when it's inadequate to the task at hand.

      Delete

Post a Comment

Popular posts from this blog

Reverse Engineering Life

Cancer update: precision oncology comes into its own