More Recent Comments

Saturday, August 11, 2012

Is "Unguided" Part of Modern Evolutionary Theory?

Creationists are unhappy with the claim that the evolution of life on Earth appears unguided and purposeless. Casey Luskin would like to think that it's a fundamental part of "the theory of Darwinian evolution" [Unguided or Not? How Do Darwinian Evolutionists Define Their Theory?].
An argument we are increasingly hearing from theistic evolutionists is that the "unguided" or "random" aspects of Darwinian evolution are merely "philosophical gloss" or an "add-on" promoted by new atheists who use bad philosophy. Jay Richards covered this question in his recent dialogue with Alvin Plantinga--see here, here, here, and here for the series. While many new atheists undoubtedly make poor philosophers, the "unguided" nature of Darwinian evolution is not a mere metaphysical "add on." Rather, it's a core part of how the theory of Darwinian evolution has been defined by its leading proponents. Unfortunately, even some eminent theistic and intelligent design-friendly philosophers appear unaware of the history and scientific development of neo-Darwinian theory.
This is wrong in many ways.

Here's the simple explanation ... so simple that even an IDiot should be able to understand it.

Modern evolutionary theory consists of many parts including the mechanisms of evolution. The main mechanisms are natural selection and random genetic drift and those two mechanisms act on populations containing variation. The variation is due to the presence of mutations and mutations arise "randomly" with respect to ultimate purpose or goal.

There are tons of experiments proving that mutations are essentially random. (Let's not get into quibbling about the meaning of "random.")

Now let's look at the history of life on Earth. This is a completely separate subject from evolutionary theory. It's like the difference between the theory of gravity and how and when our particular solar system formed.

Looking all the evidence used to reconstruct the probable history of life we see no evidence whatsoever that it was guided in any particular direction or that there was any underlying purpose. That's why we conclude that the evolution of life on Earth appears unguided. It's a tentative conclusion based on fact and observation and not on "the theory of Darwinian evolution."

While it is true that evolutionary theory doesn't allow for a "guided" mechanism, it isn't true that the history of life has to be devoid of purpose or guidance. There could well be evidence that god intervened or that particular organisms were preferred over others and the history was tilted in one direction. But there's no evidence that this is the case—with the possible exception of beetles.

The history of life looks exactly like it should if it were the result of accident, contingency, and evolution. There's no evidence of god(s). That's what makes the creationists upset, not evolutionary theory.


27 comments :

chemicalscum said...

"The history of life looks exactly like it should if it were the result of accident, contingency, and evolution. There's no evidence of god(s). That's what makes the creationists upset, not evolutionary theory."

Exactly since evolution is stochastic it does not exclude a theistic interpretation. In this case God chooses a specific path though the multitude of possible ones. Since the source of chance variation ultimately depends on Quantum Mechanics and I take realistic Everett interpretation of QM all the possible evolutionary paths exist orthogonally in Hilbert space. God is not necessary.

Le Hasard et la Nécessité

Georgi Marinov said...

There is a danger that this is going to get as cliched as its predecessor phrase, but I think this is still not the case so I will quote Micheal Lynch and repeat that "nothing in evolution makes sense except in the light of population genetics". At present population genetics is the core of the theory of evolution at present, there isn't anything guided in population genetics and there is no need to invoke any such guidance as the theory matches observations quite well. Therefore unguided is very much an inseparable part of the theory and has to be mentioned.

Now it is true that there is no way to prove that mutations are not in fact random. But if you argue that they are the product of conscious divine intervention, you are not just a creationist, but you also have an awful lot of additional explaining to do given the features of genomes and organisms on this planet. We get into the "cheater deity" territory and this is not a very interesting area to go into again.

Michael M said...

So why do popularizers of evolution got to such great lengths to insist that evolution and/or natural selection are not random?

J Thomas said...

The history of life looks exactly like it should if it were the result of accident, contingency, and evolution. There's no evidence of god(s). That's what makes the creationists upset, not evolutionary theory.

It's that they have a stupid idea what a god would look like.

They figure there ought to be one rule for inanimate stuff, and a completely different rule for life. God is supposed to mold people out of clay, and if you can't see his fingerprints left in your skull then it means there's something wrong with your faith.

The history of life is perfectly compatible with a god who could make a universe. But they think he ought to be imperfect enough that he'd leave special traces that are obviously incompatible with the rest of his work.

Not to say that the existence of a world that appears to be consistent requires a god to create it. But it doesn't require the absence of a god either.

The guys who get upset are the ones who think they understand God so well they know exactly how He would make the earth, and then it looks like they're wrong. They can't accept that they're wrong, so it must be either God is wrong or the evidence is wrong. Of course given that choice they decide the evidence is wrong.

Anonymous said...

Unguided. Not random.

Selection is not random and has a "direction",. Natural selection is not random either. But it is not guided. There is no cosmic plan to make most animals be arthropods and nematodes. It just ended up that way.

Georgi Marinov said...

I didn't use the word "random", neither did the OP. The word was "unguided".

Whether evolution is "random" or not is a philosophical debate the resolution of which depends on the semantics of the word "random". There is a very strong random element in evolution as mutations are definitely random the vast majority of the time, and when things get fixed by drift this is also definitely a random event. Given fixed environmental conditions and sufficiently strong selection, selection is a mostly deterministic process. But you can look at those environmental conditions as the product of chance. Then there is the big debate about the relative importance of selection and drift - a lot of people, including some very prominent ones do not consider drift very important, this has been the dominant view of the hardened Modern Synthesis for quite a while, and has entered a lot of popular treatments of the subject. But in recent decades it has become clear (or rather, what has been known since the very beginning of population genetics was appreciated again in the light of the data we have now) that the combination of effective population size, drift and insufficiently strong selection has shaped genomes and organisms to an extent that can not be ignored, in fact it may well be a more important factor than selection. But even then some things that seem to have happened because of neutral process can be looked at as in fact inevitable given initial conditions that reduce the power of selection.

Anyway, the discussion here is about the word "unguided" - the religious have a problem with it it because it excludes theistic evolution as a valid option; that's the important thing here

Georgi Marinov said...

The history of life is perfectly compatible with a god who could make a universe. But they think he ought to be imperfect enough that he'd leave special traces that are obviously incompatible with the rest of his work.

Every meaningful to people concept of God includes a God that cares about humans and has included them in his plan for the Universe. A God could make a Universe and let evolution run its course but unless you go into strict deism with a deity that does not care at all about humans and has nothing to do with their appearance, which is in practice pretty much equivalent to atheism in the eyes of most believer, you have to posit that God had everything rigged so that humans eventually appear.

This argument is 100% incompatible with modern understanding of evolution - mutations are random, drift not just plays a big role, but it seems that the progression of organizational complexity that ultimately resulted in us seems to be largely a product of it. If it was all about selection, it could have been said that God rigged the environmental conditions so that humans evolved. This is still essentially creationism, but it is not nearly as worse as what you need to posit to make it compatible with neutral process, which is that God rigged the trajectory of the atoms and molecules that were part of all those mutations, meiotic divisions and other molecular and cellular processes that resulted in the apparent evolution of so many things through neutral processes. This is an even stronger form of creationism and it has to be called out as such every time someone dares bring it up; for some reason I don't see it properly answered though. Yes, it is compatible with the observations. But what an elaborate scheme you have to make up out of thin air to make it so...

Unknown said...

Unguided but not unconstrained.

SLC said...

Lynch was paraphrasing Theodosius Dobzhansky who said, nothing in biology makes sense except in the light of evolution.

Alex said...

They're gonna quotemine your "beetles" gag, you know.

jaxkayaker said...

So, which type of creationist is Michael M? The dishonest type, or the type with poor reading skills?

I know that others have already stated the essence, but I'll just reiterate that natural selection isn't random and therefore evolution isn't entirely random, the key word in the latter phrase being "entirely". Mutations aren't entirely random, but seemingly are random with respect to fitness. Point mutations have different probabilities of occurrence, depending on whether the mutation is a transition or a transversion.

Richard Edwards said...

This is actually a common Creationist mistake. Random doesn't mean uniform. Transition/transversion bias does not stop them being entirely random.

If you want to argue against complete randomness, you need something that directionally changes the probability of something happening, rather than has a fixed different probability. An example of this would be an increased probability for genes being expressed to undergo mutation because they are not protected by chromatin structure. This would probably increase the likelihood of a fitness effect but, crucially, would not necessarily increase the likelihood of beneficial mutation.

J Thomas said...

Every meaningful to people concept of God includes a God that cares about humans and has included them in his plan for the Universe.

Christians etc have the problem that they observe bad things happen to good Christians. But God is supposed to care about people.

The obvious conclusion is that God cares about people's souls. People don't know what's good for them. It's OK for them to die. The suffering they go through is not evil in itself. When a good Christian bets in the lottery and loses, it doesn't mean God hates him. If he gets cancer God still loves him and cares about him. If a crazy mugger puts a knife in his heart, God still cares about him and his soul has not been damaged.

If you decide that you know what good and evil is, and it involves what happens in your life, and you believe in God, you will almost inevitably decide that God is evil. If you can solve that for your own life, extending it to evolution is easy.

Michael M said...

Where did I even come close to suggesting that evolution was "uniform" in any sense of "uniform"?

And, do pray tell, how does acknowledging that evolution, in its current formulation, is a random process (i.e., a process in which outcomes are not certain) make me a creationist?

Diogenes said...

Do we describe the theory of gravity by saying the motion of planets is "unguided"?

Do we describe the germ theory by saying infection by viruses is "unguided"?

Why should we say evolution is "unguided"? Why is evolutionary theory different?

To get technical, here is what we should say:

In evolutionary theory, when we say "random mutations" we mean mutation in which the mutated state (pre-natural selection) is not correlated with its contribution to fitness (i.e. differential fecundity).

Novel adaptive complexity--that is, systems which have multiple independent parts that contribute to the overall function-- results from multiple mutations that are random in the sense described above. However, one cannot safely conclude that, when parts are interdependent, all of them were beneficial when they happened and were filtered through NS. Some parts may be due to neutral drift and others contingent on those were fixed by NS later.

So, if the number of parts in a complex interdependent system are more common than one would expect relative to random mutation, the magnification is due to NS given the fitness in the historical context when the mutation occurred.

andyboerger said...

because it is necessary to distinguish between the randomness of mutation and the fact that organization and improvement comes out of that through a selection 'process'. It is a way of answering people who say, 'no matter how many monkeys bang on computers, you're never going to get Shakespeare!' by saying, yes, you COULD, IF you gave each monkey a banana every time they typed something that happened to make sense, and withheld the banana when they just banged out a bunch of meaningless letter combinations.

Georgi Marinov said...

When I say "cares", I don't mean caring in the sense of "taking care of", I mean caring in the sense that humans being important for him, important enough to send them his son and interact with them

Allan Miller said...

So why do popularizers of evolution got to such great lengths to insist that evolution and/or natural selection are not random?

In part, their own preferences about what the word 'random' means. This was discussed at great lengths on the now defunct Richard Dawkins discussion boards (and no great surprise that he didn't want to pay indefinitely for the servers and bandwidth to host lengthy criticisms).

A common complaint of the evolution skeptic is 'how can a random process of mutation produce adaptation?'. A superficially satisfactory answer is that Natural Selection introduces a 'nonrandom' component. But for anyone with a mathematical bent (eg the average population geneticist), that is not an accurate use of 'nonrandom'.

The problem then becomes: how to present an argument that requires mathematical conceptions to a public that would run a mile from the first equation?

I would describe it as a sampling process. Even biased sampling (eg NS) is mathematically random. Mutations blindly - randomly - sample accessible 'sequence space', generating variation among survivors. Births sample prior generations' variation, with (NS + drift) or without (drift alone) bias. Either way, variation inexorably declines due to the fundamental distortive property of sampling. The result - fixation of something, elimination of everything else - is 'almost certain' - a near-as-dammit 'nonrandom' result from a mathematically 'random' process, irrespective of bias. The tendency to 'freeze' is, of course, countered by the continual drip-feed of new mutations.

Larry Moran said...

So why do popularizers of evolution got to such great lengths to insist that evolution and/or natural selection are not random?

Creationists are fond of characterizing evolution as "random" and that really annoys adaptationists. The adaptationists respond by stating categorically that natural selection is not random.

The adaptationists conveniently forget about random genetic drift, one of the main mechanisms of evolution, and they conveniently forget about the accidental nature of mutation and it's timing, and about contingency and mass extinctions.

Evolution is accidental or unguided. "Random" isn't quite the right word but it's pretty close.

Shawn said...

...but of course, you know why evolution theory is different...much more so than even the average godly person understands it themselves! They have been misled repeatedly but cannot accept that. To do so, would be like knocking out a load bearing wall in a house...risky to downright calamitous.

Atheistoclast said...

Larry is right. Genetic drift is indeed random, and also regarded as very important in the evolution of gene duplicates because there is an appreciable relaxation of selection. It is believed that drift allows evolution leeway to mess things about which it couldn't do in the case of singleton genes.

But all this is bullshit and Larry knows it. The outcome of random and accidental processes is just what we would expect from a monkey working on a typewriter - gene duplicates with premature truncations and big deletions, pseudogenes and, dare I say, some non-functional "junk".

Larry can't have it both ways. He can't have his cake and eat it too.

konrad said...

I don't understand what is being discussed here - what is meant by the distintion between "guided" and "unguided"? If an apple falls to the earth under the influence of gravity, is this a guided or an unguided process? Would the apple fall differently if it changed from guided to unguided or vice versa?

Anonymous said...

The falling apple would have an unguided fall, but the path taken is strongly biased by the force of gravity.

If the apple fell due to a specific action of a person trying to eat the apple, or the great god of Applefall willing it to fall, etc then it is guided event.

jeff said...

"Looking all the evidence used to reconstruct the probable history of life we see no evidence whatsoever that it was guided"

Can you point me to some of that evidence, particularly something for a non-biologist? I am curious as how we could tell if a deity was loading the dice. Chaos tells us a very small change can have large effects and I would expect that biological evolution is a chaotic dynamical system. So a deity who occasionally mucked with the probabilities would be undetectable and could effectively guide the biosphere.

This is what NASA does to guide spacecraft. They make use of the chaotic but predictable nature of orbital dynamics and apply very small thrusts, thus saving fuel, at carefully chosen times to guide their spacecraft where they want them. If you observed the motion from afar you might not be able to detect the thrusts and could conclude the spacecraft were unguided but still somehow ended up in interesting places.

Is a deity who loads the dice the "cheater diety" mentioned above? I googled the phrase with no success. Any references?

To tell if the dice is loaded you need enough samples to fill out the probability distribution. The state space of the genome is so huge that I can't imagine we could ever get enough data to test how fair the evolutionary dice are.

Imagine you have 23 dice with different colors. If you could roll the dice once a second, it would take the age of the Universe to fill out the probability distribution. Of course you could assume the dice are independent and test each one in minutes. Is this what biologists do? Assume independence in some way?

Larry Moran said...

jeff asks,

Can you point me to some of that evidence, particularly something for a non-biologist? I am curious as how we could tell if a deity was loading the dice.

Hmmm ... let me see if I understand you correctly.

I say that we don't see any evidence that life was guided.

You agree with me that we don't detect any evidence of guidance.

What's the problem?

The problem is that you can IMAGINE a situation where god exists but is so clever that his handiwork can't be detected. Isn't that true of just about everything?

Here's an example. We have no evidence that the spaghetti monster exists and no evidence that it steals meatballs. However, it's possible that it exists but is just very devious and clever. It never leaves any trace of its existence and it only steals meatballs that won't be missed.

Do you believe in the spaghetti monster? Does this form of argument convince you that it exists?

Of course not. So why in the world would you even bother imagining a entity that preferred to be completely undetectable and had the power to remain undetectable? Where in the world did you even come up with the idea that evolution is guided if god is so sneaky. Did he slip up somewhere?

jeff said...

I am wondering what the scientific limits are on the abilities of a deity. Since we have no scientific evidence for any deity, then if such a deity exists we can assume they are limited, by choice or abilities, to only influence the biosphere in scientifically undetectable ways. My question, I guess, is what could such a deity do or not do?

They couldn't prevent the evolution of flying creatures. My naive understanding is that flying has independently evolved several times. So if we find a biosphere on an extraterrestrial planet that has sufficient time to evolve we expect to find flying creatures. (How long? Do we know the timescale where we expect to find flying creatures?) If a biosphere had no flying creatures, then either a deity has prevented their evolution or our (probably my) understanding of evolution is flawed. As scientists, we assume the latter and would try to improve our science. But do we expect to find butterflies or similar creatures with gossamer wings with colorful and interesting patterns? Would the lack of butterflies tell us anything about how correct our view of evolution is, or would we chalk it up to chance?

Thinking about guidance has led to me think about these kind of questions: what is expected and what is random and what are the timescales? These seem like valid scientific questions and I would be pleased if you could direct me to something to read.

And I agree that positing a deity who remains scientifically undetectable, while not inconsistent with science, would probably not satisfy anyone who believes in a traditional deity. But if someone was happy with such a deity, it would have no conflict with science. Just like the FSM.

Anonymous said...

You did great until "quantum mechanics"...