Evolution Not Random

[Skeptic Ginger]

Skeptgirl, the reason that I dislike dsecrbing evolution as nonrandom is that it implies that there was an inevitability about the evolution of the particular species that we see today.

And is there evidence yet this is not the case?

I see evidence it is the case. The flying bat and flying birds both fly by 2 wings. The Thylacine resembles predators which evolved via a separate lineage from Thylacines. There are a limited means of locomotion, of vision, of consumption, circulatory systems, muscle systems, digestion, and excretion.

If we had ten absoultely identical Earths, just before the KT impact and let evoultion run its course; due tothe fact that quantum events do seem to be truely random, and chaotic systems magnify these effects to affect macroscopic events, then you would get different species 65MY later in each "version". There might be humanity, or something akin to it, or there might not. It was not inevitable that our environmental niche was going to be filled.

Rather than looking at your example, let's ask a couple different questions. We've already said there is a random component in evolution. The question is how significant is that random component to the outcome, and does mijo's mathematical model describe evolution? I have already answered that mijo's model only describes a small piece of evolution and is not a model of actual evolution any more than a model of reproduction is a model of evolution.

So the question here is then, how predictable and how random is evolution? Is the question just a matter of ideology, or is it a matter of evidence?

First I say, forget the hypothesis we wouldn't be here if it weren't for the random event of an asteroid hitting the Earth 65 million years ago. It's just someone's hypothesis. We seem to so readily accept such hypotheses as facts when they are no such thing.

Would human intelligence have evolved if the asteroid impact had not occurred? From the evidence it is more likely than not that human intelligence would have evolved eventually. Perhaps the timing would have differed. What's a million years give or take in a 14.5 billion year old Universe and on a 4.5 billion year old Earth? Not much.

Which features of humans would have predictably evolved and which features would have been random? I suggest the evidence points to hands and brains as predictable results of evolution. Those in turn result in technology.

But there are no technological societies within radio wave range of the Earth. Maybe not but it is a very large and very old Universe relative to our short time spent in the technological era. Considering the vastness of the Universe, and the variety of conditions we have detected just in our solar system, and considering what we have recently discovered about how nonrandom the evolution of life really is, the hypothesis life will evolve anywhere in the Universe the conditions allow it is supported by the most evidence.

For so long we've been operating under the hypothesis life on Earth is so unique that it could have evolved only rarely, and development of technology must be even more rare, we've just come to accept these beliefs as if they were as factual as the laws of physics. That is as ignorant as the belief humans are more than one step further along on the evolutionary path than the other great apes. We have been raised on these egocentric assumptions as if science actually supported the conclusions.

Nonsense! Get a telescope. Look up into the night sky. You can see billions of stars and trillions of galaxies each with billions more stars in them. Planetary systems are as common as dirt. And abiogenesis and evolution are as predictable as the rate molasses runs downhill on an Earth-mass body.

Would life evolve anywhere in the Universe that the initial conditions existed? I would conclude the evidence is a definite yes on that one. Read up on the current research and hypotheses on abiogenesis. You can draw almost no other conclusion. (Unless of course, you don't read up on the current state of the science of abiogenesis.) Would intelligence and technological species evolve given the conditions? Of course they would. Even if an asteroid calamity sped the process up a tad on Earth, that is not evidence it would not have eventually occurred.

Instead of sticking with old hypotheses and old premeses, take a look at the state of the evidence today. It isn't ideology, it is evidence that supports the conclusion evolution is not a random process. Why do you think the research cited in the OP specifically makes that statement? Because they specifically proposed a question which answers that question. If evolution is random, X will occur. If evolution is not random, Y will occur.

Well, Y occurred.

 

[Kotatsu]

I've heard all the arguments that 'primitive' organisms are more complex. I understand that primitive does not denote time or genetic complexity from the original organisms which evolved. I realize what might be designated as primitive might have more genes than a 'less primitive' organism.

All those concepts are separate issues. A mammal eye is less primitive than a cell which only detects light.

No worries^^. It's just that in a debate on the semantics of "random", you suddenly use another of those words which means one thing to laymen and another thing to some experts, and yet another thing to other experts (and so on). I find it funny, but understand in what way you used it.

 

[Fredrik]

There are too many posts in this thread for me to read them all, so it's possible that all of this has been said already...

I just want to say that I find it very strange that people in this thread are claiming either that "evolution is random" or that "evolution is non-random", when they know that evolution in fact has two components, one of which is 100% random and the other essentially non-random. Why not just say that the question of whether evolution is random doesn't really make sense? I wouldn't answer that question with "yes" or "no", and I don't see why anyone would want to. I also don't see how anyone can claim that one of those answers is correct and the other is wrong.

If I had to choose one, I'd go with "yes", because I feel it's closer to the truth (because of how mathematicians define thinks like "random variable"), but in a discussion with a creationist I'd probably still say "no", because this "lie" would get them closer to the truth than the truth would.

The strangest claim I've seen in this thread is the claim that evolution can also be described as "deterministic". To me this is just flat out wrong. If evolution had been deterministic, than it would be possible (in principle) to determine exactly how the species will evolve over, say, the next 50 million years, and it clearly is not. Someone said (roughly) that it's correct to use the word "deterministic" because the outcome is inevitable. This is definitely false. "Deterministic" means that exact knowledge of a physical system at a certain time is sufficient to determine exactly what it will be like at any point in the future. This is not the case with evolution. Exact knowledge about the world right now would only be sufficient to determine roughly what sort of things we can expect to see in the future. That's not determinism.

 

[cyborg]

This is definitely false. "Deterministic" means that exact knowledge of a physical system at a certain time is sufficient to determine exactly what it will be like at any point in the future.

No it doesn't.

Look up Rule 30.

 

[Meadmaker]

Rather than looking at your example, let's ask a couple different questions. We've already said there is a random component in evolution. The question is how significant is that random component to the outcome, and does mijo's mathematical model describe evolution?

Sometimes people have discussed poker in this context. How significant is the random element in poker, vs. the non-random element?

Without the random element, you couldn't play the game.

Would human intelligence have evolved if the asteroid impact had not occurred? From the evidence it is more likely than not that human intelligence would have evolved eventually. Perhaps the timing would have differed. What's a million years give or take in a 14.5 billion year old Universe and on a 4.5 billion year old Earth? Not much.

In humans? I don't think you would find many experts in the field who would agree with this statement. The asteroid completely changed the availability of ecological niches and altered the course of evolution completely. By emptying out those ecological niches, it created opportunity for the mammals that survived the impact and its aftermath.

I suggest the evidence points to hands and brains as predictable results of evolution.

Hands and brains are predictable. Humans are not.

Why do you think the research cited in the OP specifically makes that statement? Because they specifically proposed a question which answers that question. If evolution is random, X will occur. If evolution is not random, Y will occur.

Well, Y occurred.

Have you read the research cited in the OP? It doesn't make that statement.

 

[sol invictus]

Take a green and red polka dot shirt. Is it green? Is it not green?

We seem to all agree there is a random component to evolution, namely mutations. Then there is natural selection, which is chaotic (and hence unpredictable) in its short-term details, but roughly predictable in its long-term outcome (for example wings and eyes would probably evolve in the natural environment, but the ability to digest plastic would not).

Does anyone disagree with any of that? Because if not, what is this conversation about?

 

[Belz...]

That particular event is not chaotic, and nothing is acausal, but it is random.

If it's causal then it isn't random, is it ?

There is a physical law that governs the probability the photon will be emitted, but there is no physical law that tells you with certainty what will happen. Furthermore there cannot be such a law, for a deep reason that would take us very far off track.

It works almost the same way, every time. The uncertainties in the earth's orbit due to quantum mechanics are mind-bogglingly (is that a word?) and undetectably small, but they are real.

Well, I'll take your word for it, but I'll need more information on this.

Links would be appreciated, if you've got the time.

 

[sol invictus]

If it's causal then it isn't random, is it ?

Causality has a specific meaning in physics, and QM is consistent with it. QM allows you to predict probabilities, and causality means those probabilities don't change without a cause (and in relativistic theories, that the cause must be in the past lightcone).

Well, I'll take your word for it, but I'll need more information on this.

Links would be appreciated, if you've got the time.

Hmmm... well, anything on QM will talk about this. You might start with wiki:

http://en.wikipedia.org/wiki/Quantum_mechanics#Theory
http://en.wikipedia.org/wiki/Quantum_mechanics#Philosophical_consequences

As for why there cannot be a law that makes predictions about precisely when the photon is emitted:

http://en.wikipedia.org/wiki/Bell's_theorem

 

[Fredrik]

No it doesn't.

Look up Rule 30.

Yes it does, and why do you mention rule 30? I looked it up, and that system is obviously deterministic, but that doesn't in any way imply that evolution is.

 

[Belz...]

Was it inevitable that a tool-using social biped would have evolved 65 millioon years after the KT impact? No. (See the discussion about quantum events causing macroscopic effects in nonlinear systems).

Assuming that quantum events have NO impact on macroscopic events, then the universe is entirely deterministic and there was no other way any of the events could've happened any other way.

Assuming QM events DO have an impact, then sure, there's a level of randomness in the events.

However, the selection itself is still non-random.

I would say that (because of nonlinear amplification of quantum uncertanties, that truly are random, it is in principle impossible to predict the weather, far enough ahead. However it is still possible to predict climate.

Re quantum effects, as I said I'll have to look into it, because last I checked they didn't have an noticeable effect.

 

[Belz...]

Causality has a specific meaning in physics, and QM is consistent with it. QM allows you to predict probabilities, and causality means those probabilities don't change without a cause

Then how are they "random" in the absolute sense ? Sorry for my insistence, but I genuinely don't understand: if something happens strictly because of something else, how is it random ?

Hmmm... well, anything on QM will talk about this. You might start with wiki:

http://en.wikipedia.org/wiki/Quantum_mechanics#Theory
http://en.wikipedia.org/wiki/Quantum_mechanics#Philosophical_consequences

As for why there cannot be a law that makes predictions about precisely when the photon is emitted:

http://en.wikipedia.org/wiki/Bell's_theorem

Okay, I'll look into it in a couple of hours, and then I'll wonder why I never spotted this before.

 

[sol invictus]

Then how are they "random" in the absolute sense ? Sorry for my insistence, but I genuinely don't understand: if something happens strictly because of something else, how is it random ?

Please re-read what I wrote before. Probabilities change causally, but they remain probabilities. The outcome of a given experiment is not predictable - only the probability of obtaining that outcome is predictable. However if you perform the same experiment many, many times, the average result will converge statistically to something you can compute.

Roughly speaking that convergence is why we don't notice quantum uncertainty ordinarily - even though everything in the world is quantum, the objects we interact with are composed of an enormous number of particles, and their collective behavior is almost perfectly predictable.

 

[hammegk]

.... But as we discover more and more about evolutionary processes, there is evidence accumulating that there is a lot more determinism in the system than there is randomness.

You must be wrong if you are Dawkin's disciple.

Other choose (so to speak) to believe Dawkins is missing something, which results in all our ongoing controversies here and in the world at large.

 

[Belz...]

Hmmm... well, anything on QM will talk about this. You might start with wiki:

http://en.wikipedia.org/wiki/Quantum_mechanics#Theory
http://en.wikipedia.org/wiki/Quantum_mechanics#Philosophical_consequences

As for why there cannot be a law that makes predictions about precisely when the photon is emitted:

http://en.wikipedia.org/wiki/Bell's_theorem

Okay...

But all those links talk about is the wavefunction, which, if I understand it correctly, is just that we don't know where the particle is. It doesn't mean the particle occupies the entire wavefunction. We Just don't know. QM uses probabilistic calculations because they are practical and because we can't know all the values of all the variables. But in reality, again if I understand it correctly, they have a set value which is precisely the result of the previous set value, and so on. That's clockwork, non-random.

When I think "random" in the absolute sense, I think about quantum fluctuations that, say, create particle-antiparticle pairs from nothing, or an entire universe.

Please re-read what I wrote before. Probabilities change causally, but they remain probabilities. The outcome of a given experiment is not predictable - only the probability of obtaining that outcome is predictable. However if you perform the same experiment many, many times, the average result will converge statistically to something you can compute.

Unless my understanding of QM and of the links your provided is wrong, if the experiment happens to have the exact same values to the exact same factor (something that doesn't happen in real life, of couse), the outcome will be the same. Always. That's why I said the universe is entirely deterministic. From the moment that the laws of physics were set, there was only one possible outcome.

 

[Meadmaker]

Re quantum effects, as I said I'll have to look into it, because last I checked they didn't have an noticeable effect.

Quantum effects don't have a "noticeable" effect on weather, but they do have an effect. Consider the following scenario:

A photon, caused by a quantum effect, hits a man on the forehead. Unbeknownst to him, it messes up the DNA in his skin, and cancer starts.

After noticing a mole on his forehead, he goes to the doctor. It is biopsied, and there is cancer.

Leaving the office after hearing the results, concerned, he decides to call his son. His son is sitting in a sidewalk cafe, but he has his cell phone. It is sitting on the table. It rings.

Moments earlier, a butterfly had landed on the phone. Startled, it flaps its wings and flies away.

Two weeks later, it rains in New York.

This is a string of events made up of things we have already acknowledged (or will after you finish reading.) The photon really was random, in every sense of the word. A single photon can cause cancer. Butterfly wings can affect the weather. All I did was add the sequence that links cancer to butterfly wings.

Of course, no one could connect any of those events to each other in a "noticeable" way. No one could say which photon caused the cancer, or which butterfly flap made it rain, but the theory is well established. Any perturbation of the system, no matter how small, can have macroscopic effects.

 

[sol invictus]

But all those links talk about is the wavefunction, which, if I understand it correctly, is just that we don't know where the particle is. It doesn't mean the particle occupies the entire wavefunction. We Just don't know. QM uses probabilistic calculations because they are practical and because we can't know all the values of all the variables. But in reality, again if I understand it correctly, they have a set value which is precisely the result of the previous set value, and so on. That's clockwork, non-random.

Your understanding is incorrect. The wavefunction contains all the information about the particle's position (to use that example) that it is possible to have. There cannot be a so-called hidden variable (at least not without giving up really basic things, like locality); that's what the third link I gave is about. Any theory that contains more information about the position makes predictions which are inconsistent with those of QM, and which have been ruled out experimentally.

You might also look up "double slit experiment". A single particle can interfere with itself, because its wavefunction is spread over more than one location. No theory where the particle is in fact localized (and we just don't know where) can possibly behave that way.

 

[Meadmaker]

Okay...

But all those links talk about is the wavefunction, which, if I understand it correctly, is just that we don't know where the particle is. It doesn't mean the particle occupies the entire wavefunction. We Just don't know. QM uses probabilistic calculations because they are practical and because we can't know all the values of all the variables. But in reality, again if I understand it correctly, they have a set value which is precisely the result of the previous set value, and so on. That's clockwork, non-random.

That is not the opinion of most physicists. What is known, with absolute certainty, is that one can never know the hidden variables that control the set value. The general feeling among physicists is that there is no set value, although that is both unproven and unprovable.

 

[Belz...]

Your understanding is incorrect. The wavefunction contains all the information about the particle's position (to use that example) that it is possible to have. There cannot be a so-called hidden variable (at least not without giving up really basic things, like locality); that's what the third link I gave is about. Any theory that contains more information about the position makes predictions which are inconsistent with those of QM, and which have been ruled out experimentally.

You might also look up "double slit experiment". A single particle can interfere with itself, because its wavefunction is spread over more than one location. No theory where the particle is in fact localized (and we just don't know where) can possibly behave that way.

I'm aware of that experiment.

But the uncertainty principle exists specifically because observing a particle interracts with it and causes it to change. The particle isn't at more than one place at any given time. And barring interraction with another particle, it isn't going to change its momentum.

Re the wavefunction: it contains all the possible locations of the particle, but obviously, since we don't know where the particle IS, there IS a hidden variable: the location of the particle.

That is not the opinion of most physicists. What is known, with absolute certainty, is that one can never know the hidden variables that control the set value. The general feeling among physicists is that there is no set value, although that is both unproven and unprovable.

I'm aware of that, as well. However, I wasn't talking about practical randomness, but true randomness. The fact that we don't have all the data doesn't mean it isn't there.

 

[Meadmaker]

I'm aware of that, as well. However, I wasn't talking about practical randomness, but true randomness. The fact that we don't have all the data doesn't mean it isn't there.

It isn't just that we don't have all the data. It means that the data not only cannot be measured for practical reasons, it means that no amount of measurement, by any system, conscious or unconscious, technologically advanced or otherwise, can ever find that data.

The system will behave as if it is truly random. One possible reason that it will behave that way is that it is, indeed, truly random. The general opinion of most physicists is that there is no reason to assume anything else. Most of them say that it is truly random. The rest say it might be truly random. You won't find a reputatble physicist anywhere that says it actually works, "like clockwork". The closest you might find is that we can never know whether or not it works like clockwork, and that to the limits of experimental verification, even in principle, it does not work like clockwork.

ETA: And I'm not even certain that I might be being overly generous to the "clockwork" hypothesis. The implications of the Bell's Theorem experiment make my head hurt. As best I can tell, it doesn't completely rule out clockwork, it just rules out knowing the clock mechanism or determining if there is any clockwork at all. It might be somewhat stronger than that, ruling out even the possibility of clockwork. I bow to Sol's greater knowledge on the subject.

 

[Fredrik]

The particle isn't at more than one place at any given time. And barring interraction with another particle, it isn't going to change its momentum.

Re the wavefunction: it contains all the possible locations of the particle, but obviously, since we don't know where the particle IS, there IS a hidden variable: the location of the particle.

This isn't true at all. What you're saying here (or at least statements that are equivalent to yours in every way that matters) have been proven wrong by experiments. Your claim contradicts quantum mechanics, but the experiments don't. It is completely false to say that the particle has a specific location. The wave function is the best possible answer to the question of where the particle "is". It's not just that we use a wave function when we we don't have complete information. The wave function really is all the information.

Let's assume for the moment that your view of QM is correct. It's possible to show that this implies that something called Bell's inequality must hold. What the experiments have shown is that nature violates Bell's inequality. This is what proves that your view is wrong.