Randomness in Evolution: Valid and Invalid Usage

[zosima]

You are missing the point entirely. The processes of evolution, mutation and natural selection, are random while the result of evolution, adaptive optimization to the environment, is not. Similarly, the process that make an ionization smoke detector work, radioactive decay, is random, the result, the actual functioning of the detector is not.

As I said in my previous post, we already went over this. I just wanted to clarify what you were saying about use of statistics. With the definition you're supporting now, evolution is non-random just as a smoke detector is non-random. There are definitely random components to evolution on the small scale, but just like the smoke detector they smooth out on big scales. So we're completely in agreement. I'll even admit the likelihood of selection of an individual is probabilistic in some cases, but once you get up to a very modest number of individuals that is not the case.

No, convergence does not strengthen your case that evolution by natural selection is non-random. The observation convergence happens whether or not the underlying processes are random; otherwise the functioning of an ionizing smoke detector would prove that radioactive decay was non-random, which contradicts most of quantum mechanics.

The insistence that convergence make a system non-random is what I was referring to, and, as mentioned above convergence does not imply that the underlying mechanism is non-random.

Well, as Belz noted you seem to be shifting your argument, but I'll definitely admit convergence doesn't prove that the components of a system are non-probabilistic. If you thought I meant that it did,then I'm sorry I didn't spell it out in greater clarity. But as I mentioned above, convergence is purely superlative. It strengthens the case that evolution has clearly defined rules and tendencies, that at evolutionary scales details smooth out. Also, if you feel like you were saying this the whole time, then I apologize for the misunderstanding. As I've mentioned above, it can be quite difficult, nigh impossible, to read your mind.

Even though extinction events are rare, they have been instrumental in determining the course of evolution on this planet. I am also not sure about the long-term stability of ecosytems either. We are only about 10,000 years from the end of the last Ice Age, where environments were significantly different.

Well I wouldn't really call things like ice-ages and volcanic eruptions random. I mean a glacier or volcano is about as random as a mac truck. That said I fully agree that they represent important historical contingencies, noise in the system that is crucial for making the species on this planet what they were and are. Without them, things would definitely be different.

Also, I think the speed with which traits spread through a population is often underestimated, so its easy to imagine that things are more 'random' than they really are. The spread of a trait actually follows an exponential function so in a very small number of generations everyone in a population that randomly interbreeds should have a beneficial trait(or at least the population should be at Hardy-Weinberg equilibrium). I showed(roughly) in another thread that if a trait that doubled an individual's rate of reproduction, if it showed up at a rate of one in one million, and if the population had a stable 10 billion individuals, that the trait would be in every individual after only 17 generations. And humans actually have relatively long generations, the generations of insects or bacteria are so short that a beneficial trait could spread to all in months or years. (You may quibble about the numbers, but the function itself will always be exponential in form)

Earlier than that, it seems that the Toba eruption 70,000 years ago reduced the human population dramatically (figures of between 1,000 and 10,000 breeding pairs). A quick google search suggests that this level of poplulation would qualify for "endangered species" status. Obviously humanity survived that event, but with population levels as low as that, chance begins to become more significant.

Two significant events within 100,000 years seems like quite a high rate to me, but again it depends on whether you are interested in the similarities or the differences.

I think that would definitely increase the probability that a random event could affect the course of a species, but its still not very likely. Even 1,000 individuals is enough that any statistical survival of an individual would translate into a predictable fraction of the 1000 individuals surviving some evolutionary pressure.

Now there is a question of whether the Toba event actually randomly culled certain segments of the population. I don't know for sure, I'm not really familiar with Toba, but from what I read online its effects were pretty predictable. It killed all other human species, with the exception of H.sapiens and H. neanderthalensis
Selectively, specifically, killing certain sub-populations that lack adaptive traits isn't random.

If you are talking about the evolution of humanity, which is what many people are most interested in, then random (selective) factors have been very important. This would also be the case for virtually any other particular species.

Once the hominids had begun to emerge, there might have been an inevitibility about the emergence of something akin to humanity. However I would contend that this emergence and the "opening" of the niche was a result of the environment interacting with appropriate precursor animals at the right time, and with the right traits.

Something was going to evolve, and whatever did evolve, it would have shown amazing levels of adaptation to the environment, and modulated the slective environment for other organisms. Just because humanity evolved, doesn't mean that anything special happened.

I'm not trying to say that humans or any intelligent species must necessarily evolve. I'm not trying to say that there is anything special about human existence. If the KT event had not happened, we probably wouldn't be here, or at least, not in the way that we are. But I think its a mistake to call events like that random. You're talking about 1000s of tons of rock moving with the inevitability of huge inertia, huge momentum. If you ask me, thats about as non-random as you can get.

 

[jimbob]

except that the survival must also be a result of interactions with weather events, which I think we do agree are random.

 

[jimbob]

And my other point is that with events such as the ice age ending, and near extinctions that there are new environmnets that could be exploited, so one should see an increase in the evolutionary ratre afterwards, and this initial setup would be influenced by the random events.

Like a river's course shifting chaotically, or a drunkard's walk.

 

[mijopaalmc]

zosima and Belz...-

My point is that the orderly operation of an ionization smoke detector is not non-random. The only reason that it works that the americium-241 produces enough ionizing radiation to sustain a detectable and steady voltage. It simply wouldn't work predictably if you used uranium-238 instead.

 

[Belz...]

But radioactive decay is random, right Mijo ?

 

[zosima]

except that the survival must also be a result of interactions with weather events, which I think we do agree are random.

And my other point is that with events such as the ice age ending, and near extinctions that there are new environmnets that could be exploited, so one should see an increase in the evolutionary ratre afterwards, and this initial setup would be influenced by the random events.

Like a river's course shifting chaotically, or a drunkard's walk.

Thats a fair position to take and I'm perfectly willing to agree to disagree. That said I tend to think that even things like weather smooth out in to factors that act more or less deterministically when you're talking about evolutionary time scales. I also think that scenarios where the fate of a species depends on a single weather event or one summer's melt-water end up looking like very elaborate contraptions rather than solid hypotheses. So, to me, whether or not they ever happen; whether or not they're important, seems akin to asking whether or not quantum tunneling is an important factor to consider when determining the position of my cat. That said I think we've both made these exact same arguments in previous posts, so I'm ready to leave our dispute standing.

Mijo:
Last Post:

My point is that the orderly operation of an ionization smoke detector is not non-random.

2 Posts Ago:

Similarly, the process that make an ionization smoke detector work, radioactive decay, is random, the result, the actual functioning of the detector is not.

So the behavior of a smoke detector is not non-random, but a smoke detector's behavior is not random? Two posts. Complete contradiction. I'm glad you're up to form. I guess its time to redefine 'non' in the next post.

an ionization smoke detector...simply wouldn't work predictably if you used uranium-238 instead.(Of Americum-241)

Good point, let me just add that:
If trucks had Jello for wheels we wouldn't use them for shipping.
If birds had no wings they would be unable to fly.
If my hands were made of string-beans I wouldn't typing.
If light were pasta it would travel much slower.
......
Seriously I think this argument is sooooooooooooooooooo done.

 

[jimbob]

Thats a fair position to take and I'm perfectly willing to agree to disagree. That said I tend to think that even things like weather smooth out in to factors that act more or less deterministically when you're talking about evolutionary time scales. I also think that scenarios where the fate of a species depends on a single weather event or one summer's melt-water end up looking like very elaborate contraptions rather than solid hypotheses. So, to me, whether or not they ever happen; whether or not they're important, seems akin to asking whether or not quantum tunneling is an important factor to consider when determining the position of my cat. That said I think we've both made these exact same arguments in previous posts, so I'm ready to leave our dispute standing.

Fair enough..
I would actually take a different view as to timescales though.

Two generations:
Little difference between the two, we agree here I think.
ETA: Many potentially beneficial traits won't survive at this stage, but enough overall will provided the population survives...

many generations in stable environment (thousands of years+)
There will be adaptations to the stable parts of the environment, e.g whatever predators there are in a hot dry desert, any prey will still be adapt ed to the heat and aridity.

Many features will be predictible, i.e. randomness would be less important


Geological timescale:
Large unpredictible events are likely to occur within this timescale and set things partly to "year zero", when a different set of random patterns are likely to emerge. There would have been no way to predict what would have arisen 2million yars after the end of the Jurassic, ditto the Cretaceous, or the Cambrian. I would argue that this would not be simply due to a lack of information but to the essentially random nature of the initial evolutionary directions in the initial new environment.

In other words I am saying that the physical environment will have a "fitness" landscape which might be fairly stable, but the interactions between the organisms in the new environment is plastic, and is so until they begin to fill the niches. ETA: when randomness becomes less important again though the "directions" have been partly "set" randomly. (once there is a large carnivore, and a large herbivore, the herbivore is not going to evolve into a competing carnivore.

 

[zosima]

Fair enough..
I would actually take a different view as to timescales though.

Two generations:
Little difference between the two, we agree here I think.
ETA: Many potentially beneficial traits won't survive at this stage, but enough overall will provided the population survives...

many generations in stable environment (thousands of years+)
There will be adaptations to the stable parts of the environment, e.g whatever predators there are in a hot dry desert, any prey will still be adapt ed to the heat and aridity.

Many features will be predictible, i.e. randomness would be less important


Geological timescale:
Large unpredictible events are likely to occur within this timescale and set things partly to "year zero", when a different set of random patterns are likely to emerge. There would have been no way to predict what would have arisen 2million yars after the end of the Jurassic, ditto the Cretaceous, or the Cambrian. I would argue that this would not be simply due to a lack of information but to the essentially random nature of the initial evolutionary directions in the initial new environment.

In other words I am saying that the physical environment will have a "fitness" landscape which might be fairly stable, but the interactions between the organisms in the new environment is plastic, and is so until they begin to fill the niches. ETA: when randomness becomes less important again though the "directions" have been partly "set" randomly. (once there is a large carnivore, and a large herbivore, the herbivore is not going to evolve into a competing carnivore.

That definitely clarifies the point you're making. I'd appreciate it if you would elaborate. Why would we expect a random path to be taken after a major ecological event? While the change would be quick, dramatic, and quite probably unpredictable, why does a random element also come in? Can you give an example of a random scenario after a large event?

 

[articulett]

He thinks the unpredictable event IS "random" and this makes selection random... because he's looking at from before the event occurs. I know it's garbled, but to him it makes sense.

A scientist would say, a creature evolves to fit it's environment... and including the assorted events in that environment... if they do not, they perish. If you belong to a species that can't do well when an asteroid hits the part of the planet, you die out... other life that was better adapted survives and evolves in your place.

We see all of the environment including tornadoes and such as "part of the environment" that "selects"-- part of the culling process winnowing what evolves. Jim and Mijo teleport themselves back in time before any "hypothetical" events so they can call selection "random" just like they call mutation and then conclude that "evolution is random" by virtue of being probabilistic (whatever the hell that means).

They speak of mutation in regards to evolution as a concept... but in order to focus on unpredictability and "randomness" they turn to specifics when it comes to selection--suddenly we are out of the realm of abstraction and into the realm of specific hypotheticals just so they can feel like it's meaningful to call evolution random. It's a true mind numbing game, zosima-- just like mijo's very elastic definition of random. You will feel like you are being drawn into crazy land, and there is no exit.

And they will do this as long as they both have an audience... and a couple of more of same folks will regularly chime in with meandering off topic pedantry to confuse the issue. It's not you. And it's not fixable.

 

[jimbob]

Whilst articulett thinks that aircraft evolved*...

I don't like consistently attacking what I think is someone's view (and unlike articulett, I actually don't have anyone on ignore) rather than their arguments, so I'll leave my comments at that for now.

ETA:
I'll explain the rest of my point later, but interactions between organisms whilst being hard to model in their completeness, do show chaotic behaviour, indeed we made a simple model of one type in our physics lectures when I was at university.

In the new environment, the "interactions" landscape is fairly flat, so, just as a river's drainiage system and its initial course is chotic, and thus significantly affected by quantum events, so the interactions between the emerging (and initially less-well adapted) organisms is going to be too.

At the beginning of unicellular life there was no inevitibility about the evolutionof any ecology like ours, it wasn't merely unpredictible but that it was not yet determined, as it would be significantly affected by quantum events. I am unsure whetther articulett would even accept the question as to whether ecosystems like ours were inevitable at this stage, or even later.



*maybe OK in a general diacussion, but this "development" is not similar to biological evolution, although William Dembski (prominanet ID proponent) likes to equate them.

The evolution here would be reconceived not as blind evolution but as technological evolution. Nor would it be committed to Darwin’s idea of descent with modification. But, hey, it would still be evolution

 

[cyborg]

What the **** is your obsession with quantum events all about? Seriously.

 

[mijopaalmc]

What the **** is your obsession with quantum events all about? Seriously.

Uh....they provide the basis for the variation that is necessary for evolution to occur, and, even though they are, to the best of humanity's collective knowledge, random, they converge to describe the deterministic behavior of Newtonian mechanics.

 

[cyborg]

Uh....they provide the basis for the variation that is necessary for evolution to occur,

Uh no.

and, even though they are, to the best of humanity's collective knowledge, random, they converge to describe the deterministic behavior of Newtonian mechanics.

Dare I state the simple fact that variation is not dependant in any way whatsoever on "randomness"?

 

[mijopaalmc]

Uh no.

Uh....mutation, which provides the variation for the process of evolution observed here on the planet Earth, is a quantum mechanical phenomenon and therefore, to the best of our knowledge, random.

Dare I state the simple fact that variation is not dependant in any way whatsoever on "randomness"?

You could certainly say that, but you would be dead wrong because of the reason given above.

 

[cyborg]

Uh....mutation, which provides the variation for the process of evolution observed here on the planet Earth,

Goal-post shift.

Answer the question:

It is a necessary property of variation that the variation is random in all cases?

 

[mijopaalmc]

Goal-post shift.

Answer the question:

It is a necessary property of variation that the variation is random in all cases?

I see the problem here is that you want to redefine the parameters of the discussion in mid-stream. We have always discussed evolution as the process that we have observed here on Earth, not some "abstraction" that covers all possible cases of evolution. That is why the discussion has been predominately about genes, phenotypes, and natural selection. In essence, it is you who is shifting the goal posts to talk about variation in general and selection in general.

And no, neither variation in general nor selection in general for evolution in general to occur, but mutation and natural selection can both be random, and evolution by natural selection will still occur.

 

[cyborg]

I see the problem here is that you want to redefine the parameters of the discussion in mid-stream.

No. The problem is that if you define something as being necessary for evolution then it implies that if it does not have that property then the conditions for evolution are broken.

Abstract or not this applies: if it is merely variation only that is necessary for evolution then the type of variation is not contigent.

Theories also tend to have to cope with the abstract because of this little thing called "the prediction" - e.g. what is the prediction of the theory when the nature of variation is changed?

Continually stating things such as "QM is random, QM is small, therefore QM rules them all!" isn't interesting or useful in the least.

 

[mijopaalmc]

cyborg-

Does natural selection divide populations into two mutually exclusive groups of phenotypes: one where all of the individuals who possess its constituent phenotypes always survive and the other where none of the individuals who possess its constituent phenotypes ever survive?

If it does, then evolution by natural selection is deterministic.

If it doesn't, then evolution by natural selection is random.

 

[Dancing David]

Uh....they provide the basis for the variation that is necessary for evolution to occur, and, even though they are, to the best of humanity's collective knowledge, random, they converge to describe the deterministic behavior of Newtonian mechanics.

Really?

No they don't.

You can have natral selection through other means, random mutation is not the only way.

There is variation the expression of traits, say the length of a bone. For various reasons the length of the bone will vary, it could be from the genes that appear in five other expressed areas of the genome. So say that each individual has this trait but that the length of the bone will vary by 5% around the mean. So the bone will tend to have x length but will vary by some amount determined by the alleles at the five sites. (Length=x of prior parent varied by +/- 5%)

An indiviual can have a shorter bone or a longer bone or a bone on the mean. Now say that for reasons of contingent history a population of the species becomes isolated from the rest of the population and that for some reason they all have the longer set of the five alleles. So what will happen over time, due to the breeding they will tend to have a similar bone length but thier mean bone length will be higher than the original bone length. They will also have a five percent variation around the mean but thier mean bone length is longer than the original pool.

Now say that there is some selective benefit to having a longer bone, and the isolated population begins to have greater reproduction in the memebers of the population. For a while the population will have greater than 5% variation around the mean , yet over time the population is likely to stablize at a new mean with 5% variation around that mean. But the mean of this isolated population will be significantly higher than the original population.

This does not require random mutation.


Nor does random mutation depend upon quantum mechanics, nor does QM converge to describe Newtonian mechanics.

 

[Dancing David]

Uh....mutation, which provides the variation for the process of evolution observed here on the planet Earth, is a quantum mechanical phenomenon and therefore, to the best of our knowledge, random.

Really?

Mutation is the only way the that natural selection can occur?

Why then do artic animals have white fur? Are you sure that in all cases it is due to mutation?

What if it is from the expression of double recessive traits?

How did mutation become QM?

Copying and other errors can occur at the classical level without QM.