What evidence is there for evolution being non-random?

[Meadmaker]

In what way is it narrow minded to be angry with people for convincing others, deliberately, that an established scientific field is wrong, not with contradictory evidence, but with lies and misinformation?

It isn't.

When did you stop beating your wife?


Dawkins was saying that there is one way in which the term ought to be used, and in so doing he excluded the use of the term that is taught to anyone who studies engineering, advanced mathematics, or the physical sciences. Furthermore, the context made it seem as if those people who use the term in that precise, technical, manner in which they were educated were not merely making a mistake, but they were at best duped by a group of liars, or at worst lying themselves. That strikes me as rather narrow minded.

For more, read Schneibster's commentary, including his comments in post 1000. He's quite good.

ETA: FWIW, my comments about Dawkins and being narrow minded were meant to refer to his discussion of "chance" and evolution, not the book as a whole. The book as whole was better than I expected. If anyone is really interested, they could read my thoughts in the Religion thread that started this whole snowball on the subject of randomness.

 

[Schneibster]

Thanks, MM, I appreciate the compliment.

My take on the overall situation is that the cretinists have managed to create a controversy between the physical sciences and biology, which is precisely what they set out to do. I think it's unfortunate that it has worked this way. I'd like to see it change.

 

[mijopaalmc]

I have a question for all the randomites:

What place do you think the philosophy of science has in discussing the randomness (or lack thereof) of evolution?

 

[ImaginalDisc]

Dawkins was saying that there is one way in which the term ought to be used, and in so doing he excluded the use of the term that is taught to anyone who studies engineering, advanced mathematics, or the physical sciences. Furthermore, the context made it seem as if those people who use the term in that precise, technical, manner in which they were educated were not merely making a mistake, but they were at best duped by a group of liars, or at worst lying themselves. That strikes me as rather narrow minded.

For more, read Schneibster's commentary, including his comments in post 1000. He's quite good.

Where on Earth are you getting this? He means, quite clearly, discussions about the subject of evolution. What random means in other scietific contexts is a completely unrelated issue.

 

[Schneibster]

Where on Earth are you getting this? He means, quite clearly, discussions about the subject of evolution. What random means in other scietific contexts is a completely unrelated issue.

The same argument can be applied to biology, or the common definition. I think it's time to recognize that by putting it this way before this audience, you are misleading rather than informing.

 

[jimbob]

Where on Earth are you getting this? He means, quite clearly, discussions about the subject of evolution. What random means in other scietific contexts is a completely unrelated issue.

But this is wrong information for so many people, when "not disordered" would be correct.

The "commonly accepted" scientific usage of the term random is also used within biology. For example Ronald Fisher.

Genetical Theory of Natural Selection

Fisher was an ardent promoter of eugenics, which also stimulated and guided much of his work in genetics of man. His book The Genetical Theory of Natural Selection was started in 1928 and published in 1930. It contained a summary of what was already known to the literature. He developed ideas on sexual selection, mimicry and the evolution of dominance. He famously showed that the probability of a mutation increasing the fitness of an organism decreases proportionately with the magnitude of the mutation. He also proved that larger populations carry more variation so that they have a larger chance of survival. He set forth the foundations of what was to become known as population genetics.

My italics and emphasis

As Fisher was a founding figure in both statistics and evolutionary biology, one can be fairly sure that he would have used the definition of random shared by the physical scientists and statisticans.

"Nonrandom" is confusing to too many, and being a slight snob here, it is confusing to those who are more likely to be able to understand evolution.

ETA:

Seconded Schneibster's post #1005

Jim

 

[Meadmaker]

Where on Earth are you getting this? He means, quite clearly, discussions about the subject of evolution. What random means in other scietific contexts is a completely unrelated issue.

When used in mathematical models, it means the same thing to biologists as it does to engineers. Not every biologist does mathematical modelling, so a lot of them would use the common definitions.

A fair number of biologists study sampled data, where the phrase "random sampling" is used frequently, and that might cause some confusion. That term, too, has a very specific meaning. It isn't just the word "random" stuck onto another word.

 

[Meadmaker]

I have a question for all the randomites:

What place do you think the philosophy of science has in discussing the randomness (or lack thereof) of evolution?

I'm generally not interested in philosophy of science as such. However, some of the papers you cited looked interesting for their scientific content, not necessarily their philosophical content.

 

[Schneibster]

I have a question for all the randomites:

What place do you think the philosophy of science has in discussing the randomness (or lack thereof) of evolution?

No place at all. This is, in my eyes, purely a technical terminology issue, not a philosophical one.

 

[Ivor the Engineer]

Perhaps the problem randomites have with understanding non-random with respect to evolution is they think the fact a random model can be used to model a real world process means the real world process is random?

Random models are used when we have incomplete information. The more incomplete our knowledge, the less confidence we can have in the estimate a model produces. This does not mean there is anything random happening in the real world.

Everything we measure has a random component to it. In mechanical engineering it's called tolerancing. No engineer thinks of 10mm dia. rod of silver steel as having a random diameter. They implicitly acknowledge that the random component is small with respect to the measurement they're interested in.

In evolution it's the mutation of DNA that has the rapid and large random variation in a population, compared to natural selection, which normally is slowly varying. In fact, when a "random" step-change does occur in the environment, many species die out. Most (all?) adapted species depend on tomorrow being much like today.

Thus it makes sense to emphasize the "random" in mutation and down-play it for natural selection, since that is how the real world works.

 

[mijopaalmc]

I'm generally not interested in philosophy of science as such. However, some of the papers you cited looked interesting for their scientific content, not necessarily their philosophical content.

No place at all. This is, in my eyes, purely a technical terminology issue, not a philosophical one.

What did you think of the quotation that I posted from "The Trials of Life: Natural Selection and Random Drift"?

It, and the rest of the article, address why it was reasonable to describe evolution by natural selection as arising from the statistical nature of populations rather than the dynamical nature of the "forces" of evolution in a way that would intelligible to anyone who had taken an algebra-based statistics course. Now, of course, not every philosopher of science agrees with this position, as is obvious from reading the abstracts of the articles that cite "Trail", but it still seems that if what the authors of "Trials" are right and scientists are using the dynamical and statistical description of evolution by natural selection interchangeably when they are not, then it becomes necessary to address the issue as it would that it would muddy the pedagogy of evolution by natural selection and make it more difficult for the people who most need to understand it to understand it.

 

[Meadmaker]

No engineer thinks of 10mm dia. rod of silver steel as having a random diameter.

Maybe not until they get to graduate school.

 

[jimbob]

Re: Ivor, #1010

There is "random" and "pseudorandom" and most stuff is actually pseudorandom, but quantum-mechanics? See some of the work on entanglement, it really does seem that these events are truely random.

This is one for the "free-will vs determinism debate" and could easily derail this thread. I pretend to myself that I have free will, even if I suspect that at best it means that my brain is somehow at the mercy of random quantum effects magnified to affect the firings of my synapses, as opposed to being determined from birth... (e.g. thermal noise).

If this is the case, and synaptic firings can be affected by rando events, then not only will radiation-induced mutations be random, but some behaviours and thus some of the predator-prey environment for animals, and
their selection. To say nothing of alterations to the weather affecting loations of wind-blown seeds.

I can see that one could accept that a meteor strikes as the result of newtonian mechanics, but even if one knew the behaviour of every particle in the universe, and replicated it at the origin of the Earth, the random nature of mutations means that humanity would have zero chance of arising in this duplicate universe. Something similar might arise, and might even be probable, but certainly not inevitable.

For my purposes, a meteor strike is a random event in the context of evolution.

The offspring of the last common ancestor of humanity and chimps would probably have all been fairly similar, especially those that survived, which would have included both the chimp ancestor and the human ancestor.

They obviously differed slightly in traits, which led their decendants to occupy different ecological niches.

A chance laming of one of these descendants could have stopped the evolution of humanity, although a similar animal might have appeared. This chance laming could occur by treading on a randomly placed thorn from a branch that blew down in a random storm, from a tree that grew there due to previous random weather (ranin and wind)...

The actal selection process has a lot of chance in it, and the survivors are generally not only fit, but lucky. The unfit *are* doomed, but so are most of the "otherwise fit".


Jim

 

[Meadmaker]

Perhaps the problem randomites have with understanding non-random with respect to evolution is they think the fact a random model can be used to model a real world process means the real world process is random?

The only problem I'm having a hard time understanding is why people think that a phenomenon as complex as the evolution of life should always be described in the same way, regardless of the aspect of the problem under consideration, or the target audience.

 

[mijopaalmc]

The only problem I'm having a hard time understanding is why people think that a phenomenon as complex as the evolution of life should always be described in the same way, regardless of the aspect of the problem under consideration, or the target audience.

Because evolution is the same regardless of the process under consideration and the or the target audience. I will justify this later.

 

[articulett]

articulett, you're ignoring the point. The point is, by saying "evolution is not random," you're communicating with six-year-olds, not adults with training in the physical sciences. You'd be far better served by a thoughtful and thorough approach to understanding what your audience hears when you say that, than by excoriating them because they don't think like six-year-olds.

And you are missing my point. "What is the evidence for the evolution of this thread being non-random"? I would say that it's the part that "sticks'--the words that end up posted are "selected" (and though there are random components as to who reads and posts, the posts themselves are not random nor is the process involved--hence the evolution of this thread is not random.) I contend that calling the evolution of this thread random is meaningless for convey anything about anything. I don't care if you are 6 or 60 with 6 scientific degrees...it still conveys no information--particularly as to the "non-random" elements of this thread.

What is your answer to my question involving your definitions or Mijo's of this question: B]"What is the evidence for the evolution of this thread being non-random"[/B]? What is it Jim-Bob? Meadmaker? Mijo?

Whatever the answers are, I suspect they apply equally to evolution in general.

If per your meaning of the word "random", the evolution of this thread is "random"--then there will be no evidence sufficient enough to convince you that evolution is not random--hence evolution is random per your definition--as meaningful or as meaningless as that may be. I cannot understand this vague definition of random until someone with mijo's defintion explains it to me using the evolution of this thread as an example; and I suspect the same is true for anyone outside whatever field is defining random in this context.

 

[articulett]

Because evolution is the same regardless of the process under consideration and the or the target audience. I will justify this later.

Correct. The Priniciple is the same. The facts are the same. Your confusion is using personified examples. (e.g. Identical twins don't always have the same number of grandchildren...some can die or get lucky more or less etc.). This changes nothing about the process of evolution. The random factors are already part of the equation. It's the fact that some things will stick around that is important in the evolutionary process just like the fact that words on this thread will be posted and that drives the evolution of this thread. The fact that some people can't post because they never saw this thread or their computer is down and so forth is not relevant to understanding the evolution of this thread--just as your examples are not relative to understanding the process of evolution. The random is easily understood and not as important as the "sticking" factor...that which sticks around to be built upon. The fact that some people could have written stellar (more fit per mijo) responses is not relevant to understanding "the evolution of this thread".

Many random factors are in play in the evolution of this thread...but what sticks (what is posted and built upon) is not "random". This thread is not "random" (or it would be meaningless to describe it that way) and it would be equally uninformative to describe the evolution of this thread as "not non-random" or "stochastic" or "random". This is especially true if a person was actually interested in understanding the "non-random" aspects of evolution or understanding what Dawkin's is talking about.

The non-random aspects of the evolution of this thread are of the same kind of non-randomness in selection. Only some stuff sticks around to be build upon.

 

[articulett]

Re: Ivor, #1010


They obviously differed slightly in traits, which led their decendants to occupy different ecological niches.

A chance laming of one of these descendants could have stopped the evolution of humanity, although a similar animal might have appeared. This chance laming could occur by treading on a randomly placed thorn from a branch that blew down in a random storm, from a tree that grew there due to previous random weather (ranin and wind)...

The actal selection process has a lot of chance in it, and the survivors are generally not only fit, but lucky. The unfit *are* doomed, but so are most of the "otherwise fit".
Jim

And I want to add on to this. Identical twins (which would be equally fit per mijo's defintion) are not identically fit in reality. Because the environment affects the fitness from prenatal environment on. Oftentimes one twin gets more nutrients in the womb. But even then twins are born at different times and have differing experiences a tiny "random" mutation can cause the death of one twin and not another...etc. Moreover, the brain of more evolved animals...including humans has evolved to "fill niches"--identical twins will drive each other's differentiating...such that identical twins reared together are often less similar to identical twins reared apart. So, the inputs cannot be identical unless you are talking about the same organism and the exact same set of circumstances regarding that organism...in which case, I'd expect the same result. (Determinism)

Yes--the fit are those that survives due to fitness and luck--that doesn't mean that some of those who are identically fit from a genomic perspective will have the same "luck". Of those who pass any elimination round--there will only be the "fittest"--and not all of those or even most of those who are the "fittest" genomically will pass the elimination round. Every thing that affects DNA or an organism containing DNA is part of the selective forces that drive evolution. Of all the sperm that could have made you, the one that did make you is not necessarily the "most fit" in the genome department by any definition mijo might give it...but it was "fit enough" in the genome department to swim fast and find the egg. There were many contenders that never came close due to being "unfit". Anything in the genome of that sperm that helped it fertilize that egg will be passed on to the resulting zygote. (Except for mt. DNA from your fathers mom which drives the flagellum "motor"...but that's for another time.) The sperm that made you ends up being the "most fit" because it got itself into an organism whereby it can get itself copied-- partially through being amongst the fittest--but also largely due to "luck". It happened to be in a fertile environment (along with hundreds of millions of others) and was the first to penetrate the egg in it's entirety (lots of sperm help digest the outer layer...but only one gets in.)

The hoops that must be jumped through to get to the egg have also evolved to make it more likely that the "fittest" will get through...There is a great recent article about the evolution of duck reproductive tracts and how they evolved to make some really wild genitalia in an "arms race"; whereby females evolve tracts that allow them to select and keep rapists from fertilizing their eggs, and males evolve tricky genitals that keep other male sperm away from fertilization.

 

[articulett]

Is this random per physicists definition?

http://www.nowpublic.com/long_duck_dongs_evolution_and_sexual_organs
http://www.nytimes.com/2007/05/01/s...=8a3af3b12c59780dei=5088partner=rssnytemc=rss
http://www.pandasthumb.org/archives/2007/05/twisty_maze_of.html
http://www.plosone.org/article/fetchArticle.action?articleURI=info:doi/10.1371/journal.pone.0000418

I mean would you really want to sum up the evolution of duck genitalia as a random process? This is why I am saying, having random components does not make a process random. I'm not saying you should call it "non-random"--I'm saying it's SELECTION...and random is about chance--not being "selected" in any way. I cannot imagine a more useless way to describe the evolution of this thread, ducks' genitals, or evolution itself that to sum it up with "it's a random (or stochastic) process". If someone wanted to know what was non-random about it, I would hope that every scientist and person of intelligence would have more to say than "there is no evidence that the process is non-random".

Moreover, appealing to the "philosphy of science" is another ploy often used by creationists, and I think you should avoid doing it, mijo.

Natural Selection is entirely determined by what came before--what is in the pool to select from. Therefore it is deterministic. Random, on the other hand, is a term used to describe something not connected to what came before...not "determined" and so it's a generally good descriptor for mutation (even though mutations not truly random in the strictest sense of the word.) To apply the some vague meaning of the word random so that it applies equally to mutation and selection seems bizarrely misleading and uninformative, and I cannot imagine any reason anyone would do so if they actually had the goal of understanding the non-random aspects of evolution and what it is biologists are trying to convey.

 

[articulett]

duplicate