Evolution Not Random

[Apathia]

Good point. I was going to say that the motion of the ball is random, but of course it depends on the time scale involved.

Let us use the language of a mathematician to describe the motion.

Let us assume that the ball is launched at t=0, and t, representing time, goes upward. i.e. t=1 second is one second after the ball is launched.

Let us also assume that the maximum amount of time the ball spends in the pinball machine before draining is 10 minutes. This value is determined using empirical observation.

The "motion of the ball" can be represented by describing the position of the ball at time t. We'll call that p(t).

For t<10 minutes, p(t) is a random variable. For t>10 minutes, p(t) is not a random variable.

So, for a mathematician, the motion of the ball may be considered random, or non-random, depending on exactly what question related to motion he wants to answer.





Nor should you. You should use the best language possible to describe to your target audience the phenomena in question. If you happen to be describing evolution at the time, that might mean "random" or it might be "nonrandom", depending on exactly which aspect of evolution you are describing, and to whom you are describing it.


The article from the OP did a real disservice, in my opinion. It's title somehow implied that this study confirms Darwin's theory. Is that correct? I certainly hope not, because the finding of the study was not what the article claimed. The article found that most traits evolved deterministically, but some traits evolved stochastically. If one or the other finding would disprove Darwin's theory, then the theory would have been disproved.

Fortunately, we don't have to rewrite biology textbooks just yet. Darwin's theory was not being tested in this study. This study was simply refining a description of the mechanism involved. If you were to read the journal article, you would find no reference at all to any question about verifying evolutionary theory, because that wasn't under study. That question has been settled long ago. The biologists are just fine tuning their understanding of the details at this point.

Meanwhile, the paper defined the terms "stochastic" and "deterministic" in a particular way, and then tried to determine whether traits evolved stochastically or deterministically. We could argue with their definitions if we were so inclined, but that would be foolish. The words have ambiguity in their definitions, but the paper's authors defined exactly what definitions they were using, and how they were applying them to their specific aspect of evolution. Therefore, whether or not we like the way they defined their terms, those terms were defined precisely within the paper, and anyone reading the paper can see what they meant, and learn something about evolution from the paper.

 

[Skeptic Ginger]

And I don't know anyone who would not consider the weather random. It's almost as if we are using the same word, but speaking a different language.

Out of respect for the professors of mathematics and engineering who taught me a specific use of the word random, I will continue to call weather random. If you are using some other definition of the word, you can call it non-random if you wish. We will both be right, because words can have multiple meanings.

Just to clarify the term, random....

If one tosses the equally weighted die with 6 sides the result is random. The weather is only unpredictable because we have poor models. Perhaps I am wrong on that, I am certainly not a meteorology expert. In that case it isn't the definition of random, it is my misunderstanding about the weather. So I'll not use it as an example here.

 

[chulbert]

No one that's ever struggled to keep the ball in play in pinball would agree that its motion is random.

Every ball is exactly like the ball before it. Where's the genetic mutation in this metaphor? How would you characterize the game if each "ball" had a different size, weight or texture?

 

[articulett]

Every ball is exactly like the ball before it. Where's the genetic mutation in this metaphor? How would you characterize the game if each "ball" had a different size, weight or texture?

Each flip of the battle determines the balls path... not randomly... each play is a new "test" of ball flipping procedure and one "evolves" in keeping the ball in play...

Processes may contain randomnesses, but the fact that they are processes makes them non-random...

Stochastic processes are sometimes called random processes because they are based on random variables... but algebra is based on random variables too-- but the processes themselves are not random.

When you call such things random... even the weather... you would need to say it's random in respect to something else. The weather is determined by multiple inputs... it's random in regard to our ability to control it or even predict it to any great extent... we look for "non random" patterns to understand it. It's not random the way freak accidents are random.

Poker is not random the way roulette is random. Using random in both instances is misleading and uninformative as to the processes of how the games are played.

 

[joobz]

If one tosses the equally weighted die with 6 sides the result is random.

Even this wouldn't be random if you could model the exact trajectory that you threw the dice, you know the exact function of the fluid the dice if moving in, the landscape surface on the table, density of dice, it's elasticity, loss of energy due to bouncing/friction, ....

There's no reason not to be able to know the outcome of the dice through if you can model the system exactly.

Same is true for really anything (except the quantum level, stupid heisenberg).

The level of detail you need to know to predict the outcomes seems to be the drawing line on whether we call the system random or not.

Like in the pinball example, to a person used to a machine and knows all of the machine imperfections (where does the ball slip, what part of the flippers are warn, which ball has a chip, ..) You can then start to guess trajectories more accurately. However, even then there are uncertainties. Will this sticky bumper click now, or later? Is this guy standing next to me going to bump the table? there are factors not accounted for and therefore make the trajectory a random event.


My favorite example is and always has been diffusion.
we can model diffusion perfectly from a macroscopic level, but we can't tell you the molecular motion of every single molecule. I wouldn't call diffusion random(because we can predict the outcome exactly), but I would say it is randomly driven.

Anyone who attempts to label evolution as impossible because it has inherently random events built into it is simply, completely and totally wrong. They make themselves look like fools to insinuate as such.

However, It's equally wrong to pretend that evolution doesn't have an element of chance in it. Unlike diffusion, we don't know nor can we predict the exact outcome in an evolutionary adaptation. We know that adaptation will occur, but we don't know in what form. For instance, nylon eating capacity of bacteria can could be hypothisized to be possible, but you wouldn't be able to tell me which bacterial enzyme would be used to evolve this trait.

 

[mijopaalmc]

No one that's ever struggled to keep the ball in play in pinball would agree that its motion is random. Why? Because it's not, at least not by the common definitions of the word:

It is obvious that neither evolution by natural selection nor the motion of a pinball are random by the first definition. By the second one would have to qualify the statement "evolution is random" by specifying precisely in what sense, and what the distribution is.

One can say "evolution is not random" and be perfectly correct. One could also say the contrary, with some qualification - which just illustrates the utter futility of arguing over semantics.

The problem is that when I asked a question way back in May about what evidence there was that evolution was "random" is the sense of "[o]f or relating to a type of circumstance or event that is described by a probability distribution" or stochastic in the sense of "nvolving or containing a random variable or variables" or "nvolving chance or probability". I got answers explaining to me why evolution was not random in the sense of "[o]f or relating to an event in which all outcomes are equally likely" or stochastic in the sense of "[o]f, relating to, or characterized by conjecture; conjectural". In other words, those who argued that evolution is non-random did so by equivocation and, most frustratingly of all, disseminating disinformation like the definition I was using "makes everything random" or that "algebra is random" by it.

In sum, seven months after posting my question I still haven't gotten a straight answer as to what evidence there is of evolution being non-random*?

*in sense of not "[o]f or relating to a type of circumstance or event that is described by a probability distribution"

 

[Skeptic Ginger]

Because "mixed" processes ended up being either random or determistic.

For example, supposed I have a random number generator and I take the result and multiply by 2. We have a random process followed by a determistic one, but the process taken as whole is random, as knowing the additional state of the system will not allow a predictable outcome.

Supposed in stead of just doubling the number produced, I double the number then add 8, divide the result by 2, and then subtract the original number. If I combined the random number generator and this process I get a determistic result. No matter what number the generator produces, I get 4 out. So this process is determistic.

Both examples are "mixed" by your definition, but one is obviously a random process (identical starting points can yield different results) and the other is obviously determistic even though it has a random sub-process.

Walt

OK then, let's look at your example a different way. Suppose when you use your random number generator you only use even numbers when they come up and you ignore all odd numbers. Now you always get an even number. You have lots of possibilities which are still random, but they all have a determined property, they are all even numbers. They have a random property, the actual number and they have a determined property, they are all even numbers. Depending on which property you look at, you have mixed random and non-random outcomes.

If mathematically you want to call that a determined outcome because you determined which numbers that were generated you would keep, then you have something more akin to evolution.

 

[Meadmaker]

In sum, seven months after posting my question I still haven't gotten a straight answer as to what evidence there is of evolution being non-random*?

*in sense of not "[o]f or relating to a type of circumstance or event that is described by a probability distribution"

I beg to differ. You have received many, many, straight answers to your question.

Of course, not everyone has agreed with those straight answers, but there's nothing anyone can do about that. The fact that some people have never accepted those straight answers and still continue to beat the dead horse is an interesting phenomenon from the field of psychology.

 

[mijopaalmc]

I beg to differ. You have received many, many, straight answers to your question.

When I ask for evidence that evolution is "random" by one definition and receive evidence that evolution is "random" by another definition, that is not a straight answer. In fact, equivocation is one form of fallacious argumentation that people here like to slam creationists for.

 

[Skeptic Ginger]

People often have this Mendelian simplistic idea about how evolution actually works. They picture random nucleic acid substitutions producing one beneficial change out of every million or so which benefits the organism and is selected. That just isn't how evolution works.

There are now over 6 billion people on the planet and all but the identical twins have slightly different genomes. That means at least 6 billion different human genetic blueprints all work, and most of them work equally well. There may be a random distribution of genetic variations in that population. Actually, once a mutation occurs then it frequently only exists in the offspring. Having or not having a particular ancestor is deterministic, the original mutation was random.

That could be fit into mijo's math model which Walter concurs with (I think) were that all there was to evolution theory. But that isn't all there is. We are discovering all sorts of things about how evolution works that we had no idea about only a couple of decades ago.

For example, once you have this random generation of genomes you don't just introduce the equivalent of multiplying by 2 as in Wayne's example, instead, you discard altogether some of the genomes, you select some to multiply at greater rates, and most you don't do anything with. And, you give everyone 2 copies of almost all the genome allowing the superior versions of each segment the chance to override any inferior segments.

In the microbe genomes there is even more going on. The organisms have multiple ways of sharing whole genes and it appears they share specific genes not just random segments of genetic code. The organisms have means of increasing and decreasing mutation rates depending on need including such things as varying the surface proteins while leaving the core of the organism intact. They can incorporate the genetic code of the host making proteins that fool the host's immune response. These are deterministic processes.

That's fine if you want to label single portions of evolution processes random. But to oversimplify the entire process and claim it is no more than a math formula which must be random if it starts with random mutation is a gross misrepresentation of actual evolutionary processes.

 

[joobz]

When I ask for evidence that evolution is "random" by one definition and receive evidence that evolution is "random" by another definition, that is not a straight answer. In fact, equivocation is one form of fallacious argumentation that people here like to slam creationists for.

I'm not certain what your point is.

Feel free to call evolution random based upon the most strictest sense of the word. it's accurate and fine.


If you got everyone to agree to this clear definition of random, what are you left with? What do you prove?


How would getting everyone to agree to this change the fact that evolution is deterministic, in that we know that the populations will become more adapted to thier environments over time?

 

[Skeptic Ginger]

Even this wouldn't be random if you could model the exact trajectory that you threw the dice, you know the exact function of the fluid the dice if moving in, the landscape surface on the table, density of dice, it's elasticity, loss of energy due to bouncing/friction, ........

It's a hypothetical die used in math. It isn't a real die in the real physical world. It's just used to represent true randomness.

 

[Skeptic Ginger]

Why do you hate typography?
Why do you hate good taste?

In the world of things I choose to fight battles over, comic sans isn't one of them.

 

[joobz]

It's a hypothetical die used in math. It isn't a real die in the real physical world. It's just used to represent true randomness.

sure, but we're talking about real proccesses and how they relate.

I wanted to demonstrate that even our example of a ideal random system is deterministic in practice.

 

[cyborg]

Every ball is exactly like the ball before it. Where's the genetic mutation in this metaphor? How would you characterize the game if each "ball" had a different size, weight or texture?

How would you characterise the game if these variations were deterministically decided?

 

[mijopaalmc]

How would you characterise the game if these variations were deterministically decided?

Do you have evidence that mutation is deterministic?

 

[mijopaalmc]

How would getting everyone to agree to this change the fact that evolution is deterministic, in that we know that the populations will become more adapted to thier environments over time?

Because knowing "that the populations will become more adapted to thier [sic] environments over time" doesn't make evolution "deterministic".

 

[Ichneumonwasp]

Oh, good Lord. It's back from the dead. Again.

Die, vampire/zombie/undead wretch, die.

 

[Skeptic Ginger]

....
The article from the OP did a real disservice, in my opinion. It's title somehow implied that this study confirms Darwin's theory. Is that correct? I certainly hope not, because the finding of the study was not what the article claimed. The article found that most traits evolved deterministically, but some traits evolved stochastically. If one or the other finding would disprove Darwin's theory, then the theory would have been disproved.

Fortunately, we don't have to rewrite biology textbooks just yet. Darwin's theory was not being tested in this study. This study was simply refining a description of the mechanism involved. If you were to read the journal article, you would find no reference at all to any question about verifying evolutionary theory, because that wasn't under study. That question has been settled long ago. The biologists are just fine tuning their understanding of the details at this point.

But the research was addressing a specific aspect of Darwin's theory. (From artic's 3 links) Verdict in: Darwin's evolution theory confirmed

The team found that the selection of successive traits occurred naturally in a process that Darwin called deterministic inheritance, instead of by seemingly random providence.

For years, scientists opposed to Darwin's theory have proposed their own theory that evolution could have taken place by pure providence—randomly inherited traits only some of which endure by surviving.

ScienceDaily is reporting a team of biologists has demonstrated that evolution is a deterministic process, rather than a random selection as some competing theories suggested.

Here's an updated link from the OP article and the actual title there: Evolution Is Deterministic, Not Random, Biologists Conclude From Multi-species Study

....Meanwhile, the paper defined the terms "stochastic" and "deterministic" in a particular way, and then tried to determine whether traits evolved stochastically or deterministically. We could argue with their definitions if we were so inclined, but that would be foolish. The words have ambiguity in their definitions, but the paper's authors defined exactly what definitions they were using, and how they were applying them to their specific aspect of evolution. Therefore, whether or not we like the way they defined their terms, those terms were defined precisely within the paper, and anyone reading the paper can see what they meant, and learn something about evolution from the paper.

From mijo's post #11, the original article: Trends, Stasis, and Drift in the Evolution of Nematode Vulva Development

Conclusions: We propose that developmental evolution is primarily governed by selection and/or selection-independent constraints, not stochastic processes such as drift in unconstrained phenotypic space.

I was going to address this in more detail but I have to go. So I'll get back to this.

 

[Ivor the Engineer]

I'm sorry, but how the **** do you get someone to give you money to study nematodes' vulvas?