Randomness in Evolution: Valid and Invalid Usage

[Ivor the Engineer]

From a statistics book I have:

Random. Nondeterministic, occurring purely by chance, or independent of the occurrence of other events.

 

[mijopaalmc]

From a statistics book I have:

Random. Nondeterministic, occurring purely by chance, or independent of the occurrence of other events.

So is P(pair of jacks|jack of spades) not random?

 

[Walter Wayne]

Here's a typical definition of random: "Lack of predictability, without any systematic pattern."

Something like that is what most people mean when they use the word. And that clearly does not apply to evolution - evolution is very predictable in many ways, and it follows patterns that we are quite capable of understanding. Of course its details are not predictable (which is the case for ALL OTHER physical processes too), but many broad patterns are - increase in complexity over time, survival of the fittest rather than the least fit, general characteristics of responses to changes in the environment, etc.

Well two of the patterns you mentioned aren't real patterns. Increase in complexity over time just isn't the case. In Gould's Full House, he points out that the idea that evolution is "progressive" is largely a myth. Lineages both increase and decrease in complexity and there doesn't appear to be a bias either way. (In Dawkins's review of Full House, he disagrees with Gould's idea that "progressiveness" is a myth, not because of the complexity argument, but linking complexity to progressiveness is an anthropocentric view. If one was a bacteria you might laugh at all the crap we go through to pass on our genes.)

Survival of the fittest is not really a pattern at all. The only way to define the terms "survival" or "fittest" consistently with natural selection is to define fitness in terms of survivability. The expression is circular.

Not that there aren't patterns. Species which become parasitic tend to simplify over time, since their host performs several functions for them (one reason increased complexity isn't the norm).

As for predictability, it exists in the short term but not so much in the long term. Humans are considered a bit player compare to bacteria, and even among animals insects and other creatures are far more dominant. But we have had a large affect on numerous species, but would an outside observer 70000 years ago predict that we would be where we are today. At the time it is believed the human population was a few thousand, and 70000 years isn't much in evolutionary time.

Walt

 

[Wowbagger]

Wowbagger and Sol Invictus... mark my words... you are having a discussion with people who have a vested interest in not understanding. No progress will be made. It's not you... it's not your explanations; it's them. These guys were saying the exact same nothingness over a year ago on Mijo's "evolution is not nonrandom" thread"... truly... nothing has changed. Nothing.

I say a few things have changed. Both of us are being enlightened on using the term "random" more carefully. Progress is slow, to say the least, but it is being made. (I think.)

To get Mijo to enter a thread all you need to do is to say evolution is not random... (or to criticize creationists)--

There you go, accusing people of being creationists, again!
I honestly don't think mijo is a creationist, nor a troll. I think he is over-trained in the realm of mathematics, and thus fails to recognize the importance evidential support plays in establishing working definitions in scientific models. Maybe.

• Deterministic evolution by natural selection divides the phenotypes in a population into two mutually exclusive groups individuals in which either
  1. all produce reproductively viable offspring or
  2. all produce fail to reproductively viable offspring.

Whole life forms might be divided into those two groups: those who reproduce viable offspring, and those that do not. I think it confuses the issue when you talk about phenotypes, in that way. I very much doubt phenotypes would fit into "mutually exclusive" categories, anyway.

• Stochastic evolution by natural selection allows each phenotype to confer a probability of producing reproductively viable offspring on each individual that possesses it.

That sounds fine, for the most part. So, what is the problem? I already said stated "stochastic" is perfectly fine for us humans to model evolution. All you need to understand is that it is only a model.

(I would place a Monty Python joke, in here, but frankly I was never enough of a fan to do that sort of thing.)

Seriously some of the people who are arguing that evolution is not random are more skilled at equivocating than creationists.

Scientists build general models, based on what the evidence shows us. Creationists do not. That is the difference.

The problem here is that scientists and others who argue for evolution get all up in arms when creationists use the common definition of "theory" to imply that evolution is just a guess or when they use their owm ifiosuncratic definition of "transitional form" to declare that evolution has not produced any evidence of such, but have not problem using the common definition "random" to declare that evolution in non-random.

Actually, they do not normally use the common definition at all. Scientists normally use one of the first two bullet points to describe mutations (which are random in that way), and then say non-random to describe selection (which is not random in any way).

When a scientist uses the common definition, he or she normally say "pure random" or "random chance", etc. And then, of course, would go on to say that NO PART of evolution is random in that sense.

6. Arthur Dent falling, missing the ground and flying!

If that was random, it would be covered by bullet point 4: A rather unlikely series of quantum fluctuations. Though, it might have all been a plot, by the Guide Mk II, in which case it would not be. Do try to pay attention!!

QM might be
-random
-chaotic-stochastic-deterministc

I don't recall seeing anything that said that he process underlying QM was random, just that it follows some sort of proability that at this time we say 'is considered to be random'.

I would say that QM is just about the closest thing we actually have to something being purely random. Maybe it is, or maybe it is not. But, for the sake of argument, I will allow it to be called "random", because it does not even make much of a difference at large scales, anyway.

No. The point is: because scientists use a particular definition of the word "random" in the context of evolution, evolution can be said to be random... even if the word "random" has a slightly different meaning to the general public.

A good summary.


Just a heads up: I will probably be phenomenally busy, for the next couple of days, so I might not be able to respond to anyone else until the end of the week or so. Thanks!

 

[Earthborn]

Richard Dawkins is a scientist. When he says "evolution is not random", he too is not using the word the way mijo wants.

And maybe once in a while Richard Dawkins is wrong.

because every single physical process in the world is random according to it.

Yes, so? I fail to see why that should be a problem.

The statement "evolution is random" is therefore misleading to the point of being simply false.

I fail to see why it be "misleading to the point of being simply false" if it means that it is a physical process like any other physical process.

 

[sol invictus]

So popular dictionaries are your "technical" sources? How interesting. Very well, let's give you every benefit of the doubt and have a look:

That's funny that you say you couldn't find any source for the definition I provided as The American Heritage Dictionary,

False. Here's their definition:

1. Having no specific pattern, purpose, or objective: random movements. See synonyms at chance. 2. Mathematics & Statistics Of or relating to a type of circumstance or event that is described by a probability distribution. 3. Of or relating to an event in which all outcomes are equally likely, as in the testing of a blood sample for the presence of a substance.

Only definition 2 is similar to yours, but definition 3 disagrees, and is technical and (given their example) more relevant to biology. So you lose that one.

The American Heritage Science Dictionary (both at dictionary.com),

ran·dom /ˈrændəm/ Pronunciation Key - Show Spelled Pronunciation[ran-duhm] Pronunciation Key - Show IPA Pronunciation
–adjective
1. proceeding, made, or occurring without definite aim, reason, or pattern: the random selection of numbers.
2. Statistics. of or characterizing a process of selection in which each item of a set has an equal probability of being chosen.
3. Building Trades.
a. (of building materials) lacking uniformity of dimensions: random shingles.
b. (of ashlar) laid without continuous courses.
c. constructed or applied without regularity: random bond.
–noun
4. Chiefly British. bank3 (def. 7b).
–adverb
5. Building Trades. without uniformity: random-sized slates.
—Idiom
6. at random, without definite aim, purpose, method, or adherence to a prior arrangement; in a haphazard way: Contestants were chosen at random from the studio audience.

(my bold)

Not a single one of those coincides with yours, and the technical one explicitly disagrees and supports mine. That's two lies.

and Merriam-Webster's Online Dictionary all cite the definition I use.

Here's what M-W online has:

random:
a haphazard course
— at random
: without definite aim, direction, rule, or method <subjects chosen at random>

Nothing like your definition.

Three lies and you're out.

Goodbye, little troll.

 

[sol invictus]

Well two of the patterns you mentioned aren't real patterns. Increase in complexity over time just isn't the case.

I see. So you regard it as a coincidence that mammals with a hundred trillion cells exist now, whereas 3 billion years ago when life started there were only single-celled organisms?

Survival of the fittest is not really a pattern at all. The only way to define the terms "survival" or "fittest" consistently with natural selection is to define fitness in terms of survivability. The expression is circular.

So you regard it as a coincidence that, if I have a population of bacteria some of which are resistant to penicillin and some of which aren't, when I introduce penicillin into the environment the resistant group will dominate the population after a few generations?

Don't be ridiculous.

These are obvious patterns, and they are very predictable. They are not random at all.

 

[sol invictus]

And maybe once in a while Richard Dawkins is wrong.

Great argument: "Yeah, but he might be wrong."

Yes, so? I fail to see why that should be a problem.

Because it makes the statement as informative as picking up an apple and saying "this is something".

I fail to see why it be "misleading to the point of being simply false" if it means that it is a physical process like any other physical process.

If that's hard for you to see, I don't think I can help. Good luck.

 

[articulett]

And maybe once in a while Richard Dawkins is wrong.

Yes, so? I fail to see why that should be a problem.

I fail to see why it be "misleading to the point of being simply false" if it means that it is a physical process like any other physical process.

Yes, and then it becomes as useful or true as saying poker is random or frying eggs is random or algebra is random or childbirth is random. Processes, by definition, are generally not random. When they have random elements, they might be called "random processes" or stochastic processes, but that doesn't make the processes themselves random. Algebra has random variables... algebra is not a random process. If someone kept insisting it was... you might wonder why. You might wonder why if the only other people who did it were people who were purposefully trying to obfuscate understanding... self-appointed experts such as Behe. If your goal is to be unclear, why would you such lame language? Why would anyone call poker a random game? it's surely not random like roulette. It sure doesn't tell you anything about the game... that is what mijo is doing with evolution.

Dawkins is not wrong--he's clear. Mijo is unclear to the point of being misleading and dishonest. If your goal is to share information so that others understand it, he fails. If your goal is to be taken seriously by those who actually understand and teach the process to others, then you should use language the way they do.

If your goal is to prove to yourself that "evolution is really random" whatever vague thing that might mean-- then continue on as Mijo does. Sure, there's no problem in communicating that way... if your goal is not to convey anything useful or substantive about evolution--or if your goal is to sound like "scientists think this all happened by chance".

 

[Earthborn]

Nothing like your definition.

So does evolution occur with definite aim, direction, rule or method?

Only The American Heritage Science Dictionary as quoted by you provided definitions that were not applicable to evolution: 2 en 6. Number 2 is simply wrong. Throwing two dice will get you a process of selection in which numbers of the set {2 to 12} does not have an equal probability of being chosen. Which number comes up is still called "random" AFAIK.

Number 6 mentions something better: "adherence to a prior arrangement", which obviously does not apply to evolution. In evolution "prior arrangement" is rather significant. Not that this means evolution is entirely immune to events that are random even according to this definition, such as asteriod impacts.

 

[mijopaalmc]

So popular dictionaries are your "technical" sources? How interesting. Very well, let's give you every benefit of the doubt and have a look:

<snipped extraneous definitions and insults>

Now you're deliberately equivocating and trolling. I never said that some of the definitions that others had used weren't valid; I just said that the definitions that others were using weren't consistent with all of the ways in which randomness is used in biology.

Moreover, you are deliberately ignoring that the definition to which I referred you do actually exist.

Here is the definition of "random" from The American Heritage Science Dictionary as it appears on dictionary.com:

The American Heritage Science Dictionary - Cite This Source - Share This
random (rān'dəm) Pronunciation Key

1. Relating to a type of circumstance or event that is described by a probability distribution.
2. Relating to an event in which all outcomes are equally likely, as in the testing of a blood sample for the presence of a substance.


The American Heritage® Science Dictionary
Copyright © 2002 by Houghton Mifflin Company.
Published by Houghton Mifflin Company. All rights reserved.

And here is the definition for the adjective entry for the Merriam-Webster's Online Dictionary:

Main Entry:
2random
Function:
adjective
Date:
1632

1 a: lacking a definite plan, purpose, or pattern b: made, done, or chosen at random <read random passages from the book>2 a: relating to, having, or being elements or events with definite probability of occurrence <random processes> b: being or relating to a set or to an element of a set each of whose elements has equal probability of occurrence <a random sample>; also : characterized by procedures designed to obtain such sets or elements <random sampling>
— ran·dom·ly adverb
— ran·dom·ness noun
synonyms random, haphazard, casual mean determined by accident rather than design. random stresses lack of definite aim, fixed goal, or regular procedure <a random selection of books>. haphazard applies to what is done without regard for regularity or fitness or ultimate consequence <a haphazard collection of rocks>. casual suggests working or acting without deliberation, intention, or purpose <a casual collector>.

Again, it is not the only definition, but the definition is the only one that is consistent with all implicit assumptions made by biologists in their day-to-day work.

In other words, your entire argument is a straw man. Try harder next time.

 

[mijopaalmc]

Algebra has random variables... algebra is not a random process.

Algebra does not contain random variables as they are defined in probability theory, but articulett had bothered to learn anything about probability theory she would know that.

What is it with those who argue that evolution is non-random that predisposes them to equivocation and straw men?

 

[T'ai Chi]

Evolution is a stochastic process, because at least some of the mechanisms involved are stochastic (that is, random). However, it is misleading to call evolution a "random process, full stop" because selection is not random with respect to the local environment.

Wrong. It is more like "evolution is a stochastic process, full stop" is the correct statement.

Notice how you start talking about evolution, then by the end you are talking about selection. You know evolution is more than just selection, right? (hint: the "more than" is the random stuff).

 

[mijopaalmc]

Here is the a summary of what a stochastic process is that you will get if you read any textbook on probability theory and stochastic processes:

Basically, the axiomatization of probability theory begins with the creation of a mathematical space called a probability space designated by the ordered triple [latex]$(\Omega, \mathcal{F}, \mathbb{P})$[/latex]. [latex]$\Omega$[/latex] is a set, which, for our purposes, is almost any collection of objects. If [latex]$\Omega[/latex] is finite or countably infinite, the probability space is called discrete. If [latex]$\Omega[/latex] is uncountably infinite, the probability space is called continuous. [latex]$\mathcal{F}$[/latex] is a family of subsets of [latex]$\Omega$[/latex] called a [latex]$\sigma$[/latex]-algebra whose members share certain attributes and are known as "events". In particular, a [latex]$\sigma$[/latex]-algebra is closed under complementation (i.e., the "opposite" of every event is also an event) and countably infinite unions (i.e., a collection of events is also an event). [latex]$\mathbb{P}$[/latex] is a function called a probability measure that maps each event in [latex]$\mathcal{F}$[/latex] to the interval [0,1] in such a way that [latex]$\mathbb{P}(\Omega)=1$[/latex] and the probability of the union of mutually exclusive events is the sum of their respective probabilities.

A random variable is, by definition, a function, [latex]$X$[/latex] from the probability space [latex]$(\Omega, \mathcal{F}, \mathbb{P})$[/latex] to another measurable space, most often [latex]$(\mathbb{R}, \mathcal{B}(\mathbb{R})$[/latex] , the real numbers and the Borel [latex]$\sigma$[/latex]-algebra on [latex]$\mathbb{R}$[/latex] (i.e., the smallest [latex]$\sigma$[/latex]-algebra form by the open intervals of [latex]$\mathbb{R}$[/latex]). The random variable defines a probability measure, known as a pushforward probability measure, on the measurable space [latex]$(\mathbb{R}, \mathcal{B}(\mathbb{R})$[/latex] such that [latex]$(X_*(\mathbb{P}))(B)=\mathbb{P}(X^{-1}(B))[/latex]. In other words, that is the pushforward probability measure of [latex]$B\in\mathcal{B}(\mathbb{R})$[/latex] equals the probability measure of [latex]X^{-1}(B)\in\mathcal{F}[/latex], the preimage of [latex]$B$[/latex] in [latex]$\mathcal{F}[/latex] (i.e., [latex]$\{A\subset\Omega}\}\in\mathcal{F}$[/latex] that maps to [latex]$\{B\subset\mathbb{R}\}\in\mathcal{B}(\mathbb{R})$[/latex]). This pushforward measure is called a probability distirbution.

A stochastic process is a family of random variables [latex]$\{X_{t}\}_{T\in{t}}$[/latex] for an index set [latex]$T$[/latex] If [latex]$\Omega[/latex] is finite or countably infinite, the stochastic process is called discrete. If [latex]$\Omega[/latex] is uncountably infinite, the stochastic process is called continuous.​

In terms of evolution, the population is [latex]$\Omega$[/latex], the reproduction events are the elements of [latex]$\mathcal{F}$[/latex], and natural selection is the measure [latex]$\mathbb{P}$[/latex].

By the way, the algebra to which articulett refers is not random because its function are not necessarily defined are a probability measure.

 

[articulett]

Earth born-- there's an aim... pass on your genes... those who don't, don't get to be a part of evolution... bummer. The genes that are best at getting themselves passed on drive evolution and make the process far from "random"...
repeat after me: random components do not a random process make.

ETA-- great Tai Chi' is here... you know you have a winner of a definition and explanation when the omnipresent Tai Chi' weighs in.

If your goal is to sound like Tai Chi and Behe-- just keep insisting that evolution is random and that somehow to someone you make more sense than Dawkins and all those who teach the subject to actual other people.

Natural selection is the "derandomizer"-- the mechanism that gives the appearance of design and the look that things somehow "knew" what features to evolve.

 

[Dancing David]

I swear i just read 'survival of the fittest", there is no such thing, it is survival of the reproductive.

 

[Earthborn]

Earth born-- there's an aim... pass on your genes...

Is that an aim, or is it just a result? And if it is an aim, whose aim is it? Is it God's aim? Is the aim of the organism carrying the genes who does not even know what genes are? Is it the aim of the genes, that don't have the mechanism to have 'aims' at all?

By introducing an 'aim' you are making a teleological argument instead of a cause and effect argument, which is generally frowned upon in the natural sciences. But who knows, maybe stones do fall back to Earth because they tend to move towards their natural place...

those who don't, don't get to be a part of evolution... bummer.

Untrue. They are part of evolution, they just don't pass on their genes. But they do shape the environment in which others do pass on genes.

The genes that are best at getting themselves passed on drive evolution and make the process far from "random"...

Which genes are best at getting themselves pass on depends on the environment, which is constantly changing.

repeat after me: random components do not a random process make.

What if all components are subject to random changes?

 

[UnrepentantSinner]

An example:

Was it random when a V2 rocket hit London? No

Was it random whether it hit a particular house and not another? Essentially yes.

First, excellent analogy for the difference between mutation and natural selection. Second, don't bother trying to trying to get Arti to accept any shades of gray. In her world, there's only crazed Creationist/Apologist fundies and logical, Vulcanesque strong atheists. And irony is apparently as foreign to her as the concept of a spectrum.

And mijo, you never responded to my question...

Let me be perfectly clear. Unless everyone reject reality and substitute your own, you are unwilling to accept the simple, straight foreward statement that "mutations are random but evolution is not"?

As jimbob's analogy noted, if, when and where in the DNA a mutation occurs is unequivocably random, but whether that mutation confers a suvival advantage is not because it is filtered through natural selection. This is an undeniable fact in the observation and a conceptual half of evolutionary theory. Unless you are equivocating, how can you deny that "mutations are random but evolution is not"?

 

[mijopaalmc]

As jimbob's analogy noted, if, when and where in the DNA a mutation occurs is unequivocably random, but whether that mutation confers a suvival advantage is not because it is filtered through natural selection. This is an undeniable fact in the observation and a conceptual half of evolutionary theory. Unless you are equivocating, how can you deny that "mutations are random but evolution is not"?

You are the one who is equivocating. Long-term orderly behaviors, such as the adaptive patterns we see in evolution by natural selection, does not imply a non-random process. Even if evolution was random in the sense of being equiprobable for all phenotypes, population would still evolve by regression to the mean.

 

[articulett]

That doesn't make the process random. If you aim to sound clear then use the words of those who convey the concept to others. If you aim to sound like T'ai-- refer to evolution as random. Processes are "steps in a procedure" one built upon the other... randomness is not related to past or future... Theoretically all things made of matter have random electron spins in their atoms... but who declares every thing random? And why would they?

Aim or direction applies to things without consciousness-- like wind or shooting starts or galaxies...it doesn't need to be conscious... it's just that physical laws make it so. Water and cold weather make lattices that form ice. It's a process. No one would call it random.

The same with evolution. Genes don't choose to get passed on... but those that do manage to do so drive evolution... the organism copying the DNA dies... but the information lives on to be copied again and possibly pick up a beneficial mutation in the process. It isn't random. No one who wants to convey what evolution actually is or help anyone else understand it, wouldn't describe it that way. It's a useless definition.

As Sol said, it's like picking up an apple and saying "this is something"... it's true... but uninformative. It's less useful than saying "evolution is a 4 syllable word"-- because while also being true... that doesn't have nearly the power to mislead or to be interpreted as "scientists think this all came about by random chance." Mijo's blather does. If your goal is to discuss evolution with Mijo, T'ai and Behe-- then by all means... you are doing a fabulous job. And an apple truly is something. And poker IS random per Mijo's definition.

Just don't expect anyone intelligent to think you actually understand the process or could correctly teach it to anyone else. You can't. You couldn't anymore than the person calling an apple a "something" could teach about apples. For the same reasons. It's equally uninformative and misleading. You end up wondering the motives of the person doing the describing -- if not their mental capacity. Really.

You can argue all you want and ask all the questions that you have... but that's the bottom line.