Thursday, November 29, 2007

What Is Evolution?

Greg Laden has just posted this video on his blog [Evolution ... Its for real]. I assume Greg endorses the definition of evolution in the opening minutes.

On Tuesday and Wednesday evenings I gave lectures on Evolution Is a Theory and a Fact. The first part of the lecture was devoted to explaining What Is Evolution. Here's the minimal scientific definition of evolution.
Evolution is a process that results in heritable changes in a population spread over many generations.
I explained to the audience that the definition of evolution has to be neutral with respect to mechanism. Evolution could occur by Lamarckian or Darwinian mechanisms, or something else entirely different.

The purpose of a definition is to define what we mean by the word. If we were to define evolution as that form of change caused by meiotic drive then change by natural selection wouldn't be evolution. How silly is that?

It's just as wrong to define evolution in terms of natural selection. That means random genetic drift is not evolution, by definition. Is that right?

The opening screen of the video states "The world is full of misinformation." The second screen says, "Many of those who claim evolution is wrong have no understanding what evolution really is."

This is exactly how I begin my talk. However, from that point on my lecture diverges considerably from the video that cdk007 created. Do you agree with the definition in the video? (Natural selection + time = EVOLUTION)


  1. I wonder if this isn't a case of an equation being correct but not definitive. For example, 1 + 3 = 4 is correct. But 4 = 1 + 3 is too limited, because 2 + 2 = 4 also.

    So, Natural selection + Time = Evolution is not incorrect, because if you add those two you will get evolution as you define it. But Evolution = Natural selection + time would make me cringe, as it does for you. The question is, which did the video's producer intend?

  2. "It's just as wrong to define evolution in terms of natural selection."

    How about if we instead define "interesting evolution" in terms of selection?

    (I kid--really!)

  3. I think a good definition of "heritable" is needed for that defition of evolution to be adequately understood.
    It is not as simple as some would think. Mst people would siomple equate heriyable with genetic, and take the defintion as a rehash of Dobzhansky's defintion.

    Plus, I myself think that not any population frewuency change is an evolutinary change, specially if no new organic, phenotypic change is involved. I don't think that will always be evolution, but according to your dfinition, it always is.

  4. Incidentally, half the sorting in the population in the simulation results from drift, so he clearly does not think selection is the only important mechanism.

  5. Yes, I noticed at the end that he has a random factor in his simulation. That makes it even more surprising that he didn't create a better definition in the beginning.

    Why all that stuff about infinite population growth and environmental influences? What does that have to do with a definition of evolution?

    You don't need to have more births than deaths in order to have evolution.

  6. I'm not sure it was a definition, only a sum. The other way 'round would have been closer to a proposed definition, but it wasn't presented that way.

  7. The video begins with the statement that "Many of those who claim evolution is wrong have no understanding what evolution really is."

    It's reasonable to expect that what follows is an attempt to explain what evolution really is. And it sure looks to me like that was the intent.

    Maybe you're right. Maybe they changed their minds after two or three minutes and decided to forget about explaining "evolution" in favor of just describing one of the many mechanisms of evolution.

    Do you actually believe that or are you playing devil's advocate?

  8. I am just noting that it was never clearly offered as a definition per se, that an obvious alternative interpretation presents itself, and that the model included exactly as much drift as selection. I don't have any particularly strong opinion either way beyond that. Overall, I think the simulation is useful, but I agree that they should take special steps to avoid the appearance of defining evolution as natural selection. Unless, of course, that is how they define it, in which case you and I will be on exactly the same side.

  9. "Evolution is a process that results in heritable changes in a population spread over many generations."

    A couple of things that I don't understand about this minimal definition.

    Why do you say "spread over many generations"? Doesn't evolution also occur in just a few generations?

    My second one is probably just a matter of style, but why say "a process that results in"?

    In other words, what would be wrong with this:

    "Evolution is heritable change in a population over generations."

  10. Larry Moran seems to oppose "Natural Selection + time = evolution" because it leaves out sampling error (genetic drift).
    Does this alternative definition satisfy you Larry?
    Natural selection superimposed on Variation (mutations, variation produced by selective mating and by sampling error) + time (generations of selection) = ADAPTIVE evolution.
    Darwin replaced the theological argument from "design" to God with a scientific argument from "design" (adaptations) to the natural processes that produce the adaptations.
    The really "interesting" questions for most of us are the theories about the natural processes that bring about the adaptive evolution of "designs" rather than questions about the role of chance in evolution.