Sunday, March 31, 2013

Do I have a "seriously confused picture of how evolutionary biology works"?

David Friedman seems to think so. I made this point:
"You keep speaking as if evolution is a numbers game in the sense that the genes with the most copies win. I don't think that's quite right. There is no winning evolution or if it is it's that your genes survive. There are many more bacteria in the world than there are humans or sharks but I personally feel that humans should feel pretty good about themselves and sharks oughta feel really good about themselves. I don't think more copies is the right metric (if we have to choose metrics at all)."
I was sure he'd disagree on the grounds that I was overstating his point about "winning", but apparently he had no concerns about that part:
"You keep speaking as if evolution is a numbers game in the sense that the genes with the most copies win. "

Yes. That's what reproductive success aka fitness is. It wins in the sense that its genes become more frequent in the population, with the result that later generations are more like it.

Suppose someone who has the normal inclination to truck and barter also has the objective of maximizing the number of children he produces and rears as productive individuals capable of themselves producing and rearing children, instead of the objective of maximizing his own utility with a reasonably conventional utility function. Further suppose that this objective is hardwired into his genes, so that his descendants are likely to have the same objective.

My claim is that, over time, the number of people with that gene, hence the number who behave that way, will increase. Do you disagree?

You might find it worth actually reading _The Selfish Gene_, which I gather from your earlier comment you haven't done. I may be mistaken, but it sounds as though you have a seriously confused picture of how evolutionary biology works."
This seems surprisingly wrong to me and weirdly teleological. First, I don't have any particular disagreement with his supposition, although I would note that the brute force method of simply maximizing reproduction can introduce obstacles to survival of the gene too (think about cancer - it does pretty well for itself until it kills its host). But sure, brute force is one way to make more of yourself. That was never really the issue.

The issue is whether that's some kind of Darwinian golden ticket. Since natural selection hasn't selected that particular trait, that ought to give us pause I think.

Now as I noted to Friedman earlier I'm perfectly familiar with the argument of The Selfish Gene although I haven't read it. I don't deal with evolutionary biology day in and day out so I may have a slip of the tongue on genetic vs. species level phenomena every once in a while - particularly when talking more casually - but that seems minor compared to what Friedman is doing here. Evolution by natural selection is simply the point that the genes that are able to survive will survive. Persistence is the fundamental question. Will a gene persist or not? Prevalence is obviously relevant to persistence to a certain extent, but natural selection has little to say about what level of prevalence is good or bad except insofar as it contributes to persistence.

This is Dawkins's perspective, as far as I'm aware.

He repeatedly refers to both organisms and genes as "survival machines". The point is to survive, not to make more copies of yourself than anything else.

Dawkins writes:
“We are survival machines – robot vehicles blindly programmed to preserve the selfish molecules known as genes. This is a truth which still fills me with astonishment.”
Or:
“Was there to be any end to the gradual improvement in the techniques and artifices used by the replicators to ensure their own continuation in the world? There would be plenty of time for improvement. What weird engines of self-preservation would the millennia bring forth? Four thousand million years on, what was to be the fate of the ancient replicators?

They did not die out, for they are past masters of the survival arts. But do not look for them floating loose in the sea; they gave up that cavalier freedom long ago. Now they swarm in huge colonies, safe inside gigantic lumbering robots, sealed off from the outside world, communicating with it by tortuous indirect routes, manipulating it by remote control.

They are in you and in me; they created us, body and mind; and their preservation is the ultimate rationale for our existence. They have come a long way, those replicators. Now they go by the name of genes, and we are their survival machines.
Or:
“Prediction in a complex world is a chancy business. Every decision that a survival machine takes is a gamble, and it is the business of genes to program brains in advance so that on average they take decisions that pay off. The currency used in the casino of evolution is survival, strictly gene survival"
Now I'm just utilizing google here, but I didn't find anything suggesting that Dawkins thinks that you win evolution by making more copies of yourself than anyone else. Repeatedly the concern is with persistence and survival. Indeed in a weird Paul Ehrlich type moment Dawkins gets very worried about human overpopulation and the threat it poses to the survival of the species. That's hardly the sort of view that Friedman is promoting.

Perhaps I have badly misunderstood evolution and Dawkins, but I really don't think so (at least as far as it concerns the subject here - I'm sure there's plenty else I misunderstand). I don't usually feel comfortable making the sort of sweeping pronouncements that David made about me, but I'm beginning to wonder if he fundamentally misunderstands it.

24 comments:

  1. I'll condense this.

    I think David Friedman is saying "fitness means a gene makes more copies of itself"

    I am saying "fitness means that a gene survives or persists - "more copies" only matters insofar as prevalence might serve the purpose of survival"

    ReplyDelete
    Replies
    1. I'm a little puzzled by your "survive."

      Suppose we start with a population half of whose members carry a particular gene. For some reason, this gene reduces the fitness of those who carry it, in the ordinary sense--carrying it results in decreasing the number of your genes in later generations.

      After a few generations, 10% of the population carry the gene. After a few more, 1% do. The gene has survived--there are still copies out there. But it is having less and less effect on the characteristics of the population because it is becoming less and less common.

      So do you say that the gene is entirely fit, since it still survives--it has failed to be fit only if frequency goes to zero? If not, how do you disagree with my version, which judges fitness by success in getting copies of itself into later generations?

      Delete
  2. Actually, in The Selfish Gene, Dawkins flat out says that he defines gene as what selfishly reproduces itself, so the "selfish gene" is tautological. Literally, he says that.

    But I am trolling, so carry on.

    ReplyDelete
    Replies
    1. Oh it's definitely tautological. The interesting stuff is what that implies about evolution of course. The idea of replicators (genes), and the idea of fitness itself is totally tautological.

      But what is "more fit" - a gene that increases its prevalence twofold in 100 years or a gene that increases its prevalence fivefold in 100 years? It strikes me as being completely irrelevant EXCEPT insofar as one or the other growth rates has an effect on whether the gene is still around 200 years from now and 500 years from now and 1000 years from now.

      Otherwise who cares? Are we jealous of bacteria that they outnumber us so much? Do we give a damn? Of course not.

      Delete
    2. It's precisely that you can't state Friedman's position tautologically that troubles me.

      I can say "a gene that's best at surviving is more likely to be the gene to survive".

      I'm not sure at all if it's necessarily true that "a gene that produces the most copies of itself will survive". Maybe it will. Maybe it won't.

      Delete
    3. "EXCEPT insofar as one or the other growth rates has an effect on whether the gene is still around 200 years"

      Which is precisely the issue we are discussing--not just whether it exists but how common it is. You seem to have somehow tried to convert my point about evolutionary biology into some weird imagined claim about moral philosophy.

      If genes for having and rearing lots of kids become much more common in the human population, the human population will be different than if they don't. Is that so hard to understand? And if that pattern of behavior can be produced genetically and leads to greater reproductive success, in the conventional evolutionary biology meaning of the term, it will become more common. Is that hard to understand?

      Delete
  3. Though "fitness" is a bit outdated and imprecise.

    But the way Friedman is framing everything is the more traditional Neo-Darwinian way of thinking about the issue (which I essentially agree with).

    ReplyDelete
    Replies
    1. Maybe. I don't delve into the literature on this one.

      But that leaves me a bit befuddled as to why Dawkins seems to talk so much about survival of genes and (as far as I can tell) so little about how important high levels of population growth are.

      Delete
    2. Remember - Friedman embraced the idea that it's a numbers game and that we can talk about "winning". I was worried I'd get smacked down for putting it so crudely, but apparently he was fine with that.

      Delete
    3. The point of Dawkins' revolting robots metaphor, part of the argument which you believe you perfectly understand without having read it, is that we are not behaving in the way that maximizes reproductive success--we are the robots, who are revolting by maximizing our utility instead of what our creators, our genes, "want" us to maximize, namely the number of copies of them in future generations. That presumably reflects some combination of the difficult of constructing a phyloprogenitive gene and the slow pace of human evolution. If the genes ever overcome those problems, population growth rates will be very high--for a while.

      Arguing with ideas without having first made the effort necessary to understand them---i.e. being befuddled about what Dawkins says without having bothered to read his book—doesn't work. Your attitude here reminds me of my son's conversation on literature which he hasn't read but knows from the TV Tropes web site.

      The sense of "winning" that is relevant for predicting what will happen to the genetic characteristics of humans is precisely reproductive success. That has nothing to do with whether it is a win in terms of the welfare of humans, the good of the universe, or any other normative criterion.

      Delete
    4. I'm really getting sick of getting lectured by you. I presented you with ideas. You can discuss them or not.

      You've added addendums to what you said yesterday that completely change the meaning of your argument. Fine. As far as I'm concerned the conversation is done, but don't take it out on me that I took your at your word with what you said yesterday.

      Delete
    5. You wrote:

      "I did notice rearing made it into your most recent comment (before this post), but it had never been mentioned before"

      As best I can tell, the post of mine that set off this (multi post) thread was the one that contained the following:

      "Natural selection, after all, is selecting for reproductive success--success in getting copies of your genes into later generations. The simplest way of doing so is to produce as many offspring as you care capable of rearing to the point where they, in turn, can produce lots of offspring."

      You will note that "rearing to the point" was in that post. That's precisely the same point you just claimed had never been mentioned before

      Hence I am not adding addenda--I am merely repeating things you seem to have missed. I am indeed lecturing you--and trying to convince you that you deserve to be lectured. You keep arguing with an imaginary view of my position inconsistent with what I wrote from the very beginning--and then objecting when I point the fact out to you.

      Delete
    6. I stated the argument I thought you were making, for clarification and you agreed - enthusiastically.

      If you didn't think that, don't agree. It's as easy as that. It's not like I was inventing it - I quoted you David.

      I really prefer my commenters not lecture each other - I prefer that they actually hold a discussion. If what you want to do is lecture, I hope you'll respect that and not bring it here.

      Delete
  4. “First, I don't have any particular disagreement with his supposition, although I would note that the brute force method of simply maximizing reproduction can introduce obstacles to survival of the gene too (think about cancer - it does pretty well for itself until it kills its host). But sure, brute force is one way to make more of yourself. That was never really the issue.”

    I’m not talking about “brute force.” I specified someone who produced as many offspring as he could successfully rear to a condition where they were capable of doing the same. “Reproductive success” isn’t about how many children you have—that’s only one mechanism. It’s about how many copies of the gene get into future generations.

    Indeed, reproductive success doesn’t even require reproduction—that’s the simplest and most direct way of getting it, but promoting the survival of your close kin, who share many of your genes, is another mechanism.

    “Now as I noted to Friedman earlier I'm perfectly familiar with the argument of The Selfish Gene although I haven't read it.”

    How can you know you are perfectly familiar with the argument in a book you haven’t read? I think you already said that you were not familiar with the particular bit I was referencing—the revolting robots passage, which is about precisely the question we are arguing.

    “but natural selection has little to say about what level of prevalence is good or bad except insofar as it contributes to persistence.”

    I have said nothing about good or bad—insofar as there was any implication on the subject suggested by my posts, it was that the consequence of evolution to fit current technological circumstances would be undesirable. The question is what characteristics will or will not become more common in the species.

    “He repeatedly refers to both organisms and genes as "survival machines". The point is to survive, not to make more copies of yourself than anything else.”

    The point is not to survive—no multicellular organism succeeds in doing so.

    “Now I'm just utilizing google here”

    You don’t learn a science by googling for random quotes by an author. Read the damn book.

    “Indeed in a weird Paul Ehrlich type moment Dawkins gets very worried about human overpopulation and the threat it poses to the survival of the species. That's hardly the sort of view that Friedman is promoting.”

    It is precisely the point I was making—that if humans evolved fast enough to adapt to the evolutionary implications of current technology, and if a phyloprogenitive gene is something that can be produced via mutation, we would end up with a population of people all of whom were having as many children as they could produce and successfully rear. As it happens, I don’t think population is a serious concern in the world as it is, in part because human evolution is slow due to our long lifetimes. But in that world it would be.

    This is very frustrating. You seem to have invented an argument you think I am making without paying any attention to the argument I have repeatedly made, as well as inventing your own version of evolutionary biology a la Dawkins without having actually read Dawkins’ book. In addition to reading _The Selfish Gene_ you might also want to reread my posts in this thread.

    ReplyDelete
    Replies
    1. I did notice rearing made it into your most recent comment (before this post), but it had never been mentioned before so I wasn't sure how essential it was to your point.

      Your point does seem to be much closer to mine if you condition on that.

      re: "The point is not to survive—no multicellular organism succeeds in doing so."

      Not organisms - genes. Organisms are the vehicle for gene survival.

      re: "You don’t learn a science by googling for random quotes by an author. Read the damn book."

      Oh give me a break David. Respond to my arguments or don't, but but don't lecture me.

      re: "This is very frustrating. You seem to have invented an argument you think I am making without paying any attention to the argument I have repeatedly made,"

      I find it frustrating too - you now seem to have moved into arguing my position and abandoning mere numbers (which you said quite clearly was how you "win" just yesterday), all the while condescending to me. If you didn't mean what you said initially, fine. After reading this clarification you seem much closer to what I've been saying. But I've been quoting you throughout precisely not to invert anything.

      Delete
    2. "you now seem to have moved into arguing my position and abandoning mere numbers (which you said quite clearly was how you "win" just yesterday)"

      Where the form of winning being discussed was reproductive success, that being what is relevant to the effect of technology on human evolution, which is the subject we were discussing. As I explained.

      You wrote:

      "re: "The point is not to survive—no multicellular organism succeeds in doing so."

      Not organisms - genes. Organisms are the vehicle for gene survival."

      Correct. But I was responding to your writing (about Dawkins)

      "“He repeatedly refers to both organisms and genes as "survival machines". The point is to survive, not to make more copies of yourself than anything else.”

      You were the one who was talking about organisms as survival machines, and I was pointing that they were not.

      And, as I've just pointed out in a response to a different one of your posts, rearing was in the very first post of mine (I believe it was the first) in this particular subthread.

      Delete
    3. Yes. And Dawkins calls them "survival machines". They are the vehicle for genes, which he terms "survival machines".

      Organisms are survival machines.

      Genes are the unit of analysis, and their persistence in a population is the fitness we are concerned with. Go back to Dawkins. Organisms are the "survival machines" that genes use.

      You keep thinking I'm making this basic gene/organism confusion. I have no idea where you're getting that or why you think you need to explain that to me.

      re: "And, as I've just pointed out in a response to a different one of your posts, rearing was in the very first post of mine (I believe it was the first) in this particular subthread."

      OK, but can't you understand my confusion if you say "yes" to and proceed to adamantly defend that it is just about making copies and that you can "win" evolution. It's the maintenance of the next generation of organisms carrying a particular gene that I have been mentioning the whole time, after all. If you agreed with that the whole time, can't you understand how it would be confusing for you to argue vigorously against me when I was making those points?

      Delete
    4. You agree that the point you have been reproaching me with not making was in the post of mine that started the discussion. That's progress. It follows that every post of yours which assumed I was talking only about number of offspring produced was mistaken, along with posts of yours that accused me of changing my position, which ought to explain what I have been irritated about and justify my claim that you were attributing to me a position out of your imagination.

      I don't know what you meant, or thought I meant, by "winning" evolution. What I have been saying throughout is that getting more copies of a gene into later generations results in the characteristic produced by that gene becoming more common. That's the sense of winning that is relevant in a discussion of the consequences of technology for evolution. And that is simply a matter of numbers--as I agreed. If each generation has a higher relative frequency of a phyloprogenitive gene, then the number of people who act accordingly will be increasing. The relevant numbers aren't the numbers of offspring produced but the numbers of copies of the gene that end up in future generations.

      So no, I don't understand the source of your confusion. You seem to have started with the idea that I was confusing number of offspring produced with reproductive success, despite my initial post making the relevant distinction (a point repeated several times thereafter), and with some independent confusion about what conclusion I was arguing for. I've been talking throughout about the implications of evolution, not about what is or is not good for humanity, which is the best guess I can make (not a very good one) about what you thought I was arguing about.

      You might want to reread the thread if you want to figure out how and why your perception of what I was saying diverged from mine. It might even be sufficient to reread my first post in the thread, in the light of the current exchange.

      Delete
    5. I've looked at the thread. You notice in reading the conversation that the disagreement started when you were solely talking about making lots of copies (the technology discussion was a different disagreement). I couldn't believe you were making a "lots of copies" point, looked for clarification, and you embraced it.

      I can't read your mind if you tell me that's what you mean.

      I can't see why it's hard for you to understand why this is confusing, but you're not pushing that point now and honestly I'm sick of this. As I noted - I suppose we're on the same page, but it took an awful lot of grief to get there.

      An important point is I wasn't just making stuff up about you David. That's why I asked you and then quoted you. You can't very well get upset with me when I'm just sharing your exact quote.

      Delete
    6. I can get upset with you when you repeatedly claim I have just introduced something that I actually introduced in my first post of the discussion. And when you finally discover that what you have been claiming isn't true, your response is to try to find some way of attributing your misunderstanding to me.

      I'm sure you weren't deliberately making stuff up about me--there would be no point to that. But you were, as best I can tell, attributing views to me that were not consistent with what I had posted. That's confusion, not deliberate dishonesty, but enough of it is still irritating.

      I did tell you what I meant, over and over again. Eventually you decided that meant that I had changed my position and now agreed with you.

      I don't know where you think that the disagreement started. Reading the thread is a little complicated since it appears in several different comment threads. But you seem to think I somehow abandoned the distinction between reproducing and reproductive success when you wrote:

      "You keep speaking as if evolution is a numbers game in the sense that the genes with the most copies win."

      and I replied:

      "Yes. That's what reproductive success aka fitness is. It wins in the sense that its genes become more frequent in the population, with the result that later generations are more like it. "

      You didn't say "the gene that produces the most copies wins." If a gene produces lots of copies all of which die two hours later, it doesn't end up with lots of copies. And my answer started with "yes" and then specified in precisely what sense it wins.

      My explanation, in the post in question, included "rears as productive individuals capable of themselves producing and rearing children," which surely made it clear that I wasn't talking about producing children who then couldn't reproduce.

      I notice, just after that, your writing: "Bacteria haven't "won" evolution ...," which looks as though you are thinking of winning evolution in terms of the interest of the species, which is not what I ever said or implied.

      It's true that in the post before your "numbers" post I referred to passing on copies of genes without explaining what that meant, but I had already earlier specified producing and rearing and I specified it again in the post responding to your numbers post--the one which you seem to think confirmed your view of my position.

      The whole thing was about technology's effect on evolution. The post to which your numbers post responded contained the phrase: "and if humans evolved fast enough to reflect the implications of modern technology."

      To avoid further confusion, what I refer to as your numbers post is:
      http://factsandotherstubbornthings.blogspot.com/2013/03/technology-and-future-human-evolution.html?showComment=1364672743914#c3139185474942422555

      The post of mine it responded to was:

      http://factsandotherstubbornthings.blogspot.com/2013/03/technology-and-future-human-evolution.html?showComment=1364664253233#c4508594295216975161

      The post mine that responded to it was:

      http://factsandotherstubbornthings.blogspot.com/2013/03/technology-and-future-human-evolution.html?showComment=1364697501827#c2354534707497208513

      Delete
  5. I think I get your point. It's an odd angle to approach the problem from - definitely unconventional.

    You're saying that if, for example, gene A forms only 2% of the population and survives for 100 years and gene B forms, on average, 20% of the population but dies out after only 10 years, over this 100 year period it would be odd to say B is the more fit despite it being more frequent in the population (at certain times).

    I've done some reading of evolutionary mathematics. "Evolutionary Dynamics" by Martin Nowak is a fantastic way to get an idea of the mathematics of evolution. And Evolutionary game theory is a large part of that apparatus.

    In all cases that I can think of, fitness is defined as the frequency of a gene within a population. Maybe some work has been done along the lines of how you're thinking, but I don't think it's the convention.

    I can think of a few reasons why the convention is what it is. First, it's just more intuitive to think of fitness as the frequency of a gene. If I pick it out a time period, how do I say which gene is more fit? The one that's more frequent. There's no way to do this under your definition unless you look at "all of time".

    It also makes more sense because the actual selection is taking place contemporaneously.

    And last, how do you weight prevalence vs. persistence? Would you give all of the weight to persistence?

    But all together, I think there's something to your point. I guess I'd just say it's semantics. "Fitness" in evolutionary jargon is not interchangeable with "success" as conventionally thought of.

    ReplyDelete
    Replies
    1. re: "In all cases that I can think of, fitness is defined as the frequency of a gene within a population. Maybe some work has been done along the lines of how you're thinking, but I don't think it's the convention."

      I'm not sure I understand your point - this is my understanding of fitness too.

      The point is there's no particular advantage to high levels of reproduction, because it is the of the gene frequency within the population (not the size of the species's population) that matters. High population growth is irrelevant - what matters is the differential survival rate of the gene within the population. Small differences in these rates will shift the frequency of the gene in the population over time.

      I'm not saying this as if you don't know - I'm making the point that this is my point.

      We can imagine how high birth rates can prevent a gene from successfully being passed on, which is exactly why in this comment thread David is adding that he also cares about whether offspring could be successfully reared and able to successfully rear subsequent offspring.

      He didn't mention anything about this yesterday, which I think was the source of the confusion.

      Delete
    2. "He didn't mention anything about this yesterday, which I think was the source of the confusion."

      As I have pointed out in response to other places where you say this, "rearing" was in my very first post in this subthread.

      Delete
    3. I just noticed that you specified "yesterday." My post responding to your numbers post, which was March 30th, which was "yesterday" when you posted on March 31, contains:

      "Suppose someone who has the normal inclination to truck and barter also has the objective of maximizing the number of children he produces and rears as productive individuals capable of themselves producing and rearing children"

      So I not only mentioned something about it, I specified it in full detail.

      Delete

All anonymous comments will be deleted. Consistent pseudonyms are fine.