Guy hoelzer 2007-10-29 07:57:23
in article email@example.com, John Edser at
I have explained my position in a number of sbe threads, but I will do it
again here. The multilevel selection selection model is more parsimonious
than the single-level selection model, because it does not require the
additional factors that must be invoked to limit the process of natural
selection to action at only a single level. What is it that you invoke to
prevent natural selection from occurring at other levels? I have not
perceived any logic in your arguments to date, so an straightforward answer
to this question might help me.
As an aside, I don’t like the slam on McGinn and I don’t agree that more
parsimonious models are inherently more easily tested than less parsimonious ones.
I have already responded to all of these points of disagreement.
It seems futile at this point to continue correcting your interpretation of
Hamilton’s variables, but… I would be surprised if you have ever found a
single individual with even the slightest familiarity with Hamilton’s rule
to agree that it assumes or predicts ANY particular relationship at all
between “c” and “rb”. It emphatically does not. The rule represents a
threshold of this quantitative relationship between selection against and
selection favoring a mutation causing altruistic behavior. It is precisely
analogous to the following: d(you) < d(water), where d(you) is your density and d(water) is the density of water. This is a rule for the conditions under which I expect you to float. This rule does not predict that you will float, nor that your density will be less than that of water. In all the years you have been sounding this alarm, you have never explained it in a way that gave it any credibility IMHO. It seems more like hot air every time. I agree that what you call "fitness mutualism" has often been confused with altruism in uncritical interpretation of observations, but that seems besides the point here. I also think that your notion of "fitness mutualism" fails to adequately account for the risk associated with waiting for payback in systems of reciprocation or community "fitness banks." Then it is incumbent on you to explain what it is that prevents the process of natural selection from manifesting under those conditions. I don't have to do this any more than you have to show that fitness heritability is not independent at every other level. Nevertheless, I have provided solid arguments for the autonomy of selection at levels other than the individual, and it seems to have fallen on deaf ears. Why should I shout louder when you are not trying to listen? Well, as long as you are sure. [Sorry - I couldn't resist the sarcasm.] I looked into it a bit further and found that Darwin did invoke group selection to explain eusociality among insects, and it is a central theme of his book on human evolution, because he explicitly acknowledged that humans and eusocial insects often act in selfless ways. The "i.e." bit is about you, not about Darwin. He clearly argued that species are real entities and NOT just "human statistical" artifacts. What? You can't have both heritability and plasticity? My own view is that species are often not as discrete as Darwin thought. Well, I don't think that organisms are as discrete as Darwin thought, either. The process that defines your existence, for example, requires the involvement of many individuals and species. Consider lichens as perhaps a more concrete example. Both. All manifested processes are both autonomous and open, thus allowing influences in and out to the rest of the universe. The same way we measure heritability at the individual level. You typically measure the extent of phenotypic correlation among relatives in a population. In the case of species heritability, which is certain to be extraordinarily high for many traits when speciation occurs through fission, you would examine phenotypic correlations (try to think beyond the phenotypes of individuals here) among closely related species. Note that speciation sometimes happens through hybridization, which is analogous to sexual reproduction at the species level and would greatly reduce species level heritability just as sexual reproduction does at the individual level. First, I argue that selection at level B has resulted in the evolution of mechanisms undermining, but not entirely preventing, selection at level A. I agree with those who go on to argue that selection at levels higher than B (e.g., the species level) is increasingly resulting in the evolution of mechanisms that undermine selection at level B. As I have explicitly argued to you before, I see no temporal sequencing of selective consequences among levels, so your selection first arguments make no sense to me. If that were generally true, then selection at level A would never have been undermined by selection at level B, so your theory seems self-inconsistent. The degree of independence achievable at levels A and C allow them to oppose the effects of selection at level B, even if level B is currently the dominant level of selection in most biological systems. It is not (e.g., meiotic drive). There is no first place in a cycle. Effects flow both up and down in the biological hierarchy. Group selection effects on individuals would not cause the extinction of individuals. It would influence their phenotypes, and could result in a protection of individuals that individual level selection itself might tend away from. "Winning over" does not mean absolute victory to any multilevel selectionist that I know, so your description of multilevel selection theory is a straw man. Note also that "agency" is largely limited to within level effects. You typically have very little agency through interactions with cells, or genes, or quarks, or species, or ecosystems, or... However, you frequently have strong agency in your interactions with organisms on a scale similar to your own. Because models can be true in part, and are always false in part. Also, no particular point of refutation would constitute all possible points of refutation, so failing to find such a point in no way supports the model. I have explained before why I think the notion of refutation in science is flawed. I know we have been through this before, but please give me an example of a point of refutation for your theory as an example, even though I think it is a pointless exercise. Since I don't think that anything in the universe is entirely independent of anything else in the universe (not even any indirect effects), this would be pointless. I do agree with McGinn on this point, and I think it is the ONLY logical position to hold regarding selection at different levels. I have never seen an argument from you or anyone else supporting the bottom-up sequencing of selection effects. If you have such an argument, please share it with us. The argument advanced by McGinn and I (and all multilevel selectionists AFAIK) is simple and makes perfect sense. Natural selection happens when and where there is heritable variation for fitness. It does not depend on the nature or timing of selection elsewhere or at other times, although selective effects from other levels can influence the conditions that are manifested at a particular level. Thus, bottom-up and top-down effect cascades can affect the potential for selection at a particular level, although I think that these cascades usually happen without regard to influence over the potential for selection at other levels. When you feed your cat you reduce its potential to live "in the wild", but this is not why you feed your cat. That is not a fair appraisal. I agree with Popper on some points and disagree with him on others. I am even able to make the arguments of Popper with which I disagree to help others understand his position. I don't think that I have not been inconsistent in the philosophy of science that I have advocated. If you want to argue otherwise, then show me my inconsistent statements. Actually, I have posted enough on sbe that there must be some inconsistencies out there that could be found, and I would be happy to either own up to them or explain why they are not really inconsistent. Well then, you had better tell the theoretical physicists of today who are now considering non-stationarity in the speed of light. I think the speed of light constant is an example of an assumption that was never really a conclusion, and which has come to seem like Truth after a sufficiently long period of time without questioning it. If it was not testable, then how could it have been refuted? While I obviously do not fall into the fixity of species camp, even this idea has a toe-hold in the observation of punctuated equilibria. Refutation has not happened here in a way that allows us to dismiss the idea altogether. So, I am still looking for a single example of true scientific refutation. I am sure that racism would happen with or without science. It is not a consequence of scientific practices. Cheers, Guy