Evolution for Everyone

In T&R XIV I showed that prejudice against group selection is impervious to evidence from laboratory experiments. It is also impervious to evidence from the wild.

I will focus on one of many examples that can be provided. In 1995, Robert Heinsohn and Craig Packer published an important paper on territorial defense in lions in the journal Science. As good experimental field biologists, they had played recordings of lions from neighboring territories to observe how females of the focal territory responded. They discovered that the same individuals consistently arrived first at the scene while others consistently lagged behind. There seemed to be bravehearts and cowardly lions within the same pride.

Heinsohn and Packer looked for an advantage to counteract the cost of territorial defense for the bravehearts within their own pride and couldn’t find it. The bravehearts weren’t socially dominant, they didn’t have more offspring, and they didn’t punish the cowardly lions, who simply seemed to be cheating and getting away with it. The bravehearts were providing a public good at their own expense, an animal version of the tragedy of the commons made famous by Garrett Hardin in the 1960′s. Here is how Heinsohn and Packer described the situation to the best of their knowledge:

Female lions share a common resource, the territory; but only a proportion of females pay the full costs of territorial defense. If too few females accept the responsibilities of leadership, the territory will be lost. If enough females cooperate to defend the range, their territory is maintained, but their collective effort is vulnerable to abuse by their companions. Leaders do not gain “additional benefits” from leading, but they do provide an opportunity for laggards to gain a free ride.

Let me pair this passage with the canonical passage by Darwin that I also quoted in T&R II:

It must not be forgotten that although a high standard of morality gives but a slight or no advantage to each individual man and his children over other men of the same tribe, yet that an increase in the number of well-endowed men and advancement in the standard of morality will certainly give an immense advantage to one tribe over another. There can be no doubt that a tribe including many members who, from possessing in a high degree the spirit of patriotism, fidelity, obedience, courage, and sympathy, were always ready to aid one another, and to sacrifice themselves for the common good, would be victorious over most other tribes; and this would be natural selection.

I hope you can see the similarity between these two passages. Darwin was identifying what I call the original problem; traits that benefit the whole group are often disadvantageous within the group. The counterbalance for cheating does not reside within the group; it resides in the process of more cooperative groups outcompeting less cooperative groups . Darwin’s example was hypothetical. Heinsohn and Packer seem to be providing a real-world example with territorial defense in lions. They even stress the importance of between-group competition as the primary influence on the evolution of lion sociality.

Before continuing, let me issue two caveats. First, I have the highest respect for Heinsohn and Packer. They are topnotch scientists who can only be admired, not only for conducting such arduous research but also for attempting controlled experiments in the wild. Second, the last word has not been written on lion social behavior. Perhaps they or someone else will find a within-group advantage for bravehearts in the future. I’m interested in how they interpret their current data. They don’t interpret it as a provisional example of group selection. They don’t even mention group selection as a possibility. I doubt that it even occurred to them to regard group selection as a viable possibility, even though a description comparable to Darwin’s flowed from their own pen!

For those who feel impelled to shout “kin selection!” because lion prides are composed of related females, my reply is the same as for the chicken example discussed in T&R XIV. Genetic relatedness explains why bravehearts are clustered in some prides and cowards in others. Cowards still have the advantage within each pride, the observation that Heinsohn and Packer find so puzzling. If they had a better understanding of kin selection seen through the lens of multilevel selection theory (see T&R XIII), they wouldn’t be so mystified by their own data.

This example illustrates a problem that pervades the evolutionary literature. Group selection became such a pariah concept that most people don’t look for it. They haven’t looked for it for so long that they forgot what it looks like and can’t even recognize it when it bites them in the butt! Even when they do recognize it, they are tempted to describe it in a way that doesn’t use the G word to make it more palatable to their peers. In this fashion, our students continue to learn that the rejection of group selection was a triumphant advance for evolutionary theory while the evidence for group selection lies all around us.

Science is no better than political revisionism if this situation is allowed to persist. One reason that I am writing this series of blogs is because I am an idealist about science. I regard it as the best cultural system we have for holding people accountable for what they say. Scientists have a responsibility to keep track of the history of their ideas and to acknowledge mistakes from the past, no matter how large. Unfortunately, like religion, science as practiced often falls short of science as idealized. The group selection controversy is an embarrassment for science and the sooner its shortcomings are corrected, the better.

To be continued.

Comments

  1. #1 Bob O'H
    November 8, 2009

    Sorry, I’m not buying the argument. Your only evidence is to say that two passages have some similarity. But you have done nothing to show that there is a fitness cost for the lions that defend. More lions = better defence, but one might imagine that the marginal benefit of adding another lion decreases. Depending on how the lions make their decisions about how to defend, it might be possible that when a lion choses, it chooses the behaviour that is individually optimal.

    In a system like this, I would also expect policing to have evolved: freeloaders are punished somehow (e.g. by being pushed to the back of the food queue). Or there could be specialisation occurring: the freeloaders give to the group in another way. If there is no punishment for misbehaving, it might be that there isn’t freeloading but something else.

    I don’t know the system well enough to judge, so maybe thus is genuine group selection. But if you’re going to claim it, you have to present the evidence that the conditions are met. Otherwise you’re not doing science, but propaganda.

    As with many of the posts in this series, I come away disappointed by the lack of evidence for the claims made. I too am an idealist about science but I think that evidence and justifying one’s statements is part of the ideal.

  2. #2 piker
    November 8, 2009

    Animals within groups are cooperatively competitive, using conditional strategies appropriate to their individual status, that status also being conditional on a variety of capabilities.
    Groups of such animals then use cooperatively competitive strategies when dealing with different and in some cases kin-selected groups of others in the species.

    Strategies evolve. Big whoops. Time to look now at how this happens. How the lessons from experience become instinctive. No mention of that so far in this series at all – unless lack of response to commentary on the subject counts as a form of mention.

    Check this out: http://www.ncbi.nlm.nih.gov/pubmed/9782102

    J Theor Biol. 1998 Jun 21;192(4):445-453.
    Stability with Inheritance in the Conditional Strategy.
    Gross MR, Repka J.

    Department of Zoology, University of Toronto, 25 Harbord Street, Toronto, Ontario, Canada M5S 3G5
    The conditional strategy is a theoretical framework that explains the existence within populations of individuals that express alternative behavioral, physical or life history tactics (phenotypes). An example is fighters and sneakers in many animal mating systems. In the conditional strategy the alternative tactics are chosen by individuals based on their state, for example large or small bodied. Since state is often heritable, due for example to additive genetic variance, the alternative tactics may also have inheritance. As the tactics do not have equal fitnesses, it is generally believed that any such inheritance would prevent the evolutionary stability of the conditional strategy. However, in previous work we introduced an Inheritance Theorem and were able to prove that a conditional strategy with tactic inheritances can have a unique equilibrium proportion of the tactics. We now prove a second property of our Inheritance Theorem, namely the stability of the equilibrium. This means that if the tactics are perturbed from their equilibrium proportions, they will return across generations to their equilibrium proportions. An example is provided in mites. We have therefore established an Inheritance Theorem which includes both the existence of an equilibrium and its stability for alternative tactics in a conditional strategy.Copyright 1998 Academic Press Limited

    PMID: 9782102 [PubMed - as supplied by publisher]

  3. #3 InfuriatedSciTeacher
    November 8, 2009

    Bob>
    The cost for lions acting in defence is energetic, firstly, and the increased potential for bodily damage secondly. That didn’t require a significant amount of imagination to produce, its plausible, and it’s backed by study of other animals (I’ll go look for references if you insist).
    As for your statements on possible punishment of freeloaders or them giving to the group in some other way, that is a possiblity… it still needs to be documented as well. The point of the post seems to be that group selection isn’t being considered as a possible explanation, rather than that it’s an ironclad example of group selection in action. Perhaps we’ve interpreted that differently, and conveniently the author is available to determine who’s perception is correct.

  4. #4 Bob O'H
    November 8, 2009

    InfuriatedScienceTeacher – first, I am aware that there are costs associated with defence, my point is that there are also benefits, and one has to properly account for them all.

    On whether DSW is saying that this us just a possible explanation, I think the first paragraph pretty clearly sets this up as evidence for group selection.

  5. #5 Jamie
    November 8, 2009

    I have no opinion on whether group selection receives due consideration as an explanatory hypothesis in general. But clearly there are other possible hypotheses that this author does not consider. Until someone isolates the lion ‘boldness’ gene, such field experiments must control for birth order, early developmental nutrition, maternal relationship, early developmental experience etc. Before arguing that a particular genetic trait might or might not be selected at the group level it seems to me that one ought to establish that one is actually discussing a genetic trait. I have no problem inferring territorial behavior itself has a genetic basis, but subtle variation in territorial behavior may also have other sources. The success of sociobiology may mean there is no longer a “nature/nurture controversy” but it does not mean that nurture no longer plays a role in animal behavior.

    By all means call for due consideration of your theories. But before claiming the rational/moral high ground you must give due consideration in return.

  6. #6 piker
    November 8, 2009

    Are groups of prey a selective factor with regard to the composition of groups of predators?

    Are groups of prey by their natures determinant in the evolution of single predators? Do single predators in turn provoke individual adaptations in those groups in turn?

    Etc., etc.

    Singling out group selection as some sort of primary cause in an evolutionary chain of effects is like putting blinders on the rider instead of the horse.

  7. #7 David Sloan Wilson
    November 8, 2009

    Let me make a few comments about Bob’s post. When he says “it might be possible that when a lion chooses, it chooses the behaviour that is individually optimal” the metric seems to be absolute fitness. If anything should come through in the T&R series, it is that natural selection is based on relative fitness and within-group selection must be evaluated on the basis of relative fitness within groups. Everything known about the example points to a fitness difference within prides favoring the cowards over the bravehearts. The investigators looked hard for a compensating benefit for the bravehearts within prides and came up empty. They also stress the paramount importance of competition among prides. Based on the available information it looks like a classic case of multilevel selection.

    The main point of this installment, as InfuriatedScienceTeacher notes, is that the investigators failed to see group selection as a possibility, even when it was staring them in the face. I myself stressed that the final word hasn’t been said on lion social behavior and perhaps within-group benefits for bravehearts will be found in the future. The question was what the investigators did with the available information.

    It might sound scientifically rigorous to list a bunch of additional factors that weren’t tested, but science would grind to a halt if such rigor were applied to all hypotheses being tested. Take Jamie’s list: “Until someone isolates the lion ‘boldness’ gene, such field experiments must control for birth order, early developmental nutrition, maternal relationship, early developmental experience etc.” If you think that all of these things have been determined in the field for any trait that evolved by any level of selection, think again. Science involves making judgement calls on the basis of available information. If a hypothesis is deemed extremely unlikely, like a paranormal phenomenon, then extremely convincing proof is required. I have already shown that between-group selection is fully plausible. Saying that between-group selection requires more proof than within-group selection is like saying that competition requires more proof than predation. They are both plausible, so we need to make judgment calls on the basis of available information.

    If you want an example of group selection in the wild that is supported by more data, I offer female biased sex ratios. See chapter 2 of Unto Others for a review. This example is instructive because the evidence was accepted under the name of “local mate competition” and it was a struggle to establish that group selection was involved, even though Williams (1966) used the absence of female-biased sex ratios in nature as his best empirical test against group selection. If the data was good enough to be accepted under the name of local mate competition, it’s good enough to be accepted under the name of multilevel selection.

  8. #8 piker
    November 8, 2009

    Again from: J Theor Biol. 1998 Jun 21;192(4):445-453.
    Stability with Inheritance in the Conditional Strategy.
    Gross MR, Repka J.
    “The conditional strategy is a theoretical framework that explains the existence within populations of individuals that express alternative behavioral, physical or life history tactics (phenotypes).”

    You seem to pick and choose items to respond to that are easy pickins’. When it comes to the hard stuff, from a respectable source, you often have nothing to say. Is it because you can neither agree or disagree without disrupting the flow of the preconstructed series, and the preselected conclusions? Is that what your silence is telling?

  9. #9 daedalus2u
    November 8, 2009

    piker, if there is a conditional strategy, that conditional strategy had to evolve. Presumably before there was an organism that could choose between two strategies there was a more primitive organism that had only one strategy, and before that an organism that had no strategies.

    How did the ability to exhibit a cooperative strategy evolve? Cooperation is only advantageous when coupled with one or more cooperating partners. Making cooperation conditional doesn’t solve the problem of how it evolved in the first place.

    The conditionality of strategy may be “hard wired” and not a “choice”. One of the things I am working on is the physiology behind the “cycle of violence”. There is a very robust observation that individuals who are abused and subjected to violence in utero, as infants, as children, as adolescents, as young adults become more violent. The physiology of this change need not be conscious, and is likely not. Epigenetic programming of multiple physiological systems is known to happen. It would be surprising if epigenetic programming of physiological systems responsible for behavior did not happen also.

    Physiology equivalent to the “cycle of violence” is observed in animals also. Are those organisms making “choices”? I suspect that at least some of the behavior observed in the water-striders is due to epigenetic programming of gametes depending on the conditions their parents experienced. There are indications that parental imprinting can have endocrine effects in mammals.

    Plasticity of behavior doesn’t solve the problem of how the behavior evolved in the first place. There is no “default” unevolved behavior. Every behavior evolved from organisms that didn’t exhibit it before they did exhibit it. How did those things happen? We don’t know, but we know that they did happen. As DSW says, unless you are prepared to posit a paranormal (i.e. supernatural) event, the path we are convinced these behaviors developed on is through evolution.

  10. #10 piker
    November 9, 2009

    daedalus2u, are you speaking for DSW because if so, you’re not doing him any favors. There’s no question hanging in the air here as to behaviors having evolved without supernatural assistance in the process.
    You write about: “a more primitive organism that had only one strategy, and before that an organism that had no strategies.” There was never a living organism that had no strategy that was also able to evolve into one that did. There is no conflict between being hard wired and conditional. There IS one if the hard wired strategies prevent the organism from having the benefit of experiences so that the hard wired conditional will also be adaptable.
    The paper I referenced above deals with the question of how this adaptation can have happened, and among books and writers others have pointed to, in Evolution in Four Dimensions, authors Jablonka and Lamb have much of their book devoted to the subject.

    You’d do well to read some of this before proceeding with your research. DSW has already done so, I’m sure. Which may be why he’s avoiding any mention of a competing theory that he hasn’t yet found a way to convincingly counter.
    .

  11. #11 David Sloan Wilson
    November 9, 2009

    To Piker: Again and again you raise the issue of conditionality as if it’s a game changer, when it’s not. There are closed and open forms conditional behavior. The closed forms are sensitive to very specific environmental inputs and result in specific behavioral outputs, such as the fight or flight response to a loud noise. The open forms rely upon variation and selection processes, in which selection is guided by genetically evolved reinforcers, as Skinner called them. Then we have a range of inheritance mechanisms, such as standard Mendelian, new-fangled biological inheritance mechanisms discussed by Jablonka and Lamb, and cultural inheritance mechanisms discussed by Richerson and Boyd.

    In all cases, selection operates on phenotypes and phenotypic fitness can be partitioned into within and between group components. The mechanism of inheritance changes the details but not the basic framework. For example, one cultural transmission rule is to copy the majority behavior of one’s group, which causes groups to become phenotypically uniform for whatever behavior is initially in the majority. If a population follows this transmission rule, selection will be concentrated entirely at the between-group level because there is no phenotypic variation within groups. We still must consider the genetic evolution of cultural transmission rules; what happens if a mutant genotype refuses to copy the majority? This will result in a new pattern of phenotypic variation within and between groups, which maps onto genetic variation within and between groups, and so on. It gets a bit complex, but it doesn’t challenge the basic framework of multilevel selection. Here’s an example from my own work:

    Wilson, D. S., & Kniffin, K. M. (1999). Multilevel selection and the social transmission
    of behavior. Human Nature, 10, 291-310.

    Finally, you haven’t been questioning anyone’s basic intelligence lately, but you still have a penchant for scolding them for ducking the tough issues, not being sufficiently well read, etc. There is wisdom in proverbs, including “people in glass houses shouldn’t throw stones.”

  12. #12 Chester Burton Brown
    November 9, 2009

    As someone joining the discussion late, I’d like to point out that it would be much easier to catch up and follow along if the referenced previous entries were hyperlinked.

    Yours,
    CBB

  13. #13 Omar Tonsi Eldakar
    November 9, 2009

    Bob O’H, considering I have studied punishment for a bit, I caution using punishment as an alternative to group selection. The punisher provides a public good by keeping freeloaders in-check but at a personal cost. Non-punishers benefit from the punisher but do not share in the costs (i.e. non-punishers are freeloaders to punishment). So a punisher has a lower fitness than non-punishers within that group and therefore fizzle out. If you state that maybe others are greatful to the punisher and reward him, well then they lose to those that do not reward the punisher and follow the punisher to extinction as well. I don’t mean this as a hostile comment but any illusions of punishment regulating cooperation without invoking group selection is just not true. Punishment itself is a means to minimize phenotypic variance within the group and increase that groups’ fitness compared to other groups. By the way, bringing up selfish-punishment won’t change things as group selection decreases its effectiveness so supports group selection in its own right

    Also David should have posted this paper in clearing up the absolute fitness argument.
    http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VJ1-4BVPSGJ-3&_user=10&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_searchStrId=1084732369&_rerunOrigin=scholar.google&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=efc109b52ff3c38f7d847f8fbf39dbe2

  14. #14 piker
    November 9, 2009

    Dr. Wilson, thanks for responding. But isn’t stone throwing one way to determine the glassiness of the structure? And my challenge isn’t to your framework, it’s to the buttressing, and especially to the strength of the foundation. I don’t think the glass house metaphor is apt in any case as it’s not I who have built one here and opened it up to the public for critical evaluation.

    As to ducking the tough issues, I still see some bobbing and weaving here, like dismissing the content of my referenced material as simply “new-fangled biological inheritance mechanisms.”

    And I’ve read your principal writings and papers, and will read the one just cited. In turn, I don’t see where you get off saying I was scolding anyone by citing things to read as well.
    If you’re referring to Daedalus2u as having been scolded, it was he who challenged me directly to answer some questions, and I think I went out of my way to answer most of them. Some were best answered by recommended reading. Hardly the same as telling him he was not sufficiently well read per se. Did I imply he lacked a sufficient understanding of what he’s read on this subject so far? Sure. I think he does. I think we all do.

    And what you have just written in your response tells me a lot about how and why you’ve gone wrong. You’ve provided a litany of your premises as sufficient rationale for your hypotheses. Such as “there are closed and open forms conditional behavior,” as if there’s a distinction there that defines the differences. You posit arbitrary dichotomies all over the place as if that’s all one needs for an explanation.
    There are more of such assumptives here than I had expected, and that are in dire need of re-examination. The old-fangled as opposed to the new-fangled would be the distinction that ill defines the difference.

    But no more throwing stones here today. Feels too much like beating my head against a stone wall.

  15. #15 daedalus2u
    November 9, 2009

    piker, no, I speak for no one but myself. You say

    ”There was never a living organism that had no strategy that was also able to evolve into one that did.”

    Neglecting that it is populations that evolve and not organisms (and assuming you simply miswrote), do you have any data or theory to back that up? It sure seems to me that you are saying that the ability to have a “strategy” is an example of irreducible complexity. If so, then how did the first organism with a “strategy” come into existence other than by supernatural poofing?

    I really don’t understand how you reconcile the terms you are using with what I understand to be non-supernatural evolution. Are you using idiosyncratic definitions of “living organism”, “strategy” and/or “evolution” such that “living organism” is defined to be equivalent to “organism with a strategy” and that all “evolution” only came after “living organisms” had “strategies”? If so, then we are talking past each other and talking about completely different things.

    As I understand evolution, many of the steps in abiogenesis can be thought of in evolutionary terms. The puddle of peptides that was able to reduce nitrite to ammonia and so generate more amino acids and more peptides in a sense “evolved” in that it was more “fit” than the puddles of peptides that couldn’t reduce nitrite to ammonia.

    It turns out that cytochrome c is an excellent nitrite reductase, even better when it is denatured by boiling, or cleaved via proteases to 9 or 11 peptides. It is the only heme enzyme where the heme is covalently bonded to the peptide chain. Are we all descended from puddles that had a primitive cytochrome c in them? If so was that group selection or individual selection?

  16. #16 daedalus2u
    November 9, 2009

    OTE, that is very interesting. As I was going through PubMed, I found this:

    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1691731/?tool=pubmed

    and it occurred to me that if males adopted a cooperative strategy as a disadvantageous trait to signal male fitness (as the peacock’s tale is thought to indicate), and if females then chose such males, then a stable equilibrium might result.

    An individual female might have better reproductive success when paired with a cooperative male (because to be cooperative and successful he needs better genetic fitness elsewhere), but her children might have better reproductive success if they were defectors and received defector-type genes.

    This might play out as changes in mate choice over time, choosing defector-type males when young (and foolish) and reliable cooperative-type males when older (and wiser).

  17. #17 piker
    November 9, 2009

    Daedalus2u, I promised not to throw any more stones today, but not to avoid answering questions, so here goes.
    First, read the sentence you quoted again a bit more carefully. The key phrase is “also able to evolve into one who did.” It’s not simply that living organisms are equivalent to organisms with a strategy, it’s that surviving organisms are those that had developed a strategy. Any prototypical organism that arose from your proverbial puddle without a strategy would not have survived. Replication is a strategic process. Cell division is a strategic process (first cooperative one as well perhaps). One more likely to have involved accident than intention, but life itself may represent nature having taken advantage of accident. Arguably the basic strategy behind natural selection.

    Was that group selection or individual selection, you ask? The question assumes the selection process had to make that distinction. But replication strategy itself involves the separation of individual groupings. So in my view trying to make such distinctions as essential to an understanding of the selection process overall is like hitching the horse to the wagon bassackwards.

    At bottom however, you can’t separate life from strategy. Essentially it’s nature’s plan of action. It’s the plan that evolves, structures and tactics accordingly. In concert.

    You say it’s populations that evolve and not organisms? But which is the chicken and which is the egg? The answer is neither.

  18. #18 piker
    November 9, 2009

    daedalu2u, I should add that when you say “it is populations that evolve and not organisms,” you seem to be echoing what Dr. Wilson had just alleged, that “If a population follows this transmission rule, selection will be concentrated entirely at the between-group level because there is no phenotypic variation within groups.”

    Except that there clearly are phenotypic variations within groups, physically, culturally, strategically, genetically. So perhaps I don’t really get what you two are talking about, but I do get what the excerpt taken from this site is talking about, and it’s not what I get from you.

    From an apparently reputable site on How Humans Evolved at:
    http://www.wwnorton.com/college/anthro/evolve4/ch/15/welcome.shtml

    “Understanding Human Variation
    In the past, the study of human phenotypic variation has often been the victim of gross misinterpretation. The distinction made about differentiating within group variation from between group variation addresses this issue. When the subject of interest is a trait in a population, the fact that variation of that trait exists within a group does not say anything about the variation of that trait between groups, and vice versa.
    Now, however, the study of human genotypic variation, and how such variation influences phenotypic variation, has significantly expanded our understanding of human variation as a whole. Some genetic diseases are maintained within groups because possessing one copy of a deleterious allele is actually advantageous. Sickle-cell anemia, for example, is a debilitating condition when a person carries both copies of the gene that misshapes their red blood cells, but carrying a single copy of the gene—the heterozygous condition—confers resistance to the most severe type of malaria, falciparum malaria. This example of a balanced polymorphism is joined by others (such as the possible resistance to tuberculosis granted by having one copy of the allele that causes the always fatal Tay-Sachs disease) in different regions with different selective pressures on human populations.
    Other genes, such as the variants that cause lactose tolerance—typically primates lose their ability to digest lactose, a sugar found in mammalian milk soon after they are weaned—may be quite new in human populations. Joel Hirschhorn and his coworkers at the Harvard Medical School have gone deeper into the gene that controls the ability to digest milk. They found 101 nucleotide sites where the bases that make up the gene (single nucleotide polymorphisms) have the possibility to vary within different groups. Finding one of these variants actually allows researchers to predict the pattern of variation at the rest of the sites. This work corroborates earlier estimates hypothesizing that the ability to digest lactose among Middle Eastern and North African-descent peoples evolved as recently as 7,000 years ago, when these groups first domesticated livestock.”

  19. #19 daedalus2u
    November 9, 2009

    piker, an individual organism, lets say a mouse, is conceived, develops in utero, is born, matures, reproduces (or not) and dies. The genome of that organism does not change over its lifespan. An organism does not “evolve” in the sense that is being talked about here (mostly, discounting culture evolution stuff) during its lifespan. There can be some epigenetic programming of gametes, but those are (usually) not heritable because that programming (usually) gets erased at some point (so that new programming can be rewritten).

    A group “evolves” to the extent that there are heritable changes in its collective genomes, or in its “culture-ome”, to the extent that there are heritable changes that are passed down to descendents of the group. A group “genetically evolves” to the extent that there are changes in the frequency of genes in the collective genome. A group “culturally evolves” to the extent that there are changes in the frequency of socially determined behaviors in the collective culture. (I don’t like to use the term “evolution” when applied to cultural behaviors, I appreciate that it is difficult to distinguish behaviors that are cultural from behaviors that are genetic, i.e. nature vs. nurture, and there can be epigenetic changes that occur to an individual that change that individual’s behaviors, i.e. I think the cycle of violence is mediated in part by epigenetic reprogramming. I think the “capacity” for a whole range of behaviors are already present in the genome, the environment simply causes the phenotype to develop such that those behaviors are present. I think discounting this plasticity of the genome is a meme that those in power adopt to maintain their position relative to those not in power).

    A group of dogs that is selectively bred to have certain characteristics is “evolving” to acquire the characteristics that are being selected for. No single dog is “evolving”, it is only the breeding population of dogs that can be said to be “evolving”. The group is “evolving” by changes in the relative frequency of genes that confer the phenotypes being selected for and against.

    Supplying a selective pressure that leads to specific epigenetic programming of gametes might accelerate the fixing of mutations in those genes that facilitate the epigenetic programming and might make the epigenetic programming superfluous in future generations. I would not term that as the selection “causing” the genetic changes, rather as selecting for them once they happened.

  20. #20 piker
    November 9, 2009

    Daedalus2u,
    Assuming the particular process involves mutations, the mutations are in the individuals, the phenotypes. The selection process is not independent of the individual. It’s a cooperative process. Without other group members, kin or not kin, the individual will not effect a long term change in its descendants. Without the individuals’ help. there’s no group dynamic than can be termed selective. No progress of the changes from the phenotype to genotype.

    Breeding dogs of course is directed selection involving deliberate choice of selected individuals! In other words it’s not simply a breeding population of dogs, it’s one of selected dogs. Huge difference there. Nevertheless both processes select for individuals to affect the groups. Group contributions may differ, but cannot bypass those of the individual.
    By the way, the dogs are selected for their particular strategic behaviors. To say it’s only the breeding population that’s evolving misses the whole point of the exercise, which is to produce evolved traits in individual dogs.
    Natural selection serves the same purpose but takes a hell of a lot longer, at least with evolving its wolf populations. Best it seems to have done was with coyote production.

    It’s kind of laughable to hear you recite the canon that no single dog is evolving. Does that also mean there’s a single dog in the group that is not evolving?

  21. #21 daedalus2u
    November 9, 2009

    piker, it is not mutations in phenotypes that are transmitted, it is mutations in germ cells. Those germ cells then form gametes with the mutations and those mutations express themselves (or not) in the individual organism that derives from those gametes. The phenotype develops from the genotype, not the other way around.

    It is only the breeding population that comprises that specific breed of dog that develops the specific traits that that breed of dog was developed for. “All dogs” don’t develop the specific traits of that breed, only the breeding population of dogs that has their gene pool modified through selection such that they express those specific traits.

    A single organism evolving is like one hand clapping. It is a concept that doesn’t make any sense because it takes a population of organisms to form a gene pool. If the genes don’t change, there is no evolution. In an individual organism, the genes don’t change over the individual’s lifespan. The genome that forms the germ cells of an individual organism is (in most cases) identical to the genome that forms the phenotype of that organism. The “selection” that can occur in gametes is extremely limited (i.e. is sperm able to make enough ATP to swim) and is very different than the “selection” that can occur in that organism’s phenotype.

    So no, a single dog can’t “evolve” but that dog can form part of a breeding population that is evolving.

  22. #22 piker
    November 9, 2009

    daedalus2u, I knew you would answer that the phenotype develops from the genotype and not the other way around, BUT the mutation or mutations were NOT initially IN the genotype, now were they? So, where do your random mutations actually start, in some platonic genome in the sky? This is where your traditional theories of selection get vague and take on the aura of the magical. The mechanism is described most often as a sieve that nevertheless makes selective choices, exact location of that sieve being kept a secret from us, because as long as nature knows its location, we don’t need to. Because if we can put in a model, why then its physicality becomes of no importance.
    Can you tell me where the physical starting point of any mutation is, if not in an individual? Of course we say that individuals reflect their genome, but they are not in and of themselves the genome from which come the distribution of other individuals.
    Same goes for gene pool. There is no physical pool out there where the random mutating forces can find and mutate the germ cells. It starts with the individual and takes a very long time before these individual changes can find common cause enough to pool themselves in the general population.
    if the breeding population of dogs is evolving, where is the process physically taking place except in the dogs? Is it not a pair of dogs that physically link together that actuate the process? If the evolution starts with those dogs it occurs as a process within a succession of those dogs. Who is to be the first in the succession that has evolved, if there are those before it that have not? Evolution includes the process, not simply the fait accompli.

    You also write: “The “selection” that can occur in gametes is extremely limited (i.e. is sperm able to make enough ATP to swim) and is very different than the “selection” that can occur in that organism’s phenotype.” Those sperm are swimming where? In a metaphorical pool common to all, or in one common only to it’s physically closest kin?
    And I’m puzzled as to what you mean by “selection that can occur in that organism’s phenotype.” Did you mean to write genotype instead? And if so, or even so, do you see no necessary connection between the processes?

    Or don’t you think about what you read and study in these matters with a critical mindset? If you do, don’t you see some gaps or inconsistencies in the official doctrine or preceptives?
    Do you then simply presume that if they could have been filled by an explanatory process, excluding the supernatural, then they would have been already?

    There are gaps here in the selection theories as an aggregate whole. They won’t be filled by labeling them as “natural.” They are potholes in the theoretical road. Temporarily filled with material formed from a loosely compacted mass of fragments or particles – aggregate ersatz.

  23. #23 Bob O'H
    November 10, 2009

    DSW @7 –

    When he says “it might be possible that when a lion chooses, it chooses the behaviour that is individually optimal” the metric seems to be absolute fitness

    I don’t know where you get that idea from. I wasn’t thinking about relative/absolute fitnesses, but I don’t see that it matters: the argument still works with relative fitness.

    The investigators looked hard for a compensating benefit for the bravehearts within prides and came up empty.

    Can you give the citation for this? It’s not in the Science paper.

    The main point of this installment, as InfuriatedScienceTeacher notes, is that the investigators failed to see group selection as a possibility, even when it was staring them in the face.

    Again, you assert this, but don’t give any evidence to back it up. Yes, it might be plausible, but that doesn’t mean there is evidence for it. The best you seem to have is this:

    If anything should come through in the T&R series, it is that natural selection is based on relative fitness and within-group selection must be evaluated on the basis of relative fitness within groups. Everything known about the example points to a fitness difference within prides favoring the cowards over the bravehearts. The investigators looked hard for a compensating benefit for the bravehearts within prides and came up empty. They also stress the paramount importance of competition among prides.

    But absence of evidence is not evidence of absence. This is only little better than a common argument used by creationists: “we demonstrate that this evolutionary model can’t explain the results, therefore it must have been a designer”. You don’t even present evidence on fitness effects: the paper only looks at behaviour, and makes the point that lead lions will be the ones that are wounded, but that’s only a small component of fitness.

    Omar @ 13 –

    Bob O’H, considering I have studied punishment for a bit, I caution using punishment as an alternative to group selection.

    Just to be clear, I wasn’t suggesting it as an alternative. I was suggesting that, if DSW’s model is correct, we would expect that the lionesses that do defend their territories might punish those that don’t. This is one thing that puzzles me about this: the lions who do defend apparently don’t gain anything from those that don’t. So why are they happy for these loafers to be around? Why hasn’t punishment evolved? It’s something that the authors of the paper look at, and don’t find evidence for (again, this is only evidence during the experiments: there may be more going on at other times, of course).

  24. #24 daedalus2u
    November 10, 2009

    The starting point (and ending point) of mutations is in germ cells. It is only germ cells that become the next generation. The phenotype and the germ cells both derive from the same genome that is first put together in the fertilized egg that derives from the two gametes which derive from the germ cells of the two parents.

    The genotype comes first, in the fertilized egg, then the phenotype comes later. The fidelity of copying the DNA is not 100%, so there are errors introduced as the cells divide. Only errors that end up in germ cells give rise to heritable mutations. Mutations in somatic cells may lead to cancers, to cells dying, or to no change in phenotype of that cell (because the mutation is in DNA that is inactive in that cell).

    Sperm are haploid, they have half the genome. They have to swim some distance so as to reach the egg and fertilize it. Sperm that are unable to make ATP because their haploid genome contains a defect in a part of the mitochondrial respiration chain can’t make it to the egg, so that particular defect can’t be passed down.

    How is a single dog “evolving”? Its genome stays the same over its lifespan. What did it “evolve” from and what did it “evolve” into?

    Each cell has its own genome, mostly the same, but sometimes slightly different (due to mutations, copy errors, epigenetic programming). Saying an organism has a genome is a short-cut. Mostly right, but not always.

  25. #25 piker
    November 10, 2009

    daedalus2u, that’s all very nice, but your errors in the germ cells that give rise to heritable mutations are in germ cells in individual entities, not yet in the population of those individuals. The genome has not been changed for the rest of the population. You kind of left out the part about how that happens.

    How is the single dog evolving? As you said at the outset, the starting point is in its germ cells. They are a part of the individual dog, whether you designate it as a phenotype carrying a copy of its genome or use some other labeling system. And you’re confusing evolving with evolved, as if the process has to be finished before it can be said that any of the dogs have evolved. So that in retrospect you theorize that something was evolving, but it wasn’t the dog.
    Let’s see now – there were dogs in the past that were different from dogs in the present, the past having morphed into the present, but the “evolving” didn’t happen with or to any of the dogs in particular. Because if it did, we ought to have been able to see the morphing function in action, and yet we couldn’t. So we’ll leave out that part when constructing our explanatory theses. Because, look, ma, we can make it all work on the computer!

  26. #26 Jamie
    November 10, 2009

    It might sound scientifically rigorous to list a bunch of additional factors that weren’t tested, but science would grind to a halt if such rigor were applied to all hypotheses being tested.

    If all you wish to claim is that the data from this observation do not falsify your theory, I have no problem with that. But you seem to be claiming more. It does raise the question of what sort of observation would falsify the theory. If the speed of the process means more to you than its rigor then it seems to me you will be vulnerable to error. Science, I would suggest, will not be delayed by being cautious. True, grand theorizing may be restrained. It seems to me to be a matter of whether one wants to know what is actually going on, or whether one wants a theory that is useful for some purpose regardless of whether it is correct. Is it your contention that your theory has as much utility as competing theories, but that none of the theories has any claim on reality because the matter it too complex and there are too many confounding factors that cannot be controlled? For this is what you seem to be telling me. Yes, science is hard and we are impatient to move forward. But let’s not substitute the illusion of progress for the real thing or sacrifice knowledge for utility.

  27. #27 daedalus2u
    November 10, 2009

    piker, it happens the same way and the only way that all individuals are added to any population, by being conceived from gametes, developing in utero, being born and growing into an adult and reproducing. How do you think or propose that the genome gets distributed among the population? only through phenotypes reproducing and producing more genotypes with the selected for gene/trait/features.

    I am not confusing evolving with evolved. You are saying that a single dog can evolve and is evolving.

    ”It’s kind of laughable to hear you recite the canon that no single dog is evolving. Does that also mean there’s a single dog in the group that is not evolving?”

    It doesn’t come from “canon”, it comes from the definiton of “evolving”.

    ”In biology, evolution is change in the genetic material of a population of organisms from one generation to the next.”

    http://en.wikipedia.org/wiki/Evolving

    Do you have a different definition of evolving? If so, I would like to hear it.

  28. #28 Guy
    November 10, 2009

    I’ve read many reasonable and thoughtful comments here about how science should work. I particularly appreciated the post by Jamie. I certainly wouldn’t argue against these statements of scientific principles. Of course, they are rarely, if ever, perfectly applied and this can lead to bias in the degree to which they are implemented. One of DSW’s main points in this blog has been that there has and continues to be such a bias against the idea of group selection. My personal opinion is that some degree of bias against new or unconventional ideas is prudent in science, because I think validity is usually more concentrated in conventional thinking than it is in “fringe” alternatives. I accept that it should be a hard fight to overturn convention, even though I have personally been frustrated by the higher bar this presents. BUT (this is a big but), the extent of bias can be amplified beyond reason. In my opinion, this has happened in the case of group selection. It is easy to criticize any idea in science by holding its feet to the fire of scientific principle. It is harder to appreciate the validity or utility of ideas outside of the mainstream. Sometimes ideas like group selection become pariahs and are aggressively attacked in a knee-jerk fashion. That can be scientifically counter-productive, so preventing or fixing this sort of thing is another kind of scientific principle. I hope we can try hard to maintain a balance in scientific rigor when comparing ideas like group selection with the mainstream.

  29. #29 piker
    November 10, 2009

    daedalus2u, doesn’t the change in the genetic material start with what you already described as errors in the germ cells of an individual? And do you realize what you’re implying when you say” that all individuals are added to any population, by being conceived from gametes, developing in utero, being born and growing into an adult and reproducing.”

    If this is your answer to my question about how mutations are spread so that the general population evolves and the genome is changed accordingly, you’re implying that evolution is starting with the actual evolution of the individual as a requirement before it then passes on the changes by that species replicatory process. Gee, isn’t that pretty much what I said, except that’s even more radical than my version of evolution as beginning at the individual level.
    But even so, you don’t say with any exactness how one such mutation in one individual turns into a change that can be passed or morphed into the genome from which a population with the new phenotypes can form. If an individual mutation has to be replicated by spreading from kin to kin, how long can this be expected to take? Millenniums or the obligatory billeniums?

    Because in a great number of species, such changes are close to being generational and that doesn’t seem to jibe with the replication requirement as proposed. And hey, that generational time line seems to be what your Wikipedia definition of evolution requires.
    In short, I think you have again left out the description of a workable evolutionary selective mechanism, at least for the mutational version of the theory. Which, if generational, would require one humongous number of coincidentally similar mutations to accidentally or otherwise occur at around the same time in a critical mass of individuals within the genomic population. Which of course nobody really expects is happening, but then if not, what else could it be? Some say it’s a process directed by an environmental change that a critical mass of he population picks up on. But no, that can’t be it, because that would mean life participates in its own evolution.

    That gets us into science fiction something like the following which I quote from memory so I’m sure I’ve taken some liberties with the text:
    “Directed mutations based on preselected or pre-adaptive algorithmic response options could be the mechanism that allows real time adaptations, genome wide, based on a quorum sensing mechanism that recognizes a critical mass forming in the population as to the felt need for relief of the pressures. Such adaptations don’t have to be a perfect fit to be an improvement of overall fitness. Culturally based transmission components might account for the type of signaling that would enable the quorum sensory apparatus.”

    There was more but that’s all I remember..

  30. #30 daedalus2u
    November 10, 2009

    piker, I think you need to be more careful in keeping science fiction out of your understanding of how populations evolve.

    The mutation passes by direct descent and by direct descent only. That is the entire basis for phylogenetic analysis. If there was any other common type of “descent” or transmission of genes, then phylogenetic analysis wouldn’t work.

  31. #31 piker
    November 10, 2009

    daedalus2u, You miss the point that your failure to come up with any mechanism that accounts for generational change in the genotype, leaving us with the magic of nature that seemingly mutates large numbers of individuals in a simultaneous burst of exactitude, leaves the door open to any fanciful conception of the wizardry that must have been involved.

    But you could try not ducking the question and make the wizard step out from behind the curtain. Metaphorically speaking of course.

  32. #32 bob koepp
    November 10, 2009

    I’ve been puzzling for a couple days now about what piker objects to in the “canonical” account (‘splanation) of evolution as a population level phenomenon. From a logical point of view, there’s nothing objectionable about the idea that non-evolving individuals comprise evolving groups. Can the objection be stated in a way that makes the “problem” clear?

  33. #33 piker
    November 10, 2009

    Well, yes, you can join a population and cause it to evolve accordingly. The nature of your presence could change the group dynamic and over time force the members that stuck around for a few hundred generations to come up with revisions in their genotype. I don’t expect that’s in your catechism however.

    Of course if it’s just about evolving the group culture, why that shouldn’t take long at all. But catch 22 would be that some individuals would inevitably adapt to that culture before it reaches its optimal level.

  34. #34 bob koepp
    November 11, 2009

    Sorry, piker, but I’m not gettin’ it. Could you please indulge me by presenting your case in the form of an argument where you explicitly state your premises, and then employ valid rules of inference to arrive at the conclusion that there’s something “wrong” with the canonical account of evolution as a population level phenomenon? I’ve tried to “reconstruct” such an argument based on your comments, but have not met with success.

    Thank you in advance for your consideration.

  35. #35 Guy
    November 11, 2009

    piker:

    I am struggling with the same issue described by Bob Koepp and look forward to reading your argument. I also want to respond to you post (#33) where you claim that an individual joining a group can affect that group now and into the future. I agree that this can happen, but it is only one side of the story. I can’t tell if you merely highlighted the bottom-up path of causation in this context, or if you would deny a top-down effect. I personally find it useful to think about multilevel selection as an instance of complex system dynamics leading to and constrained by structural hierarchy. A balance between bottom-up and top-down causation is expected from this viewpoint. So I would argue that when an individual joins a new group it also changes the individual. This effect might, for example, take the form of entrainment by group norms. I think a role for top-down causation is central to the claim that groups exist as functional entities that might be subject to natural selection. It would help me to understand your objections if you denied the possibility of top-down causation in this context.

  36. #36 piker
    November 11, 2009

    Sorry back there bob, but to use the deductive method here you would have to supply me with a different foregone conclusion, because I’d otherwise have to prove the impossibility of your particular one having been reached. And yet clearly you have reached it. And done so by induction. Except it was from your well that you found and drew the inference, not mine.

    May I suggest that the true or false question should be whether the canonical account is correct in its understanding of the function that activates the phenomenon. The inference I had drawn was that it wasn’t. I can’t then prove myself false at that level.
    And since my argument from that inference has been that the canon has been lacking in specificity here, the accuracy of what it lacks is not yet determinable by deduction.

    The problem is that I haven’t been able to find anyone in service of the canon that can explain why specificity is lacking, or even agree with my perception that functional specificity has any importance here.

  37. #37 piker
    November 11, 2009

    As to guy’s intervening commentary about there needing to be a balance between bottom-up and top-down causation, I agree. Note my reference to such causation when I wrote that “catch 22 would be that some individuals would inevitably adapt to that culture before it reaches its optimal level.”

    And as to what I’m objecting to, it’s not to causation per se, it’s to the nature of some of the argumentation here as to how cause leads directly to effect without the necessity of our understanding of the mechanism. All that seems to be necessary is a simulation of the circumstances, and from that a determination that that the causative connections exist.

    Guy wrote: “I think a role for top-down causation is central to the claim that groups exist as functional entities that might be subject to natural selection.” I think so as well.
    But it’s the nature of that selection process that determines causation. Causation is by definition the action of causing something. There is no action that is without a specific mechanism. The role of science is to understand the specifics involved, not simply to determine that a mechanism has to exist and identify it by suppositions as to its purpose. The causative nature of group selection can’t be understood without specificity as to its mechanistic function. To demonstrate that that the function is purposive is only that – a demonstration. Specificity is clearly lacking as to the physical process of the proposed phenomena.

  38. #38 daedalus2u
    November 11, 2009

    Guy (#28), I completely disagree with you about your personal opinion. We should not hold new ideas to more stringent criteria than were used to adopt the old ideas in the first place. An idea is not “more correct” simply because it is old and has been the conventional wisdom for a long time. All ideas, new, old, those not yet thought of, must stand or fall upon the totality of the data and the logic that holds those facts together. In the words of the aphorism, we need to adopt the new as soon as it is better but keep the old as long as it is good (i.e. useful); as in the discussion on the other thread “all models are wrong, some are useful”.

    One of the ideas that I have been trying to overturn is the idea of homeostasis; an idea which has no data supporting it. It was adopted ~70+ years ago before very much was known about cells and the physiology that was going on inside them. It was adopted because it intuitively felt “right”, that physiology had to keep things constant in order to sustain the organism as a living entity. There were very few things that could be measured at the time; virtually all physiological parameters were regulated by physiology to greater accuracy than they could be measured by the then state of the art instrumentation. Now, with better instrumentation we know that nothing in physiology is kept constant. Why is homeostasis still held up as a unifying principle in physiology? Is there any data to support it? No, there isn’t. Every time new complexities are understood, the idea of homeostasis has more epicycles added to it. Now there is “dynamic homeostasis”, a complete oxymoron; a dynamic process of changing internal conditions so as to keep internal conditions constant. Huh? Just like the geocentric model of the solar system had epicycles upon epicycles added to it.

    At one time, all ideas that are now generally accepted were fringe and were actively fought against. That is the nature of scientific paradigms and scientific revolutions as Thomas Kuhn discusses. Most scientists can only work within the scientific paradigm that they are familiar with. Ideas that fall outside that paradigm are rejected as wrong because they are not explainable with the scientific ideas that the scientific paradigm of the day has available.

    Bob O’H wants the lion data to (essentially) “prove” that the group selection hypothesis is correct. No data can “prove” that a hypothesis is correct; data can only prove a hypothesis is incorrect. I am sure that Bob knows that as do all scientists, but usually people look for data to “confirm” their hypothesis because it is much easier (and actually humanly possible) to search the existing and potential idea and data-space for instances of positive confirmation, rather than to do an exhaustive search and confirm that there is the complete absence of disproving ideas and data.

    Bob rightly states that “absence of evidence is not evidence of absence” (#23), which is correct, but DSW’s point is not about the correctness of the hypothesis of group selection in lions, but rather a hypothesis as to why the authors of the article on behaviors of females in lion prides didn’t bring up group selection as a possible mechanism for the evolution of the braveheart behavior. The selection mechanisms that were brought up were insufficient to explain the behavior.

    The lion behavior is well documented, so it is very likely correct data. The hypothesis of group selection is not unknown, Darwin even mentions it, it is not a difficult concept, but group selection was not mentioned as a possibility even though it does have the potential to explain the observed behavior. Absence of the mention of group selection is evidence that the researchers, authors, reviewers and editors thought it important to not include mention of group selection.

    DSW’s hypothesis (as I understand it and which he is trying to address in these many posts) is that there is a non-scientific bias against group selection which is clouding the scientific judgment of researchers, authors, reviewers and editors such that group selection can not be directly invoked as a plausible mechanism of evolution.

    I think that understanding why there has been the adoption of the “no group selection paradigm” in the absence of data demonstrating that there is no group selection is more important than correcting the evolution paradigm to include it. The correction of the evolution paradigm is going on by slowly morphing the definitions of the terms such that they mean different things than when the “no group selection paradigm” was initially adopted. It would be better to do so explicitly and better still to recognize explicitly when such things are going on in other fields (such as my pet example of homeostasis).

    Back to my first paragraph, we should not require more stringent criteria to reject the “no group selection paradigm” than was used to adopt it in the first place. If the data and logic that was used to adopt the “no group selection paradigm” was flawed, then the paradigm should be rejected even in the absence of new data.

    In reply to Jamie (#26), correcting errors in the status quo is much more important than trying to cover new ground. There is nothing “new” about group selection.

  39. #39 abb3w
    November 11, 2009

    daedalus2u: The mutation passes by direct descent and by direct descent only.

    In eukaryotes, close enough to “only” for this discussion. Lateral transfer mechanisms mostly can be neglected unless you’re dealing with bacteria and such.

    piker: You miss the point that your failure to come up with any mechanism that accounts for generational change in the genotype, leaving us with the magic of nature that seemingly mutates large numbers of individuals in a simultaneous burst of exactitude

    The mutation occurs in one germ cell, giving rise to a mutated population member; that individual has descendants in quantity which tends approximately exponential with time, with the degree of benefit/harm of the mutation controls the sign and magnitude of the exponential constant. (The discrete approximation model is more complex than the continuous approximation model, which is how for small enough magnitudes, drift can dominate.) The burst is only “simultaneous” in the sort of geologic terms that make each of us a contemporary of George Washington (or even Paul of Tarsus).

    Over time, it’s somewhat of a “Ship of Theseus” problem again. The “population” refers to the organisms that are members at any particular time t; the properties of any individual member do not change over time. However, the population also has departures (deaths) and arrivals (births) — with arrivals having the potential to have properties different from those previously part of the group. So while the individual members do not change over time, the differences between the new arrivals and the existing population enables the properties of the group to change, making the property of group a non-constant function f(T), even while the type of organism remains “the same”.

    “This, milord, is my family’s axe. We have owned it for almost nine hundred years, see. Of course, sometimes it needed a new blade. And sometimes it has required a new handle, new designs on the metalwork, a little refreshing of the ornamentation . . . but is this not the nine hundred-year-old axe of my family? And because it has changed gently over time, it is still a pretty good axe, y’know. Pretty good.”

          –from Terry Pratchett’s The Fifth Elephant

    More subtly, of course, “the same kind” over time becomes less absolutely “the same” as the starting kind; and the various speciation modes limiting the degree mutations diffuse between sub-populations means what was one kind may split into two sub-kinds (with essentially all life on earth being (sub-)nkinds of the original kind: “cellular”).

  40. #40 bob koepp
    November 11, 2009

    piker -
    I asked you as nicely as I am able to provide an argument for a claim you have repeatedly made. True, I did phrase my request in terms of “valid rules of inference,” which you rightly associate with deductive argument forms. But I’d be happy to consider an inductive argument, if you can present one. Can you?

  41. #41 piker
    November 11, 2009

    bob, I also mentioned that I can’t adapt as my premise the inference you incorrectly drew as to my previous conclusions. And induction still requires the proposition that its premises are factual.

    As to presenting one based on premises of my own choosing, I thought I just did. On the other hand it’s not your particular satisfaction that I perceive as being a gauge of my performance.

    Can I present an inductive argument with the goal of making you happy? Not in any way that would make me happy in the bargain.

  42. #42 bob koepp
    November 11, 2009

    piker – I haven’t asked you to provide an argument that would make me happy. I have asked you to provide an argument that might help me and several others here to understand certain statements you have made “on the record.” If you can’t provide such an argument, then on what basis have you made those statements? Outside academia, we would say that it’s time to put up or shut up.

  43. #43 piker
    November 11, 2009

    abb3w, I like a lot of what you are saying, except that you haven’t accounted for the differences in the speed of the process over time, and the apparent fact that the process speeds up exponentially with some meaningful correlation to its prior success. Undirected or unassisted or otherwise spontaneous mutations as proposed by daedalus2u just don’t cut it as far as I can determine. Neither does the idea that individuals don’t change in any way that contributes to the evolution of the species. If the replicative mechanism is in the individual, any changes that can be expected to eventually alter the genotype will occur prior to or during that replication rather than after the process is completed. Except of course with changes that occur in the replicated product as to it’s developmental progress. Yet if those resultant changes are passed on without further change from experience or mutation, evolution in that part of the population isn’t happening. Somewhere at some time, some individual is changing in some significant and directly causative respect for it to happen.

  44. #44 piker
    November 11, 2009

    Again Bob, I don’t care if it’s your opinion that I haven’t made an understandable argument. At least you recognize that I have made an argument. Or series of arguments if you prefer. Many are able to respond to them quite well, even as we speak. They use the process of counter argument. You don’t seem to know how that’s done. Maybe it’s the dysrationalia syndrome. Some argue that it’s infectious, especially in the cyberspace environment.

  45. #45 bob koepp
    November 11, 2009

    No, piker, I don’t recognize that you’ve made an argument. What are your premises? What inferences do base on those premises? And how do those inferences lead, whether deductively or inductively, to a statement about the inadequacy of the canonical view that it’s populations rather than individuals that evolve? Until you get explicit about such things, you haven’t produced any arguments, just made a lot of provocative noise. Dysrationalia syndrome, indeed.

  46. #46 daedalus2u
    November 11, 2009

    More a case of not even wrong.

  47. #47 piker
    November 11, 2009

    Gee bob koepp, I wonder what all these other people, including Dr. Wilson, were talking to me about, if they haven’t recognized that in some sense there was an argument they could respond to? You may not understand this, not being all that interested in the process of independent thought, but most here comment with the purpose of learning what they can from the feedback.
    Answering an intelligent question or responding to an intelligent counter argument makes you think and discover new angles to a problem, and with that new solutions, that you may not have come up with otherwise. Feedback is thus defined as information used as a basis for improvement.
    But one learns early on that an argument from ignorance is not in the feedback category.

  48. #48 piker
    November 11, 2009

    OK daedalus2u, just for that, no more soup for you.

  49. #49 mollyrogers
    November 11, 2009

    piker, I think people are writing to you because you clearly have very strong opinions. But it’s hard to tell exactly what you are arguing, let alone what you are basing your arguments on. That makes it difficult for anyone else to take part, even if they would enjoy a good debate with you.

  50. #50 piker
    November 11, 2009

    The problem with the blog commentary process is of course that you end up debating in a forum where the participants are seldom the same – and if they are, may have only vague memories of what any of the others has said in the past. And it’s what we have said in the past that for the most part will show where we are basically coming from. The catch is that if we have to repeat a lot of that as a basis for each later post, no-one will read the stuff, and if we don’t, fewer of those that we get to read it will understand it.
    The questions that I’m raising are not new to evolutionists. Thus the basis for raising them will also not be new. The people that sense intuitively what that basis is are the people that I seem to end up talking to. But I do try to put a new twist on things, because that’s how we learn and adapt our theories to new discoveries, etc.
    Further, if you place yourself in a category as a prelude to presenting your positions or questions, the category itself will seem to represent a bias, which then will be addressed instead of the particular argument you’d like feedback on.
    Such as when I refer to ideas I may have in common with people like Jablonka or Margulis, the response will be to their ideas instead of mine. Such as DSW’s response that such ideas were “newfangled,” so that’s all that needed to be said about that. Or if I echo some of the theories first put forth by Baldwin, I become a Lamarckian. And so on. I see little point in preaching to the choir, yet sometimes it’s only the choir that can be persuaded to listen. And of course this is not my blog, so I have no obligation to educate the uneducated accordingly. It might at times be my pleasure, but to insist that it’s somehow my duty is not persuasive.

  51. #51 Bob O'H
    November 12, 2009

    Bob O’H wants the lion data to (essentially) “prove” that the group selection hypothesis is correct. No data can “prove” that a hypothesis is correct; data can only prove a hypothesis is incorrect. I am sure that Bob knows that as do all scientists, but usually people look for data to “confirm” their hypothesis because it is much easier (and actually humanly possible) to search the existing and potential idea and data-space for instances of positive confirmation, rather than to do an exhaustive search and confirm that there is the complete absence of disproving ideas and data.

    That first sentence is simply false, and I rather objection to people putting words into my mouth. I’m simply asking for the positive evidence that this is group selection: the whole point of the post seems to be that this is evidence from the wild, but DSW doesn’t provide any and has been avoided the question.

  52. #52 daedalus2u
    November 12, 2009

    Bob, I don’t mean to put words into your mouth and apologize if that is how what I said came across. As I understand it, asking for positive evidence is asking for “proof” of a sort. It is different than asking for evidence that disproves alternative hypotheses.

    Maybe I am misunderstanding what you are asking for when you ask for “positive evidence”. I understand that to mean evidence that specifically supports group selection; that is evidence that is consistent with group selection rather than evidence that disproves alternatives to group selection, i.e. evidence that is inconsistent with individual selection and other non-group selection mechanisms.

  53. #53 Guy
    November 12, 2009

    piker:

    You wrote: “The role of science is to understand the specifics involved, not simply to determine that a mechanism has to exist and identify it by suppositions as to its purpose.”

    This is the strongly reductionistic view that has traditionally been the focus of most scientific thinking. It is the idea that the way you learn how a car engine works is to take it apart, learn the mechanisms, and put it back together again. This approach has served science well for a couple hundred years, but the approach that looks past mechanism has quietly served as a cornerstone of scientific thinking all along. For example, the laws of thermodynamics are all about outcomes, rather than mechanism. In fact, the theory of natural selection has been mechanism-free in this sense from the beginning. Selection favors mutations that improve fitness, regardless of the mechanism. If faster running speed is an effective way to improve fitness, it doesn’t matter to selection how that is achieved (ignoring correlated fitness effects). It can be worthwhile to explore mechanisms as examples of how these outcomes are achieved in particular instances, but the mechanisms do not define the model or theory. I personally think there are critical places in science for both mechanism-free theory/models and the exploration of mechanism. However, I think an exclusive focus on mechanism would greatly reduce our ability to be scientifically inductive.

  54. #54 Guy
    November 12, 2009

    daedalus2u:

    Wow. That was quiet a response given that we are rather close to agreement. The main message of my post to which you responded was that the bar has been raised to an inappropriate height regarding the idea of group selection. I don’t think the ideal of a perfectly level playing field you argue for is realistic. I do think it is an admirable goal for scientists. I still think that having a somewhat higher bar for ideas outside the mainstream works for science (not for scientists) because mainstream ideas have generally passed the test of time. The scientific rigor of this testing varies tremendously, but I think it is fair for the scientific community to regard older and persistent ideas with some respect.

  55. #55 Guy
    November 12, 2009

    daedalus2u: I should have added that I have spent most of my scientific career struggling to bring new ideas into the mainstream of scientific discussion. I know the frustration of this struggle more than most scientists, I think. Still, all I expect from my colleagues is an honest effort on their part to understand my arguments. It is uncritical dismissal that gets my goat.

  56. #56 piker
    November 12, 2009

    Guy,
    Note that I said “not simply to determine that mechanism has to exist.” There’s no exclusive focus recommended by that notion – what is missing from much of the theorizing here is any focus at all on the selective mechanism. And when such focus is attempted it seems to come up with a very blurry image. That blurring can do more damage to induction than no picture at all. If you take a metaphor such as gene pool and over time come to regard it as a physically intact pool somewhere, you’re in danger of directing your thirst for knowledge to an oasis mirage out there in the metaphorical desert.

    And reductionism shouldn’t be seen as a bad word accordingly.

  57. #57 daedalus2u
    November 12, 2009

    Guy, my hope is that you are educable, and that I can turn you away from the dark side ;). The only reason I can think of to not have a perfectly even playing field is laziness, physical and intellectual; an unwillingness to actually examine the data critically and honestly and then see which ideas fit the data best. That and because individual scientists may benefit by being intellectually dishonest and so get more funding and undeserved kudos (even if that is unconscious behavior, (but I think that is relatively minor, but it is individual selection over group selection so it is relevant to this blog and appropriate for the Science as a contact sport thread)).

    Uncritical dismissal gets my goat too.

    There are many examples that I can name where “conventional wisdom” is wrong and is impeding progress in the respective fields. Another important one is the extreme focus on the genetics of diseases, in particular on diseases with “complex genetics” (essentially all of the degenerative diseases of modern living, autism, Alzheimer’s, obesity, heart disease, diabetes, kidney failure, the metabolic syndrome, immune system deviation, and so on).

    No one has suggested how a single common polygenetic disease can evolve, let alone dozens of them with extremely similar incidence in many different populations. I can understand how a disease that is an emergent property of many genes could happen, but not how such a disease could be common. If it is an emergent property of many genes, then it should be an idiosyncratic property of many genes, unless those many genes code for a single trait, but an adverse trait can’t be selected for and so can’t become common (and redundantly coded by many genes) unless it is coupled to something positive.

    My hypothesis is that they are not “diseases” at all, but “features” (as in the Bill Gates quote, “it’s not a bug, it’s a feature”). Features from deep evolutionary time, where the adverse effects of too much of the trait are balanced with the adverse effects of too little of the trait (via the many interacting genes). From deep evolutionary time what has been minimized is the sum of non-fitness due to too much of the trait and from too little of the trait.

    My hypothesis is that the “control point” of those control systems that regulate the trait has been shifted to a control regime where the adverse physiological responses are not balanced by adversity from the environment. I think that the basal NO level is the control parameter that has shifted. Virtually all of these conditions are made worse by “stress”. Stress causes low NO, low NO activates many stress responses, many stress responses are from deep evolutionary time and are extremely conserved. Inappropriate activation of a stress response will cause adverse effects (the normal stress response balances what ever environmental adversity caused the stress in the first place).

    I tried to submit something on this as it pertains to autism genetics and the manuscript wasn’t even sent out for review, the editor wrote back saying “no one would be interested”.

    A (relatively) easy to appreciate example is Alzheimer’s which I think is due to ischemic preconditioning gone on for too long. Ischemic preconditioning (temporary reduction in ATP demand following brief ischemia) is a pretty obvious survival feature. ATP isn’t stored; it must be made as it is needed. If there is insufficient ATP supply, the only options are making more or using less, or both. Can this state of using less ATP be a permanent state? No, if it could be, then cells would be in that state all the time so that more ATP could be devoted to reproduction and there would be no such thing as ischemic preconditioning. Obviously not all physiological processes have equal time urgency, some low priority tasks can be put off during an ATP crisis until times of higher ATP. Time prioritizing ATP demand allows for reduced “overhead” in ATP production capacity, allowing greater resource allocation to reproduction.

    Ischemic preconditioning is a stress response, it is triggered by oxidative stress (i.e. low NO or low ATP) and can be observed in essentially all tissue compartments in all organisms. What would happen if ischemic preconditioning went on for too long in the brain? We would expect to see the longest time constant processes shut down first. Long distance transport down axons takes a long time, so can’t help in a short term crisis and axonal transport consumes ~half of neuron metabolic capacity. Mitochondria biogenesis takes a long time, and first involves dismantling functional mitochondria; can’t do that during an ATP crisis. Disposal of damaged proteins takes time, consumes ATP (via the proteasome or via autophagy) and all you get is some substrate; can’t do that during an ATP crisis.

    What is observed in Alzheimer’s? Reduced brain metabolism (reduced ATP production and consumption), reduced blood flow (vasodilation is regulated by NO), white matter hyperintensities (reduced water movement in axons via MRI), accumulation of damaged proteins (amyloid, Lewy bodies, lipofuscin), reduced ATP levels, reduced mitochondria turnover, increased oxidative stress.

    The “conventional wisdom” that everyone is chasing is amyloid even though removal of amyloid via vaccination didn’t improve symptoms (it caused neuroinflammation which made it worse).

  58. #58 Guy
    November 13, 2009

    piker:

    I think we greatly overlap in our views. The importance of mechanism in theory may be an exception. I tend to view that more as an empirical question.

  59. #59 piker
    November 13, 2009

    Guy, I can agree with that view. But a big problem I have with this series is that the mechanism that it proposes as an operational necessity begs the empirical question as to its material existence.

  60. #60 piker
    November 13, 2009

    I propose instead that selection in evolution ultimately involves biological computations that are necessarily made by a biological computer.

  61. #61 bob koepp
    November 13, 2009

    piker -
    What is the mechanism proposed as an operational necessity in DSW’s T&R series? And how is the question as to its material existence being begged?

    Are you talking about natural selection? If so, I hope that you aren’t simply ignoring the ongoing discussion among philosophers of biology about what it means for a process to be “a mechanism”, and whether natural selection could itself be a mechanism. These are not simple questions by any means. (And I don’t think they raise any special problems for group selection that wouldn’t apply with equal force for individual selection.)

    Also, what does it mean to say that selection “ultimately involves biological computations?” Are you saying, for example, that an organism (or group) outcompeting another organism (or group) for limited resources is a computational process?

  62. #62 daedalus2u
    November 13, 2009

    piker, I think you are neglecting in your formulation (i.e. model) that virtually all traits are polygenetic, that is they are the emergent property of many genes. The selective breeding that resulted in many different breeds of dogs did not (for the most part) depend on new mutations, it depended on the segregation of genes and polymorphisms of genes that were already present in the founder population.

    A trait such as body size is determined by many genes. Some make the body larger, some make it smaller, and some do different things under different circumstances. Every gene does multiple things depending on how you look at it. If you combine all the genes that make a dog big, you will get a really big dog. If you combine all the ones that make a dog small, you will get a really small dog. If you make a mix of them you will get something in between. If you have a mix and you segregate them, you can get larger and/or smaller dogs with no new mutations.

  63. #63 piker
    November 13, 2009

    Dog breeding is selective, initial computations made by the biological and computational mechanism of the breeder – additional biological computation directing the physical acts of dogs selecting others of their kin for coupling, etc. Now if you want to dispute that there’s a relationship between kin selection and mutation, google kin selection and mutation.

    If the selective process is “natural” then the computational mechanism will be in the organisms involved. You can posit if you like that nature is the breeder, but that gets a little too close to seeing nature as having calculative purposes – teleological as it were. Yet even if that were, the computational machinery of the organism would still be a requirement. And all organisms do compute.

    I doubt if either of you who have asked will be satisfied with that answer, but based on your previous commentary there’s nothing I could say, or want to say, that would.

  64. #64 mollyrogers
    November 13, 2009

    piker: Thanks for being specific. I don’t think anyone would disagree with your statement “selection in evolution ultimately involves biological computations that are necessarily made by a biological computer.” The disagreement might be over what “involves” means.

    If I understand correctly, you position is that the biological mechanism is itself carrying out a “selection” computation. How do you envision this? What are the evaluation criteria or features for the objective function that is being optimized or satisficed?

  65. #65 piker
    November 13, 2009

    I envision it as an assessment process, trial and error methodology, planning for incremental effects as a shortest time goal. Strategies have evolved from the most basic to the extremely complex. Attainable goals accordingly – strategies selective accordingly. Long term survival through evolution has emerged from the process.

    Some argue that evolution is the main reason for the growth of complexity in the natural world. I would argue that the growth of complexity of biological strategy is the main “reason” for evolution.
    Is evolution then, put simply, a successful biological strategy? Books are being written to test that presumption. Not going to attempt that here.

  66. #66 daedalus2u
    November 13, 2009

    piker, where do you envision this biological computation takes place, in the somatic cells, in the germ cells, in the gametes, or in the fertilized egg?

  67. #67 piker
    November 14, 2009

    daedalus,
    Why don’t you tell me where selective computation doesn’t? Because I’m still waiting for you tell me why it’s not biological. Or why you believe cells are not capable of computation. Or why you haven’t considered that function selects form and has thus allowed protein molecules to wire the reactions of a cell in any particular way the function pleases?

    Don’t bother to answer. It’s clear that your intent is to bait rather than learn, and I won’t be baited by such as you in the future.

  68. #68 abb3w
    November 14, 2009

    piker I like a lot of what you are saying, except that you haven’t accounted for the differences in the speed of the process over time, and the apparent fact that the process speeds up exponentially with some meaningful correlation to its prior success

    The technical term is autocatalysis; “nothing succeeds like success”. As for speed differences, perhaps you missed “the degree of benefit/harm of the mutation controls the sign and magnitude of the exponential constant”?

    piker Somewhere at some time, some individual is changing in some significant and directly causative respect for it to happen.

    No; it is sufficient that somewhere in time, some individual is a change in some significant respect from previous individuals, and existence of this change is causative of the change developing exponentially.

    piker I envision it as an assessment process, trial and error methodology, planning for incremental effects as a shortest time goal.

    You might want to look into the mathematics of random walks; the typical K-12 education doesn’t cover them, although they really do deserve passing mention.

  69. #69 piker
    November 14, 2009

    Nice little snark there about K-12 when you don’t seem to know the meaning of increment. Which in mathematics is a small positive or negative change in a variable quantity or function. Didn’t they teach you that at Podunk JC?
    And when doing random walks on adaptive landscapes the idea seems to be that small steps are stabilizing and large ones are not. Couldn’t you get that from what you quoted?

    You ask, ‘perhaps you missed “the degree of benefit/harm of the mutation controls the sign and magnitude of the exponential constant”?’ There’s nothing there to miss that contradicts what I wrote. Although there’s nothing much there that explains how harm versus benefits are assessed – but apparently you didn’t need such things explained for multiple choice exams.

    Saving the best for last, I note that you state “it is sufficient that somewhere in time, some individual is a change in some significant respect from previous individuals, and existence of this change is causative -” Yet at the same time you deny that some individual was changing in the process. Some individual “is a change” but not even that individual was at any time changing? Some sort of magical selection process at work? But then nature sure is full of those tricks isn’t it.
    Pathetic.

  70. #70 daedalus2u
    November 14, 2009

    piker, I don’t understand what you mean by “biological computation”. I am trying to understand what you are saying and am not having much luck.

    I understand physiology, I understand control systems, I understand computation. I understand DNA. I am trying to understand how you think they all fit together to produce “biological computation” in the sense that you are meaning.

    If you don’t want me to understand what you mean, why are you posting here?

  71. #71 Alex
    November 14, 2009

    “I note that you state “some individual is a change in some significant respect from previous individuals”… … at the same time you deny that some individual was changing …”

    (Bold added for emphasis)

    It seems amazing that you think there is a contradiction between these two statements.

    If I paint a series of pictures of a teapot, each based on the previous, but with an increasingly long spout each time, we can reasonably say that “each individual is a change in some significant respect from previous individuals” and at the same time deny that any individual painting was changing.
    The situation is the same for dogs or people, individual dogs and people don’t change, but they are different to previous individuals.

  72. #72 piker
    November 14, 2009

    Daedalus2u, are cells capable of computation? Answer yes, no, or maybe.

    Alex, that example was even sillier than were abb3w’s examples of how to think without thinking. Another proposal that there’s something or somebody out there painting the dog. A great evolutionary transistor in the sky perhaps.

    A great preserver of the cultural artifacts of science in the bargain.

  73. #73 daedalus2u
    November 14, 2009

    piker, first you have to tell me what the term “computation” means in the sense that you are using it. Do individual cells compile a FORTRAN program and then run it as a batch file? No, they do not. “Computation” means different things depending on the context, if you tell me what you mean, I can tell you if I think that cells can do it.

  74. #74 piker
    November 14, 2009

    daedalus2u, as usual you’re ducking the question. Your “bait” strategy seems to be to get someone to describe some function in enough detail that you can then say, no, that’s not accurate. Then that someone says, well then what is accurate, and you say it depends on the context. Or conversely, given the context, you say, refine the definition.

    Here’s another question to duck. Take the book Wetware which has as a subtitle, A Computer In every Living Cell. What do you think, surmise or expect that the subtitle is referring to, and do you think that reference is accurate?
    Or if you can’t deal with that, try to answer this: What are the ontogenetic choice making functions of all living organisms?

    Or if that’s so general that you will say none that all have in common, take any cellular organism of your choice and tell me in even the vaguest detail something about it’s choice making apparatus and function.
    Then I’ll tell you whether or not I think that’s computation in any sense of the term.

    But please try to answer or duck these questions in the order presented.

    And if you’re depending on one of your iconodule defenders here to jump in and deflect the questions, I won’t take that bait either.

  75. #75 daedalus2u
    November 14, 2009

    piker, you really are projecting that everyone wants to play the same semantic games that you do. Some of us are really just trying to understand things and not trying to beat everyone else over the head.

    Cells do not perform the same types of computations that Turing machines make. As was said earlier, life always operates at the machine code level. Some aspects of physiology can be considered to be analogous to some types of computation. The problem is, that abstraction is a human construct and because “computation” is such a generic term, essentially any operation can be said to be a computation of some sort.

    When a protein folds, is it useful to consider it to be an analog computer that is doing a protein folding computation? When a peptide docks with a receptor, is it useful to consider that the receptor is calculating the goodness of fit and balancing the hydrophobic, hydrophilic and electrostatic interactions? When a water molecule diffuses, it is useful to consider that the water molecule is calculating the water chemical potential and causing itself to migrate down that gradient?

    According to Wikipedia, computation is

    “Computation is a process following a well-defined model that is understood and can be expressed in an algorithm, protocol, network topology, etc.”

    According to that definition, a protein folding itself is not a computation. Some DNA interactions could be used to do computations, but implicit in this definition is that the computation is done according to an understood model. To the extent that what DNA in “the wild” does is not well defined and not well understood, it cannot be considered to be a “computation”.

    I suspect that what you are trying to say is that the living physiology of an organism is a “computation”, and the output of that “computation” is whether the organism survives and reproduces. That conceptualization of cell metabolism as computation is incompatible with the Wikipedia definition.

  76. #76 piker
    November 15, 2009

    daedalus2u,
    Well, at least you tried. Your Wikipedia definition is woefully inadequate. I suspect you chose it because it suits your purpose. Ironically, cell computations are there to suit their purposes, which are to aid them in the choice making process.
    What you suspected as my conceptualization of cell computation as devoted to such an indirect consequence as survival is not likely within their computational capacities. Choice making devoted to an immediate purpose is.

    A protein folding itself can be reactive or the result of an assessment of necessity, or a combination of such. There are examples of such folding that may point to the type of directed “mutation” we find in e. coli for example.

    Selective mechanisms heretofore described inadequately as simply “natural” may also share the use of the same functional apparatus and same computational apparatus.

    I suspect you don’t accept that cells can operate with purpose, so until you do, there’s not much point in turning this into a dissertation on the subject. But if you’d like to read one on your own, check this out:

    http://www.basic.northwestern.edu/g-buehler/summary.htm

  77. #77 bob koepp
    November 15, 2009

    daedalus2u can surely speak for himself. Speaking for myself, I neither affirm nor deny that “cells can operate with purpose.” But I do operate with a methodological rule of thumb that counsels the “default assumption” that where purposeless processes are adequate to the understanding of a phenomenon, purposes should not be posited. I don’t think the need for positing cellualr purposiveness has been established. (But what any of this has to do with DSW’s series on group selection is beyond me.)

  78. #78 daedalus2u
    November 15, 2009

    piker, I think I now understand your purpose in trying to characterize what cells do as “computation”. I think you want to make a logical argument something like this:

    Cells do computation
    Cells evolve
    Cells do computation as they evolve
    Evolution is a form of computation
    Evolution has the properties of computation
    Therefore evolution proceeds along a path of computation
    Therefore evolution has a purpose, the purpose being accomplishing the computation

    The problem with this approach is that the term “computation” means different things in each line.

    The problem with arguments like this is that they are badly formed. They are essentially semantic trickery. The definitions of the terms morph slightly as one proceeds through the argument and mean something completely different at the end. Terms have to retain their unique meaning or the argument is badly formed. If you want to use the same term to apply to different things at multiple scales, you need a really good and clear definition. So far you haven’t given us one for computation or purpose.

    You say the Wikipedia definition of computation is inadequate, ok, what do you mean when you use the term? If you are unable to articulate what the term means, then reframe your argument in terms that you can define so the terms in the argument are unambiguous.

    I completely agree with Bob that imputing purpose to inanimate or non-human agents should be a last resort assumption, not a default first resort assumption.

  79. #79 piker
    November 15, 2009

    Daedalus2u, I’m giving up on you. You attribute a logical argument to me that is far from logical and far from my intent and then conclude that since that was in fact my argument it is badly formed and proves I’m wrong about cells and their computational abilities and the purposes they are put to. But all you have done is manufacture and engineer your own derailable logic train to justify the totality of your ignorance when it comes to the purposive nature of cell behavior, and the behavior of all living organisms.
    Life computes with a purpose, not for a purpose. The within life purpose is to exercise the making of its own choices.

    By the way, life is not inanimate. That’s the entire point that you miss. It is animate precisely because it can exhibit purposive behaviors. Cells and the other life forms they animate are not non-human agents. Human life represents a complex arrangement of the very cells that you would have us regard as the non-purposive inanimate.

    That’s it. If you don’t see that, you won’t see anything else that I have in my bag of tricks to show you.

  80. #80 daedalus2u
    November 15, 2009

    piker, I still don’t understand what you mean, which is not a surprise to me because you won’t define the terms you are using and you are using them in non-standard ways.

    You use the term “purpose” several times. Could you give me a definition for each of those instances of use? For example you say:

    “Life computes with a purpose, not for a purpose. The within life purpose is to exercise the making of its own choices.”

    Does the string “purpose” mean the same in all three instances of its use in that sentence? If so, what is that meaning, if not, could you tell me what each meaning is?

    If I try to apply the meaning of purpose in the second sentence to the first I end up with this:

    [Life computes with the exercise of the making of choices, not for exercise of the making of choices.]

    Which doesn’t make any sense to me.

  81. #81 piker
    November 15, 2009

    Contemplate the following definitions and see if you can find the inference to be drawn.

    Otherwise I can’t help you.

    1) purpose |ˈpərpəs|
    noun
    the reason for which something is done or created or for which something exists : the purpose of the meeting is to appoint a trustee | the building is no longer needed for its original purpose.
    • a person’s sense of resolve or determination : there was a new sense of purpose in her step as she set off.
    • (usu. purposes) a particular requirement or consideration, typically one that is temporary or restricted in scope or extent : pensions are considered as earned income for tax purposes.
    verb [ trans. ] formal
    have as one’s intention or objective : God has allowed suffering, even purposed it. See note at intend .
    PHRASES
    on purpose intentionally.
    to no purpose with no result or effect; pointlessly.
    to the purpose relevant or useful : you may have heard something from them that is to the purpose.
    ORIGIN Middle English : from Old French porpos, from the verb porposer, variant of proposer (see propose ).

    2) preposition |ˌprepəˈzi sh ən|
    noun Grammar
    a word governing, and usually preceding, a noun or pronoun and expressing a relation to another word or element in the clause, as in “the man on the platform,” “she arrived after dinner,” “what did you do it for ?”

  82. #82 piker
    November 15, 2009
  83. #83 daedalus2u
    November 15, 2009

    piker, so you are saying that cells have intention or were intentionally created by something else so as to have intentions? If I apply that definition to your sentence:

    “Life computes with a purpose, not for a purpose. The within life purpose is to exercise the making of its own choices.”

    Are you saying that evolution is intention driven? That evolution has a purposeful intention that is manifest in the evolving group so as to achieve the purpose that evolution intends?

    Is evolution aware of this intention?

  84. #84 piker
    November 15, 2009

    daedalus2u,
    You got it backwards, perhaps deliberately. “For” a purpose would mean that the purpose was other than life’s. Not what I said or intended to say. “With” a purpose means life forms its own purposes which are to satisfy immediate needs. . Life has no intent to evolve, it intends to solve its immediate problems, initially involving what some have referred to as seeking pleasure and avoiding pain. Its evolution has been an unintended consequence. But life adapts its purposive behaviors as well. It refines its strategies. It extends its goals. Goals represent what it intends to seek, almost never exactly what it actually finds. Yet the seeking necessarily continues. If you see no computational or choice making process involved, by all means continue to think so.

    Evolution has no intention and no awareness, and you should know that’s not what I’m arguing, or have ever argued. None of my referenced material have presented such an argument. Your questions verge on the insulting and have tried my patience. Clearly you have already made a study of these matters and don’t need to learn anything additional from me. So your motives at best are suspect. I’m not taking this any further with you.

  85. #85 daedalus2u
    November 16, 2009

    piker, the definition of purpose is inextricably linked to intention. If there is no intention behind an action, there can be no purpose behind it. Intention is linked to awareness. If there is no awareness, there can be no intention. If life is intending these actions, are you saying that life is self-aware? I don’t understand how you can be, but the terms you are using lead me to that conclusion.

    “With” a purpose means life forms its own purposes which are to satisfy immediate needs. Life has no intent to evolve, it intends to solve its immediate problems, initially involving what some have referred to as seeking pleasure and avoiding pain. Its evolution has been an unintended consequence. But life adapts its purposive behaviors as well. It refines its strategies. It extends its goals. Goals represent what it intends to seek, almost never exactly what it actually finds. Yet the seeking necessarily continues.

    How do you explain apoptosis? How does a cell triggering its own death serve an “immediate problem”? Or does “immediate problem” have a different meaning than what I think it means?

    How did “seeking pleasure and avoiding pain” mechanisms evolve? Pleasure and pain are complex states of a nervous system. How can single cell organisms feel pleasure or pain? Or do those terms have different meanings when applied to single cells?

    I think what you are describing is a metaphysical philosophical approach, something to explain “why”, not something to explain “how”. I am not concerned with why, only with how even if I may use why-type language. That is sloppy writing on my part. I recognize and appreciate that every time I use “why-type” language, in that instance it is a post-hoc rationalization and not something that provided an a priori purpose or intention for the event.

    I think your approach is more of a redefinition of “life” as “something which functions with a purpose”. Is that correct?

    I think this is a reformulation of what is meant by “homeostasis”. With your reformulation being something like that “life is something that functions with the purpose of maintaining homeostasis”. Is that correct?

  86. #86 piker
    November 16, 2009

    Daedalus2u,
    Pleasure-pain are metaphorical terms in this context. I have other versions of initial motivating factors that I use in my research that wouldn’t interest you because, as you say, they are more about why than how. I don’t think you can know the how without knowing the why, but that’s just me. (And of course, as you noted, you don’t get there by simplistic post hoc rationalization.)
    Homeostasis would not be a purposeful goal, but a consequence of other problem solving choices. An equilibrium relative to achieving a satisfactory balance or place on or in the pleasure to pain spectrum. As to single cells, they clearly avoid what they have learned are destructive elements, and an anticipation of such a possibility is where the “pain” signaling seems to have sprung from. All of this involving the computational process of not repeating acts that have lead to error. The why and the how being the essential elements of any such strategy – conceptualization not a prior requirement. Cellular algorithms evolve that are conceptual structures in and of themselves. Symbol driven holography.
    I’m not going to go into detail on this. I’m neither your teacher nor your student. Open your mind, read Damasio, Margulis, Jablonka, Boehm. Read “On Deep History and the Brain” by Smail, a marvelous piece of work.

  87. #87 abb3w
    November 16, 2009

    piker: Nice little snark there about K-12 when you don’t seem to know the meaning of increment.

    You misunderstood me twice there. First, I’m griping that Random Walks are sufficiently fundamental that they ought to be in K-12, so that most people could have a few years of familiarity with them. If you took that to imply that you have a K-12 education, that was entirely not the intent. Rather, the implication was that since they are not a standard part of the general K-12 education, you might not have had much experience during the more specialized framework of college education; it’s far easier to avoid than even partial differential equations. It’s a gripe about modern education as I see it, not about you.

    More to the point of the discussion, I’m not objecting to “increment”, but to “planning”. The properties of non-random walks don’t seem to require “planning” to obtain much the same results as a directed search.

    piker: Although there’s nothing much there that explains how harm versus benefits are assessed

    Probability of survival long enough to reproduce offspring that are viable in turn.

    piker: Yet at the same time you deny that some individual was changing in the process. Some individual “is a change” but not even that individual was at any time changing? Some sort of magical selection process at work?

    Selection and reproduction are separate.

    The child is different from the parents. During the processes of mitosis of pre-germ cells, meiosis of germ cells, and fusion, some mutations occur; typically about a hundred for humans, IIR. The individual is a change from the prior population, even though it need make no changes over the course of its life as a discrete individual.

    Selection occurs subsequently, over the course of the new individual trying to reproduce in turn.

  88. #88 piker
    November 16, 2009

    abb3w, OK, I will then have to retract my own retributory snark.
    A couple of other responses:
    1) Purposive behaviors are in that sense planned, and I believe trial and error is a purposive strategy.

    2) The individual is a change because the change began in a preceding individual. Mutation is change whether or not it’s a self directed or self-organized process. In my view individuals always will have taken advantage of whatever was the proximate cause of a mutation. Natural selection theory as it stands, or has stood until recently, does not consider that aspect of selection as necessary to the selective function.
    Additionally, I don’t see how one can propose that selection must occur subsequent to replication simply as being a valid logical proposition. It stands the meaning of replication on its head. If we’re dealing with adaptation via mutation, it’s the mutated gene that is replicated. Otherwise the replicated product cannot be “a change” under any known interpretation of biological similitude.

  89. #89 abb3w
    November 17, 2009

    piker: Purposive behaviors are in that sense planned, and I believe trial and error is a purposive strategy.

    Trial and error can be a purposive strategy; this is different from saying that it necessarily is one in a particular case. Philosophically, purpose requires a choice (with more than one option in the set) and intentionality.

    Showing the latter is where the heart of the difficulty lies in your apparent position. It’s what distinguishes deliberate design as in Engineering from “accidental” evolution in biology.

    piker: The individual is a change because the change began in a preceding individual.

    However, the change is usually not considered part of the preceding entity, but the demarcation to a subsequent entity. This is in part because such genetic changes in the germ cells do not express in any phenotypic change of the progenitor.

    Essentially, it’s a question of where you draw the space-time demarcation boundary of “individuals”. From a philosophical sense, one can consider any boundary to define “individual”; for example, one could demark as “an individual” each human female from immediately prior to her first ovulation, to the moment of her death, PLUS all her male children over their entire lifetimes, and all of her female children from conception to immediately prior to their first ovulation or demise (whichever comes first). This, however, is not what I understand is the usually used demarcation of an “individual” for biology; nor do most human cultures behave quite like that.

    piker: In my view individuals always will have taken advantage of whatever was the proximate cause of a mutation.

    Effect, yes, in so far as is possible to take advantage; but cause? What you mean by that is not clear.

    piker: I don’t see how one can propose that selection must occur subsequent to replication simply as being a valid logical proposition.

    Usually it is subsequent to being born (replicated) that it is answered whether an individual will or will not die before reproducing in turn (selection).

    Although, if you want to be technical about it, replication is actually at the point the initial blastomeric cell forms from germ cell fusion.

    So, it depends which direction you’re looking at the replication from: having been produced, or producing others in turn. “Which came first, the phoenix or the flame?”

    piker: If we’re dealing with adaptation via mutation, it’s the mutated gene that is replicated. Otherwise the replicated product cannot be “a change” under any known interpretation of biological similitude.

    Not all mutative genetic changes are per se “adaptive” in and of themselves. Some are neutral; genetic drift comes into play there, behaving closer to a uniform random walk than a non-uniform one.

  90. #90 piker
    November 17, 2009

    abb3w, a whole lot of sophistry going on there.
    Trial and error is a biological strategy. The term has no epistemological value otherwise. It’s not that it “can be” a strategy. It is nothing else but a strategy.

    And no matter where you draw your space time boundary, you can’t take the change out of its individual location. You want to draw the line exactly where change is, except that change in biological entities is not all of a piece.

    This whole demarcation argument is silly. Just admit that and get it over with.

    And you write: “This is in part because such genetic changes in the germ cells do not express in any phenotypic change of the progenitor.” Actually there’s reason to believe that they do as far as the selective process is concerned. Does not an organism also need to test the initial results of its own mutations, accidental or otherwise, to make the process work? To avoid the passing of the more defective if not lethal mutations? Another reason why the selection process is not floating out there somewhere in the sky.

    Do not express in any phenotypic change? They express an operative change based on a physiological alteration of one of their parts.. What more do you need to determine where the change begin to have its effect, if effect of change must precede your required line of demarcation.

    Then you say this: “Usually it is subsequent to being born (replicated) that it is answered whether an individual will or will not die before reproducing in turn (selection).” Even if that made some relevant sense, does “usually” then equate to always? Of course not. And you’d have selection as turning on a dime as part of your imaginary mechanism.

    And note that I said IF we are dealing with adaptation via mutation its the mutated gene that is replicated.

    You then respond that not all such changes are adaptive. No, they aren’t, but IF they were, etc. Put the goal post back where it was please.

  91. #91 abb3w
    November 18, 2009

    piker: Trial and error is a biological strategy. The term has no epistemological value otherwise. It’s not that it “can be” a strategy. It is nothing else but a strategy.

    I’m not objecting to “strategy”; I’m objecting to “purposive”.

    piker: Does not an organism also need to test the initial results of its own mutations, accidental or otherwise, to make the process work?

    The testing need not be in its own phenotype, no; if organism “A” produces germ cells with mutations, those mutations need only be tested in its offspring “B” that results.

    For example: suppose an improper meiosis results in my producing a sperm cell with a full Chromosome 21 pair, rather than the usual singleton. That happening will only affect that cell, and not any of the others in my body; and furthermore, does not even affect the function of the sperm cell as a sperm cell. However, if that sperm cell goes on to fuse with an egg cell, that mutation will be expressed in the offspring’s phenotype… which, unfortunately, is Down’s Syndrome in this case, which tends to be detrimental.

    Similarly, a redundant copy (inherited from my father without obvious effect) of the gene for SWS1-Opsin might undergo the single-amine mutation to a UV-Opsin during the mitosis that creates the somatic cell immediately precursor to that sperm. If instead this sperm fuses with an egg, mutation instead leads to an offspring with the ability to see beyond the “normal” human range to a full octave of colors… which might be slightly advantageous.

    piker: Then you say this: “Usually it is subsequent to being born (replicated) that it is answered whether an individual will or will not die before reproducing in turn (selection).” Even if that made some relevant sense, does “usually” then equate to always? Of course not.

    Err… once you include the more technical demarcation of “replication is actually at the point the initial blastomeric cell forms from germ cell fusion” rather than the more poetic “being born”… yes, it would seem to be “always”; “usually” was an attempt to convey dry understatement.

    Or do you have a counterexample in mind?

    piker: And note that I said IF we are dealing with adaptation via mutation its the mutated gene that is replicated.

    However, many (most?) mutations are first introduced in the meiosis process; the mutation enters the population due to that imperfection of the reproduction process.

  92. #92 piker
    November 18, 2009

    abb3w, there are those who can run but not hide, but then there are those who hide but can’t run. You’re hiding (yet in rather plain sight) any glimmer of acknowledgment that the selection process has a biological component that involves change within individuals rather than somewhere in between the changed and the unchanged. You want to feel that you’ve won an argument, and that even though my view is essentially correct, if you can find some aspect of my dissertation that isn’t, then the truth of the matter at hand will remain irrelevant.
    Sophistry, pure and simple, the use of fallacious arguments, esp. with the intention of deceiving.

    Object as you will, but If a thing is, as you’ve conceded, a strategy, then by definition it’s purposive.

    “those mutations need only be tested in its offspring “B” that results.”
    But if that were the only need, then why is the initial testing in the parent of that offspring? And needed or not, the fact remains that with reference to your particular random mutation scenario, that’s where the selective assessments actually begin. Your litany of exceptions, whatever you seem to think they evidence, only go to prove that rule.

    “Usually it is subsequent to being born (replicated) that it is answered whether an individual will or will not die before reproducing in turn (selection).”
    Except that you omit any reference to the nature of the question that was first asked and wherefrom came the asker. Because the initial questions would include whether this mutation is worth keeping around to begin with. Worth its initial selection. No yes or no answer required as to its eventual longer term success.

    Of course many mutations are first introduced in the meiosis process. Which tends to make my case rather than yours. I’ll leave it to you to tell me why it doesn’t.

  93. #93 piker
    November 18, 2009

    Addendum to the above:
    You don’t give life its proper credit for the grasp of its collective intelligence.
    From Jablonka’s writings:
    “In a multicellular organism there are a lot of constraints on all types of variation–any variant has first to pass through the bottleneck of development to produce a viable organism.”
    Even the eventuality of death as a selective consideration requires that an individual biological consequence be a factor of the process.

  94. #94 abb3w
    November 18, 2009

    piker: Object as you will, but If a thing is, as you’ve conceded, a strategy, then by definition it’s purposive.

    That is incorrect. I repeat: Philosophically, purpose requires a choice (with more than one option in the set) and intentionality. Showing the latter is where the heart of the difficulty lies in your apparent position. It’s what distinguishes deliberate design as in Engineering from “accidental” evolution in biology.

    piker: But if that were the only need, then why is the initial testing in the parent of that offspring?

    Your question presupposes all mutations are tested in the parent. EG, the above example of Down Syndrome, where the mutation is not “tested” in me, because I do not myself exhibit the phenotype.

    piker: Except that you omit any reference to the nature of the question that was first asked and wherefrom came the asker.

    The nature of the question is “will it reproduce?” It’s a “question” in the sense of a binary boolean test having consequences, just as “did the uranium atom fission” has a consequence to “are there going to be more free neutrons”.

    piker: Because the initial questions would include whether this mutation is worth keeping around to begin with.

    Which is answered experimentally (so to speak): does the organism survive to reproduce? This is all that’s needed.

    piker: Of course many mutations are first introduced in the meiosis process. Which tends to make my case rather than yours.

    Sigh… you’re probably trying to argue these mutations are introduced deliberately. This is philosophically possible, but that’s not enough. There is no evidence to support this in general, nor is it necessary to the results observed; branching non-uniform random walks are fully capable of mapping search spaces.

    Now, if you can find evidence of such a mechanism operating, that would change things. It could probably be a Nobel caliber discovery.

    However, meanwhile, it requires positing an entity (and mechanism) without evidence, increasing description length beyond the competing alternative of evolution; and thus, probably wrong.

  95. #95 piker
    November 18, 2009

    “Philosophically, purpose requires a choice (with more than one option in the set) and intentionality. Showing the latter is where the heart of the difficulty lies in your apparent position.”
    Trial and error as a strategy represents a choice among limited options. There would of course be nothing to continue to try of there were only one option. There would be no need for anything that we are defining as strategy. The intent is to test the efficacy of the initial and then succeeding choices between or among the options. intentionality is of course defined as the fact of being deliberate or purposive. (And if you can define life as something exclusive of the need for choice, be my guest.)

    Everything else you’ve added deserves the sigh that you’ve applied to your deliberate (and up to your old tricks) misconstruction of what would be my argument from the “melosis” standpoint. Because it wasn’t to introduce the deliberateness of mutation, or posit the evidence of the mechanism. (Which others have already done to some extent, and which in any case if philosophically possible would also be scientifically possible – science being in the end a philosophical methodology.)

    And the entity without evidence that’s under review here is your posited natural selection mechanism that seems to exist without known location, structure, mechanistic detail, or philosophically defendable purpose.

    If by your standards I’m required only to prove mine exists by demonstrating that yours doesn’t, then I think my work here is done.

  96. #96 abb3w
    November 19, 2009

    piker: The intent is to test the efficacy of the initial and then succeeding choices between or among the options.

    That is an effect of the strategy (or not obviously far enough from being an effect to be worth digression from the fundamental question). However, showing an effect is not the same thing as an intent.

    For instance, my intent in bringing up the lack of coverage of random walks in K-12 was to highlight what I consider a gross inadequacy of elementary school mathematics education that could be readily remedied; the initial effect was rather different.

    Your problem isn’t in showing “choice”; in mathematical terms, a “decision function” qualifies, which is easy to show. The hard part is showing a purpose.

    piker: Because it wasn’t to introduce the deliberateness of mutation, or posit the evidence of the mechanism.

    Then I guessed wrong as to the intent behind your choice.

    piker: And the entity without evidence that’s under review here is your posited natural selection mechanism that seems to exist without known location, structure, mechanistic detail, or philosophically defendable purpose.

    I would agree that there is no “purpose” attributed; purpose is not, however, necessary to existence.

    The questions of “location, structure, [and] mechanistic detail” look to be trying to read more into “mechanism” than was intended to be conveyed by that fuzzy bit if English. To the extent they make sense to me, I think others would consider they’re pretty much covered above sufficiently. I don’t see need for a recap.

  97. #97 piker
    November 19, 2009

    abb3w
    You don’t seem to get it that if an effect of a strategy is to answer a question, whether that answer was in line with the prior expectations or not, there had to at least be a question behind the strategy. And that the intent behind the question was to get an answer.

    And your random walk, if it’s to no purpose, with nothing or no-one there to observe where it leads or ends, no-one to feel its effects, or discover its effects, or glom onto them as answers to any problems, or fill any hereto unfillable holes in its path, then to have taken such a walk would not have been in any sense a strategy, or have been to any purpose, or served any purpose in the biological world.
    So whether you had been exposed to the concept in K-12 or later, such exposure would have been to no purpose if the aspect of it or its walker having any purpose at all was overlooked.
    There’s a common thread to all of your argumentation, and it’s the one that posits a lack of need for purpose to the mechanisms of life. A lack of understanding that life depends on an ability to make its own choices from its recognition of its own options.

    Because you say in the end that “purpose is not, however, necessary to existence.”
    Yet it is fundamental to an understanding of biological behavior and motivation that all of life’s functional apparatus be operated with a purpose.
    You don’t have to know the purpose beforehand to accept that there is one. That acceptance is the philosophical basis for all intellectual curiosity in the biological sciences.

  98. #98 bob koepp
    November 19, 2009

    I think the discussion would benefit from working definitions for some key terms, such as ‘function’ and ‘purpose’, and some indication of how issues being raised intersect with views that have veen articulated by philosophers and theoretical biologists who have studied these matters. It should certainly give us pause that what purposes are “accepted” for various biological traits has changed along with our theories. And the “adaptationism debates” should have sensitized us to the possibility that some traits have no purpose whatever.

  99. #99 piker
    November 19, 2009

    There is no answer without a question. There is no question without a questioner whose intent and purpose were to ask it. Even if the particular questioner could not be expected to recognize the purposive nature of its curiosity. Or understand that any of its genetically determined characteristic were meant to fulfill a functional purpose. Let alone that all of them were.

  100. #100 bob koepp
    November 19, 2009

    According to one prominent view, biological functions are to be identified with adaptations, i.e., traits that have been “selected for.” Selection is assumed to be a “bllind” process that requires no purpose or motive or intention of any sort. From such a perspective, adaptations are not answers to questions or solutions to problems — the questions or problems are metaphoric, not actual factors in biological evolution.

  101. #101 piker
    November 19, 2009

    The terminology is metaphorical. The problems to be solved are real. Prominent views not withstanding.
    Creationism is a more prominent view than the one that sees life as having formed its own purposes. In fact the one view here is the most reasonable alternative to the other. Yet prominence should not be the decisive factor. Life forming its own purpose makes infinitely more sense.

    And any view that holds the selective process has no purpose is basically holding that it had no usefulness. Yet even usefulness unfulfilled would not negate the fact that usefulness was the intended purpose of the function.
    Another reason why a selection process operating blindly, without assistance from the organism that receives the benefit of a function with neither intent or purpose, is a nonsensical concept.

    Life acts with intent and purpose. How or why purpose arose in a universe that seemingly had or has none without it is a subject for debate. That life has received the benefit (or some say curse) of purposive functionality is not.

  102. #102 bob koepp
    November 19, 2009

    piker –
    1) My reference to “one prominent view” was meant only to note that the view is fairly widely known — not to convey any privileged epistemic status. If you want to suggest some other set of “working definitions” that’s fine with me; just be specific, please.

    2) You say, “any view that holds the selective process has no purpose is basically holding that it had no usefulness.” I don’t think the concepts of ‘purpose’ and ‘usefulness’ are related in the way suggested by your statement. Perhaps a discussion of so-called spandrels would bring issues into clearer focus.

  103. #103 piker
    November 19, 2009

    Spandrels, if selected for, would be useful for a number of purposes. But otherwise I see no point in discussing the subject further. If you don’t agree that life acts with intent and purpose, then that’s fine with me.

  104. #104 piker
    November 19, 2009

    Although it would seem you’ve noted that life appears to have become quite proficient in the use of computational geometric algorithms. Functional strategies selecting for and forming their own structural components. Strategy and structure becoming interdependent. Allowing adaptive pressures to effect change in a structure while the organism retains the essentials of its strategies – as well as the reverse, or some combination thereof. But I digress..

  105. #105 mollyrogers
    November 20, 2009

    piker:
    You wrote “As to single cells, they clearly avoid what they have learned are destructive elements, and an anticipation of such a possibility is where the “pain” signaling seems to have sprung from. All of this involving the computational process of not repeating acts that have lead to error.”

    I do not reject the notion of life acting with intent or purpose. But I do not yet understand what you are proposing as the concrete biological computations that would constitute the “assessment of error”. On what time, space and identity scales are those assessment made? What are the criteria or standards against which “error” is determined?

    I gather you are speaking of a proactive assessment by an individual for itself, rather than the hindsight assessment by another observer that failure to have survived indicates “error” on the part of some individual or group.

  106. #106 piker
    November 20, 2009

    The determination of error is not a simple or immediate process. Error includes efforts that lack success, measured in part by the unexpectedness of consequences. Cooperative efforts that don’t achieve their purpose. Competitive efforts that reveal weakness in strategy. The point would be that trial and error has arguably been every organism’s fundamental strategy since life first developed the capacity for forming expectations.

    Observations regarding another’s failure to have survived would involve a capacity for abstractions a bit above a single cell’s pay grade. But observation would be theoretically possible by one cell of another’s destruction, tending to change the nature of their cooperative strategies accordingly. In fact there is evidence to indicate a capacity for such an analytical assessment tends to accelerate the evolutionary process of certain cooperative groups as well.

  107. #107 mollyrogers
    November 20, 2009

    “In fact there is evidence to indicate a capacity for such an analytical assessment tends to accelerate the evolutionary process of certain cooperative groups as well.”

    Could you point to further information on this?

  108. #108 piker
    November 20, 2009

    You should find something on that here:
    The Life and Behavior of Living Organisms: A General Theory ~ Elliott Jaques
    Also see Evolution in Four Dimensions – Jablonka and Lamb for some supporting material.
    And Wetware – Dennis Bray

    Also see some of Carl Zimmer’s writings on bacteria and parasites.

  109. #109 bob koepp
    November 20, 2009

    “Spandrels, if selected for, would be useful for a number of purposes. But otherwise I see no point in discussing the subject further.”

    Further discussion seems absolutely necessary since spandrels are, by definition, not selected for; they are artifacts or side-effects of the selection process. This seems to me to illustrate very clearly the need for shared working definitions and a working knowledge of existing theories if discussion is to be fruitful. Can we agree at least to strive for fruitful discussion?

  110. #110 piker
    November 20, 2009

    You don’t believe that life acts as its own engineer and I do. So I presumed you were referring to biological structures as adaptive by accident while I believe they are selected with and for a purpose. A spandrel to me is a fractal form, useful to life as a geometric aspect of a strategic formation.
    As to having a fruitful discussion with you, that’s what I was referring to as pointless. Thought I’d expressed that view before. Nevertheless, I made an observation about usefulness.. My mistake. Should just have let things go.

  111. #111 abb3w
    November 22, 2009

    Hmmm… yanking a hyperlink or two to try and avoid getting stuck forever in the moderation queue….

    piker: You don’t seem to get it that if an effect of a strategy is to answer a question, whether that answer was in line with the prior expectations or not, there had to at least be a question behind the strategy. And that the intent behind the question was to get an answer.

    Essentially, yes. I don’t see that, as used in this context, the existence of a “question” implies that it was “asked” by an entity with intent.

    From where I sit, it looks like you are implicitly assuming the conclusion, resulting from an implicit redefinition on the nature of “question”.

    piker: And your random walk, if it’s to no purpose [...] then to have taken such a walk would not have been in any sense a strategy

    Much the same here, only on “strategy”.

    piker: There’s a common thread to all of your argumentation, and it’s the one that posits a lack of need for purpose to the mechanisms of life.

    More exactly, that “purpose” is not necessary to explain the evidence; that the current best (minimum description length) hypothesis explains without positing an external purpose; and that thus, such purpose probably doesn’t exist.

    Put another way: you’re making a positive assertion to existence. The philosophical default is that burden of proof lies with the positive claim; nonexistence is valid by default, until shown otherwise. The formalities of defining “positive” versus “negative” claims and of justification for why the burden of proof falls this way both lie well beyond the scope of this blog.

    piker: There is no question without a questioner whose intent and purpose were to ask it.

    If one apple falls from a tree, landing next to another already on the ground, it gives an answer of “two” to the question “What is one plus one?”, so… who asked?

    I’m not motivated enough to try and properly delineate the difference in philosophical status of existence between the actual and the potential. However, that looks to be the neighborhood of the difficulty; mistaking human interpretation of the world as question-and-answer for implying an external inquirer.

    piker: As to having a fruitful discussion with you, that’s what I was referring to as pointless.

    Eh. While I don’t think you’ll change the mind of anyone who’s bothered to participate in the discussion, the discussion might have influence on what any others who read it think of either side.

  112. #112 piker
    November 22, 2009

    To whom it may concern here:
    Before trying your hand at philosophy, you should learn something about the use of metaphor – especially in the use of mathematical symbols as metaphor. If any one did ask what one plus one meant, it would never be the exact equivalent of any two apples.

    Biologists use metaphor, often to a fault, in their description of life and its mechanisms. The term mechanism itself being a metaphor for the philosophical concept that all natural phenomena, including life and thought, allow mechanical explanation by the sciences. Question, intent, and purpose in that context are all metaphorical terms for natural phenomena.

    Good book to read on that subject would be Philosophy in the Flesh, Lakoff and Johnson.

  113. #113 bob koepp
    November 23, 2009

    to whom it may concern:
    I’m not at all shy about “trying my hand” at philosophy. Lakoff has done some good work in linguistics, and Johnson has done some good work in philosophy of mind. Peace be upon them. I don’t know of any philosophers of science, however, who think L&J’s work on metaphor has illuminated any issues in philosphy of science generally, or philosophy of biology in particular. For my money, what Mary Hesse has to stay about the use of metaphor in science is much more insightful, even if a bit dated.

  114. #114 Amplexus
    March 29, 2010

    As just an undergraduate I’ve been skeptical of this group selection thing as i’m coming to learn about it. You know the way I treat anything i’m learning. I’m just struck by
    how important social context is though. See I was fed this propaganda of science as totally objective because that’s the goal.

    I have to just say that from the outside looking in: I see an enormous inconsistency with the opponents of group selection.

    For example: Theories of the non-random dispersal of physical trait phenotypes like lets just say in the harmonia beetle were welcomed as a breakthrough in population genetics. The formulas and such are roundly used and even us undergrads have to memorize them in upper-div classes.

    It’s widely accepted that genotypes of populations are disturbed and even shifted due to dispersal. That’s not controversial. They even go so far as to point out the role of dispersal in frequency dependent fitness.

    But paradoxically when you talk about the non-random distribution of altruistic behavioral phenotypes there’s apparently some reason why we can’t think of non-random dispersal as a mechanism for the evolution of altruism.

    So in one case one can accept that dispersal changes the scenario under which something evolves and frequency dependent fitness and the small barriers to gene flow.

    But they fail to put it all together. A genetic basis for altruism can theoretically simultaneously be due to inclusive fitness and between group selection. The goal ought to be to tie all these notions of reciprocal altruism, inclusive fitness, and even group selection into a
    new synthetic theory. I mean, from what I understand that’s what theoreticians do.

    I think the resistance is like you said due to the 60′s backlash against niave group selectionism but just went too far.