Evolution for Everyone

Obsessing about different accounting methods runs the risk of turning evolutionary theory into a kind of postmodernism in which there are no facts, only different ways of viewing. In his book Fear of Knowledge: Against Relativism and Constructivism, philosopher Paul Boghossian shows, for anyone who needs reminding, that nothing has happened in philosophy to prevent us from making factual statements such as “mountains existed before people”. So it is with the factual claims associated with inclusive fitness theory.

How can we distinguish fact from perspective? It’s easy when you get the hang of it. Merely state your factual claim in a way that can be evaluated by any perspective. Here’s an example involving the first and (at the time) the most important claim associated with inclusive fitness theory–that it explains altruism without invoking group selection.

The phrase “without invoking group selection” isn’t used much any more, but it was once mandatory for authors who wanted their ideas about social evolution to be taken seriously. Their scrupulous avoidance of group selection was not a statement of preference for a particular accounting method but a factual claim–that their theory invoked a certain process, different than the process invoked by group selection, such that one could be just plain right and the other just plain wrong. That’s how science is supposed to work. Different hypotheses invoke different processes that make different predictions about the real world. Some predictions conform to reality better than others, qualifying them as provisional facts. Perspective has nothing to do with it.

Here’s how to state the factual claim in a way that can be evaluated by any perspective: “According to group selection theory, altruism is selectively disadvantageous within groups of individuals who are socially interacting with each other and can evolve only on the basis of a selective advantage at a larger scale (between groups). My theory of social evolution explains altruism in a way that does not invoke group selection in this sense.”

This way of stating the factual claim forces us to make certain comparisons. In particular, it forces us to compare selection differentials within single groups to selection differentials in the total population. All theories of social evolution include the information for making this comparison, but some retrieve the information more easily than others. In my previous post, I compared two banks that both record the same information, but one reports it according to date and the other according to tax-exempt status. Let’s say that you want to know the tax-exempt status of a certain check. If you belong to one bank, this information is immediately available to you. If you belong to the other bank, you have to work harder to retrieve the information because of the way it is categorized, even though it is available.

So it is with multilevel selection theory and inclusive fitness theory. The former is designed to compare selection differentials within and between groups, but the latter accounts for the costs and benefits of altruism in a different way. In its original formulation, it counts the effect of an altruistic gene on itself in the actor (a negative effect) and on copies that are identical by descent in the recipients (a positive effect). When the net effect is positive, altruism is assumed to evolve in the total population.

All well and good, but what is the fitness of an altruist compared to a non-altruist within the same group? Unless we can answer this question, we can’t say whether inclusive fitness theory counts as an alternative to group selection. If you find it hard to answer this question based on my description of inclusive fitness theory, don’t feel bad. Neither could Hamilton or anyone else at the time. The Price equation was required to show Hamilton that altruism is selectively disadvantageous within groups, even when the groups are composed of kin, and evolves only by virtue of groups with more altruists contributing more to the gene pool than groups with fewer altruists. The significance of relatedness is that when groups are composed of kin, altruists and non-altruists are segregated into different groups more than when the groups are composed of non-relatives.

In short, Hamilton had discovered a nifty way of calculating when altruism evolves in the total population, but he had not discovered an alternative to group selection as the process whereby altruism evolves. It is remarkable how easily Hamilton accepted this conclusion once he “saw the light” through the lens of the Price equation. In his own words:

I was on the phone telling him enthusiastically that through a “group-level” extension of his formula I now had a far better understanding of group selection acting at one level or at many than I had ever had before (see my previous post on Hamilton for more).

In other words, Hamilton now regarded his earlier belief that inclusive fitness theory explains the evolution of altruism without invoking group selection as just plain wrong. Perspective didn’t have anything to do with it, any more than when he regarded the claim as factually correct. Just because the role of group selection wasn’t obvious on the basis of his earlier formulation did not prevent it from being a fact. His earlier belief was like saying “my check is not tax deductable because it was cashed on July 18”.

All of this took place in the 1970’s. Historians of science are needed to explain why Hamilton’s new interpretation didn’t quickly become the consensus view of all evolutionary biologists. Instead, the majority continued to treat his original formulation as an alternative to group selection just because the role of group selection is opaque.

The same story can be told for every other theory of social evolution that claimed not to invoke group selection. All of them assume that social interactions take place among sets of individuals that are small compared to the total population. They must, because it is a biological fact of life. All of them assume that most traits labeled cooperative or altruistic are selectively disadvantageous within groups and evolve only by virtue of the differential contribution of groups to the total population. The only reason that they were ever regarded as alternatives to group selection is because their way of accounting for evolutionary change in the total population makes the local disadvantage of altruism opaque.

Against this background, we can return to the current controversy over the rightness and wrongness of inclusive fitness theory. Is it a nifty way of accounting for what evolves in the total population? Perhaps. Does it explain altruism in a way that does not invoke group selection? The answer to that question is just plain no. Mountains existed before people and altruism evolves by group selection. Perspective has nothing to do with it.


  1. #1 Ben Allen
    September 13, 2010

    Are you arguing that group selection is responsible for all evolved altruism?

    If so, that’s a tricky claim to make. First you’d have to define altruism in such a way to exclude both reciprocal altruism and tag-based altruism are excluded. Then you’d have to come up with a definition of “group selection” that is general enough to include assortment based on spatial location, relatedness, or phenotype.

    Let’s say you were able to do this by inventing a mathematical formalism in which some notion of “groups” could be identified in any structured population, and the evolution of altruism could be explain in terms of selection of such groups. But then you’d be making the same mistake as the kin selection people. You would have invented an alternative accounting method that emphasizes a particular perspective, in order to claim that a single mechanism is responsible for all instances of altruism.

    Kin selection proponents argue that interactions between relatives explain all evolved cooperation. You seem to be arguing that group selection explains all evolved cooperation. Both claims can only be justified by adopting a warped and limited perspective that emphasizes a single mechanism while obscuring others.

    Why not instead study the evolutionary process directly, and see which mechanism is the most natural for any given instance?

  2. #2 Graves
    September 18, 2010

    This argument can be completely inverted, as I know you realize but don’t mention here. By committing what you’ve called the “averaging fallacy”, it is possible to see from a gene’s eye view that what is evolving has the highest inclusive fitness (when averaged across the global population). Therefore, you don’t need selection working across levels in order to get the evolution of altruism. Selection at the level of the gene creates patterns at higher levels, and this includes greater reproductive output for some groups than others. This means that on the accounting side of this issue (individual versus group selection), you are right–explanations for the evolution of altruism based on inclusive fitness can be translated into group terms with ease. There is another part to this question, though, and that is whether or not selection should be properly conceived as a force acting across multiple levels. This is a deeper question, a partly philosophical one, and one of the big successes of inclusive fitness is that a selective process at the genic level has thus far proved sufficient.

  3. #3 Matt Zimmerman
    September 29, 2010

    Ben Allen

    I think you are missing the point.

    Multilevel selection models and inclusive fitness models are different modeling frameworks (or accounting methods to use DS Wilson’s terms). They are not different mechanisms to explain the evolution of altruism. They account for the effects of specific behaviors on reproductive success.

    Tag-based and reciprocal altruism are specific behaviors that can be modeled in an inclusive fitness framework or a multilevel selection framework.

    I think your confusion, and something I think is rampant in the Nowak et. al. paper and this debate generally is a confusion of modeling styles vs mechanisms.

    To extend DS Wilson’s analogy, if multilevel selection and inclusive fitness theory are accounting methods, tag-based and reciprocal altruism are the contents of the account.

  4. #4 Troy Camplin
    September 30, 2010

    There are appropriate uses of perspectivism and inappropriate versions, as I argue here:

    The Modern Era split the world asunder. First Descartes split us in two: a body and a soul. And that appeared to be enough of a division for philosophers to elaborate on for centuries, whether they called the split body and soul, the noumenal and the phenomenal, the body and the will, or Being and Becoming. But Nietzsche introduced the West to even more divisions with perspectivism, and postmodernists have taken up this idea, and have divided the world into a plurality of perspectives, all unconnected and unable to communicate. Postmodernism sees the world through multiple perspectives only, rejecting any unity. But this gets only half of Nietzsche’s thought. Perspectivism is a way to get to knowledge, so the drive for a perspectivism-only view is the unlimited knowledge drive Nietzsche criticizes. Nietzsche does not favor taking up perspectivism for perspectivism’s sake. Nor does he find equal value in all perspectives. By considering different perspectives, we can then more ably judge how each of these perspectives aid us in our understanding, helping us uncover more about a subject. Different perspectives give us different ways of knowing about a thing, but bringing together as many perspectives on the thing will help us to better understand it. “The sea is the purest and foulest water: for fish drinkable and life – sustaining; for men undrinkable and deadly” (Heraclitus, K. LXI). Art and literature and philosophy are as valid ways of knowing and understanding the world as are the sciences, psychology, economics, history, and sociology. Each provides us with perspectives that together only work to help us better understand the world.

    But some contemporary philosophers have abandoned some of these perspectives, saying they are not legitimate ways of knowing. Many postmodern thinkers have been very anti-science (another misreading of Nietzsche), and the only “legitimate” economic theory for such scholars of literature and philosophy has been Marxism. But science is a legitimate part of philosophy – it was once known as natural philosophy. Why have philosophers abandoned natural philosophy, leaving it to the scientists, whose theories (the Big Bang, quantum, string, relativity, natural selection, sexual selection, etc.) are in fact philosophies – natural philosophies? The only difference – which should be no difference – is that these theories are in constant negotiation with the discoveries being made by various scientists using the various methods of science. Philosophers could make considerable contributions to science through theoretical work that takes into consideration both the findings of science and the wisdom of the arts and philosophy. It is the philosophers in particular who should be bringing knowledge together. To the extent a person’s work does not bring knowledge together, that person cannot legitimately be called a philosopher – as philosophy comes from the Greek for “love of wisdom,” and wisdom is the unifying of knowledge, or seeing the world as a whole. If one simply studies and writes upon the works of other philosophers as philosophers, that person is a scholar of philosophy, not a philosopher. Philosophers cannot abandon any perspective on the world. Nor can the artists. Artists and philosophers must consider as many perspectives as possible, if they are going to make truly great works of lasting quality. In the realm of the novel, this is the novelist using Bakhtin’s dialogics. Consider the following poem:

    Perspectives on the Setting Sun


    The blue retreats into the dark –
    The planet’s orbit makes the sun
    Appear to disappear beyond
    The edge of the earth until one
    Last sliver slips from its blue bond
    To the sky, the moon its one mark.


    The lovers sit beside the lake
    And watch the sun slip behind trees
    That border the distant shore.
    The colors dazzle them and please
    Them into love – they each adore
    The other – their hearts rise, awake.


    The sun sets on the horizon,
    The extra atmosphere the light
    Must shine through now stretches and spreads
    The light from white to shades of bright
    Rose, spreading the sky in light reds,
    Clouds in purples by the dozen.


    The sunset on our love today.
    You left me before you left me,
    I know you never had your heart
    With mine. Why did you want to flee
    From all I wanted to impart
    To you? Why can’t we find a way?


    The scattered sunlight shines too bright –
    The setting sun is all I see
    Through the windshield of my old car.
    I cannot see for the bright glare
    I clipped the runner at the knee –
    Her blood was redder than the star
    That killed her with my blinded sight.


    I fear the coming setting sun,
    The darkness that it will herald
    In, my darkest night, a new moon
    That, on the other side, has pulled
    Down my tides to their lowest tune –
    The moon and sun eclipsed, as one.


    How do you skip from orange to blue?
    The atmosphere’s not like a cloud –
    The spectrum’s missing the middle.
    An absent center’s been allowed
    To make this evening a riddle
    And a mirror of all that’s true.


    Blessèd Ra once rose high above
    The Nile flowing brown and green
    Past the sphinx and the pyramids.
    But Ra has set, too pale and lean,
    As all the gods who lost their bids
    To the hawk disguised as a dove.


    From the space station window I
    Look out at the planet I love
    Now that I see it from this place
    We built to orbit high above
    The rest of man, first foot in space –
    The sun sets, a bright, sparkling eye.

    Which of these perspectives is “correct”? The scientific? The romantic? The tragic? The one that expresses fear? The one that sees a riddle? The religious one? The one from space? Each perspective is legitimate. But all of them together create a whole that is itself a different view than each of them separately creates. I am not going to claim my poem is one of lasting quality just because it takes into consideration multiple perspectives (many poems especially take only one perspective into consideration at a time – though many poets do take multiple perspectives into consideration in their body of work – and the poetic perspective, the perspective of any one poem on its subject(s) is itself valuable). The value of my poem is for others to judge. I included it as an example of multiple perspectives one could take on one thing, unified through and as a poem.

    The divisions in contemporary studies of art and literature show the extent perspectivism has been taken to heart. Art/literary theory has been divided into New Criticism/close reading, history of art and literature, culture studies, structuralism, poststructuralism and deconstruction, genre studies, and psychoanalyical (of artist and fictional characters), philosophical, biographical, political, economic (especially Marxist), philological, rhetorical, linguistic, hermeneutical, cognitive, etc. approaches. Recently chaos theory and complex systems theory, information theory, and game theory have been applied to art and literature, contributing to the perspectives. But each perspective alone cannot tell the whole story about a work of art or literature any more than one can understand an organism by knowing about it through only one of the perspectives through which we study organisms: biochemistry, biophysics, molecular biology, cell biology, evolutionary biology, ethology, ecology, population biology, genetics, anatomy, physiology, etc. If we are to more fully understand a work of art or literature, we cannot abandon any perspective any more than we can claim to understand an organism if we abandon any one of the biological perspectives – say, ethology. “Just as the fact that an élite has played a vital part in the history of civilization is distasteful to many people today, so the fact that the laws of supply and demand have a bearing on the history of poetry may be found disconcerting by readers whose aesthetic is more romantic than they know” (Ian Jack, The Poet and His Audience, 2). To understand art and literature better, we have to study more economics than Marx (who is not an economist). Nor can we take an ahistorical approach to studying literature. “When we read a poem in an anthology or in the artificial situation required by the demands of ‘Practical Criticism’, it is reduced to the condition of a cut flower. It we wish to understand the poem it becomes necessary to try to see it, in the manner of an ecologist, in its natural habitat” (Jack, 3). Audience and patronage should be taken into consideration because “the history of European music, painting and sculpture cannot be intelligently studied without reference to the overwhelmingly important part played by the patronage of Church and aristocracy” (Jack, 3).

    By taking into consideration as many different perspectives as possible, we can come closer to having a more objective understanding of works of art and literature. Though “Objectivity” [ought to be] understood not as “contemplation without interest” (which is a nonsensical absurdity), but as the ability to have one’s For and Against under control and to engage and disengage them, so that one knows how to employ a variety of perspectives and affective interpretations in the service of knowledge. Henceforth, my dear philosophers, let us be on guard against the dangerous old conceptual fiction that posited a “pure, will-less, painless, timeless knowing subject”; let us guard against the snares of such contradictory concepts as “pure reason,” “absolute spirit,” “knowledge in itself”; these always demand that we should think of an eye that is completely unthinkable, an eye turned in no particular direction, in which the active and interpreting forces through which alone seeing becomes a seeing-something, are supposed to be lacking; these always demand of the eye an absurdity and a nonsense. There is only a perspective seeing, only a perspective “knowing”; and the more affects we allow to speak about a thing, the more eyes, different eyes, we can lend to the thing, the more complete will our “concept” of this thing, our “objectivity,” be. (Nietzsche, GM III: 12)
    This is the approach we should be taking when studying art and literature, a historical figure, or an object of scientific interest. This is the approach we should be taking when studying anything. The more perspectives we can take into consideration, the better. Of course, not everyone can do this. We still need those who more fully develop one particular perspective on a given object of study. What we need are those who can both see these multiple perspectives, and yet bring them together – even when those perspectives come into conflict. In fact, it is when those perspectives come into conflict that the philosopher and the artist becomes creative.

    A theory of art and literature should be built upon as many perspectives as possible. One must establish a (meta)physics, ethics, epistemology, etc. if one is to develop a theory of art and literature relevant for humans, who exist in a field of changing objects, ideas, and perspectives. A good theory of art and literature cannot exist in isolation – it cannot be decoupled from any of the perspectives one sees and can see the world through. This has not stopped many 20th Century philosophers from trying. Heidegger thought he could do away with metaphysics and develop a theory of art and literature without it. But Heidegger attempted to do away with metaphysics by being more metaphysical than anyone else. Which shows us what happens when one adopts a particular metaphysics – in Heidegger’s case, making a metaphysical connection to the German language that led to his becoming a Nazi, and in postmodern theorists, adapting his philosophy of language to language as a whole, developing political correctness. To avoid such problems, I find it necessary to establish a (meta)physics so my ideas on art and literature can be better understood and seen in proper relation to my views on (meta)physics, ethics, and epistemology.

    Such work to demonstrate unity in diversity is hardly unique to me. There have been a number of recent works by people attempting to demonstrate the unity of knowledge: Consilience by biologist E. O. Wilson, The Blank Slate by linguist Steven Pinker, The Culture of Hope by poet Frederick Turner, The Power of Limits by architect Gyorgy Doczi, and the works of the philosopher of time J. T. Fraser. I also argue in this work that Nietzsche himself, though in great part responsible for the fracturing of scholarship into unconnected perspectives, is interested in creating unity of perspectives, as the quote above suggests. These works (and other works by these same authors) act as answers to contemporary postmodern perspectivism – including Nietzsche’s works. It is appropriate that Nietzsche, having indirectly created the problem of disconnected perspectivism, acts as the answer to the problems created by 20th century thought as a whole, and postmodern thought in particular.

  5. #5 Glauber
    November 15, 2010

    I would like to suggest this paper:

    Cheating yeast help group.
    New results show that yeast populations grow better when a few individuals cheat the system

    Yeast colonies with mooches, thieves and cheats actually grow faster and larger than colonies without these freeloading individuals, according to a study published 15th September in PLoS Biology, challenging the widely held belief that cheaters bring only bad news to cooperating populations.

    Researchers found that when some yeast cheat their neighbors out of glucose, the entire population grows faster. Image: Eric Miller, Max Planck Institute of Evolutionary Biology “This is a most surprising result,” said Laurence Hurst of the University of Bath in the UK, who coauthored the study. “The theory of cooperation was one of the best worked theories in all of evolution. Everyone assumed that it had to be the case that the world is better off when everyone cooperates.”

    The results may explain why yeast populations tolerate the presence of cheaters, added Michael Travisano, a biologist at the University of Minnesota, who was not involved in the research — “because a mixed strategy is to everyone’s benefit.”

    Most yeast secrete invertase, which hydrolyzes sucrose into fructose and glucose, their preferred food. However, some yeast are known to cheat the system. Cheater yeast don’t secrete invertase and therefore don’t contribute to the glucose production, yet they still eat the glucose that is generated by the rest of the population.

    According to the theory of cooperation, which states that organisms are better off when everyone cooperates, yeast populations should be best off when all the yeast produce invertase. This would maximize the availability of glucose, which should enable more yeast growth. But when Hurst and his colleagues grew yeast populations with both producers and non-producers of invertase, this is not what they saw. Instead, the yeast grew the fastest and saw the highest population numbers when a proportion of the population was cheating.

    One reason populations with cheaters grew better has to do with the yeast’s inability to efficiently use abundant resources, Hurst said. “If you can hop down to your local McDonald’s for a Big Mac, and it’s very easy and very cheap, then you don’t mind if you eat half of it and throw the rest away,” Hurst said. “If you were starving in Africa, you wouldn’t even imagine doing that.” With the cheater yeast using up some of the available glucose, the cooperators are able to use the remaining resources much more efficiently, he said, allowing the population to grow larger and more quickly.

    The team also modeled the experimental results in an effort to see whether their findings were specific to yeast growing on a petri dish, or whether they might apply to other organisms as well. The results showed that their experimental outcomes could be generalized — cheats would benefit a population whenever certain criteria were met. These results imply that cooperation isn’t always the most beneficial path for a population, Hurst said. Instead, the benefits of cooperation depend on the characteristics of the population itself. Under certain conditions, some amount of cheating is likely beneficial.

    But the story is not a simple one, said Jeff Gore, a biophysicist at the Massachusetts Institute of Technology, who did not participate in the research. For example, “if the cheaters and cooperators are growing at different rates, the ratio of cooperators to cheaters won’t be stable,” Gore said. Thus, the population may be changing, and “you still have to ask what [it] is going to evolve to,” and not just look at where it is now.

    From: http://dialeticsofdisorder.blogspot.com/
    Originally from: The Scientist

  6. #6 tütüne son
    November 18, 2010

    Yeast colonies with mooches, thieves and cheats actually grow faster and larger than colonies without these freeloading individuals, according to a study published 15th September in PLoS Biology, challenging the widely held belief that cheaters bring only bad news to cooperating populations.

New comments have been disabled.